Shaping Graft Immunity

Shaping graft immunity:
Prevention of relapse, viral reactivations and graft-versus-host disease after allogeneic stem cell transplantation
Suzanne van Dorp
Shaping graft immunity:
Prevention of relapse, viral reactivation and graft-versus-host disease after allogeneic stem cell transplantation
Vormgeving van het transplantaat ter voorkoming van relapse, virale reactivaties en graft-versus-host ziekte na allogene stamceltransplantatie
(met een samenvatting in het Nederlands)
Proefschrift ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnicus, prof. dr. G. J. van der Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op dinsdag 4 juni 2013 des ochtends te 10.30 uur
door
Suzanne van Dorp

geboren op 8 juli 1982 te Haarlem
Promotor: Co-promotoren:
Prof. dr. H.M. Lokhorst Dr. J.H.E. Kuball Dr. E. Meijer
iv
Voor Henk van Dorp v
Contents
1 Introduction and scope of the thesis 1
Therapeutic potential of an allogeneic stem cell transplantation
2 Single-centre experience with nonmyeloablative allogeneic stem cell transplantation in patients with multiple myeloma: prolonged remissions induced
11
II Graft-versus-Host Disease and mechanisms of disease

3 Rituximab treatment before reduced intensity conditioning transplantation associates with a decreased incidence of extensive chronic GVHD 4 The immunological phenotype of rituximab-sensitive chronic graftversus-host disease: a phase II study
25
41
III Towards new effectors with the potential to control viral infections and leukemia
5 9 and 2CDR3 domains regulate functional avidity of T-cells harboring 92T-cell receptors T-cells elicited by CMV-reactivation after allo-SCT crossrecognize CMV and leukemia General discussion 57
85 119 129
Bibliography vi
8 Nederlandse samenvatting Curriculum Vit List of publications Dankwoord
155 165 167 169
vii
viii
Chapter 1
Introduction and scope of the thesis

1.1 Therapeutic potential of an allogeneic stem cell transplantation
Allogeneic stem cell transplantation is used as curative treatment for a wide variety of hematological malignancies, such as multiple myeloma [5, 13] and acute myeloid leukemia [66, 185]. The treatment effect is partially due to the conditioning regimen given before transplantation, but mainly to the graftversus-tumor effect that is seen after infusion and repopulation of the new immune system. Graft-versus-tumor (GvT) effect is induced by competent donor lymphocytes and is clearly illustrated by sustained remission after infusion of donor lymphocytes after allo-SCT [120, 138, 228]. Donor T cells are key players in the curative GvT effect. In an HLA-matched setting donor T-cells can react against polymorphic peptides, minor histocompatibility antigens (mHags), presented by HLA-molecules to donor T cells. mHags can be expressed on all cells, but can also be specic to hematopoietic cells and tumor cells [99, 206]. T cells directed against mHags, specically presented on tumor cells, can induce GvT effect. Despite the GvT effect, relapse is not uncommon. 10 45% of patients will ultimately develop relapse of disease [5, 31, 43, 171]. This is the main cause of treatment failure. Furthermore non-relapse mortality (NRM), mainly caused by graft-versus-host disease (GVHD) and infections, such as viral reactivations, gravely impairs the success rate of allo-SCT. After myeloablative and nonmyeloablative allo-SCT NRM of respectively 32% and 20% was reported [59, 204]. Thus, there is still a substantial need to improve allo-SCT. This can be achieved 1
either by modulating the transplantation in order to minimize the risk of GVHD. However, this strategy is usually hampered by an increase in relapse [81, 91]. Alternatively, achieving a better control once GVHD occurs represents a valuable option. Finally, understanding the rather complex molecular mechanisms for virus and tumor control after allo-SCT could allow the development of new treatment strategies with improved efcacy in terms of tumor control and limited toxicity. Therefore, we will briey introduce so far know mechanisms of GVHD (Section 1.2) and introduce potentially valuable new effector cells in order to control leukemia as well as viral infections after allo-SCT (Section 1.3).
1.2 Graft-versus-Host Disease as major challenge after allo-SCT

The presence of mHags on normal cells can also induce a severe and possibly lethal complication of allo-SCT, GVHD [57]. Acute GVHD is seen mostly in the rst three months after allo-SCT and is characterized by dermatitis, hepatitis and enteritis. Acute GVHD is seen in up to 60% of patients and morbidity and mortality are high [80]. T-cell depletion of the graft is associated with lower incidence of acute GVHD, however also impairs the GvT effect and relapse of disease is more often seen in patients treated with a T-cell depleted graft [91, 148]. This again emphasizes the importance of donor T cells in the pathogenesis of acute GVHD. Acute GVHD is treated with immunosuppressants, such as high dose prednisone, ciclosporin-A and mycophenolate mofetil (MMF) [80]. Promising results are also seen with mesenchymal stromal cells (MSCs), which immunomodulatory effect could taper the allo-reactivity of donor T cells [112, 129, 215]. Chronic GVHD is the major long-term complication of allo-SCT. Up to 70% of patients surviving 100 days post-SCT will develop chronic GVHD [14]. This disease is invalidating and decreases quality of life enormously. Moreover it is the major cause of late mortality after allo-SCT [131, 151]. Chronic GVHD can affect various organs and has a heterogenous phenotype [70]. In the majority of patients the skin is involved, either locally or generally [71]. Chronic GVHD can cause sclerosis of the skin, with similarities with systemic sclerosis (SSc). This 2
CHAPTER 1. INTRODUCTION AND SCOPE OF THE THESIS
can inuence movability and chest wall compliance. Moreover chronic GVHD causes ulcerations of the skin, which are, besides painful and dysmorphic, a port d entre for pathogens [201]. Other often affected sites are mucosal surfaces of eyes, mouth, vagina and gastro-intestinal tract. Furthermore chronic GVHD can cause bronchiolitis obliterans syndrome, hepatitis and nephritis [71]. The general nding in affected organs is brosis. The pathogenesis of organ damage and brosis in chronic GVHD is poorly understood. In both mouse and human studies brosis was not only restricted to the affected sites, but was generally seen in various organs, even in the spleen [van Dorp, unpublished data] [100]. Activation of broblasts, causing brosis, can be mediated by several cytokines and growth factors, such as IL-6 [15], TGF [104, 127] and PDGF [1]. A role for T cells in this process is undoubted. Epithelial damage and Tcell inltrates at the affected sites are seen in both acute and chronic GVHD. These inltrates are, when present, however less fulminant in chronic GVHD [35]. The addition of anti-thymocyte globuline (ATG) to the conditioning regimen, causing an in vivo depletion of both host and donor T cells, lowers the incidence of chronic GVHD, however not in the same extent as seen in acute GVHD [114, 122]. Furthermore the differences in phenotype between acute and chronic GVHD suggest a different pathogenesis. The similarities in phenotype with auto-immune diseases like SSc, suggest a role for auto-antibodies and therefore B cells in development of the brosis seen in chronic GVHD. The ndings of plasma cell inltrates and depositions of immunoglobulins in tissues affected by chronic GVHD also contribute to this hypothesis [36]. Most importantly however, promising results are seen in patients with chronic GVHD, refractory to steroids, when treated with B-cell depleting therapy [36, 51, 179]. However, so far limited data on efcacy and potential mechanisms of action of B-cell depleting therapies have been reported in men. Therefore, we investigated in more detail the potential of B-cell depleting therapies in preventing or controlling chronic GVHD.
1.3 Towards new effectors with the potential to control viral infections and leukemia
Viral infections and relapse of the underlying disease are the other two challenges after allo-SCT. Control of leukemia has been reported to be mainly mediated by NK cells [174] and T cells [150]. NK cells are known to recognize tumor cells due to loss of inhibitory signals, caused by downregulation of MHC molecules on tumor cells. Furthermore several activating NK receptors, such as NKG2D and NKp30, contribute to this recognition by binding ligands upregulated on malignantly transformed cells, i.e. the MHC-like molecules MICA and MICB and ULPs [141]. T cells can recognize tumor specic antigens presented by MHC molecules through their-T cell receptor (TCR) [19, 160]. In patients with leukemia the control mechanisms are not sufcient enough to clear tumor cells. To overcome this problem and to optimize the anti-tumor response in vivo, it has been suggested to generate substantial amounts of tumor specic T cells by adoptive transfer of engineered T cells bearing a tumor specic TCR [106, 150, 218]. Although promising, this technique has several limitations. First, a distinct TCR is restricted to an HLA type, therefore the amount of patients that can be treated with one TCR is limited. Secondly, redirecting T cells, with a high afnity TCR against tumor specic antigens, can induce on-target auto-reactivity, since these specic antigens are often also expressed on normal tissues, although to a lesser extent [107, 162]. However, even when an TCR directed against antigens uniquely expressed on tumor cells is used, autoreactivity can occur. Introduction of a specic or engineered TCR can induce pairing with the endogenous TCR. This mispairing induces T cells with unknown specicity. When these unwanted TCRs gain specicity towards normal tissues, this can lead to off-target auto-reactivity [20, 67]. The major challenge in adoptive TCR transfer is to select TCRs that can be broadly used, with high tumor specicity and afnity, while on- and off-target auto-reactivity are prevented. An attractive option to meet all these conditions at once is the use of a TCR. T cells are a minor population in the peripheral blood and more widespread in epithelial tissues, such as the skin, lung, gastro-intestinal and reproductive 4
tract [89]. T cells express a heterodimer composed of rearranged and chains, comparable to the TCR. T cells rearrange and express clonally diverse TCRs, however the variability of the variable (V) and joining (J) elements in the TCR and TCR loci is more limited in number than of the TCR and TCR . In humans there are at this moment 6 known V genes and 5 V genes [12, 89, 95]. The diversity of the TCR is enhanced by a special feature of the TCR locus. The TCR locus contains, like the TCR locus, diversity (D) elements. TCR chains frequently use both of their D elements to create more diverse complementarity determining regions (CDR3). The TCR locus is located within the TCR locus and the TCR locus is located in the proximity of the TCR locus. Despite the proximity of their locations rearranging of genes of the TCR and TCR, or TCR and TCR, is very rare. Transfer of a TCR into T cells could prevent mispairing between the endogenous and the introduced TCR and thereby formation of TCRs with unwanted specicity. Using the T cells as carrier of the TCR also solves the problem that function and proliferation capacity of T cells can be impaired in cancer patients [234], whereas T cells can, even in end stage cancer, induce potent immune responses [126]. TCRs, like the TCR, use the signalling domain of the CD3 molecule after activation [161]. However they recognize antigens in a major histocompatibility complex (MHC) independent matter [89, 110]. It is thought that not one, but a various number of ligands causes activation of the TCR. They specically recognize both self- and non-self ligands, upregulated on stressed cells. Selfligands, upregulated on infected or malignantly transformed cells and activating T cells, are endogenous low molecular mass phosphoantigens, which are metabolites of the mevalonate pathway, such as isoprenylpyrophosphate (IPP) [34, 109, 214], and the mitochondrial F1ATPase [198]. Furthermore, T cells recognize known ligands of NK-receptors, upregulated on infected or transformed cells, such as the MHC-like molecules MIC-A and MIC-B [87, 238], and ULBP4 [119]. This recognition is mediated by both the TCR as NKG2D, which is also expressed by most T cells [110, 184]. Moreover CD1c, a MHC-like molecule, presenting self lipids, is upregulated on inammation induced monocyte differentiation into dendritic cells (DCs) is shown to activate T cells [190]. A broad spectrum of bacterial species and parasites are described to induce proliferation and cytokine production by T cells. They are recognized by non-self 5
ligands, mainly exogenous phospho-antigens, like HMBPP [17, 42]. These are products of the equivilant of the mevalonate pathway, as used by different microbes [109]. In this context, we investigated in this thesis the role of T cells and dened TCRs to control leukemia. However, as T cells are also activated by cytomegalovirus (CMV) and other herpesvirus antigens we investigated also the role of T cells and dened TCRs after allo-SCT and CMV-reactivation and focused on mainly V1pos T cells as CMV infection induces proliferation of mainly V1pos T cells. Healthy individuals, who are CMV seropositive, have higher percentages of circulating V1pos T cells, with an effector memory phenotype [176]. This suggests a role for V1pos T cells in controlling CMV virus after primary infection. Proliferation of V1pos T cells has also been described upon reactivation of CMV disease after kidney transplantation [55]. After allo-SCT reactivation of CMV, contributes to the non-relapse mortality and morbidity, causing pneumonia, retinitis and most frequently colitis [28]. In immunocompetent individuals there is a tight balance between the virally infected cells and the host immune system, mainly virus specic T cells. This ne balance is disturbed in patients with a compromised immune system as seen after intensive chemotherapy and allo-SCT. The virus is no longer controlled by the immune system and can reactivate from its latency [28]. Interestingly, there is recent evidence that virus specic T cells in the graft can protect against severe CMV reactivations [175]. It is hypothesized that, since the graft of a seronegative donor does not contain virus specic T-cells, a competent immune response against CMV will take even longer than in patients treated with the graft of a seropositive donor. However in many cases immune responses early after allo-SCT are not yet adequate enough to control the virus, even after tapering of immunosuppressants. Therefore, we asked as to whether individual T-cell lines, clones or even individual receptors might not only provide tools to improve control of leukemia after allo-SCT but also to further prevent viral reactivations. In the context of above described major challenges during an allo-SCT we dened the scope of this thesis: In Part I the potential of allo-SCT in multiple myeloma patients has been addressed. In Part II (Chapters 3 and 4) the effect
of both host and donor B-cell depletion on the development of chronic GVHD has been evaluated. In Part III (Chapters 5 and 6) the potentially therapeutic efcacy of T cells for the control of leukemic and viral infections has been investigated. This includes the impact in the context of an allo-SCT as well as the development of next generation immune therapies for the clearance of leukemia and infections. Part I. Therapeutic potential of an allogeneic stem cell transplantation Chapter 2. Single-centre experience with nonmyeloablative allogeneic stem cell transplantation in patients with multiple myeloma: prolonged remissions induced. Part II. Graft-versus-Host Disease and mechanisms of disease Chapter 3. Rituximab treatment before reduced-intensity conditioning transplantation associates with a decreased incidence of extensive chronic GVHD. Chapter 4. The immunological phenotype of rituximab-sensitive chronic graft-versus-host disease: a phase II study. Part III. Towards new effectors with the potential to control viral infections and leukemia Chapter 5. 9 and 2CDR3 domains regulate functional avidity of T-cells harboring 92T-cell receptors. Chapter 6. T cells elicited by CMV-reactivation after allo-SCT crossrecognize CMV and leukemia
Part I
Therapeutic potential of an allogeneic stem cell transplantation
Chapter 2
Single-centre experience with nonmyeloablative allogeneic stem cell transplantation in patients with multiple myeloma: prolonged remissions induced
Suzanne van Dorp1 , Ellen Meijer1 , Niels W.C.J. van de Donk1 , Adriaan W. Dekker1 , Karel Nieuwenhuis1 , Monique C. Minnema1 , Eefke Petersen1 , Roger Schutgens1 , Leo F. Verdonck1 and Henk M. Lokhorst1
Netherlands Journal of Medicine, 2007
1 Department
of Haematology, University Medical Centre Utrecht, Utrecht, the Netherlands
11
Abstract Background: The role of allogeneic stem cell transplantation in multiple myeloma is not yet established. Methods: We retrospectively evaluated the outcome of nonmyeloablative allogeneic stem cell transplantation (NMA) in patients with multiple myeloma treated at the department of Haematology of the University Medical Center Utrecht. Thirty-six patients received NMA as part of the rst-line treatment; 23 patients as part of salvage therapy. Conditioning regimen was low-dose total body irradiation (TBI, 2 Grays) only; udarabine was added in patients without previous autologous stem cell transplantation and patients with matched unrelated donors received antithymocyte globulin in addition to udarabine and TBI. Results: Following NMA overall response increased from 84 to 90%, complete remission rate from 15 to 32%. As part of rst-line treatment NMA induced complete remission in 50% of patients versus one patient (4%) treated for relapsed multiple myeloma. Median progression-free survival was 26 months (13 months for the salvage group, 38 months for the upfront patients). Median overall survival has not been reached yet. The achievement of complete remission following NMA as part of rst-line treatment was associated with prolonged progression-free and overall survival. Major toxicities were acute and chronic graft-versus-host disease occurring in 64% (23% grade 3-4) and in 54% (49% extensive) patients, respectively. Seven patients (12%) died from nonrelapse mortality (NRM), ve patients (9%) directly related to toxicity of NMA. Conclusion: NMA in multiple myeloma is feasible, is associated with acceptable NRM and may induce prolonged complete remission. In pretreated patients the result of NMA is disappointing which urges new strategies.
2.1 Introduction
Allogeneic stem cell transplantation (ASCT) is probably the only treatment with a curative potential for multiple myeloma. This is due to the graft-versusmyeloma effect, mediated by immune competent donor lymphocytes, best illustrated by the induction of sustained remissions following donor lymphocyte infusions after ASCT [5, 13, 181]. However, the necessity of performing ASCT in multiple myeloma is disputed as no survival advantage has been obtained compared with autologous SCT, in particular when myeloablative conditioning 12
CHAPTER 2. NONMYELOABLATIVE ASCT IN MULTIPLE MYELOMA
for the ASCT is applied [24]. An important factor for this is the high nonrelapse mortality associated with myeloablative conditioning [24, 139]. In an attempt to lower nonrelapse mortality and make ASCT available to more patients, nonmyeloablative conditioning was introduced. Nonmyeloablative ASCT (NMA) is associated with reduced acute toxicity, while antitumor activity is probably maintained [6, 76, 147, 204]. In this retrospective single center study we show that NMA in multiple MM is feasible, with acceptable NRM and that prolonged remissions may be induced in patients who received NMA as part of rst line treatment and achieved a complete response following SCT.
2.2
Patients and methods
2.2.1 Selection of patients. Patients with MM, who received a NMA at the University Medical Center Utrecht, The Netherlands, between September 2001 and September 2005 were included in this retrospective study. In this period in all newly diagnosed patients younger than 66 years and their siblings the human leucocyte antigens (HLA) class I (HLA-A, HLA-B, HLA-C) and class II (HLA-DR, HLA-DP, HLA-DQ) were typed in the rst three months after diagnosis. If a HLA matched sibling donor was available (1 factor class I or class II mismatch was allowed), patients could proceed to NMA between 2 and 6 months after high dose melphalan (HDM) 200 mg/m2 and autologous stem cell rescue, that followed 3 courses of induction therapy with Vincristine, Adriamycin, Dexamethasone (VAD) or Thalidomide, Adriamycin, Dexamethasone (TAD) [77]. Also patients with a relapse after preceding treatment, but responsive to salvage therapy, and having an HLA-matched related or unrelated donor were also allowed for subsequent NMA.
2.2.2 Conditioning. The conditioning regimen before allogeneic stem infusion for the patients with completely matched HLA identical siblings donors consisted of 1 course of low dose Total Body Irradiation (TBI) (2 Gy) only, in case they had received HDM 200 mg/m2 within the preceding 2 and 6 months (tandem auto-NMA). Fludarabine 30 mg/m2 intravenously for 3 days was added in case no preceding autologous SCT was performed. The conditioning regimen 13
before allogeneic stem infusion for the patients with an HLA mismatched or unrelated donor consisted of anti-thymocyte globulin (ATG; 2 mg/kg/day for 4 days) followed by udarabine 30 mg/m2 intravenously for 3 days and 1 course of low dose TBI (2 Gy).
2.2.3 Immune suppression. In the post-transplantation period all patients were treated with the immunosuppressive drugs cyclosporine A (CSP) and mycophenolate mofetil (MMF). Patients received 30 mg/kg/day MMF for 60 90 days and 2 4.5 mg/kg/day CSP for 3 6 months according to the Seattle scheme [147].
2.2.4 GVHD grading and treatment. For diagnosing and grading acute Graft versus Host Disease (GVHD) the Gluckberg criteria [77] were used. Chronic GVHD was graded according to the Seattle classication [202]. Time of onset of acute and chronic GVHD and grade of GVHD were monitored. Acute GVHD > grade I was treated with prednisone 1 2 mg/kg/day and when necessary topical prednisone treatment was applied. In these cases the doses of CSP and/or MMF were increased or continued. In case of steroid refractory acute GVHD other drugs were used, such as sirolimus, tacrolimus, rituximab or more experimental drugs, such as alemtuzumab and dacluzimab. Chronic GVHD of the skin was treated with topical prednisone. In severe cases of extensive chronic GVHD prednisone 1 mg/kg/day was given.
2.2.5 Denitions. Response and progression were determined according to the European Group for Blood and Marrow Transplantation (EBMT) criteria [25]. In short, a partial response (PR) was dened as 50 reduction of serum Mproteine or 90% reduction in 24 hour excretion of Bence Jones proteinuria in case of light chain disease (LCD). A complete response (CR) was dened as complete disappearance of serum and urine M-proteine as determined by immune xation of serum and 10 times concentrated urine. In addition monclonal myeloma cells as determined by immune phenotyping had to be absent in a representative bone marrow aspirate or biopsy. Non relapse mortality (NRM) was dened as any death not related to progressive or relapsed myeloma. 14
2.2.6 Statistical analysis. For the statistical analysis SPSS 12.0.1 for Windows (SPSS Inc., IL, and USA) was used. Overall survival (OS) was measured in months and dened as the time from the date of transplantation until date of death or last follow-up. Progression-free (PFS) survival was measured in months and dened as the time from the date of transplantation until the date of progression or death from any cause or last follow-up. Time to acute or chronic GVHD was calculated from the date of transplantation until occurrence of acute or chronic GVHD. Probabilities of overall survival, progression-free survival, and NRM were calculated using the Kaplan-Meier method. KaplanMeier curves were generated to illustrate survival and the log-rank test was used to compare survival curves between sub-groups. Univariate Cox regression analysis was used to determine the prognostic value of various variables for overall survival and progression-free survival. The predictive value of acute and chronic GVHD for OS and PFS was calculated using a time-dependent univariate Cox regression analysis.
2.3
Results
2.3.1 Patient characteristics. Fifty-nine patients were included in this study. The median age was 55 years (range 35 to 67). There were 42 males (71%) and 17 females (28%). The median follow-up duration of survivors was 25.2 months (range 6.8 to 54.6) (Table 2.1). In 36 patients (61%), NMA was part of rst-line treatment and in 23 patients (39%) it was part of salvage treatment. At the time of transplant, nine patients (15%) were in complete remission and 40 patients (68%) were in partial remission. Forty-four patients (74%) had a matched related donor, four patients (7%) had a partially matched related donor and six patients (10%) had a matched unrelated donor, and ve patients (9%) had a partially matched unrelated donor. In 16 cases (27%) there was a female donor and a male recipient. Thirty-ve patients (59%) were conditioned with TBI only (2 Gy) and 24 (41%) with TBI and udarabine (30 mg/m2 /day for 3 days) [6]. Fifteen patients (25%) received ATG as in vivo T-cell depletion. At the time of diagnosis 21 out of 50 patients (42%) had chromosome 13 abnormalities in FISH analysis and 20 out of 44 patients (46%) 15
had an elevated 2-microglobulin ( 3.0 mg/l).

No. of patients (%) Sex Male Female Age (years) Median Range Median follow up1 (months) Median Range Extent of prior therapy First-line treatment Relapse treatment Donor MRD PMRD MUD PMUD Conditioning regimen TBI TBI and udarabine Donor sex match Female to male Other Deletion of chromosome 132 Presence of deletion of chromosome 13 Absence of deletion of chromosome 13 2-microglobulin3 21 (42.0) 29 (58.0) 16 (27.1) 43 (72.9) 35 (59.3) 24 (40.7) 44 (74.6) 4 (6.8) 6 (10.2) 5 (8.5) 36 (61.0) 23 (39.0) 25.2 6.8 54.6 55 35 67 42 (71.2) 17 (28.8)
< 3 mg/L > 3 mg/L

Status at the time of Allo-SCT CR No CR
24 (54.5) 20 (45.5)
9 (15.3) 50 (84.7)
Table 2.1: Patient characteristics ( N = 59). Allo-SCT indicates allogeneic stem cell transplantation; CR, complete response; MRD, matched related donor; MUD matched unrelated donor; PMRD, partial matched related donor; PMUD, partial matched unrelated donor; TBI, total body irradiation. 1 Follow-up duration of survivors. 2 Determined in 50 patients (84.7%). 3 Determined in 44 patients (74.6%).
16
Total No. of patients (%) Remission state before allo-SCT CR PR NR Remission state after allo-SCT CR PR NR 19 (32.2) 35 (59.3) 5 (8.5) 9 (15.3) 40 (67.8) 10 (16.9)
First-line treatment n (%)
Relapse treatment n (%)
p 0.007
9 (25) 24 (66.7) 3 (8.3)
0 (0) 16 (69.6) 7 (30.4)
< 0.001
18 (50) 17 (47.2) 1 (2.8) 1 (4.3) 18 (78.3) 4 (17.4)
Table 2.2: Response rates to non-myeloablative allo-SCT in rst-line and relapse treatment. Allo-SCT, allogeneic stem cell transplantation; CR, complete response; PR, partial response; NR, no response; Differences in categorical variables were determined with the Pearson 2 -test.
2.3.2 Response and survival. Total response rate following NMA increased from 83% (n = 49) to 92% (n = 54); complete response rate increased from 15 to 32%. NMA as part of rst-line treatment induced a complete remission in 50% of patients, as compared with achievement of a complete remission in one patient (4%) treated for relapsed multiple myeloma. An ongoing response, dened as improvement of partial to complete response and from no response to partial or complete response occurred in 24% of patients; 28% in patients who received NMA as part of rst-line treatment and 17% in patients who received NMA as part of relapse treatment (Table 2.2). Twenty-ve patients (42%) relapsed or progressed after NMA, two from complete remission and 23 from partial remission. At the time of analysis 48 patients were alive. Eleven patients (19%) had died, four from progressive disease and seven (12%) from nonrelapse mortality. The estimated overall survival of the whole group of patients at two years was 84% (Figure 2.1). Median progressionfree survival was 23.5 months (range 1.0 to 38.0 months; Figure 2.2). In patients who received NMA as part of rst-line therapy, overall and progression-free survival at two years were 88.9 and 68.9%, respectively (Figures 2.3A and B). The achievement of complete remission after NMA in this group of patients was associated with superior overall survival and progression-free survival (Figures 2.4A and B). Also the presence of complete remission before NMA was 17
Figure 2.1: Overall survival following nonmyeloablative allogenic stem cell transplantation.
Figure 2.2: Progression-free survival following nonmyeloablative allogenic stem cell transplantation.
associated with prolonged progression-free survival ( p = 0.037), but not with prolonged overall survival ( p = 0.234). The occurrence of chronic GVHD was associated with prolonged overall survival ( p = 0.012) but not with progressionfree survival ( p = 0.3). The 11 patients with acute GVHD grade III and IV had inferior overall survival due to fatal outcome of this complication in ve patients ( p = 0.001). No fatal deaths were observed in the patients with acute GVHD grade 0 to II. None of all other factors tested including age, gender of recipient or donor, conditioning regimen, use of ATG, family or a matched unrelated donor, deletion of chromosome 13 (FISH), 2 microglobulin 3 mg/ml, had an impact on overall or progression-free survival. In the patients who received NMA as part of the treatment for relapsed myeloma overall survival and progression-free survival at two years were 77.5 and 23.9%, respectively (Figure 2.5). None of the factors tested including age, gender as described above had an impact on progression-free and overall survival. It should be mentioned, however, that the statistical analysis must be interpreted with caution due to the small number of patients.
2.3.3 Toxicity. Nonrelapse mortality at 12 months was 12% (Figure 2.6). Five patients (9%) died from acute GVHD grade III to IV. One patient died from 18
Figure 2.3: (A) Overall survival in patients who received NMA as rst-line treatment. (B) Progression-free survival in patients who received NMA as rst-line treatment.
Figure 2.4: (A) Overall survival in patients who recieved NMA as rst-line treatment and did reach complete remission afterwards and patients who did not reach complete remission. (B) Progression-free survival in patients who recieved NMA as rst-line treatment and did reach complete remission afterwards.
19
Figure 2.5: (A) Overall survival in patients who recieved NMA as treatment for relapsed myeloma. (B) Progression-free survival in patients who recieved NMA as treatment for relapsed myeloma.
complications occurring after heart catheterisation and one relapsed patient refused further treatment, including a stem cell boost for secondary aplasia and ultimately died from overwhelming septicaemia. Acute GVHD following NMA occurred in 38 patients (64%): grade I in 12 (20%), grade II in 12 (20%), and grade III or IV in 14 patients (24%). Chronic GVHD following NMA occurred in 32 patients (54%), with three patients (5%) experiencing limited disease and 29 patients (49%) extensive disease. NMA as rst-line treatment was associated with a higher incidence of grades II to IV acute GVHD, when compared with NMA as relapse treatment (56 versus 26%; p = 0.034). The use of ATG signicantly reduced the incidence of chronic GVHD (20 versus 66%; p = 0.003). This may explain the lower incidence of chronic GVHD in patients with an unrelated or mismatched donor. All other factors tested were not associated with occurrence of acute or chronic GVHD.
2.4 Discussion
Several conclusions can be drawn from this retrospective study. The rst one is that NMA is feasible in multiple myeloma, even in heavily pretreated patients. Nonrelapse mortality after NMA compares very favourably with nonrelapse mortality after myeloablative ASCT [24, 139]. What is remarkable is the absence 20
Figure 2.6: Nonrelapse mortality.
of nonrelapse mortality in the patients receiving a transplant from a matched unrelated donor, probably due to the administration of ATG. The second observation is that NMA as part of rst-line therapy results in a high percentage of complete responses which seems to be predictive for prolonged progressionfree and overall survival, while all patients not achieving a complete response, including the vast majority of the relapsed patients, have remissions of short duration. Longer observation, however, is needed to determine the quality and durability of these complete remissions. Late relapses from complete remissions are not uncommon after ASCT for multiple myeloma [44]. The third conclusion is that overall survival is remarkably good even in the pretreated patients. This may be due to the efcacy of novel agents such as thalidomide, bortezomib and DLI given to the patients who relapsed after NMA [220, 221]. Acute and chronic GVHD were the most important toxicities and responsible for the fatal outcome in ve patients (9%). Nonrelapse mortality percentage may still increase due to the considerable number of patients with chronic extensive GVHD. Chronic GVHD, the most important negative factor for quality of life after NMA with full stem cell grafts, is however a signicant factor for prolonged progressionfree and overall survival. Although our results and results from other studies are encouraging, the role of NMA for myeloma is not yet established [147, 243]. In the recently published prospective study by the French IFM, high-risk myeloma patients with an HLA21
identical family donor and treated with tandem autologous/NMA-ASCT had comparable progression- free and overall survival to the patients with no donor who were treated with double autologous SCT [74]. In this study in vivo T-cell depletion was performed with high- dose ATG as part of the nonmyeloablative conditioning regimen in all patients. The benecial effect of in vivo T-cell depletion is the low incidence of acute and chronic GVHD; the detrimental effect is the elimination of the graft-versus-myeloma (GvM) effect [137]. The importance of immune-competent donor T cells for GvM effect is illustrated by responses to DLI and the occurrence of chronic GVHD [140]. European study groups, including the Dutch Haemato-Oncology Association (HOVON), Spains Programa para el estudio y tratamiento de las hemopatias malignas (PETHEMA), and the European Group for Blood and Marrow Transplantation (EBMT), are performing comparable prospective donor versus no-donor studies. The results of these studies have to be awaited for more denite conclusions about the value of NMA in multiple myeloma. In anticipation of the outcome of these studies it is necessary to explore new strategies, which are aimed at stimulating the cytotoxic efcacy of the donor T cells towards the residual myeloma cells without enhancing GVHD. The suggestion that the novel antimyeloma agents such as bortezomib, thalidomide, and lenalidomide may preferentially stimulate the graft-versus-tumor effect and not GVHD is fascinating in this respect [121, 221]. In conclusion, NMA ASCT as part of rst-line treatment of multiple myeloma is feasible, is associated with acceptable transplant-related mortality and may induce a high percentage of complete remissions of good quality and prolonged duration. The outcome of prospective donor versus no-donor studies, however, has to be awaited to better dene the role of this treatment for multiple myeloma. In extensively pretreated patients response rate and progression-free survival are disappointing and in this category of patients new strategies need to be explored. These strategies should be aimed at enhancing the graft-versustumor effect, probably by incorporating novel agents.
22
Part II
Graft-versus-Host Disease and mechanisms of disease
23
Chapter 3
Rituximab treatment before reduced intensity conditioning transplantation associates with a decreased incidence of extensive chronic GVHD
Suzanne van Dorp1,2 , Floor Pietersma2 , Matthias Wlfl3 , Leo F Verdonck1 , Eefke J Petersen1 , Henk M Lokhorst1 , Edwin Martens4 , Matthias Theobald1 , Debbie van Baarle2 , Ellen Meijer1 , and Jrgen Kuball1,2
Biology of Blood and Marrow Transplantation, 2009
1 Department 2 Department 3 Childrens 4 Julius
of Hematology and Van Creveld Clinic, UMC Utrecht, The Netherlands of Immunology, UMC Utrecht, The Netherlands
Hospital, University of Wrzburg, Germany
Center for Health Sciences and Primary Care, UMC Utrecht, The Netherlands
25
Abstract Chronic graft-versus-host-disease (cGVHD) is the major cause of late morbidity and mortality after allogeneic stem cell transplantation. B cells have been reported to be involved in mediating cGVHD. To assess whether preemptive host B-cell depletion prevents extensive cGVHD after allogeneic reduced-intensity-conditioning-transplantation (RICT), 173 patients treated with RICT for various haematological diseases, who have or have not received Rituximab (Rtx) within 6 months prior to RICT, were analyzed retrospectively. Rtx-treatment within 6 months prior RICT reduced extensive cGVHD signicantly from 45.8% to 20.1%. We hypothesize that most likely host B cells initiate cGVHD and, thus, a systematic B-cell depletion prior to RICT by Rtx might be a valuable strategy in order to reduce extensive cGVHD after RICT.
3.1 Introduction
Chronic graft-versus-host-disease (GVHD) is the major long term complication of allogeneic stem-cell-transplantation (allo-SCT) as up to 70% of all survivors of allo-SCT beyond day 100 develop chronic GVHD (cGVHD) [10, 130, 187]. Several lines of investigation indicate that B cells are involved in the development of cGVHD [130, 210] and B-cell depletion using the monoclonal anti-CD20 antibody Rituximab (Rtx) has demonstrated benet in the treatment of steroidrefractory cGVHD with a success rate of up to 70% [36, 51, 179, 242]. Given poor clinical responses in patients with steroid-refractory cGVHD [130] a potential prophylactic value of Rtx in cGVHD warrants further pursuit. To date, the answer to this question has been approached by others, but remains unanswered. A study of patients after myeloablative allo-SCT suggested no difference in cGVHD in 35 patients treated with Rtx as part of the allo-SCT conditioning regimen, but indicated a possible decrease in overall acute GVHD (aGVHD) incidence [113]. Therefore we studied, whether depletion of host B cells prior to transplantation by Rtx can reduce the incidence of this quality-of-life and life-threatening complication in patients who underwent allogeneic reducedintensity-conditioning-transplantation (RICT). Patients with and without Rtxtreatment within 6 months prior to RICT were analyzed retrospectively and were compared for the incidence of acute and cGVHD, graft-versus-leukemiaeffect (GVL) and overall survival (OS). 26
CHAPTER 3. DUAL ROLE FOR HOST B CELLS IN RICT
3.2
Design and methods
3.2.1 Power analysis and selection of patients. To calculate the minimum amount of patients needed to see a signicant difference in extensive cGVHD between the Rtx pre-treated and non-Rtx-group an effect size, thus the expected difference between two groups was set at 25%. The Chi-square calculation indicated that 18 patients per group were needed to achieve a power of 80% ( = 0.05). 173 patients with various hematological diseases (Table 3.1), who received a RICT at the University Medical Center Utrecht, from September 2001 until April 2007, were included in this retrospective study and allowed to further increase the number of Rtx treated patients to 29 Rtx patients and 144 control patients, which increased the power from 80% to > 99%. One patient in the Rtx-group has been treated with alemtuzumab 6 months prior to RICT. RICT was given as curative or as rescue treatment to patients younger than 70 years with an available HLA-matched related or unrelated donor (1 HLA class I or class II mismatch was allowed). Patients were treated according to clinical protocols approved by the local ethics board and gave their informed consent.
Total population N (%) Median age (years) (range) Median follow up (months) (range) Sex % (female/male) Disease, n ALL AML MDS CLL CML MM NHL SAA Myelobrosis Other Donor, n (%) MRD PMRD MUD 111 (64) 10 (6) 38 (22) 6 29 8 13 6 62 32 6 2 9 173 (100) 56 (20 69) 28 (1 40) 34/66
Rtx 29 (17) 58 (29 67) 21.9 (1 40) 28/72 0 0 0 8 0 0 18 0 0 3 19 (66) 1 (3) 7 (24)
No Rtx 144 (83) 56 (20 69) 28.8 (1 40) 35/65 6 29 8 5 6 62 14 6 2 6
p-value
0.662 0.011 0.417
< 0.001
0.925 92 (64) 9 (6) 31 (22) Continued on next page
27
Continued from previous page Total population PMUD HLA-mismatch, n (%) Patient male/donor female, n (%) ATG as part of conditioning, n (%) RT as part of pre-treatment, n (%) Remission before RICT, n (%) CR PR NR/progression Median time of Rtx treatment before RICT (months) (range) Duration of IS, n (%) Short Long Organs affected by limited cGVHD, n (% of limited cGVHD) Skin Liver Organs affected by extensive cGVHD, n (% of extensive cGVHD) Skin Liver Oropharynx Eyes Lungs Gut Kidneys Tendons/Joints Genitals Pericard 50 (78) 17 (27) 56 (88) 30 (47) 12 (19) 7 (11) 2 (3) 2 (3) 3 (5) 1 (2) 4 (80) 2 (40) 3 (60) 2 (40) 1 (20) 0 (0) 0 (0) 1 (20) 1 (20) 0 (0) 46 (78) 15 (25) 53 (95) 28 (47) 11 (19) 7 (12) 2 (3) 1 (2) 2 (3) 1 (2) 0.916 0.479 0.053 0.748 0.941 0.414 0.676 0.024 0.092 0.769 16 (100) 3 (19) 5 (100) 0 (0) 11 (100) 3 (27) 0.195 89 (51) 84 (49) 12 (41) 17 (59) 77 (54) 67 (46) 55 (32) 74 (43) 44 (25) 4 (14) 16 (55) 9 (31) 2.0 (1 6) 51 (35) 58 (40) 35 (24) 0.235 14 (8) 24 (14) 37 (21) 70 (41) 25 (15) Rtx 2 (7) 3 (10) 5 (17) 11 (38) 5 (17) No Rtx 12 (8) 21 (15) 32 (22) 59 (41) 20 (14) 0.547 0.551 0.761 0.639 0.1 p-value
Table 3.1: Patient characteristics. p-values: Mann-Whitney U test for age and followup; 2 -test for other factors; ALL: acute lymfoblastic leukaemia; AML: acute myeloid leukaemia; ATG: antithymocyte globulin; CLL: chronic lymfocytic leukaemia; CML: chronic myeloid leukaemia; CR: complete remission; Disease: disease for which RICT was given as treatment; cGVHD: chronic graft-versus-host disease; HLA: human leukocyte antigen; IS: immunosuppression; MDS: myelodysplastic syndrome; MM: multiple myeloma; MRD: matched related donor; NHL: non-Hodgkin lymphoma; No Rtx: No rituximab treatment within 6 months prior to RICT; NR: no remission; PMRD: partial matched related donor; PR: partial remission; RICT: reduced intensity conditioning transplantation; RT: radiotherapy; Rtx: Rituximab treatment within 6 months prior to RICT; SAA: severe aplastic anemia.
28
3.2.2 Determination of B-cell counts. Absolute B-cell counts were determined in whole blood using TRUcount tubes R (Becton Dickinson), according to the manufacturers protocol. In brief, whole blood samples were incubated with anti-human CD19 (Fluorescein Isothiocyanate (FITC)-labeled, Becton Dickinson) in TRUcount tubes R . Erythrocytes were lysed with lysing buffer (Becton Dickinson). The samples were acquired on LSR-II (Becton Dickinson) ow cytometer. Results were analysed with FACS DIVA software (Becton Dickinson). Absolute B-cell numbers were calculated according to manufacturers protocol.
3.2.3 Conditioning regimen before RICT. For patients with a matched related donor the conditioning regimen consisted of udarabin (30 mg/m2/day i.v. for 3 days) and of one fraction low dose total body irradiation (TBI) (2 Grays). The transplantation with GCS-F mobilized peripheral blood hematopoietic stem cells was performed after TBI. In the case of a HLA-mismatched family donor or any unrelated donor rabbit antithymocyte globulin (ATG; 2 mg/kg/day for 4 days) was added to the regimen and infused before udarabin was given. Patients with multiple myeloma, who were treated with RICT, within 3 months after high dose melfalan (200 mg/m2) and autologous SCT, only received TBI.
3.2.4 Immunosuppression after allo-SCT. In the post-transplantation period all patients were treated with the immunosuppressants cyclosporin A (CSA) and mycophenolate mofetil (MMF). Patients received 2 4.5 mg/kg/day CSA until day +84 (short) or day +120 (long). Hereafter CSA was tapered if no GVHD was present. CSA dose was lowered in the case of raised creatinin levels or severe side effects. Patients received 15 mg/kg/day MMF (maximum of 3 g/day) until day +28 (short) or +84 (long), also followed by tapering in the absence of GVHD.
3.2.5 Acute and chronic GVHD. aGVHD was diagnosed and graded according to the Glucksberg criteria [77], cGVHD was graded according to the Seattle classication [71, 132]. aGVHD > grade 1 was treated with prednisone 1 2 mg/kg/day, in case of skin localization, topical prednisone treatment was applied. Additionally CSA and MMF doses were increased or continued. Steroid29
refractory aGVHD was treated with sirolimus, tacrolimus, Rtx or more experimental drugs, such as alemtuzumab and dacluzimab. cGVHD of the skin was treated with topical prednisone. In severe cases of extensive cGVHD prednisone 1 mg/kg/day was given. Time to acute and cGVHD was calculated from the date of transplantation until occurrence of acute or cGVHD.
3.2.6 Statistical analysis. Progression free survival (PFS) was dened as the probability of being alive with no indication of disease progression. Overall survival (OS) was dened as the probability of survival without considering the occurrence or non-occurrence of relapse. PFS and OS were measured in months and calculated from the date of transplantation until the date of rst signs of progression or last date of follow-up. Incidence of acute and cGVHD, EBV and CMV reactivations, 95% donor chimerism and probabilities of PFS and OS were calculated using the 1-Kaplan Meier method. Kaplan Meier curves were generated to illustrate survival and the log-rank test was used to compare survival curves between subgroups. Univariate Cox regression was used to determine the prognostic value of various variables for the development of acute and cGVHD, EBV and CMV reactivation, 95% donor chimerism, PFS and OS. These variables included sex, age, disease type, HLA-mismatch, sex-mismatch, antithymocyte-globuline as part of conditioning regimen (ATG), radiotherapy as part of the treatment regimen of the prior disease (RT), conditioning, remission state prior to RICT, aGVHD grade II-IV or III-IV, limited or extensive cGVHD, status of T cell and non-T cell chimerism and application of Rtx. For acute and cGVHD a univariate time-dependent Cox regression analysis was used to assess its predictive value. Variables that had a p-value 0.10 in univariate analysis were included in a multivariate Cox regression analysis, because the number of available potential predictors exceeded the maximum allowed number. Rtx treatment was always included in multivariate analysis. E.g., the effect of disease stage before RICT was assessed rst with univariate Cox-regression analysis and when a p-value < 0.10 was reached, in multivariate Cox-analysis together with Rtx. We thereby corrected for stage of disease. Not more than 7 variables were used at once in multivariate analysis. Thereby it was guaranteed that the compared groups would not be too small. Variables were entered all at once in the Cox model. We used a backward logistic regression model to calcu30
late the effect hierarchically. Statistical analyses were performed with SPSS 15.0 for Windows (SPSS Inc., IL, and USA). The signicance of the difference between the B-cell counts on time points preand post-transplantation of patients who received Rtx and of patients who did not, was assessed by using a Mann-Whitney U test. Analyses were performed with GraphPad Prism 4.0 for Windows (GraphPad Software Inc., San Diego, CA, USA). A probability level of 5% ( p < 0.05) was considered signicant in all analyses.
3.3
Results
3.3.1 Inuence of Rtx-treatment prior to RICT on B-cell counts post-RICT. 173 patients with hematological malignancies who underwent RICT were analyzed retrospectively (Table 3.1). Patients were divided in a group that received (Rtx, n = 29) or did not receive (no-Rtx, n = 144) Rtx within 6 months prior to RICT as B-cell depletion has been reported to last for up to 6 months [7, 22, 37]. The median time of the last administration of Rtx before RICT was 2.0 months (range 1.0 6.0 months). Patients received 375 mg/m2 rituximab per application and in 27 cases this was administered in combination with chemotherapy (R-CHOP/R-PECC; 6 8 cycles). One patient with EBV reactivation post-solid organ transplantation received three single applications of Rtx, one patient received twice Rtx for chronic active EBV infection. To investigate whether B-cell depletion with Rtx within 6 months prior RICT effects B-cell counts pre- and post-RICT, B-cell counts from patients who did (n = 5) and did not receive Rtx (n = 6) within 6 months prior to RICT where investigated from the study cohort when B-cell counts were available prior and after RICT at least until 9 months post-RICT. B-cell counts from patients were compared to B-cell counts from 14 healthy controls (median: 212.5/l; range 88 418/l). Median B-cells counts pre-RICT (13.0/l; (n = 4)) and 3 months post-RICT (23.5/l; (n = 6)) were signicantly lower ( p < 0.01) in patients who received Rtx, as compared to a healthy control group (n = 14) (Figure 3.1). In contrast, median of B-cell counts in patients who did not receive Rtx prior to RICT differed not signicantly before (182/l; (n = 3)) and 3 months after (148.5/l; (n = 6)) RICT as compared 31
Figure 3.1: B-cell counts pre- and post-RICT in patients who received (white bars) and did not receive (black bars) Rtx within 6 months prior to RICT as compared to a healthy control group (grey bar). Statistical analyses compare B-cell counts of patients of each time point to B cell counts of the healthy control group. Counts before RICT (within 1 week), or months after RICT are indicated. Bars indicate median and range. Numbers indicate the number of samples measured per time point. Only signicant changes ( p < 0.05) are indicated. Statistical analyses were performed with a Mann-Whitney U test.
to the healthy control group (n = 14). 6 and 9 months after RICT, no signicant difference in B-cell counts was observed for both, the Rtx- and the no-Rtx-group, when compared to the healthy control group (Figure 3.1). Thus, Rtx-treatment within 6 month prior RICT reduces total number of B cells post-RICT at least until 3 months after RICT.
3.3.2 Impact of Rtx-treatment prior to RICT on aGVHD and cGVHD. The incidence of aGVHD was analyzed by both univariate and multivariate analysis. There was no signicant difference in the incidence of grade II-IV and grade IIIIV between the Rtx- (48.2% grade II-IV (n = 14); 17.9% grade III-IV (n = 5)) and the no-Rtx-group (45.7% grade II-IV (n = 64) and 15.6% grade III-IV (n = 22)) (Figure3.2A and B). However, the median onset of grade III-IV aGVHD was signicantly earlier in the Rtx- as compared to the no-Rtx-group (Rtx-group 0.60 months, no-Rtx-group 1.60 months ( p = 0.013); Figure 3.2C). Only ATG prevented grade III-IV aGVHD signicantly ( p = 0.041, HR 2.57 (95% CI: 1.04 6.38)) in multivariate-analysis with Cox regression (Table 3.2). 32
In order to assess whether Rtx-treatment prior to RICT prevents the development of limited and extensive cGVHD, rst the overall frequency of limited and extensive cGVHD was determined and not signicantly different between the Rtx- and the no-Rtx-group (42.5% (n = 10) versus 54.9% (n = 69), p = 0.468), in uni- nor in multivariate analysis (Figure 3.3A; Table 3.2). However, in the Rtxgroup 20.1% of patients developed extensive cGVHD (n = 5), while in the noRtx-group 45.8% of patients developed extensive cGVHD (n = 58) ( p = 0.053) (Figure 3.3B). The main sites affected were skin, liver, mucosae and lung with no statistical difference between the Rtx- and the no-Rtx-group (Table 3.1). After multivariate analysis signicantly less extensive cGVHD was detected in patients treated with Rtx ( p = 0.035, HR 2.67 (95% CI: 1.07 6.68)) (Table 3.2). Only the application of ATG as part of the conditioning regimen ( p < 0.001, HR 4.35 (95% CI: 2.26 8.38)) but no other analyzed factors also decreased the incidence of extensive cGVHD in a multivariate analysis. Vice versa limited cGVHD was increased in the Rtx-group (25.6% (n = 5)) as compared to the no-Rtx- group (12.9% (n = 11), p = 0.088) (Figure 3.3C). Again this difference reached signicance after multivariate analysis ( p = 0.040, HR 0.33 (95% CI: 0.11 0.95)) (Table 3.2). In summary, Rtx-treatment prior to RICT did not affect the frequency of grade II-IV or III-IV aGVHD but attenuated substantially extensive cGVHD and increased thereby presumably the frequency of limited cGVHD.
3.3.3 Rituximab prior to RITC did not associate with a difference in either PFS or OS. Frequently, manipulations which result in a reduction of GVHD are also associated with a reduction of GVL and vice versa [10, 130]. Therefore it was assessed whether host B-cell depletion by Rtx-treatment might not only impair GVHD but also inuence GVL. As surrogate marker for GVL, progression free survival (PFS) was assessed separately in the Rtx- and no-Rtx group. PFS was decreased by HLA mismatch ( p = 0.038, HR 0.51 (95% CI: 0.27 0.96)) and radiotherapy as part of the treatment regimen prior to RICT ( p = 0.009, HR 0.43 (95% CI: 0.23 0.81)). Although Rtx-treatment was associated with a signicantly reduced incidence of cGVHD, PFS was not inuenced ( p = 0.306) (Figure 3.4A; Table 3.2) by any other analyzed factors including type of disease. Also OS at 40 months differed not signicantly ( p = 0.159) between the 33
Figure 3.2: 1-Kaplan Meier curves of acute GVHD grade II-IV (A) and grade III-IV (B) in patients who received Rtx prior to RICT and in patients who did not. (C) Onset of acute GVHD grade III-IV in patients who received Rtx prior to RICT and in patients who did not.
34
Acute GVHD grade II-IV Disease No HLA mismatch No Rtx Acute GVHD grade III-IV No Rtx No ATG Totalchronic GVHD No HLA mismatch No sex mismatch No Rtx No ATG No RT Acute GVHD Extensive chronic GVHD Disease No HLA mismatch No Rtx No ATG No Remission No aGVHD Limited chronic GVHD Male sex Disease No Rtx Progression free survival No HLA mismatch No Rtx No RT Non relapse mortality Male sex Age < 55 yrs No Rtx
Univariate p-value 0.051 0.058 0.61 Univariate p-value 0.702 0.034 Univariate p-value 0.093 0.036 0.468 0.001 0.099 0.014 Univariate p-value 0.054 0.069 0.053
Multivariate p-value 0.806 0.065 0.621 Multivariate p-value 0.733 0.041 Multivariate p-value 0.895 0.132 0.528 0.002 0.34 0.026 Multivariate p-value 0.16 0.727 0.035
HR 1.022 2.189 0.864 HR 0.844 2.574 HR 1.066 0.664 1.241 2.216 1.515 1.658 HR 0.846 0.801 2.673 4.354 0.99 1.515 HR 2.423 1.469 0.328 HR 0.507 1.652 0.434 HR 2.312 0.378 0.449
95% CI 0.861 1.212 0.951 5.034 0.485 1.541 95% CI 0.320 2.230 1.039 6.379 95% CI 0.409 2.779 0.391 1.130 0.634 2.427 1.339 3.667 0.645 3.560 1.063 2.586 95% CI 0.670 1.068 0.232 2.773 1.070 6.675 2.261 8.384 0.487 2.013 0.921 2.493 95% CI 0.671 8.746 1.047 2.060 0.113 0.950 95% CI 0.267 0.963 0.700 3.901 0.232 0.814 95% CI 0.866 6.174 0.150 0.950 0.187 1.081
< 0.001
0.07 0.046 Univariate p-value 0.075 0.011 0.088 Univariate p-value 0.006 0.306 0.001 Univariate p-value 0.097 0.042 0.055
< 0.001
0.978 0.102 Multivariate p-value 0.177 0.026 0.04 Multivariate p-value 0.038 0.252 0.009 Multivariate p-value 0.094 0.038 0.074
Table 3.2: Multivariate Cox regression analysis of outcome in terms of acute GVHD grade II-IV, acute GVHD grade III-IV, total chronic GVHD, extensive chronic GVHD, limited chronic GVHD, progression free survival, non relapse mortality. p-values < 0.05 are considered signicant. ATG: anti-thymocyte globuline; CI: Condence Interval; GVHD: graft-versus-host disease; HLA: human leukocyte antigen; HR: Hazard Ratio; RT: radiotherapy; Rtx: rituximab within 6 months prior to RICT.
35
Figure 3.3: 1-Kaplan Meier curves of total chronic GVHD (A) extensive chronic GVHD (B), and limited chronic GVHD (C) in patients who received Rtx prior to RICT and in patients who did not.
36
Figure 3.4: Kaplan Meier curves of progression free survival (A) and overall survival (B) in patients who received Rtx prior to RICT and in patients who did not.
Rtx-group (n = 18; 61.4%) and no-Rtx-group (n = 103; 67.4%) (Figure 3.4B). This observation is surprising as PFS and OS [8], in contrast to aGVHD and cGVHD [50, 51], might have been heavily inuenced by the imbalance in the type of disease in this study cohort. However, the low number for certain entities in this study cohort might have hampered the analysis. In summary, our data demonstrate a reduced cGVHD in Rtx-treated patients, while GVL is not affected.
3.4
Discussion
Discussion B cells play an important but yet unclear role in the pathogenesis of cGVHD [130, 210] and B- cell depletion of donor B cells after allo-SCT with the monoclonal anti-CD20 antibody, Rtx, has been shown to improve steroid refractory cGVHD [36, 51, 179, 242]. We asked whether B-cell depletion by Rtx prior to RICT could decrease the incidence of cGVHD and pre-emptive B-cell depletion could therefore be a strategy to prevent this major complication after allo-SCT. Our data indicate indeed that Rtx-treatment prior to RICT decreases the incidence of extensive cGVHD while relapse rate was not inuenced. This result is in contrast to a previous study which retrospectively compared acute and cGVHD in 35 leukemia patients, who received Rtx, and 31 control pa37
tients [113]. The authors reported no inuence of rituximab on the incidence of chronic GVHD and a possible but not signicant decrease of the incidence of acute GVHD. The main explanation for this difference might be the myeloablative conditioning regimen which depletes the majority of immune cells, thus also B cells. Consequently, the reported overall incidence of cGVHD was in both arms very low (11%). A signicantly shorter follow-up period in Rtx patients than patients who did not receive Rtx might have inuenced the incidence of cGVHD. However, the latest time point of onset of extensive cGVHD was 18.2 months (median 5.0 months (range 1.5 18.2 months)) and the median follow-up was in both groups
> 20 months. So even though the follow-up of the Rtx group was shorter than follow-up of the no-Rtx group, it was still long enough for extensive cGVHD to develop.
B cells have been suggested to be nal effector-cells in cGVHD by secreting IL6, a known broblast-growth-factor [237]. However, in patients of whom B-cell counts were available at longer follow up (6 or 9 months post-RICT), no signicant differences could be observed in total B-cell counts at the onset of cGVHD and in most patients a full donor chimerism was reached (data not shown). This suggests that the early depletion of most likely host B-cells rather than donor B-cells are important for the development of pathogenesis of cGVHD. Thus, alternative mechanisms must be responsible for the reduced incidence of extensive cGVHD if patients are treated with Rtx prior to RICT. Host B cells have been proposed to serve as professional-antigen-presenting-cells, present minorhisto-compatibility antigens, and induce GVL and GVHD [197]. In this light our data suggest that host antigen-presenting B cells are under certain conditions [153] required for initiating priming of donor T cells which mediate maximum GVHD but no substantially GVL. Thus, host B cells might rather inuence shaping of a self-reactive donor T cell repertoire than directly mediating cGVHD. We can only speculate as to why Rtx-treated patients who developed grade III-IV aGVHD developed it earlier and conclusions have to been drawn very cautiously due to the very low number of patients in the Rtx-arm (n = 5). One explanation for this more rapid course of aGVHD could be that B cells serve as regulatory B cells, as reported in mice, and provide IL10 to attenuate the course 38
of disease [188]. However, we cannot exclude that also other modulators of the immune system such as regulatory T cells [165, 244] or Th17 [241] cells were indirectly effected by B-cell depletion. We are aware of major limitations of this study, such as imbalance in disease, which primarily might have inuenced the analysis of PFS and OS but not the incidence of acute and cGVHD. Thus, our data indicate that most likely host B cells are important in the pathogenesis cGVHD. In order to reduce cGVHD, pre-emptive B-cell depletion might be therefore benecial prior to RICT.
39
40
Chapter 4
The immunological phenotype of rituximab-sensitive chronic graft-versus-host disease: a phase II study

Suzanne van Dorp1,2, , Henrike Resemann2, , Liane te Boome2 , Floor Pietersma1 , Debbie van Baarle1,3 , Frits Gmelig-Meyling1 , Roel de Weger4 , Eefke Petersen2 , Monique Minnema2 , Henk Lokhorst2 , Saskia Ebeling1 , Marijke van Dijk4 , Ellen Meijer2 and Jrgen Kuball1,2
Haematologica, 2011
1 Dept 2 Dept 3 Dept 4 Dept
of Immunology, UMC Utrecht, Utrecht, the Netherlands of Hematology, UMC Utrecht, Utrecht, the Netherlands of Internal Medicine and Infectious Diseases, UMC Utrecht, Utrecht, the Netherlands of Pathology, UMC Utrecht, Utrecht, the Netherlands authors contributed equally to this work
Both
41
Abstract Chronic graft-versus-host disease is the major long-term complication after allogeneic stem cell transplantation with a sub-optimal response rate to current treatments. Therefore, clinical efcacy and changes in lymphocyte subsets before and after rituximab treatment were evaluated in a prospective phase II study in patients with steroid-refractory chronic graft-versus-host disease. Overall response rate was 61%. Only responding patients were found to have increased B-cell numbers prior to treatment. B cells had a nave-antigenpresenting phenotype and were mainly CD5 negative or had a low CD5 expression. Normal B-cell homeostasis was reestablished in responding patients one year after ritxumab treatment and associated with a signicant decline in skin-inltrating CD8+ T cells, suggesting that host B cells play a role in maintaining pathological CD8+ T-cell responses. Imbalances in B-cell homeostasis could be used to identify patients a priori with a higher chance of response to rituximab treatment (Eudra-CT 2008-00412542).
4.1 Introduction
Graft-versus-host-disease (GVHD) is the most common and life-threatening complication after allogeneic stem cell transplantation (allo-SCT) [84, 200]. Bcell depletion with rituximab (RTX) has been successful in steroid-refractory chronic GVHD, showing response rates of 43 80% [36, 51, 159, 179, 216, 242]. However, the nature of B-cell contribution, as well as to what extent B-cell depletion can restore physiological conditions, has so far not been claried. Hypotheses obtained from mouse models of chronic GVHD and retrospective analysis of patient materials have been conicting. For example, a correlation was found between high levels of B-cell activating factor (BAFF) in a retrospective analysis of 45 patients with active chronic GVHD [192]. However, no signicant correlation could be found between the levels of BAFF and RTX-responsiveness in the most recently published prospective phase II study of 37 patients [115]. In active chronic GVHD prior to response to immunosuppressants, expansion of activated CD27+ B cells has been observed [192]. Another retrospective study of 35 patients suffering from active chronic GVHD, in which treatment included prednisone and calcineurin inhibitors, showed signicantly lower numbers of memory B cells (CD27+) [83]. To our knowledge, no prospective comparisons 42
CHAPTER 4. RITUXIMAB-SENSITIVE CHRONIC GVHD
have been made between immune subsets before and after B-cell depletion therapy of steroid-refractory chronic GVHD, so details on reestablishment of normal B-cell pools after RTX treatment, as well as changes in immune-pathology of the skin, still have to be claried. Consequently, the aim of this study was to demonstrate effectiveness of B-cell depletion therapy in steroid-refractory chronic GVHD and to identify potentially involved cell subsets by a comprehensive immunological analysis in a prospective clinical trial.
4.2
Design and methods
4.2.1 Patients and patient material. In the course of a prospective study (Eudra-CT 2008-004125-42), a cohort of 20 chronic GVHD patients who received allo-SCT due to various hematologic malignancies was treated with 4 weekly doses of 375 mg/m2 RTX (F. Hoffmann-La Roche Ltd., Basel, Switzerland). All patients had chronic GVHD with at least skin symptoms. Inclusion criteria were age over 18 years, life-expectancy of more than six months, and a World Health Organization (WHO) performance status of 2 or under. All patients were steroid-refractory or steroid-dependent. Refractory chronic GVHD was dened as progression of disease after at least two weeks of prednisone treatment (approximately 1 mg/kg) or no response after four weeks of prednisone treatment. Steroid-dependent chronic GVHD was dened as an inability to completely taper immunosuppressive treatment. Patients received RTX treatment at the University Medical Center Utrecht between March 2007 and February 2010 according to clinical protocols approved by the local ethics board. In cases of progression or recurrence of chronic GVHD, for which an alternative systemic therapy was needed, patients were excluded from the study and follow up was ended. Response criteria were set as the following: complete response (CR) and partial response (PR) dened according to the recently published National Institutes of Health (NIH) criteria [173]. For laboratory analysis, blood samples were taken from each patient before and after treatment, and consecutively every two months until one year after treatment. Peripheral blood mononuclear cells (PBMCs) were immediately isolated and lymphocyte numbers measured by TruCount (according to the manufacturers protocol, BD Biosciences). PBMCs were then frozen and stored in liquid nitrogen until further analysis. Plasma and 43
serum were stored at 80 C until further analysis. Furthermore PBMC, plasma and serum samples from allo-SCT recipients without GVHD (No-GVHD) and healthy donors were obtained. An informed consent was obtained for other control samples to be used for analysis. Skin biopsies were obtained before RTX therapy and ve months after treatment from adjacent sites if possible, otherwise from the most active site, and stored in 4% formalin and embedded in parafn. Eye involvement was measured before RTX, three and seven months after treatment, using a Schirmers test. FACS staining, cytokine analysis [54], Bcell clonality, chimerism [224], and auto-antibodies [103], histological stainings [36] and the control groups used are described in the Supplementary Design and Methods section and Supplementary Tables 4.3 and 4.4.
4.2.2 Statistical analysis. For prospective data analysis of the cohort of 20 patients SPSS (IBM, Chicago, USA) was used. Receiver operating characteristic (ROC) curve was used to determine a cut-off point for B-cell numbers that was predictive for responsiveness to treatment. Hazard ratios were calculated using a Coxs regression analysis. Incidence of response was calculated using the Kaplan-Meier method. A Kaplan-Meier curve was used to illustrate responsiveness and a log rank test was used to compare incidence between patients with high and low B-cell numbers. Data from FACS staining, plasma and serum analysis were analyzed using GraphPad Prism 5 for Windows (GraphPad Software, La Jolla, USA). Differences in lymphocyte subsets were compared using two-way ANOVA for normally distributed data. Gaussian-distributed groups were compared using Students t-test. Groups of data, which were not normally distributed were compared using Mann-Whitney U tests. In either case, a probability level of 5% ( p < 0.05) was found to be signicant.
4.3 Results and Discussion

In order to prospectively test clinical efcacy of B-cell depletion therapy in steroid-refractory chronic GVHD, a cohort of 20 patients presenting with at least skin involvement was treated with rituximab and followed until one year after treatment or until relapse of chronic GVHD. Two patients had to be excluded from further study; one due to an allergic reaction to rituximab and 44
Response n (%) Median TTR Months (range) OR 11 (61) OR 13 (76) 0 4 (80) 9 (75) 6 (40) 3 (38) 3 (43) 9 (50) 2 (40) 2 (50) 1 (25) 2 (40) 4 (22) 5 (28) 0 1 (25) 2 (29) 1 (14) 1 (12) 2 (25) 5 (75) 4 (57) 9 (50) 3 (60) 2 (50) 4 (27) 2 (13) 7 (47) 4 (33) 5 (42) 2 (17) 2 (40) 2 (40) 0 1 (8) 2 (13) 0 0 0 0 1 (33) 2 (67) 1 (20) 4 (24) 9 (53) 2 (12) 2 (12) CR PR SD Progression 3 (1 8) 0 11 (61) 3 (17) 4 (22) 3 (1 4) CR PR SD Progression
Response duration Months (range)
Total (n = 18)
12 (1 12)
Organ involvement
Skin (n = 18) 6 (3 12)
Erytherna (n = 17)
Ulcera (n = 3)
x
1 (1 3) 1.5 (1 3) 2 (1 6) 1 (1 6) 1 ( x)
x
10 (3 12) 8 (3 12) 6.5 (4 12) n.d. n.d. n.d
Movable sclerosis (n = 5)
Deep sclerosis (n = 12)
Eyes (n = 15)
45
Oral mucosa (n = 8)
GI tract (n = 7)
Dose reduction
Prednisone (n = 18)
x x x
3 (1 7) 3 (2 4) 3.5 (3 4)
Cyclosporine A (n = 5)
MMF (n = 4)
Table 4.1: Total response rates, response rates per organ and dose reduction of immunosuppressants after treatment with RTX. CR indicates complete response/reduction; GI: gastrointestal; MMF: mycophenolate mofetil; OR: overall response/reduction; PR: partial response/reduction; SD: stable disease/dose; TTR: time to response/reduction.
Figure 4.1: Possible predictive markers for responsiveness to RTX treatment and immunological changes after RTX treatment. (A) Absolute numbers of nave (CD27 ), and activated (HLA-DRhigh ) B cells in responding, non-responding patients, No-GVHD and healthy donor controls. Arrows indicate the control group on which statistical analysis was performed by a Mann-Whitney U test ( p < 0.05). (B) Representative dot plots of cell surface expression of CD5 and CD20 in responding patients at T = 0 and T = 12, non-responding patients at T = 0, No-GVHD and healthy-donor controls. Number of skin-inltrating (C) CD4+ and CD8+ T cells before and ve months after RTX treatment in responding patients as well as percentage of skin-inltrating CD8+ T cells before and after RTX treatment in responders and healthy controls. Statistical analysis was performed using an unpaired t test and a two-way ANOVA ( p < 0.05, p < 0.01).
46
Factor Age Sex ATG Acute GVHD B cells > 0.40 109 /L Low IL-21 Ulcerative skin
HR 0.985 0.796 0.925 1.585 10.683 3.615 0.033
95% CI 0.914 1.061 0.210 2.980 0.198 4.307 0.418 6.009 1.237 92.235 0.435 30.057 0.000 16.323
p-value 0.689 0.728 0.920 0.498 0.031 0.234 0.281
Table 4.2: Predictive value of various factors for responsiveness of RTX treatment. ATG indicates anti-thymocyte globuline as part of conditioning regimen; CI: condence intenval; GVHD: graft-versus-host disease; HR: hazard ratio. HR and p-values: univariate Coxs regression model.
one due to relapse of leukemia. Eighteen patients could, therefore, be included for further analyses. Patients characteristics are shown in the Supplementary Table 4.5. Overall response rate was 61% (n = 11). Only partial responses were seen during the time of follow up. Median time to response was three months (range 1 4 months) and 55% of responders had an ongoing response (n = 6). Median response duration, measured until last time of follow up, was 12 months (range 1 12 months) (Table 4.1). Dosage of prednisone could be reduced in 50% of patients (n = 9) and completely stopped in 4 patients (22%). Median time to dose reduction of prednisone was three months (range 1 7 months) (Table 4.1). To investigate whether the production of auto-antibodies was associated with symptoms of chronic GVHD as reported [172, 196], serum before and after rituximab treatment of patients and No-GVHD controls was tested for a panel of antibodies correlated with Systemic Sclerosis (SSc) in terms of quality (type) and quantity. Several auto-antibodies were found in serum of both responders and non-responders, as well as in serum of No-GVHD controls. However, no signicant associations between presence of antibodies and chronic GVHD could be found (data not shown) as also reported by others [203]. Conicting data have also been reported on the correlation between BAFF-levels or BAFF-to-B-cell ratios in RTX-responding patients, as the recently published prospective study of 37 patients [115] did not show a signicant correlation in contrast to a retrospective study of 20 patients [193]. Also, our prospective study of 18 patients did not show any correlation between BAFF47
levels and RTX response. However, differences in these studies and our data could also be partially a consequence of the fact that patients received different doses of corticosteroids in different studies, and high doses of corticosteroids as used in our study have been reported to partially inhibit BAFF [105]. In our study, only IL-21 was signicantly decreased in responding as compared to non-responding patients (data not shown). Peripheral blood mononuclear cells from different groups were analyzed by ow cytometry for lymphocyte subsets, and there was no signicant difference in total lymphocyte numbers between patient groups, No-GVHD and healthy donor controls. No signicant differences were observed between responders and non-responders when comparing CD8+ and CD4+ T cells in the peripheral blood. Also, regulatory T cells (Tregs, CD3+ CD4+ CD25 CD127+ FoxP3+ ), nave, effector memory and central memory, distinguished on the basis of CD62L and CD45RO expression, did not show any signicant difference between all groups at any time point, as well as T cells expressing early (CD69), intermediate (CD137) and late (HLA-DR) activation markers (data not shown). B-cell numbers in responding patients before treatment (T = 0) were increased with signicantly higher absolute numbers of nave B cells (CD19+ CD20+ CD27 ) and CD86+ and HLA-DRhigh B cells when compared to both non-responders, No-GVHD, and healthy donor controls (against all controls all < 0.05, Figure 4.1A). A cut-off point for B cells of 0.40 109 /L was calculated using a ROC curve. The relative risk of responsiveness in patients with absolute B-cell numbers more than 0.40 109 /L was estimated using a univariate Coxs regression analysis. Patients with absolute B-cell numbers of more than 0.40 109 /L had a 10.7 times higher chance of being responsive to rituximab treatment (HR 10.7 [95% CI 1.2 92.3]; p = 0.031). B-cell number was the only factor found to be of signicant predictive value (Table 4.2). Rituximab sufciently depleted all B cells in the peripheral blood of all patients as indicated by the lack of CD19+ cells at one month after the rst rituximab dose (T = 1; data not shown). B-cell reconstitution in peripheral blood could only be observed 10 12 months after start of rituximab. The MFI of CD5+ B cells was signicantly decreased in responding patients before treatment (T = 0) and normalized until T = 12 ( p < 0.05), and was, therefore, then again comparable to all controls. These differential expression patterns of CD5 in patients and all controls are shown in representative dot plots in Figure 4.1B. At T = 12, also CD5 B-cell numbers, naive B-cell numbers, and HLA-DRhigh B48
cell numbers in responding patients were again comparable to all controls, thus with a signicant difference when compared to T = 0 ( p < 0.05). Knowledge about CD5 and its signaling functions in lymphocytes is still very limited and the biological role of CD5 and CD5+ B cells differ in mice and men [52, 61]. In contrast to our study, an increase in CD5+ B cells has been observed in patients with other autoimmune disease such as lupus erythematosus [61]. However, regardless of its function, loss of CD5 expression on B cells might assist in identifying patients who will benet from B-cell depletion. To investigate whether responding and non-responding patients also show differences in inltrating immune cells of the skin, skin sections of responders, non-responders, and healthy donors were additionally stained for T- and B-cell markers. As reported [35], only minimal immune cell inltrates could be observed in patients with chronic GVHD: total numbers of T cells were within the range of T-cell inltrates usually observed in healthy individuals (data not shown). However, the percentage of skin-inltrating CD8+ T cells was signicantly higher when compared to healthy controls ( p < 0.05; Figure 4.1C). No signicant difference was observed in skin inltrating T cells before rituximab treatment between responding and non-responding patients (data not shown). There was no difference in numbers of skin inltrating B cells between responding and non-responding patients, and B cells disappeared after rituximab treatment (data not shown). After rituximab treatment, only ongoing responding patients were analyzed at ve months after study entry as nonresponding patients had already undergone other therapies and were thus no longer eligible. A signicant decrease in CD8+ T cells was observed after ve months in responding patients ( p < 0.01), whereas there was no signicant change in CD4+ T-cell numbers (Figure 4.1C and D). This resulted again in normalization of the percentage of skin-inltrating CD8+ T cells when compared to the healthy control group (Figure 4.1C). This suggests that B-cell depletion by rituximab not only reduces the number of potentially antigen-presenting B cells, but also reduces the skin inltrating CD8+ T-cell compartment. This, therefore, supports recent hypotheses of a T-cell to B-cell crosstalk in the setting of chronic graft-versus-host disease in man [203]. In summary, to our knowledge this is the rst prospective comprehensive study,
49
which describes the immunological phenotype of RTX-sensitive as compared to RTX-unresponsive chronic graft-versus-host disease in the peripheral blood and the skin. Elevation of B-cell numbers with a dominant nave, antigen-presenting phenotype, as well as skewing towards CD5 B cells and B cells with a low CD5 expression was selectively found in responding patients and was the only predictive factor of responsiveness to rituximab. Physiological B-cell homeostasis was re-established in responding patients one year after treatment and associated with a reduced skin inltration of CD8+ T cells in responding patients. This suggests that an imbalanced B-cell repertoire can contribute to chronic graft-versus-host disease by sustaining skin-inltrating CD8+ T cells. These ndings could also be useful in identifying in advance those patients who will benet from rituximab treatment and provide a basis for larger conrmatory prospective clinical trials.
4.4 Supplementary Design and Methods

4.4.1 FACS staining. For phenotypic analysis, PBMCs from patients, No GVHD controls and HD controls were stained with antibodies against the following markers, with uorescent labels as indicated: CD3-PerCP, CD4PerCP, CD80-R-PE, IFN-g-FITC, CD69-FITC, CD137-PE, CD5-PE and CD62LPE-Cy5 (all from BD Pharmingen), CD8-PerCP, CD69-APC, CD19-APC, CD138PerCP, IL-17-PacBlue (all from BioLegend), CD3-eFlour-450, CD4-PE-Cy7, CD4Alexa Flour 750, CD8-APC, CD25-FITC, CD127-PE-Cy7, HLA-DR-Alexa Flour 750,CD38-PE-Cy7, CD86-PE-Cy5, CD27-eFour 780, FoxP3-APC, CD20-PacBlue, IL-4-PE-Cy7, and IL-10-PE (all from eBioscience). For FACS analysis test samples of 300, 000 cells were analyzed. For evaluation of cytokine production capacities of lymphocytes, cells were stimulated with IL2 (20 u/ml, Novartis Pharmaceuticals) and PHA-L (30g/ml, Sigma Aldrich). After 4 hours of stimulation, cells were stained for extracellular and intracellular markers. FoxP3-staining was performed according to the manufacturers instructions for intracellular staining (eBioscience). Samples were processed with a LSR-II ow cytometer (BD Biosciences). The acquired data were analyzed using FACS Diva software (BD Biosciences). 50
4.4.2 Cytokine analysis. For cytokine analysis, plasma samples from patients, No GVHD and HD controls were examined for their content of IL-2, IL-10, IL-12p70, interferon-gamma (IFN-), tumor necrosis factor-alpha (TNF-), IL-4, IL-13, IL-6, IL-17, and IL-21 using multiplex immunoassays as described earlier [54]. Transforming growth factor-beta (TGF- ), BAFF and platelet-derived growth factor-AA (PDGF-AA) were measured with ELISA according to manufacturers instructions (BD Biosciences, Bender MedSystems (TGF- and BAFF) and Antigenic America (PDGF-AA). PDGF-AA was measured in plasma, while TGF- and BAFF were measured in serum samples.
4.4.3 B-cell clonality, chimerism and auto-antibodies. For clonality assessment of B cells, genomic DNA was isolated from patient PBMC samples using a nucleospin blood quick pure kit (Qiagen). B-cell receptor diversity was analyzed using BIOMED multiplex PCR assays as described earlier [224]. Chimerism analysis of T and B cells was performed by PCRbased amplication of short tandem repeats sequences as described earlier [103]. For analysis of auto-antibodies, an immunoblot for SSc-specic auto-antibodies was used according to the manufacturers instructions (Euroimmun Lbeck, Germany). Sera of patients and No-GVHD controls were analyzed for IgG antibodies against Scl-70, CENP A, CENP B, RP11, RP155, Fibrillarin, NOR90, Th/To, PM-Scl100, PM-Scl75, Ku, PDGFR and Ro-52.
4.4.4 Histological stainings. Skin biopsies were stored in 4% formalin and embedded in parafn. Slides were stained with hematoxylin and eosin (H&E, Klinipath), and monoclonal antibodies against CD3 (A0452), CD8 (M7103), CD20 (M0755; all from Dako), CD4 (Monosan, monx10326), CD5 (Novocasta, NCL-CD5-4C7) and FoxP3 (eBioscience, 14-4776). Slides were stained for all markers, except FoxP3, using a BondmaX stainer (Leica). Slides were stained for FoxP3 manually. Epidermal involvement and dermal sclerosis was scored as described earlier [36]. Pathologists were clinically blinded during analysis. Nine skin biopsies of the leg, arm and trunk, which were obtained from healthy donors after receiving their informed consent served as controls.
51
No-GVHD controls (n = 5) Median age (yrs; range) Sex M/F (%) Disease (n) 50 (45 63) 60/40 3 1 1 4 (80) 5 (100) 1 (20) 0
AML CML NHL

Related donor (n,%) NMA conditioning (n,%) ATG (n,%) Acute GVHD (n)
Table 4.3: Characteristics of No-GVHD control group used for ow-cytometry analyses. AML: acute myeloid leukemia; ATG: antithymocyte globuline; CML: chronic myeloid leukemia; NHL: non-Hodgkins lymphoma; NMA: non-myeloablative.
Healthy controls (n = 5) Median age (yrs; range) Sex M/F (%) 30 (23 40) 40/60
Table 4.4: Characteristics of healthy controls used for ow-cytometry analyses.
52
Total population 18 (100) 53 (39 66) 78/22 7 (1 13) 1 34 (10 61) 12 (2 51) 2 2 1 8 1 4 14 (78) 16 (89) 4 (22) 14 (78) 0 1 5 1 4 9 (82) 10 (91) 2 (18) 8 (73) 0 15 (3 51) 38 (9 77) 1 8 (4 13) 73/27 86/14 4 (1 4) 1 34 (11 62) 8 (2 43) 2 2 0 3 0 0 5 (71) 6 (86) 2 (29) 6 (86) 53 (39 64) 55 (44 66) 11 (67) 7 (33)
Responding patients
Non-responding patients
p-value
N (%)
Median age (years; range)
0.751 0.485 0.001 1.000 0.319 0.441 0.450
Sex male/female (%)
Median follow-up (months; range)
Number of pre-treatments
Median time after allo-SCT (months; range)
Months from onset of chronic GVHD until RTX-treatment (median; range)
Disease
AMI/MDS
CLL
CML
53
MM
Myelobrosis
NHL
Related donor (n; %)
0.515 0.641 0.515 0.428
NMA conditioning (n; %)
ATG (n; %)
Acute GVHD (n; %)
Table 4.5: Allo-SCT indicates allogeneic stem cell transplantation. AML: acute myeloid leukemia; ATG: antithymocyte globuline; CLL: chronic lymphocytic leukemia; CML: chronic myeloid leukemia; Disease, disease for which allo-SCT was given; GVHD: graft-versushost disease; MDS: myelodysplastic syndrome; MM: multiple myeloma; NHL: non-Hodgkins lymphoma; NMA: non-myeloablative. p-values: Mann-Whitney U test for age, follow up, months after allo-SCT and chronic GVHD; Fishers exact test for other factors.
54
Part III
Towards new effectors with the potential to control viral infections and leukemia
55
Chapter 5
9 and 2CDR3 domains regulate functional avidity of T-cells harboring 92T-cell receptors
Cordula Grnder1 , Suzanne van Dorp1 , Samantha Hol1 , Esther Drent1 , Kirsten Scholten1 , Sabine Heijhuurs1 , Wouter Scheper1 , Zsolt Sebestyen1 , Roland Strong2 and Jrgen Kuball1
Blood, 2012
1 Department
of Hematology and Immunology, University Medical Center Utrecht, Utrecht, The
Netherlands
2 Fred
Hutchinson Cancer Research Center, Seattle, United States
57
Abstract Immunotherapy with innate immune cells has recently evoked broad interest as a novel treatment option for cancer patients. 92T-cells in particular are emerging as an innate cell population with high frequency and strong anti-tumor reactivity, which makes 92T-cells and their receptors promising candidates for immune-interventions. However, clinical trials have so far reported only limited tumor control by adoptively transferred 92Tcells. As a potential explanation for this lack of efcacy, we found unexpectedly high variability in tumor recognition within the physiological human 92T-cell repertoire, which is substantially regulated by the CDR3 domains of individual 92T-cell receptors (TCR). We demonstrate that here reported molecular needs of CDR3 domains to interact with a targetcell shape the physiological 92T-cell repertoire and most likely limit protective and therapeutic efcacy of 92T-cells, the rst defense against tumors. Based on these ndings, we propose combinatorial-TCR-chain-exchange (CTE) as an efcient method for designing high-afnity 92TCRs that mediate improved anti-tumor responses, both in vitro and in vivo in a humanized mouse model.
5.1 Introduction
Immunotherapy with innate immune cells has become widely used because this approach obviates the need to match a cellular product to a dened humanleukocyte-antigen (HLA) haplotype, allowing adoptive immunotherapies to be used in virtually any cancer patient without extensive in vitro selection or manipulation of the cellular product [82, 189, 207]. 92T-cells are promising as an innate cell population for this purpose. They are usually observed at high frequencies in the human peripheral blood and provide a very strong antitumor reactivity against various solid and hematological cancers [38]. However, within 92T-cell populations, individual clones display great diversity in the repertoire due to the activating or inhibitory receptors expressed [164]. Selecting innate cell products for certain cell types, like those with a low level of inhibitory receptors, therefore seems plausible, especially considering the limited efcacy of adoptively transferred innate immune cells in clinical trials [75, 167]. An alternative proposal is to engineer cells to express dened activating innate receptors that mediate strong anti-tumor-reactivity, such as a dened 92T58
CHAPTER 5. 9 AND 2CDR3 DOMAINS
cell receptor (TCR) 8, which could pave the way for readily available and more effective cellular products. However, the molecular details of how a 92TCR interacts with its target are not fully understood, making it challenging to select dened 92T-cells or to engineer T-cells with dened 92TCRs. In classical immunoreceptors, such as TCRs or immunoglobulins (Igs), the complementary determining regions (CDR) determine afnity and specicity for a specic peptide epitope. V(D)J recombination allows the creation of a highly variable CDR repertoire ensuring recognition of an immense collection of antigens. 92T cells also possess a rearranged TCR that mediates recognition. Phosphoantigen isopentenyl pyrophosphate (IPP) has been suggested to be a key player in 92TCR mediated activation [38, 78, 111], but no direct interaction between a 92TCR and IPP or any other phosphoantigen has ever been demonstrated. It was previously suggested that positively charged residues within the 92TCR are crucial for the response to negatively charged phosphoantigens [4, 163] and a potential IPP-binding groove has been proposed [4]. Interestingly, it appeared that responsiveness to phosphoantigens depends in particular on germline-encoded residues within all CDRs apart from CDR3 [232], extending the footprint of recognition to a much larger region than initially predicted. Sequence alignment studies suggested that no dened CDR3 motif is required for recognition beyond a hydrophobic residue at position 109 (Kabatnumbering 97 [134]), which suggests a less dominant role for CDR3 [33, 157, 232]. Consequently, it is still unclear why variations in the 92CDR3 regionswhich are particularly abundant in CDR3have evolved in humans and whether this variability is important in regulating activation of a 92T cell. Understanding the reason for this variation would help to explain either the specicity or the regulation of functional avidity of a 92T cell. This would also provide insight into the role of a 92TCR during the selection process of a 92T cell and allow engineering of therapeutic cells with higher anti-tumor activity. In our study, we therefore asked the following research questions: 1) what is the clonal diversity in terms of tumorspecicity and functional avidity within 92T cells once they express an identical set of activating and inhibitory receptors, 2) what is the specic role of individual 92TCRs, and 3) can we
59
engineer a 92TCR with improved and broader anti-tumor-reactivity?
5.2 Material and Methods

5.2.1 Cells and cell lines. Daudi, MDAMB231, SW480, K562, MCF-7, BT549, Phoenix-Ampho and Jurkat cells were obtained from ATCC, MZ1851RC from Barbara Seliger (University Halle, Germany). Saos2 was kindly provided by Arnold Levine (Princeton University, USA), FaDu and SCC9 by Niels Bovenschen (UMC Utrecht, the Netherlands) and Mogens Claesson (University Copenhagen, Denmark), Psf5 and MRC5 by Bodo Plachter (Mainz, Germany). Phoenix-ampho cells were cultured in DMEM+1%Pen/Strep (Invitrogen) +10%FCS (Bodinco), all other cell lines in RPMI+1%Pen/Strep+10%FCS. PBMCs were isolated from buffy coats obtained from the Sanquin Blood Bank (Amsterdam, The Netherlands). PBMC samples from AML patients were a kind gift from Matthias Theobald (Mainz, Germany) and were collected according to GCP and Helsinki regulations.
5.2.2 TCR Mutagenesis, Cloning and Sequencing. All 92TCR modications are based on codon-optimized genes of 9- or 2-TCR chain G115 anked by NcoI and BamHI restriction sites (synthesized by GeneArt, Regensburg, Germany). To generate alanine-mutations, site-directed mutagenesis was performed by overlap extension PCR [97] or whole plasmid mutagenesis [158] as previously described, using a proofreading polymerase (Phusion, Biok). Mutated NcoI-BamHI digested 9- or 2-TCR chains were ligated into the retroviral vector pBullet and sequenced by BaseClear c
5.2.3 Retroviral transduction of T cells. 92TCRs were transduced into Tcells as previously described [149]. In brief, packaging cells (phoenix-ampho) were transfected with gag-pol (pHIT60), env (pCOLT-GALV) [209] and two retroviral constructs (pBullet) containing either 9-chain-IRES-neomycine or 2chain-IRESpuromycine, using Fugene6 reagent (Takara, Gennevilliers, France). Human PBMC activated with CD3 (30 ng/ml) (Orthoclone OKT R 3, JanssenCilag, Tilburg, The Netherlands) and IL-2 (50 IU/ml) (Proleukin R , Novartis, 60
Arnhem, The Netherlands) were transduced twice with viral supernatant within 48 hours in the presence of 50 IU/ml IL2 and 4g/ml polybrene (SigmaAldrich, Zwijndrecht, The Netherlands). Transduced T-cells were expanded by stimulation with CD3/CD28 Dynabeads
TM
(0.5 106 beads/106 cells) (In-
vitrogen) and IL-2 (50 IU/ml) and selected with 800g/ml geneticin (Gibco, Karlsruhe, Germany) and 5g/ml puromycin (Sigma-Aldrich, Zwijndrecht, The Netherlands). Following selection, TCR-transduced T-cells were expanded in vitro based on a previously described REP protocol [183].
5.2.4 Flow cytometry. 92TCR expression was analyzed by ow cytometry using the following antibodies: V9-PE (clone B3, BD), V2-FITC (clone B6, BD), pan-TCR-PE (clone IMMU510, Beckman Coulter), and TCR-APC (clone B1, BD). Fold change was calculated based on MFI values of 9- G115wt /2-G115wt transduced T cells set to 1 and mock transduced T cells to 0. The expression pattern of individual 92T-cell clones was analyzed using the following antibodies: NKG2D-PE (clone 1D11, Biolegend), NKAT2-PE (clone DX27, BD), CD158a-FITC (clone HD-3E4, BD) and NKB1-FITC (clone DX9, BD). Cell-cell conjugate assays were performed with CFSE (Invitrogen)-labelled T cells and Lavacell (Active Motif Europ)-labeled Daudi cells at room-temperature for 2 hours and analyzed by ow cytometry.
5.2.5 Functional T-cell assays. labeled overnight with 100Cu
51 Chromium-release
assay for cell-mediated
cytotoxicity was previously described [125, 226]. In brief, target cells were
51 Cr
(150Cu for primary cells) and incubated
for 4 6h with transduced T cells in ve effector-to-target ratios (E:T) between 30:1 and 0.3:1. Fold change was calculated when compared to reactivity of engineered T cells expressing unmutated 92TCR. IFN ELISpot was performed using anti-hu IFN mAb1-D1K (I) and mAb7-B61 (II) (Mabtech-Hamburg, Germany) following the manufacturers recommended procedure [23]. Target and effector cells (E:T 3:1) were incubated for 24h in the presence of pamidronate (Calbiochem, Germany) where indicated. IFN ELISA was performed using ELISA-ready-go! Kit (eBioscience) following manufacturers instructions. Effector and target cells (E:T 1:1) were incubated 61
for 24h in the presence of pamidronate as indicated. Where specied, fold change was calculated when compared to reactivity of engineered T-cells expressing unmutated 92TCR.
5.2.6 Animal models. The RAG-2-/- /c-/- -BALB/C mice, originally obtained from AMCAS b.v. (Amsterdam, The Netherlands), were bred and housed in the specic pathogen-free (SPF) breeding unit of the Central Animal Facility of Utrecht University. Experiments were conducted according to Institutional Guidelines after acquiring permission from the local Ethical Committee and in accordance with current Dutch laws on Animal Experimentation. To induce tumor xenografts, sublethal total body irradiated (2 Gy) RAG-2-/- /c-/- -BALB/C mice (11 17 weeks) were injected i.v. with 0.5 106 Daudi-Luc cells (a kind gift from Genmab Utrecht, The Netherlands) [26, 149]. Outgrowing tumors were visualized in vivo by Biospace bioluminescent imaging. Mice were anesthetized by isourane before they received an intraperitoneal injection (100l) of 25mg/ml Beetle Luciferin (Promega). Bioluminescence images were acquired and analyzed with M3 Vision software (Photon Imager; Biospace Laboratory). 107 92TCR positive transduced T-cells were intravenously injected together with Daudi-Luc cells. Mice received 0.6 106 IU of IL2 (Proleukin R , Novartis) in IFA s.c. on day 1 and every 21 days till the end of the experiment. Pamidronate (10mg/kg body weight) was applied in the indicated groups at day 1 i.v. and every 21 days i.p.
5.3 Results
5.3.1 Anti-tumor reactivity of individual 92T-cell clones. To investigate whether individual 92T-cell clones mediate differential activity against tumor cells compared to the parental 92T-cell population, 92T cells from a healthy donor were cloned by limiting dilution and tested against a broad panel of tumor cells in an IFN- ELISpot (Table 5.1). High variability in tumor recognition was observed between individual 92T-cell clones (cl) in specicity and functional avidity; compared to the original bulk population, cl5 and cl13 produced twice as many IFN- spots in response to the target Daudi and selectively generated signicant amounts IFN- when challenged with K562, BT549 62
!" T-cells PBMCs K562 Daudi MZ1851RC BT549 MCF-7 SW480 MDA MB 231 3 0 0 0 0 0 0 0
#9$2 bulk 2 14 206 205 2 2 0 4
#9$2 cl3 10 9 211 4 1 6 11 2
#9$2 cl4 2 9 55 94 0 1 2 1
#9$2 cl5 7 62 458 114 44 64 3 12
#9$2 cl7 0 181 318 216 1 1 1 1
#9$2 cl8 3 7 268 77 2 1 1 1
#9$2 cl13 12 56 500 93 41 58 4 14
#9$2 cl15 13 12 244 19 8 10 11 10
IFN-! spots/15,000 cells:
< 40
40 - 100
> 100
Table 5.1: Anti-tumor reactivity of individual 92T-cell clones.
and MCF-7. In contrast, cl3 and cl15 recognized solely Daudi cells. A variable expression of NK and KIR receptors has been reported in innate immune cells 25-27 and might have contributed to the observed differential activity of selected clones. Therefore, surface expression of 92TCR, NKG2D, CD158a, NKAT-2 and NKB-1 was examined (Figure 5.6). However, no correlation was found between receptor expression patterns and anti-tumor reactivity of tested 92T-cell clones. We hypothesized that the diversity within the 92TCR contributes to the differential activity of examined 92T-cell clones. Therefore, 92TCRs of the highly tumor-reactive cl5 and the weakly tumor-reactive cl3 were chosen for detailed analysis and compared to 92TCR G115 [4, 149].
5.3.2 Anti-tumor-reactivity mediated by individual 92TCRs. To elucidate differences among 92TCRs of tumor-reactive clones, sequences of wild type (wt) 9- and 2-TCR chains of cl3 (9-cl3wt /2-cl3wt ) and cl5 (9-cl5wt /2cl5wt ) were determined and aligned with 92TCR G115. All three 92TCRs only differed in their CDR3 domains: 1-3 amino acids between 109 and 111 63
in CDR3 and 4-8 amino acids between 108 and 112 in CDR3 (Table S1, numbering according to IMGT [134]). To determine whether distinct 92TCRs mediate differential anti-tumor reactivity, individual 92TCR chains were cloned into the retroviral vector pBullet and linked to a selection marker as described [231]. The wt-combinations 9-cl3wt /2-cl3wt , 9-cl5wt /2-cl5wt and 9-G115wt /2-G115wt were transduced into peripheral blood T-cells, selected by antibiotics and further expanded. 92TCR G115 (9-G115wt /2-G115wt ) [4, 149] served as control, as did cells transduced with an empty vector cassette (mock). First, the transductants were tested for expression of the introduced 92TCR by ow cytometry using a pan-TCR antibody and comparable receptor expression was detected (data not shown). Next, 92TCR-transduced Tcells were functionally tested against the tumor target Daudi in a 51 Cr-release assay (Figure 5.1A). T cells expressing 9-cl3wt /2-cl3wt had a 50 percent reduced ability to lyse tumor cells ( p < 0.01), whereas T cells with 9-cl5wt /2-cl5wt were nearly twice as potent ( p < 0.01) as the control 9-G115wt /2-G115wt . To determine whether the phenotypes of 92TCR-transduced cells with decreased or increased functional avidity are also present on the cytokine level, IFN- production was measured by ELISA. Daudi cells were then co-incubated with pamidronate, leading to an increased production of the phosphoantigen isopentenyl pyrophosphate (IPP) by the tumor cells. High IPP levels allowed enhanced cytokine secretion of the transductants, which was measured in the supernatant after 24h (Figure 5.1A). In line with changes observed for lytic capacity, IFN- secretion was 50 percent lower for T cells transduced with 9cl3wt /2-cl3wt , and nearly twice as high for T-cells expressing 9-cl5wt /2-cl5wt , relative to control 9-G115wt /2-G115wt . To assess whether or not this outcome depends on the differential amount substrate of the mevalonate pathway (presumably IPP), the transductants were further tested in a pamidronate-titration assay and the half maximal effective concentration (EC50) was calculated. Despite different plateaus in IFN- secretion, all selected mutants and the wt control had a comparable pamidronate-EC50 (30pg/ml) (Figure 5.1B). This indicates that distinct 92TCR clones mediate different functional avidity and the high variability among parental 92T-cell clones in tumor recognition seems to be substantially regulated by the CDR3 domains of individual 92T-cell receptors. Based on these results, the correlation between CDR3 domains and functional avidity was investigated. 64
Figure 5.1: Anti-tumor reactivity mediated by 92TCRs. Peripheral blood T-cells were virally transduced with indicated wt 92TCRs or CTE-engineered 92TCRs and (A, C) tested against Daudi in the absence (51 Cr-release assay, E:T 3:1) or presence (IFN- ELISA, E:T 1:1) of 300M pamidronate. Fold change was calculated when compared
65
to reactivity of 9-G115wt /2-G115wt engineered T-cells. IFN- secretion of indicated 92TCR engineered T-cells against Daudi was measured by ELISA (B, D) in the presence of indicated amounts of pamidronate or (E) different E:T ratios. (F) Percentages of cell-cell conjugates of Daudi and T-cells engineered with indicated 92TCR were determined by ow cytometry. Data represent the meanSD. p < 0.05, p < 0.01, p < 0.001 by 1-way ANOVA.
5.3.3 Combinatorial-TCR-chain-exchange (CTE) as rapid method to modulate functional avidity of engineered T-cells. To make the above determination, we devised a strategy named combinatorial- TCR-chain-exchange (CTE), which results in the expression of newly combined 9- and 2- TCR chains on engineered T cells. During this process, 9-G115wt was combined with 2-cl3wt or 2-cl5wt and 2-G115wt with 9-cl3wt or 9-cl5wt . These combinations were retrovirally transduced into T-cells and equivalent expression was detected by FACS using a pan-TCR antibody (data not shown). Engineered T cells were functionally tested against the tumor target Daudi in a
51 Cr-release
assay and an IFN- ELISA (Figure 5.1C). The exchange of 9- or
2-chains indeed caused notable differences. Compared to the original TCR 9- G115wt /2-G115wt , the combination of 9-G115wt /2-cl3wt , 9-G115wt /2cl5wt or 9-cl5wt /2-G115wt mediated 50 to 70 percent increased specic lysis of tumor cells (all p < 0.05) and a 70 to 100 percent increased IFN-secretion (all p < 0.001). In contrast, combining 9-cl3wt and 2-G115wt resulted in a signicant reduction in both IFN-secretion and lytic activity (both p < 0.01). Accordingly, the same magnitude of recognition was observed when IFN production of transductants was tested in a pamidronate titration assay (Figure 5.1D). Moreover, only the combination 9-cl3wt /2-G115wt led to decreased IFN production of transduced cells at all pamidronate concentrations (max. 100pg/ml), while all other CTE-92TCRs mediated an increased IFN-secretion (max.
1000pg/ml) as compared to wt TCR 9-G115wt /2-G115wt (max. 800pg/ml). Equal pamidronate-EC50s of 30pg/ml were calculated for all responsive 92TCR-transduced cells.
To determine whether cell-cell interaction inuences the response-kinetics differently than pamidronate stimulation, CTE-92TCR 9-G115wt /2-cl5wt and control TCR 9-G115wt /2-G115wt were tested in an effecter-to-target ratio (E:T) titration assay (Figure 5.1E), and an E:T-50 was calculated. Interestingly, T cells 66
with 9-G115wt /2-cl5wt responded differently with an E:T-50 of 1:1, compared to an E:T-50 of 10:1 calculated for control cells expressing 9-G115wt /2-G115wt . To test whether the interaction between different TCRs and ligandsthus the afnityis indeed increased, cell-cell conjugates between Daudi and T-cells expressing either potentially high (9-G115wt /2-cl5wt ) or low (9-cl3wt /2G115wt ) afnity TCRs were measured by ow cytometry. Signicantly more cell-cell interactions were observed when 9-G115wt /2-cl5wt was expressed as compared to 9-cl3wt /2-G115wt and mock transduced T-cells (Figure 5.1F). This effect did not depend on the presence of pamidronate (data not shown). G115wt /2-cl5wt is therefore a high afnity 92TCR. Hence, CTE appears to be an efcient method to rapidly engineer 92TCRs with increased afnity, mediating improved functional avidity in transduced T-cells.
5.3.4 Residues in CDR3 and J1 are involved in 92TCR stability and in mediating functional avidity of engineered T-cells. To elucidate the molecular requirements of CDR3 to mediate optimal functional avidity, alaninemutagenesis of a model CDR3 (clone G115) was performed including the whole J1 segment, as important residues have also been reported within J1 [157]. During an initial screening, ve sequence areas were found to either inuence TCR expression or functional avidity of 92TCR transduced T cells (data not shown). To clarify the degree to which single residues are responsible for impaired 92TCR expression and lower TCR mediated functional avidity, single alanine mutations were generated. The mutated and wt 2-G115 chains were expressed in combination with 9-G115wt in T-cells and tested for 92TCR expression using an antibody specic for the 2-chain (Figure 5.2A). Three single alanine mutations caused a 70 percent lower TCR expression when compared to the unmutated 2-G115wt , namely 2-G115L116A , 2-G115F118A and 2-G115V124A (Table 5.3). Comparable results were observed using antibodies directed against the 9-chain or the constant domain of the TCR (data not shown), indicating the importance of 2-G115L116 , 2-G115F118 and 2-G115V124 for stable TCR expression. The crystal structure of 92TCR G115 supports our ndings: 2-G115L116 , 2-G115F118 and 2-G115V 124 are located in hydrophobic cores (Figure 5.2B) and could thus be crucial for the structural stability of the 92TCR G115. 67
Figure 5.2: 92TCR expression and functional avidity of transduced T cells expressing single alanine mutated 2 chain of clone G115. Peripheral blood T cells were virally transduced with indicated 9 and 2 TCR chains and (A) analyzed by ow cytometry using a 2-specic antibody. Shown is the fold change in mean uorescent intensity (MFI) in comparison to wt control expressing 2-G115wt . (C) Lytic activity of transductants was tested in a 51 Cr-release assay against the tumor target Daudi (E:T 10:1). Specic lysis is indicated as fold change 51 Cr-release measured in the supernatant after 5h. Fold change was calculated as compared to reactivity of unmutated wt (2-G115wt ). p < 0.01, p < 0.001 by 1-way ANOVA. Arrows indicate mutations in 2-G115 that impaired receptor expression (dotted arrows) or functional avidity (continuous arrows). (B, D) Crystal structure of 92TCR G115 indicating relevant amino acids (red arrows), chain (in blue), CDR3 (in green), chain (in brown).
68
To address the impact of single alanine mutations on functional avidity, a 51 Crrelease assay was performed (Figure 5.2C). As expected, transductants with low TCR expression (2-G115L116A , 2-G115F118A and 2-G115V124A ) could not lyse tumor cells effectively, as they demonstrated an 80 percent lower lytic capacity when compared to cells transduced with 2-G115wt . However, T cells with mutation 2-G115L109A and 2-G115I117A (Table 5.3) properly expressed the TCR but showed a 70 percent reduced lytic activity when compared to 2- G115wt expressing cells. Similar results were obtained when alanine substitutions 2G115L109A and 2-G115I117A were introduced into the 2-chain of TCR clone 3 (data not shown). These results indicate that not only residue L109 [33, 157, 232] but also I117 in CDR3 is generally important for 92TCRs to mediate functional avidity (Figure 5.2D). However, sequence alignments between 2chains of clones 3, 5 and G115 indicated that L109 and I117 are conserved (Table 5.2), making it unlikely that these residues mediate different functional avidities of the 92TCR transduced cells studied here.
5.3.5 Inuence of CDR3 length on functional avidity of 92TCR transduced T-cells. Surprisingly, alanine substitutions during alanine-scanning mutagenesis of 92TCR G115 could replace large parts of the CDR3 domain without functional consequences. That raises the possibility that the crucial factor for the differing functional avidities of distinct 92TCR combinations is not a dened amino acid, but the relative length between the functionally important residues 2-G115L109 and the structurally important residue 2-G115L116 . Therefore, different 2-G115 length mutants were generated. Since the triple 2G115T113K115 is also important for stable surface expression (data not shown), nine length mutants (2-G115LM) with 0 to 12 alanine between 2-G115L109 and 2-G115T113 were generated and equally expressed in T-cells, again in combination with 9-G115wt (Figure 5.3A). To test the functional avidity of 2G115LM transduced T cells, an IFN ELISA in response to Daudi was performed in the presence of pamidronate (Figure 5.3B). Interestingly, engineered T cells expressing 2-G115LM0 and 2-G115LM1 were unable to produce IFN and T-cells expressing -G115LM4 or -G115LM12 secreted only about half the amount of IFN compared to 2-G115wt transduced cells. All other mutants (2-G115LM2,3,5,6,9 ) induced comparable amounts of IFN in engineered T cells 69
relative to transductants expressing 2-G115wt . Mutants with functional impairment (2-G115LM0,1,4,12 , Table 5.3) were further tested against various amounts of pamidronate and an EC50 was calculated. Despite different plateaus in maximal IFN secretion, all selected 2-G115LM transduced cells and the wt control had a comparable pamidronate-EC50 (30pg/ml) (Figure 5.3C). Length mutations were also studied in CDR3 of 92TCR G115 by engineering stretches of 1-6 alanines between 9-G115E108 and 9-G115E111.1 (9-G115LM16 ). However, this did not affect functional avidity (Figure 5.8A). This indicates that considerable alanine stretches within 9 and 2CDR3 domains can be tolerated, likely because CDR3 regions are relatively exposed parts of the TCR (Figure 5.3F). However, too short and very long alanine stretches between 2-G115L109 and 2-G115T113 in particular, as well as stretches with four alanines, are associated with reduced or absent function of a 92TCR (Figure 5.3B and C). Loss of binding in mutants with short alanine-stretches is most likely because the middle segment of CDR3 is crucial for binding to the ligand. That suggests the existence of an optimal CDR3 length for 92TCRs. Therefore, the CDR3 length within the 92TCR repertoire was studied.
5.3.6 Consequences for the physiological 92T-cell repertoire. The ImMunoGeneTics (IMGT) database [134] was searched for reported stretches between 9-G115E109 and 9-G115E111.1 as well as 2-G115L109 and 2-G115T113 . A preferential length for reported 9-chains was found for CDR3 regions corresponding to 9-G115LM2 and 9-G115LM3 , but shorter stretches were also reported (Figure 5.8B). In contrast, 2-chains with short CDR3 domains, such as 2-G115LM1 or 2-G115LM0 , were not reported (Figure 5.3D), in line with our observation that such chains are not functional. The majority of listed 92TCRs contain CDR3 lengths which correspond to 2-G115LM5,6,7 . These ndings support the hypothesis that positive selection favors 92TCRs with an optimal CDR3 length of 5-7 residues between 2-G115L109 and 2-G115T113 . However, the individual sequence might still play a role in 92TCR mediated functional avidity.
70
Figure 5.3: 92TCR expression and functional avidity of transduced T-cells expressing 92TCR G115 with 2-CDR3 length mutations. (A) 92TCR expression of indicated transductants was analyzed by ow cytometry using a TCR-pan antibody. Shown is the fold change in mean uorescent intensity (MFI) in comparison to wt control expressing 2-G115wt . (B) IFN- secretion of 2-G115LM transduced T cells against the tumor
71
target Daudi (E:T 1:1) was measured by ELISA after 24h incubation in the presence of 100M pamidronate. Shown is the fold change in IFN- production when compared to reactivity of transductants expressing wt 2-G115wt . (C) Transductants expressing 2G115LM0,1,4,12 were tested in a titration assay against the tumor target Daudi with increasing amounts of pamidronate as indicated. IFN- production was measured after 24h by ELISA. (D) Generated 2-G115LMs were matched in a BLAST search with 92TCRs described in the IMGT database. Shown is the number of citations compared to 2-G115LM of similar CDR3 length. (E) Transductants with 2-G115LM2,4,6 were compared side-byside to transductants expressing individual 92TCRs of the same CDR3 length. IFN- secretion of transduced T-cells against the tumor target Daudi (E:T 1:1) was measured by ELISA after 24h in the presence of 100M pamidronate. Shown is the fold change in IFN- production compared to reactivity of transductants expressing wt 2-G115wt . Data represent the meanSD. p < 0.01, p < 0.001 by 1-way ANOVA. (F) Crystal structure of 92TCR G115; the region that was used for alanine stretches within CDR3 is shown in white, residual CDR3 in green, chain in blue, chain in brown.
5.3.7 Inuence of the CDR3 sequence on 92TCR mediated functional avidity. To test the hypothesis that both the length and sequence of CDR3 can be important to mediate optimal functional avidity, 92TCR length mutants 2-G115LM2 , 2-G115LM4 , and 2-G115LM6 were transduced into T-cells in combination with 9-G115wt . IFN-secretion of transductants in response to Daudi was compared to cells transduced with wt sequences from 2-cl3wt (corresponds in length to 2-G115LM2 ), 2-cl5wt (corresponds in length to 2-G115LM4 ), and 2-G115wt (corresponds in length to 2-G115LM6 ) (Table 5.4). Although T-cells transduced with 2-G115LM6 and 2-G115wt did not differ in the amount of cytokine secretion, all other combinations of wt chains showed a more than two-fold increase in IFN when compared to the length mutant that selectively contained alanines (Figure 5.3E). These results were conrmed when the lytic capacity of transduced cells was tested (data not shown). The sequence in CDR3 is therefore also a signicant factor for the optimal functioning of a 92TCR. Accordingly, the sequential importance of CDR3 was studied. Thereby, 9G115LM13 were transduced into T-cells in combination with 2-G115wt . IFNsecretion of transductants in response to Daudi was compared to cells transduced with 9-cl3wt (corresponds in length to 9-G115LM1 ), 9-cl5wt (corresponds in length to 9-G115LM2 ) and 9-G115wt (corresponding to 9-G115LM3 ) 72
Figure 5.4: Functional avidity of transduced T-cells expressing 92TCR G115 with 9-CDR3 length mutations. (A) Peripheral blood T-cells were virally transduced with indicated 9 and 2 TCR chains. Lytic activity of transductants was compared side-byside to T-cells expressing individual 92TCRs of the same 9CDR3 length. Specic lysis is indicated as fold change 51 Cr-release measured in the supernatant after 5h. Data represent the meanSD. p < 0.01 by 1-way ANOVA. (B) Crystal structure of 92TCR G115 indicating 9CDR3 in gray including amino acids 9-G115A109 , 9-G115Q110 and 9-G115Q111 (red arrows), CDR3 is shown in green; chain in blue; chain in brown.
(Table 5.4). T cells expressing 9-cl3wt /2-G115wt selectively produced lower amounts of IFN- when compared to their wt equivalent 9-G115LM1 (Figure 5.4A). Previously, the same 92TCR combination was also found to mediate reduced functional avidity (Figure 5.1C and D). Interestingly, loss of activity could be restored to normal levels (referred to 92TCR G115wt ) by mutating CDR3E109 in 9-cl3wt to CDR3A109 , which demonstrates that a single change in the variable sequence of 9CDR3 is sufcient to regulate functional avidity of 92TCR transduced T cells tested here. In summary, the length and sequence of the 2CDR3 domain between L109 and T113 play a crucial role in 92TCR mediated functional avidity. In addition, the individual sequence between 9CDR3E108 and 9CDR3E111.1 can hamper the activity of a 92TCR, and in G115 CDR3A109 is most likely crucial for ligand interaction (Figure 5.4B). This provides not only the rationale for CTE-engineered 92TCRs but also for random mutagenesis within both 9 and 2CDR3 regions. 73
Figure 5.5: Anti-tumor-reactivity of T cells transduced with CTE-engineered 92TCRs. (A) Peripheral blood T cells were virally transduced with indicated 92TCRs and tested against indicated tumor cell lines and healthy control tissue. Transductants were incubated with target cells (E:T 1:1) in the presence of 10M pamidronate. IFN- production was measured after 24h by ELISA. (B) Peripheral blood T cells were virally transduced with clinical grade vector MP71 containing indicated 92TCR and tested
74
against primary AML blasts and healthy progenitor cells in an IFN- ELISpot assay (E:T 3:1) in the presence of 10M pamidronate. Data represent the meanSD. p < 0.05, p < 0.01, p < 0.001 by 1-way ANOVA. (C, D) The functional avidity of T-cells expressing CTE-92TCR 9-G115wt /2-cl5wt or control 92TCR (9-G115wt /2-G115wt ) was studied in Rag2-/-c-/- double knockout mice. After total body irradiation (2Gy) on day 0, mice were intravenously injected with 0.5 106 Daudi-luciferase cells and 107 CTE-92TCR transduced T-cells at day 1. Additionally, 6 105 IU IL-2 in IFA and pamidronate (10mg/kg body weight) were injected at day 1 and every 3 weeks until the end of the experiment. (C) Tumor outgrowth was assessed by in vivo bioluminescence imaging (BLI) by measuring the entire area of mice on both sides. Data represent the mean of 4 animals measured. p < 0.01 by 1-way ANOVA (day 42). (D) Overall survival of treated mice was monitored for 72 days.
5.3.8 From bench to bedside. CTE-engineered 92TCRs with increased activity against tumor cells are interesting candidates for TCR-gene therapeutic strategies. This leads to the question of whether changes in functional avidity mediated by CTE-92TCRs constitute a unique phenomenon of a dened 92TCR pair in response to the tumor cell line Daudi, or if this is a general response to most tumor targets. Therefore, CTE-92TCRs that mediated increased (9- G115wt /2-cl5wt ) or reduced (9-cl3wt /2-G115wt ) activity were tested against various tumors in an IFN ELISA in the presence of pharmacological concentrations of pamidronate (10M) (Figure 5.5A) [149]. In line with the observed response to Daudi, tumor reactivity was signicantly increased for K562 and SW480 when taking advantage of 9-G115wt /2-cl5wt , as compared to 9-G115wt /2-G115wt , but was signicantly reduced or even absent for all other targets using 9-cl3wt /2-G115wt . Furthermore, transductants expressing 9-G115wt /2-cl5wt also recognized the tumor cell lines Saos-2, MZ1851RC, SCC9, MDA-MB231, Fadu and MCF-7, which were not recognized by control T-cells expressing 9-G115wt /2-G115wt . Moreover, CTE-engineered T-cells with increased activity against tumor cells still did not show any reactivity towards healthy tissue such as PBMCs and broblasts. Therefore, CTEengineered 92TCRs can provide higher tumor reactivity against a broad panel of tumor cells while not affecting normal tissue, and thus have the potential to increase efcacy of TCR-engineered T cells. To assess the potential clinical impact of CTE-engineered 92TCRs, we tested whether an increased efcacy of CTE-92TCRs is also present when vectors 75
for clinical applications are used and primary blasts of patient cells are chosen targets. Therefore, 9- G115wt was linked via an E2F element to either 2G115wt or to 2-cl5wt (CTE-92TCR with highest efcacy) and cloned into the clinical grade vector MP71. Transduction efcacy was comparable (30%), and CTE-92TCR transduced T cells were tested against primary AML blasts and healthy CD34 progenitor cells in an IFN-ELISpot (Figure 5.5B). Transductants expressing 9-G115wt /2-cl5wt recognized AML blasts of patient 1 (AML1) equally, and were superior in recognizing AML blasts of patient 2 (AML2) when compared to cells with 9-G115wt /2-G115wt . Furthermore, CD34 progenitor cells were not recognized by T-cells expressing either 9-G115wt /2-cl5wt or 9-G115wt /2-G115wt . In light of these ndings, CTE-engineered TCR 9G115wt /2-cl5wt appears to be a promising candidate for clinical application. Finally, to demonstrate that CTE-92TCRs are safe and function with increased efcacy when compared to the original constructs in vivo, adoptive transfer of either CTE-(9-G115wt /2-cl5wt ) or wt-engineered T cells was studied in a humanized mouse model: protection against outgrowth of Daudi-luc in Rag2-/c-/- double knockout mice, where the frequency of T-cell infusion was reduced relative to our previously reported model [149]. This resulted in loss of protection with wt-engineered T-cells when tumor growth was measured by bioluminescence imaging (BLI) or overall survival was assessed (Figure 5.5C and D). However, CTE-engineered T cells expressing 9-G115wt /2-cl5wt clearly reduced tumor outgrowth in vivo (20, 000 counts/min, day 42) as compared to wt-engineered T cells (180, 000 counts/min, day 42). Furthermore, only mice treated with CTE-engineered T cells had an increased overall survival of 2 months relative to mice treated with T cells expressing 9-G115wt /2-G115wt . These results indicate that CTE-engineered 92TCRs efciently mediate antitumor reactivity in vivo, which points to CTE as a potential tool to optimize 92TCRs for clinical application.
5.4 Discussion
92T cells are innate lymphocytes that provide strong anti-tumor reactivity against solid and hematological cancers [46, 152]. However, despite positive preclinical data, adoptive transfer of 92T cells in clinical studies has 76
provided only limited tumor control [167]. In our study, we determined one factor preventing the successful translation of this strategy into humans: the strong tumor-reactive potential of 92T-cells is not a universal feature among all 92T-cells. Individual 92T-cell clones differ in their anti-tumor reactivity in specicity and functional avidity. The latter is substantially regulated by CDR3 domains of individual 92TCRs. Most likely, single amino acid substitutions in CDR3 impact the afnity of a 92TCR to its ligand. Finally, we have provided a means to engineer immune cells that harbor 92TCRs with increased anti-tumor reactivity in vitro and in vivo which are promising candidates for clinical applications. Limited evidence has been provided for an important role of the variable domains of a 92TCRs in mediating function. The requirement for germline encoded residues has been reported only within CDR3 and a hydrophobic residue at position 109 within CDR3 [33, 232]. Here, J1 residue I117 Mutation of was found to be important for tumor recognition, which shows that another germline encoded residue is mandatory in the 2TCR chain. residue 109 or I117 partially abrogates 92TCR-mediated activation, indicating that such residues may be crucial for a rst binding of a 92TCR to its target. However, we have shown that single amino acids in the highly variable part of CDR3, such as CDR3A109 , can clearly impact functional avidity. In the 92TCR G115, CDR3A109 points back towards the -chain, while CDR3Q110 and CDR3Q111 point towards the back of the TCR, but without contacting any -chain residues. In combination with our functional data, this modeling suggests that CDR3A109 in G115 is important for directly mediating ligand interactions. However, the A/E and Q/E substitutions are fairly non-conservative and could also indirectly alter the structure of the CDR3 loop and the TCR binding site through global conformational changes. Additionally, our data indicate that diverse amino acid compositions in 2CDR3 can impact functional avidity; the combination of a dened 9 and 2 chain is particularly important. We hypothesize that the highly variable parts of 9 and 2CDR3 complement each other to form a structure or allow a conformational change that is favorable or unfavorable for target recognition. Therefore, in contrast to previous reports [239], CDR3 alone does not correctly reect the full interaction of a 92TCR with its target. Receptor exibility is apparently necessary, for example to ad77
just to a variable cell surface of an antigen that might be presented in different ways [30], or to respond with different afnities, as recently demonstrated for T22 reactive T cells with variable CDR3-domains in mice [191]. This is also in line with a two-step binding model reported for TCRs, which requires a preliminary interaction and then adjustment [236]. However, these hypotheses are limited by the fact that no direct interaction of a 92TCR with a ligand has been reported, which prevents further visualization of the proposed receptorligand-interactions. Our data suggest that certain limitations in the 92TCR repertoire are mediated by the CDR3 length. If CDR3 becomes too short, 92TCRs are not functional, and such receptors have not been reported within the human 2TCR repertoire. Although the here used IMGT database for human 92TCR repertoire is certainly not complete, it is plausible that alterations in CDR3 in particular can limit the positive selection of a 92T cell. The functionally tolerated variability in CDR3 sequences and lengths reported in our study support the observation that 92T-cell responses to phosphoantigen stimulation do not further select for dened CDR3 sequences [65, 219]. In line, we observed identical dose-response kinetics in pamidronate titration experiments, although the magnitude of response differed signicantly. This indicates that 9CDR3 and 2CDR3 binding does not involve substrates directly regulated by pamidronate, which supports the hypothesis that a secondary signal is present [182] irrespective of the mevalonate pathway. Interestingly, T cells engineered to express dened 92TCRs did reect the functional avidity of the original 92T-cell clone tested here. Therefore, differences in CDR3 domains of 92TCRs can be responsible for differential functional avidities observed between individual 92T-cell clones. However, it is still likely that the functionality of distinct 92T cells is orchestrated by different stimulatory molecules, e.g. NKG2D [53], and co-stimulatory signals derived from molecules that are additionally expressed on 92T cells [182]. The plethora of specicities and functional avidities of distinct 92T-cell clones must therefore be taken into account when such cells are ex vivo expanded and adoptively transferred [82]. To generate tumor-reactive T-cells that mimic the reactivity of a 92T cell, we 78
proposed to engineer T-cells to express a dened 92TCR [149]. This allows rapid engineering of tumor-reactive T cells that are not limited by HLArestrictions and are readily available for nearly any patient with any cancer. Moreover, it allows a T-cell repertoire to be repleted. The functionality of this repertoire is usually heavily impaired in cancer patients [234]. To further improve functional avidity mediated by a TCR, laborious display strategies have been used for TCRs [101]. Based on our observation that mainly the 9 and 2CDR3 domains are involved in mediating functional avidity, we have proposed the concept of combinatorial-TCR-exchange (CTE) as an efcient method to design 92TCRs that mediate broad and strong anti-tumor response. The CTE-92TCR transduced T-cells tested remained tumor specic and did not respond to healthy tissue. This indicates that pairing distinct 9 chains or 2 chains only strengthens the response towards malignant cells instead of altering specicity, which reduces the likelihood of unwanted specicities. Hence, CTE-engineering provides an elegant strategy to redirect T cells more effectively against a broad range of tumor cells.
5.5
Acknowledgments
We appreciate the technical support by Anton Martens and helpful discussions with Charles Frink, Jeanette Leusen, Bart Nagelkerken, and Tuna Mutis.
79
!9-chain
!CDR3
J!P
104
105
106
107
108
109
110
111
111.1 112.1
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
!9-cl3wt !9-cl5wt !9-G115wt "2-chain
C C C
A A A
L L L
W W W
E E E
E I A
Q Q
E E E
L L L
G G G
K K K
K K K
I I I
K K K
V V V
F F F
G G G
P P P
G G G
T T T
K K K
L L L J"1
I I I
I I I
T T T
"CDR3
104
105
106
107
108
109
110
111
111.1 112.2 112.1
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
"2-cl3wt "2-cl5wt "2-G115wt
C C C
A A A
C C C
D D D
L A T
L L L
G K G
R M
T E
Y D Y
T T T
D D D
K K K
L L L
I I I
F F F
G G G
K K K
G G G
T T T
R R R
V V V
T T T
V V V
E E E
P P P
Table 5.2: CDR3 sequence alignment of 92TCR G115, clone 3 and clone 5.
!CDR3
J!1
104
105
106
107
108
109
110
111
111.1 112.2 112.1
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
!2-G115wt
impairment
single alanine mutants and deletions !2-G115L116A C A C D T !2-G115F118A C A C D T !2-G115V124A C A C D T !2- G115 C A C D T
deletionT113-K115A
L L L L
G G G G
M M M M
G G G G
G G G G
E E E E
Y Y Y Y
T T T
D D D
K K K
A L L L
I I I I
F A F F
G G G G
K K K K
G G G G
T T T T
R R R R
V V A V
T T T T
V V V V
E E E E
P P P P
stability stability stability stability
!2-G115L109A !2-G115I117A
C C
A A
C C
D D
T T
A L
G G
M M
G G
G G
E E
Y Y
T T
D D
K K
L L
I A
F F
G G
K K
G G
T T
R R
V V
T T
V V
E E
P P
avidity avidity
alanine length mutants !2-G115LM0 C !2-G115LM1 C !2-G115LM4 C !2-G115LM12 C
A A A A
C C C C
D D D D
T T T T
L L L L
0 1 4 12
x x x x
A A A A
T T T T
D D D D
K K K K
L L L L
I I I I
F F F F
G G G G
K K K K
G G G G
T T T T
R R R R
V V V V
T T T T
V V V V
E E E E
P P P P
avidity avidity avidity avidity
Table 5.3: Alanine mutations in CDR3 and the J1 segment with impact on receptor stability or functional avidity.
80
Figure 5.6: Expression pattern of individual 92T-cell clones. Surface expression of A) TCR B) NKG2D C) CD158a D) NKAT-2 and E) NKB-1 on individual 92 T-cell clones was analyzed by ow cytometry using the following antibodies: pan-TCRPE (clone IMMU510), NKG2D-PE (clone 1D11), NKAT2-PE (clone DX27), CD158a-FITC (clone HD-3E4) and NKB1-FITC (clone HP-3E4).
81
!9-chain
!CDR3
J!P 111 112 .1 .2 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 E L G K K I K V F G P G T K L I I T E L G K K I K V F G P G T K L I I T E L G K K I K V F G P G T K L I I T E L G K K I K V F G P G T K L I I T E L G K K I K V F G P G T K L I I T E L G K K I K V F G P G T K L I I T J"1 111 112 113 105 106 107 108 109 110 111 .1 .2 .3 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 A C D L L G Y T D K L I F G K G T R V T V E P A C D T L A A T D K L I F G K G T R V T V E P A C D A L K R T D T D K L I F G K G T R V T V E P A C D T L A A A A T D K L I F G K G T R V T V E P A C D T L G M G G E Y T D K L I F G K G T R V T V E P A C D T L A A A A A A T D K L I F G K G T R V T V E P
!9-cl3wt !9-G115LM1 !9-cl5wt !9-G115LM2 !9-G115wt !9-G115LM3 "2-chain
104 C C C C C C "CDR3
105 106 107 108 109 110 111 A L W E E A L W E A A L W E I Q A L W E A A A L W E A Q Q A L W E A A A
"2-cl3wt "2-G115LM2 "2-cl5wt "2-G115LM4 "2-G115wt "2-G115LM6
104 C C C C C C
Table 5.4: CDR3 length mutants of 92TCR G115 and length matched CDR3 domains of clone 3 and clone 5.
Figure 5.7: Functional avidity of 92TCR transduced T cells with different CDR3 length. 92TCRs with CDR3 length mutations (9-G115LM) or unmutated wt (9G115wt) were virally transduced in peripheral blood T cells. (A) Lytic activity of transductants was tested in a 51 Cr-release assay against the tumor target Daudi (E:T 10:1). Specic lysis is indicated as fold change 51 Cr-release measured in the supernatant after 5h. Fold change was calculated as compared to reactivity of unmutated wt (2-G115wt ). (B) Generated 9-G115LM were matched in a BLAST search with 92TCRs described in the IMGT database. Shown is the number of citations compared to 9-G115LM of the same length.
82
A)
B)
250
mock J9-G115wt /G 2-G115wt J9-cl3wt /G 2-G115wt
spots/15000 cells
200 150 100 50 0

*** *** ** *** * *** ** *** *** *** *** *** *** *
*
** *** *** *** *** ****** *** *** *** ***
**
**
******
36
94 LC L
37
48
22
83
61
69
LC L
LC L
LC L
LC L
LC L
LC L
70
93
12
20
Figure S3: TCR downmodulation and reduced alloreactivity of CTE-engineered T-cells. A) 92TCRs Figure 5.8: TCR downmodulation and reduced alloreactivity of CTE-engineered were virally transduced in peripheral blood T-cells and analyzed for TCR and TCR expression, respectively, T cells. (A ) 92TCRs were virally transduced in peripheral blood T cells and analyzed
for TCR and TCR expression, respectively, using an TCR-pan and a 2-chain using an TCR-pan and a 2-chain specific antibody. The mean fluorescence intensity (MFI) of TCR and specic antibody. The mean uorescence intensity (MFI) of TCR and TCR are indicated. (B) Allo-reactivity of CTE-engineered T cells were tested aginstEBV-LCLs EBV-LCLs TCR are indicated. B) Allo-reactivity of CTE-engineered T-cells were tested aginst inin an IFN- an IFN- ELISpot assay. Only EBV-LCLs which elicited allo-reactivity against mock are ELISpot assay. EBV-LCLs which elicited are shown. Data prepresent shown. DataOnly represent the mean SD as allo-reactivity compared toagainst mock mock transduced cells. < 0.05, the p < 0.01, p < 0.001 by 1-way ANOVA. meanSD as compared to mock transduced cells. *p<0.05, **p<0.01, ***p<0.001 by 1-way ANOVA.
83
LC L
LC L
LC L
LC L
LC L
LC L
LC L
LC L
96
84
Chapter 6
T-cells elicited by CMV-reactivation after allo-SCT cross-recognize CMV and leukemia

Wouter Scheper1, , Suzanne van Dorp1, , Sabina Kersting1 , Floor Pietersma1 , Caroline Lindemans2 , Samantha Hol1 , Sabine Heijhuurs1 , Zsolt Sebestyen1 , Cordula Grnder1 , Victoria Marcu-Malina1 , Arnaud Marchant3 , Catherine Donner4 , Bodo Plachter5 , David Vermijlen3,6 , Debbie van Baarle1,7 and Jrgen Kuball1
Leukemia, 2013
1 Department
of Hematology and Immunology, University Medical Center Utrecht, the Nether-
lands
2 Pediatric 3 Institute
Blood and Marrow Transplantation Program, University Medical Center Utrecht, the Netherlands for Medical Immunology , Universit Libre de Bruxelles, Belgium of Obstetrics and Gynecology, Hpital Erasme, Brussels, Belgium
4 Department 5 Institute 6 Institute
for Virology, University Medical Center of the Johannes Gutenberg-University, Mainz, Germany of Pharmacy, Universit Libre de Bruxelles, Belgium of Internal Medicine and Infectious Diseases, University Medical Center Utrecht, the
7 Department
Netherlands
Both
authors contributed equally to this work
85
Abstract Human cytomegalovirus (CMV) infections and relapse of disease remain major problems after allogeneic stem cell transplantation (allo-SCT), in particular in combination with CMV-negative donors or cordblood transplantations. Recent data suggest a paradoxical association between CMVreactivation after allo-SCT and reduced leukemic relapse. Given the potential of V2-negative T-cells to recognize CMV-infected cells and tumor cells, the molecular biology of distinct T-cell-subsets expanding during CMV-reactivation after allo-SCT was investigated. V2neg T cell expansions after CMV-reactivation were observed not only with conventional but also cordblood donors. Expanded T-cells were capable of recognizing both CMV-infected cells as well as primary leukemic blasts. CMV- and leukemia-reactivity were restricted to the same clonal population, whereas other V2neg T-cells interact with dendritic cells (DCs). Cloned V1TCRs mediated leukemia-reactivity and DC-interactions, but surprisingly not CMV-reactivity. Interestingly, CD8 expression appeared to be a signature of T-cells after CMV exposure. However, functionally CD8 was primarily important in combination with selected leukemia-reactive V1TCRs, demonstrating for the rst time a co-stimulatory role of CD8 for distinct TCRs. Based on these observations, we advocate the exploration of adoptive transfer of unmodied V2neg T cells after allo-SCT to tackle CMV-reactivation and residual leukemic blasts, as well as application of leukemia-reactive V1-TCR-engineered T cells as alternative therapeutic tools.
6.1 Introduction
Human cytomegalovirus (CMV) is a widely prevalent herpesvirus that, after primary infection, persists life-long in the human host. Although infections are asymptomatic in most immunocompetent individuals, reactivation of the virus in immunocompromised patients after allogeneic stem cell transplantation (alloSCT) can lead to life-threatening complications including colitis and pneumonia [27]. Moreover, CMV infection is associated with increased risk of acute graftversus-host disease (GVHD) [63, 213] and predisposes to secondary infections due to CMV-induced immunesuppression [166]. Paradoxically, recent evidence shows that CMV reactivation after allo-SCT reduces the risk of leukemic relapse [18, 64], suggesting an unexpected favorable association between CMV infection 86
CHAPTER 6. CMV- AND LEUKEMIA-REACTIVE T CELLS AFTER ALLO-SCT
and clearance of tumor. Multiple cell populations have been reported to be involved in clearance of CMV infection. A vast body of information has been gathered for CMV-specic T cells and NK cells [49]. For NK cells it has been hypothesized that they may cross-recognize CMV-infected cells and cancer cells by responding to CMVinfected residual AML blasts [64], which may contain considerable CMV copy numbers. An alternative population that might also contribute to a better control of leukemia after CMV-reactivation is represented by T cells. In recent years numerous studies have established the importance of T cells, a minor T-cell population in peripheral blood but prominently present at sites of CMV replication such as epithelial tissues, in both anti-viral immunity and tumorsurveillance [29]. Contrary to T cells, activation of T cells does not rely on antigen-presentation by MHC but is instead mediated by pathogen-derived antigens or self molecules that are upregulated on infected, transformed or otherwise stressed cells. In adult peripheral blood the major T-cell subset expresses T-cell receptors (TCRs) composed of V2 and V9 gene segments (therefore also referred to as V2pos T cells) and is activated by small, nonpeptidic phosphoantigens of pathogen or self origin [42, 78]. In contrast, T cells that reside in epithelial tissues express TCRs composed of mainly V1 or V3 chains paired with diverse V chains, and a proportion of these T cells (collectively called V2neg T cells) expresses CD8 [32, 58]. The involvement of T cells in the immune response against CMV has been established by studies in transplant patients as well as healthy individuals, showing that CMV infection associates with marked in vivo expansions of specically V2neg T cells that are reactive against CMV-infected cells [56, 118, 176, 178, 229]. Furthermore, expansion of V2neg T cells upon CMV infection was shown to correlate with clearance of the virus [128]. In addition to the anti-CMV response, numerous studies have implicated T cells in tumor host defense: T cells have been found to inltrate tumors of diverse origin in vivo [41, 68, 144] and both V2neg and V2pos subsets have been abundantly shown to be cytotoxic to cancer cells in vitro [45, 78, 86, 198]. Taken together these reports have strongly established the importance of V2neg T cells in the immune response against CMV and in tumor-surveillance. In 87
the present study we therefore evaluated the potential anti-leukemia capacity of T cells that expand upon CMV-reactivation in a population of patients with hematological malignancies receiving allo-SCT from either conventional or cordblood donors. We show that in this cohort CMV-reactivation after alloSCT associates with in vivo expansions of CMV-reactive V2neg T cells. These CMV-induced T cells are capable of cross-recognizing hematological cancers, and thus may explain the favorable effect of CMV-reactivation on risk of leukemic relapse. In addition, such cells can serve as tools either from third party donors to tackle CMV infection and leukemia or by taking advantage of hereidentied receptors to redirect T cells against leukemia.
6.2 Methods
6.2.1 Cell lines and antibodies. (see Supplementary Methods)
6.2.2 Patients, allo-SCT and blood sampling. A cohort of 26 patients with various hematological diseases (Supplementary Table 6.1), who received an alloSCT at the UMC Utrecht, from December 2005 until August 2008, was analyzed. Allo-SCT was given as curative or as rescue treatment to patients younger than 70 years with available HLA-matched related or unrelated donors, or with cordblood grafts. Patients were treated according to clinical protocols approved by the local ethics board and gave their informed consent. Outcome of allo-SCT of these patients was retrospectively analyzed in terms of hematopoietic recovery, viral reactivations, acute and chronic GVHD and progression free and overall survival [225, 226]. After allo-SCT, patients were weekly monitored for 3 months for CMV-reactivation by real-time automated CMV-DNA PCR using a TaqMan R probe. For patients with conventional stem cell donors, PBMCs of these time points were isolated and stored in liquid nitrogen until phenotypic analysis or expansion. Blood samples of cordblood patients were collected 50 100 days after transplantation. Absolute counts of CD3+, TCR+ and V2+ T-cells were determined using TRUcount tubes R (BD), according to the manufacturers protocol. PBMCs were stained for TCR, V2, CD3, CD4, CD8, CD16, CD25, CD27, CD45RO, CD56, CD80, and HLA-DR. The cohort of newborns with CMV infection has been described recently [229]. 88
6.2.3 Expansion and isolation of T-cell lines. T cells were isolated and expanded using a previously described REP-protocol [183] (see Supplementary Methods).
6.2.4 Functional
cell
and
DC
maturation
assays.
IFN-ELISPOT,
51Chromium-release and DC maturation assays were performed as previously described [125, 149] (see Supplementary Methods).
6.2.5 Cloning of TCRs and retroviral transduction of T cells. TCRs were isolated and sequenced as described in Supplementary Methods. Clone TCRs, V9V2-TCR clone G115 [4] and a HLA-A*0201-restricted WT1126-134-specic TCR [123] were transduced into T cells as described [149, 183, 209] (see Supplementary Methods).
6.2.6 Statistical analyses. Differences were analyzed using indicated statistical tests in GraphPad Prism 5.0 for Windows (GraphPad Software Inc.).
6.3
Results
6.3.1 CMV-reactivation after allo-SCT associates with expansion of V2neg T cells in both CMV-positive and CMV-negative stem cell donors. In order to test whether an increase in V2neg T cells is observed during CMVreactivation after allo-SCT, blood samples of 26 patients with umbilical cordblood (n = 10) and adult stem cell donors (n = 16, Supplementary Table 6.1) were collected after allo-SCT and analyzed by ow cytometry. Nine patients (56%) with adult stem cell donors developed a CMV-reactivation within 3 months after allo-SCT. In agreement with previous work in the context of alloSCT [118] but also other transplantation settings [56, 178], CMV-reactivation but not EBV-reactivation associated with a signicant increase in absolute numbers of circulating donor T cells in patients with conventional adult stem cell donors (Fig. 6.1A, left panel; Supplementary Table 6.2). Expression analysis of CD45RO and CD27 indicated signicantly lower percentages of nave (CD45ROneg CD27pos ) T cells in CMV-reactivating patients (Supplementary 89
Fig. 6.7A), suggesting expansion of effector cells in these patients. Also all patients with cordblood grafts, thus CMV-nave donors, that developed a CMVreactivation (n = 6) had signicantly increased numbers of circulating T cells when compared to time-matched non-CMV-reactivating patients (n = 4; Fig. 6.1A, right panel), although due to logistic challenges no time-course evaluation was possible in this cohort. In both patient populations the increase in T cells was due to an increase in the V2neg subset (Fig. 6.1B), while V2pos T cells did not signicantly differ between CMV-reactivating and non-reactivating patients (adult grafts: median 2.17 versus 2.39 cells/l, P = 0.37; cordblood grafts: median 4.38 versus 0.81, P = 0.38). No differences were observed in total CD3+ T-cell numbers between CMV-reactivating and non-reactivating patients (adult grafts: median 437 versus 355 cells/l, P = 0.80; cordblood grafts: median 58 versus 97 cells/l, P = 0.38). In order to assess whether the increase in V2neg T cells during CMV-reactivation was mainly driven by T cells expressing a public V8V1-TCR, which has been reported to play a substantial role in congenitally infected newborns [229], clonality of increased cell fractions was analyzed by spectratyping. However, when analyzing clonality of V1, V2 and V3 T cells no such enrichment was observed in selected patients (Supplementary Fig. 6.8). Interestingly, the increase in T cells in CMV-reactivating patients preceded the increase of T cells, as a signicant difference in T cells between CMV-reactivating and non-reactivating patients was not observed until 3 months after allo-SCT (Supplementary Fig.6.7B). Together, these data conrm in a rather small but apparently representative cohort that CMV-reactivation after transplantation of CMV-seropositive stem cell grafts associates with a signicant increase of donor V2neg T cells. Notably, V2neg T cells were also elicited during CMV-reactivation when CMV-nave cordblood grafts were used.
6.3.2 Patient-derived V2neg T cells specically recognize CMV-infected and transformed cells ex vivo. To functionally evaluate whether V2neg T cells that expanded in vivo upon CMV-reactivation could indeed contribute to an anti-CMV immune response, T cells were isolated from CMV-reactivating patients and analyzed ex vivo. Patient-derived bulk, V2neg , and V2pos T-cell subsets were coincubated with CMV-infected broblasts and T-cell activation 90
Figure 6.1: Selective expansion of V2neg T cells in patients with CMV-reactivation after allo-SCT. (A) Blood samples of patients with conventional adult stem cell donors were collected weekly after allo-SCT (left panel) or during CMV-reactivation in patients with cordblood-derived grafts (right panel), and presence of T cells was analyzed by ow cytometry. In the left panel, median values of all patients are presented. For patients with conventional donors, most CMV-reactivations were observed in the second and third month after transplantation. (B) Presence of V2neg T cells was analyzed in patients with conventional and cordblood donors by ow cytometry. In patients with conventional stem cell donors (left panel) V2neg T cells were measured in the second and third month after allo-SCT, in patients with cordblood grafts at the same timepoint as in (A). In box plots, the line at the middle is the median, the box extends from the 25th to 75th percentile, and the whiskers extend down to the lowest value and up to the highest. Mann Whitney U test was used to assess differences between CMV-positive and CMV-negative patients, and signicant differences are indicated ( p < 0.05).
91
was measured by IFN-ELISPOT. T cells isolated from patients with conventional stem cell grafts secreted signicantly higher levels of IFN upon contact with CMV-infected cells when compared to uninfected controls (Fig. 6.2A). In line with previous studies [56, 118], CMV-reactivity of patient-derived T cells was mediated exclusively by V2neg T cells, but not V2pos T cells (Fig. 6.2A). Importantly, T cells isolated from cordblood patients produced IFN in response to and were able to specically lyse CMV-infected cells (Fig. 6.2B). Previously it has been reported that T cells that expand upon CMVreactivation are able to cross-recognize solid cancer cells [47, 90], however crossreactivity with leukemic cells has not been reported. Therefore, patient-derived V2neg and V2pos T cells were coincubated with a variety of hematological cancer cell lines and primary acute myeloid leukemia (AML) blasts. Indeed, CMV-reactive V2neg T cells were able to specically recognize lymphoma (Daudi), leukemia (BV173, K562 and KCL22), and myeloma (U266) cell lines, and most importantly primary AML blasts (Fig. 6.2C). In contrast, V2pos T cells from selected patients responded to hematological cell lines but not to primary AML samples, although reactivity could be induced after treating AML cells with the bisphosphonate pamidronate, a compound that induces the accumulation of V2pos T-cell-activating phosphoantigens in treated cells [149] (data not shown). In summary, CMV-reactivation after allo-SCT associates with an increase in multipotent V2neg T-cell populations from both CMV-positive and nave stem cell donors that are able to recognize both CMV-infected cells and hematological tumor cells.
6.3.3 CMV-reactive V2neg T-cell clones cross-recognize leukemic cells, including primary AML blasts. To investigate whether cross-reactivity of V2neg T cells to leukemic blasts and CMV-infected broblasts is restricted to different clonal populations, V2neg T cells were cloned by limiting dilution. All generated clones carried V2pos TCRs and expressed the natural killer receptor NKG2D (Fig. 6.3A and Supplementary Fig. 6.9). Two clones (B11 and E1) heterogeneously expressed CD8. CMV-reactivity of generated clones was 92
subsequently analyzed by coincubation with either CMV-infected or uninfected broblasts. Two V1pos T-cell clones (B11 and E1) responded to CMVinfected broblasts by increased IFN production, while two other clones (D3 and E2) did not (Fig. 6.3B). None of the clones produced TNF in response to CMV-infected cells (data not shown). Next, CMV-reactive V1pos T-cell clones B11 and E1 were coincubated with the hematological tumor cell lines U266, T2 (T- and B-lymphoblastoid cell line), EBV-LCL (Epstein-Barr virus-transformed lymphoblastoid cell line) or primary AML blasts. Both CMV-reactive clones displayed a potent IFN-response against all (clone B11) or most (clone E1) tested tumor cell lines as well as primary AML samples (Fig. 6.3C) but not healthy broblasts. However, leukemia-reactivity was not a feature of all isolated clones, as clones D3 and E2 did not produce IFN or TNF in response to leukemic cell lines or blasts (data not shown). Together, these data suggest that here-isolated CMV-reactive clones are able to cross-recognize hematological tumor cells.
6.3.4 The interaction of V2neg T cells with DCs is clonally segregated from CMV- and leukemia-reactivity and is mediated by individual TCRs. Because isolated clones D3 and E2 did not show a cytokine response against CMVinfected broblasts nor leukemic cells (data not shown), we hypothesized that such clones elicited after CMV infection are involved in maturation of dendritic cells (DCs) [136], and thereby may aid in mounting adaptive immune responses. Therefore, monocyte-derived immature DCs were cultured alone or in the presence of V1pos T-cell clones and expression of the maturation markers CD80 and CD86 on DCs was measured after 48 hours. Selectively in the presence of V1 T-cell clones D3 and E2, but not CMV- and leukemia-crossreactive clones B11 and E1, a substantial and signicant increase in CD80/CD86+ DCs was observed compared to immature DCs alone (Fig. 6.3D), resembling the phenotype of DCs matured by the classical maturation cocktail (prostaglandin E2, IL-1 , IL-6 and TNF) [108]. However, no detectable production of IL12p70 was induced by V1 T-cell clones (data not shown). Importantly, induction of this characteristic mature phenotype of DCs by clones D3 and E2 was observed in the absence of CMV-infected cells or CMV virions in cocultures, indicating that maturation of DCs by V2neg T-cell clones is independent of CMV-antigen.
93
Figure 6.2: Specic recognition of CMV-infected and leukemic cells by patient V2neg T cells. (A) T cells isolated from patients with conventional adult stem cell donors were expanded and cultured ex vivo before MACS-sorting and use in functional analysis. Sorted V2pos or V2neg T cells were subsequently cocultured for 18 hours with CMVinfected or -uninfected human foreskin broblasts and T-cell activation was measured by IFN-ELISPOT. Results from two representative patients are shown. (B) Left panel: T cells from patients with cordblood transplantations were tested for CMV-reactivity
94
as in (A). Unsorted T-cells isolated from these patients predominantly (up to 90%) consisted of V2neg T-cells, as determined by ow cytometry. Right panel: Killing capacity of T-cells from cordblood patients against CMV-infected broblasts was determined by coincubating T-cells and CMV-infected broblasts in a 4 6hr 51Chromium-release assay. Uninfected broblasts served as control. Data from three different patients are shown. (C) MACS-sorted V2pos and V2neg T-cells from the same patients as in (A) were used to test anti-tumor recognition. V2pos or V2neg T-cells were cocultured with indicated hematological cancer cell lines or primary leukemic blasts (at 3:1 target:effector ratio) in IFN-ELISPOT. For both T-cell populations healthy unsorted T-cells served as negative control target. Error bars represent SEM. Student t test (A, B) or one-way ANOVA (C) was used to assess differences between T-cell responses ( p < 0.05; p < 0.01; p < 0.001).
To test whether induction of maturation markers on DCs by V1pos T-cell clones D3 and E2 is mediated by their TCRs, V- (V4, V8 and V9) and V1chains of here-generated clones were sequenced (Supplementary Table 6.3) and retrovirally transduced into T cells. In agreement with our previous data on V9V2 TCRs [149], all clone-derived 1-TCRs were efciently expressed in both CD4+ and CD8+ T cells and down-regulated endogenous TCRs (Supplementary Fig. 6.10). The involvement of individual 1-TCRs in the induction of the mature-like phenotype of DCs was subsequently analyzed by incubating transduced T cells with immature monocyte-derived DCs. Selectively 1 TCRs that were isolated from clones E2 and D3 but neither 1-TCRs E1 and B11 nor mock-transduced cells induced a marked (3.5 to 9-fold) upregulation of CD80/CD86 on DCs (Fig. 6.3E) and increased TNF secretion in culture supernatants (Supplementary Fig. 6.11A). In addition, a higher mean expression of CD40, CD83 and HLA-DR was detected on DCs after coincubation with E2TCR- and D3-TCR-transduced T cells, but not T cells expressing 1 TCRs E1 and B11. This phenotype depended on both CD1c, a lipid-presenting molecule previously reported to be involved in V2neg T-cell-mediated DC maturation [136], and TNF (Supplementary Fig. 6.11B). As was observed in experiments with original clones, DCs did not produce detectable levels of IL12p70 (data not shown). Taken together, these data show that distinct clonal populations within the expanded V2neg T-cell subset are responsible for CMV- and leukemiareactivity and for interacting with DCs, and that the V2neg T-cell-DC interaction involves dened 1TCRs. 95
Figure 6.3: CMV-reactive V2neg T-cell clones cross-recognize cancer cells, but do not interact with DCs. (A) V2neg T-cell clones were generated by limiting dilution and phenotyped by ow cytometry. (B) CMV-reactivity of generated clones was determined by incubating clones alone or in combination with CMV-infected or uninfec-
96
ted broblasts (at 3:1 target:effector ratio) for 18 hours in an IFN-ELISPOT assay. (C) CMV-reactive V2neg T-cell clones E1 and B11 were cultured alone, with hematological cancer cell lines or with primary leukemic blasts for 18 hours and IFN-production was determined by ELISPOT. Healthy human broblasts served as negative control. (D) V2neg T-cell clones were incubated with immature DCs for 48 hours and the percentage CD80/CD86 double-positive DCs was measured by ow cytometry. (E) TCR - and -chains of original V2neg T-cell clones were sequenced and retrovirally transduced into T-cells. Transfer of DC-interacting capacities was tested by culturing mock-transduced T-cells or T-cells expressing clone-derived 1-TCRs with immature DCs for 48 hours and measuring expression of maturation markers by ow cytometry. Error bars represent SEM. Student t test (B) or one-way ANOVA (C, D) were applied to assess differences between T-cell responses ( p < 0.05; p < 0.01; p < 0.001).
6.3.5 Cancer-reactivity, but not CMV-reactivity, is mediated by distinct TCRs. To formally conrm that individual TCRs of V1 clones B11 and E1 mediate CMV-reactivity, as reported after in primo utero CMV-infection [229], T cells transduced with clone-derived 1TCRs and the previously reported CMV-reactive V8V1TCR [229] were incubated with CMV-infected or uninfected broblasts. Surprisingly, only T cells expressing the public CMV-reactive V8V1-TCR but neither here-cloned CMV-reactive nor non-reactive 1-TCRs produced IFN after contact with CMV-infected cells (Fig. 6.4A and data not shown), suggesting that CMV-recognition by original clones B11 and E1 must rely on alternative surface receptors. In order to test the mechanism involved in tumor recognition, clones B11 and E1 were tested for expression of NKp30, a receptor recently reported to be involved in anti-tumor reactivity by V1pos T cells [45]. However, here-isolated clones did not express NKp30 (Supplementary Fig. 6.9). Thus, an alternative mechanism must mediate tumor-reactivity and could include the individual TCRs. Therefore, 1TCR- and mock-transduced T cells were cocultured with hematological cancer cell lines or primary AML blasts and T-cell activation was determined by IFN-ELISPOT. Selectively T cells transduced with 1TCRs of CMV- and cancer-reactive clones B11 and E1 but not mock-transduced T-cells were able to recognize both hematological cancer cell lines and primary AML cells, while healthy T cells were not recognized (Fig. 6.4B). Importantly, cancerreactivity of both 1TCRs could be extended to solid cancers, since pharyngeal (Fadu) and breast cancer (MDA-MB231) cell lines also activated TCR97
Figure 6.4: Isolated 1-TCRs transfer cancer-reactivity, but not CMV-reactivity, to Tcells. (A) T-cells transduced with empty vector, with a public CMV-reactive 1-TCR or with clone-derived 1-TCRs were incubated for 18 hours with CMV-infected or uninfected foreskin broblasts and IFN-secretion was measured by ELISPOT. (B) T-cells transduced with empty vector or with either the B11 or E1 1-TCR were cultured with primary AML blasts and hematological and solid cancer cell lines in an IFN-ELISPOT. Healthy T-cells were used as negative control. Error bars represent SEM. Student t test (A) or one-way ANOVA (B) were used and signicant differences are indicated ( p < 0.05; p < 0.01; p < 0.001).
transduced T cells (Fig. 6.4B). T cells transduced with 1TCRs D3 or E2 produced neither IFN nor TNF against tested cancer cells (data not shown). Thus, cancer-reactivity of selected V1pos T-cell clones is mediated by their respective 1-TCRs and this reactivity can be transferred to previously nonreactive T-cells.
6.3.6 CD8 functions as critical coreceptor for selected tumor-reactive 1TCRs. 1TCRs isolated from clones B11 and E1 can be suitable tools to redirect T-cells against leukemias as reported for V9V2-TCR clone G115, which reprograms both CD4+ helper and CD8+ cytotoxic T cells against a broad 98
panel of tumor cells [149]. Thus, we questioned whether also here-isolated tumor-reactive V1pos TCRs are able to redirect both subsets of T-cells, CD4+ and CD8+ , against cancer cells. CD4+ and CD8+ T-cells expressing tumor-reactive 1TCRs B11 or E1 were therefore separated and incubated with T2 or Daudi target cell lines. CD4+ and CD8+ T cells transduced with 1TCR E1 produced similar levels of IFN in response to tumor target cells (Fig. 6.5A). In sharp contrast, 1TCR B11 was able to reprogram CD8+ but not CD4+ T cells, even though the introduced B11 TCR was expressed at slightly higher levels in CD4+ than in CD8+ T cells (Supplementary Fig. 6.10B), as reported previously [124, 149]. This suggested that for full T-cell activation this TCR requires a molecule present on CD8+ but not CD4+ T-cells, such as NKG2D or CD8. To address this, CD8+ T-cells transduced with the B11 1TCR were preincubated with blocking antibodies against CD8, CD8 , or NKG2D and subsequently coincubated with Daudi or T2 target cells. Blocking of NKG2D, which is expressed on most CD8+ but not CD4+ T cells and can amplify - and T-cell responses [53, 85], had only an effect on target cell recognition when T-cells were transduced with a 92TCR as reported [149] (data not shown). Strikingly, blocking CD8 but not CD resulted in a marked decrease in IFN-secretion when compared to T-cells pretreated with control antibody (Fig. 6.5B). Blocking capacity of CD8 antibody was conrmed by inhibiting MHC class I-restricted T cells. CD8-blocking on CD4+ T-cells expressing the B11 or E1 1TCRs served as additional negative controls and did not inuence T-cell responses. Thus, the CD8 but not CD8 domain is important in the ligand interaction of the B11 1-TCR. These data indicate that depending on the particular TCR, tumor-reactivity is mediated by CD8-dependent and -independent mechanisms, suggesting e.g. different afnities of here-cloned TCRs to their ligands. The original clone B11 expressed the CD8 homodimer but not the CD8 heterodimer (Supplementary Fig. 6.9). To test whether CD8 was also involved in activation of the original B11 T-cell clone, clone B11 (CD8+) and clone E1 (CD8low ) were cocultured with T2 target cells in the presence of CD8- or CD8 blocking antibodies. Similar to the effect on T-cells transduced with the B11 1TCR, blocking of CD8 signicantly inhibited IFN-production by the original clone (Fig. 6.5C). However, the effect of CD8-blocking was less pronounced 99
on the original clone compared to B11-transduced T cells, most likely due to lower expression of CD8 on the parental clone when compared to CD8 expression on transduced T-cells (data not shown). Again, blocking of CD8 did not affect IFN-secretion, as expected based on the CD8-positive phenotype of clone B11. As was observed in E1-transduced T-cells, CD8-blocking did not affect activation of clone E1 (Fig. 6.5C). To corroborate these observations, additional CD8-positive V1 T-cell clones were generated from a different donor and the effect of CD8-blocking on activation of clones was analyzed. Of nine CD8+ clones tested, blocking CD8 but not CD8 inhibited activation of one clone that reacted to the colorectal cancer cell line SW480 (clone FE11), as measured by reduced IFN secretion (Fig. 6.5D). CD8-blocking had no effect on activation of the parental polyclonal V 2neg CD8+ T-cell line of this donor nor of two other donors (data not shown), suggesting that CD8-dependence of dened T-cell clones is not a general phenomenon yet observed in a substantial fraction (2 out of 10) of isolated clones. CD8 was in the majority of isolated clones not functionally involved in tumor-reactivity, questioning whether CD8 rather plays a general role in CMV-reactivity. In order to assess whether an increase in CD8 expression on T cells might be linked to CMV-infection in vivo, the cohort of conventional stem cell donors was analyzed for CD8 expression by ow cytometry. Strikingly, CMV-reactivating patients had signicantly more circulating CD8+ T cells compared to non-reactivating patients (Fig. 6.6A). This observation was conrmed in a complementary cohort of congenitally CMV-infected newborns (Fig. 6.6B). In this cohort, CD8 expression on T cells associated with a differentiated effector (CD27neg/low CD28neg ) [229] phenotype (Supplementary Fig. 6.12). Microarray gene expression proling revealed highly upregulated expression of CD8 but not CD8 upon CMV-infection (Fig. 6.6C), and ow cytometry on blood samples of infected individuals indeed showed that CMVassociated expression of CD8 on T cells is preferentially of the homodimer (Fig. 6.6D). Of note, CD8+ T cells sorted from the same CMV-infected newborns did not show increased expression of CD8 (Fig. 6.6D). To test whether CD8 plays a functional role in CMV-reactivity by V2neg T cells, clones B11 and E1 were coincubated with CMV-infected or uninfected broblasts in the presence of CD8-blocking antibody. However, blocking CD8 inhibited 100
not only the specic recognition of CMV-infected cells but also the occasionally observed background reactivity of clone B11 but not clone E1 against freshly plated broblasts (data not shown), suggesting that CD8 may interact rather with a general stress-antigen than an antigen specic for CMV infection. In summary, these data show that CD8 expressed on human V2neg Tcells associates with CMV-infection in vivo and is able to function as a critical costimulator on selected clones as well as on TCR-reprogrammed T cells when coincubated with tumor cells.
Figure 6.5: CD8 acts as a coreceptor for selected 1TCRs. (A) CD4+ and CD8+ Tcells transduced with B11 or E1 1-TCRs were sorted and subsequently cocultured with T2 or Daudi cell lines in an IFN-ELISPOT. (B) CD4+ and CD8+ transduced T-cells were coincubated with T2 target cells as in (A), but now in the presence of a control antibody or blocking antibodies against CD8 or CD8 . T cells expressing a WT1126-134specic TCR [123] that were coincubated with T2 cells pulsed with 10 6 M WT1126134 peptide served as positive control for CD8- and CD8 -blocking. (C) Original clones B11 and E1 were incubated with T2 target cells as in (B). (D) Clone FE11 was generated by limiting dilution, phenotyped by ow cytometry (left panel), and coincubated with SW480 target cells as in (B). Error bars represent SEM. Student t test (A) or one-way ANOVA (B, C, D) were used and signicant differences are indicated ( p < 0.05; p < 0.01; p < 0.001).
101
Figure 6.6: CMV reactivation after allo-SCT and congenital CMV infection associate with increased expression of CD8 on T cells. (A) The percentage of CD8+ T cells
102
of patients with conventional stem cell donors was measured in the second and third month after allo-SCT by ow cytometry. (B) Cord blood from fetuses congenitally infected (n = 11) or not infected (n = 16) was collected at term delivery and the percentage of T cells expressing CD8 was analyzed by ow cytometry. (C) Gene expression analysis of T cells derived from three CMV-infected newborns versus T cells derived from three CMV-uninfected newborns. MA plot of differentially expressed genes in T cells upon CMV infection. M (log2 of fold change) reects the differential expression of a gene. Positive and negative values indicate genes which are up- and down-regulated, respectively, upon CMV infection. A (mean expression) reects the overall expression level of a gene. Note that a similar gure, with indication of other genes, has been published before [229]. The highly up-regulated expression of CD8 RNA is indicated. The down-regulation of CD28 RNA is indicated as well for comparison. (D) The majority of CD8 on T cells of congenitally infected newborns is composed of the CD8 homodimer. Percentages of CD8+ T cells and CD8+ T cells expressing the CD8+ CD8 + phenotype were determined by ow cytometry in cordblood samples from eight congenitally infected newborns (left panel). Representative ow cytometry plots (right panel) illustrate the staining patterns of CD8 and CD8 on T cells and T cells. Mann Whitney U test (A, B) and Student t test (D) were used and signicant differences are indicated ( p < 0.05; p < 0.001).
6.4
Discussion
The contribution of V2neg T cells to controlling CMV-infection has received considerable attention in recent years, and it is now well-established that these unconventional T-cells play important roles in the immune response to CMVinfection [56, 90, 118, 176, 229]. Combined with their widely reported reactivity towards a variety of (mainly solid) tumors [41, 86, 144], this has made V2neg T cells a promising cell population for immunotherapeutic application. In the present study we demonstrate that V2neg T cells that expand upon CMVreactivation after allo-SCT are capable of responding to both CMV-infected and leukemic cells. Additionally, by demonstrating that tumor-reactivity of V2neg T cells can be transferred by TCR gene-transfer, and by identifying a novel role for CD8 in the antigen-restriction of TCRs, we provide a solid basis for the therapeutic exploration of V2neg T cells and their TCRs. Our observation that the occurrence of a single event (i.e. CMV-infection) is able to induce expansion of T-cell subsets with anti-CMV- and anti-leukemiareactivity, including reactivity against primary leukemic blasts, provides an al103
ternative explanation for recent unexpected ndings of a reduced relapse rate in patients with CMV-reactivation after allo-SCT [18, 64]. Furthermore, it is in line with a report in kidney transplant patients demonstrating that expansion of V2neg T cells following CMV-infection associated with a reduced risk of developing solid cancer post-transplantation [47]. T cells isolated from these patients reacted against both CMV-infected cells and epithelial tumor cells in vitro. Thus, although CMV-reactivation after allo-SCT is still associated with substantial non-relapse-related mortality (e.g. GVHD, colitis, and secondary infections), reactivation of the virus reduces the risk of mortality due to relapse of leukemia, and we show that one possible link is a CMV-induced expansion of leukemia-reactive T cells. This hypothesis is further substantiated by clinical data demonstrating that increased numbers of T cells after allo-SCT are associated with improved disease-free survival, without higher incidence of GVHD [79]. Mechanistically, little is known about the requirements for T-cell activation, and the identity of the molecules on CMV-infected and leukemic cells that are recognized by here-generated T-cell clones so far remain elusive. Dualreactivity of V2neg T-cell clones to CMV and solid cancer cells has been reported and has led to the hypothesis that TCRs of dual-reactive cells recognize shared antigens on CMV-infected and transformed cells [90, 235]. However, our TCR-gene transfer experiments show that cancer-reactivity, but not CMV-reactivity is mediated by 1TCRs isolated in this study, indicating that alternative immune receptors may be responsible for CMV-reactivity of the original clones or that the TCR is involved but depends on additional molecules not expressed on T cells. In line with this, it was recently reported that the TCR isolated from a CMV-reactive V4V5 clone requires costimulation by CD11a-CD18 (LFA-1) [235]. However, here-isolated CMV-reactive clones E1 and B11 as well as T cells transduced with their respective TCRs expressed high levels of CD11a (see Supplementary Fig. 6.13), suggesting that other mechanisms must be involved. Alternatively, it was recently shown that V2neg T cells could be stimulated by IgG-opsonized CMV virions via the IgG receptor CD16 (FcRIIIa), independent of TCR-engagement [48]. However, here-isolated CMV-reactive clones did not express CD16 (see Supplementary Fig. 6.9). 104
We report here for the rst time that in human T cells CD8 functions as restriction element for target recognition by distinct 1TCRs. Although long described to be expressed on T cells [32, 58], the function of CD8 on these cells has so far remained unknown. In our experiments, blocking CD8 resulted in a marked and signicant inhibition of tumor recognition by different clones and distinct 1TCRs, which was not observed when tested on the bulk population. These data put CD8 into the eld of coreceptors for 1TCRs for a dened subset of tumor-reactive T cells. Moreover, we report that CMV-infection associates with an increase in CD8-expressing T cells in both allo-SCT patients as well as congenitally infected newborns, suggesting a link between CD8 and the immune response against CMV in vivo. However, the functional involvement of CD8 in T-cell-mediated CMV-reactivity remains to be further dened. Within the T-cell compartment, CD8-positive T-cells are enriched in mucosal tissues such as intestine and these cells are described to display a characteristic innate-like phenotype [39]. However, on these cells CD8 does not function as a classical MHC class I-binding TCR coreceptor as CD8 does, but more likely serves as suppressor of TCR-mediated T-cell activation [40]. A subset of NK cells also expresses CD8, and these cells posses greater killing capacity than CD8-negative NK cells [208]. This effect was attributed to enhanced resistance to apoptosis that was specically mediated through CD8-signaling [2]. Superior cytotoxicity of CD8-expressing NK cells has been associated with clinical remission of leukemia patients [142, 143], indicating that CD8 on innate immune cells may be relevant to clinical outcome after allo-SCT. Finally, CD8 on murine innate-like intestinal T cells was shown to enhance TCR-mediated T-cell activation by binding the nonclassical MHC-I molecule thymus leukemia (TL) [135]. Thus, the expression of CD8 on innate(-like) immune cells may indicate a universal role for CD8 as regulatory receptor in innate immune responses. To tackle CMV-infections in immuno-compromised patients, several clinical trials have focused on the adoptive transfer of CMV-reactive T cells [69, 133]. However, major obstacles are presented by the MHC-restricted antigenrecognition of T cells and the challenge to generate sufcient numbers of CMV-reactive T cells within the time constraints of severe infection [98]. Our data suggest that V2neg T cells are an interesting alternative source of CMV105
reactive T cells for such patients as we observe that in vivo generated V2neg T cells react against not only CMV-infected cells but also leukemic cells in vitro. Moreover, we demonstrate that CMV-reactive T cells can also be obtained from the nave umbilical cordblood repertoire, underscoring the value of this third-party stem cell source for application in allo-SCT, in particular also for patients with CMV-negative donors. In summary, we advocate the exploration of adoptive transfer of unmodied V2neg T cells in CMV- and tumorimmunotherapies and the application of leukemia-reactive V1-TCR-engineered T cells. Clinical trials will need to be pursued in order to test efcacy and safety of the application of such strategies.
6.5 Acknowledgements
We thank the members of the stem cell facility at the UMC Utrecht for technical assistance. We also thank Margreet Brouwer for her expert technical assistance.
6.6 Supplementary Methods

6.6.1 Antibodies and ow cytometry. Antibodies used for ow cytometry included: TCR-APC (clone B1, BD), TCR-PE (clone IMMU510, Beckman Coulter), TCR-FITC (clone 11F2, BD), V2-PE and -FITC (clone B6, BD), V1-FITC (clone R9.12, Beckman Coulter), TCR-PE-Cy5 (IP26A, Beckman Coulter), CD3-eFluor450 (clone OKT3, eBioscience), CD3-pacic blue (clone SP34-2, BD), CD4-PE-Cy7 (clone RPA-T4, BD), CD8-APC (clone RPAT8, BD), CD8-PE-Cy7 (clone SFCI21Thy2D3, Beckman Coulter), CD8 -PE (clone 2ST8.5H7, BD), CD16-PE (clone CB16, eBioscience), CD27-APC-eFluor780 (clone 0323, eBioscience), CD27-APC (clone L128, BD), CD28-ECD (clone CD28.2; Beckman Coulter), CD40-APC (clone HB14, Biolegend), CD45ROPE-Cy7 (clone UCHL1, BD), CD56-PE (clone B159, BD), CD80-PE (clone L307.4, BD), CD83-FITC (clone HB15e, BD), CD86-PE-Cy5 (clone IT2.2, eBioscience), NKp30-APC (clone P30-15, Biolegend), NKG2D-APC (clone 1D11, BD), CD158a(NKAT1)-FITC (clone HP-3E4, BD), CD158b(NKAT2)-PE (clone DX27, BD), NKB1(NKAT3)-FITC (clone DX9, BD), HLADR-APC-Cy7 (clone L243, Biolegend). All allo-SCT samples were processed with FACSCanto-II or LSR-II ow 106
cytometers (BD) and analyzed with FACSDiva software (BD). Whole cord blood samples derived from infected and uninfected newborns were run on the CyAn ow cytometer and data were analyzed using Summit 4.3 (Dako).
6.6.2 Cell lines and primary acute myeloid leukemia cells. Daudi, K562, KCL22, T2, BV173, SW480, MDA-MB231, U266, foreskin broblasts and Phoenix-Ampho cell lines were obtained from ATCC. EBV-LCL was kindly provided by Phil Greenberg (Seattle, WA). Fadu was kindly provided by Niels Bovenschen (UMC Utrecht, The Netherlands). Fibroblasts and Phoenix-Ampho cells were cultured in DMEM supplemented with 1% Pen/Strep (Invitrogen) and 10% FCS (Bodinco), all other cell lines in RPMI with 1% Pen/Strep and 10% FCS. Fresh PBMCs were isolated by Ficoll-Paque (GE Healthcare) from buffy coats supplied by Sanquin Blood Bank (Amsterdam, The Netherlands). Where indicated, foreskin broblasts were infected with culture supernatants of broblasts previously infected with human CMV strain AD169 at a multiplicity of infection (MOI) of 2. After 24 hours, infected and uninfected broblasts were washed before being used in functional assays. Frozen primary acute myeloid leukemia (AML) samples were a kind gift from Matthias Theobald (Mainz, Germany) and were collected in compliance with GCP and Helsinki regulations.
6.6.3 Expansion and isolation of T-cell lines. PBMCs were stimulated for 14 days with 1g/ml PHA-L (Sigma-Aldrich), 50U/ml IL-2 (Novartis Pharma), 5ng/ml IL-15 (R&D Systems), and irradiated allogeneic PBMCs, Daudi and EBV-LCLs. Fresh IL-2 was added twice a week. After rst expansion, polyclonal T-cell lines were obtained by MACS-isolation (TCR+ T-cell isolation kit, Miltenyi Biotec) with a purity of > 90% and were further expanded using again the REP-protocol. V2pos and V2neg T-cell fractions were obtained by MACS-depleting V2pos T cells from bulk cultures using V2TCR-PE antibody and anti-mouse IgG microbeads (Miltenyi Biotec). T cells isolated from patients receiving cordblood grafts typically contained up to 90% V2neg T cells and were therefore not further MACS-sorted. V2neg T-cell clones were generated from a CMV-seropositive healthy donor by limiting dilution. All Tcell cultures were stimulated biweekly using the REP-protocol.
107
6.6.4 Spectratyping and microarray experiments. Spectratyping analysis and microarray experiments were performed as previously described [229]. Microarray data and procedures were deposited at Array Express (www.ebi.ac.
uk/arrayexpress) under accession no. E-MEXP-2055.
6.6.5 Dendritic cell maturation assay. Monocytes were isolated from PBMCs by plate adhesion and differentiated into immature dendritic cells (iDCs) by culturing for 4 days in AIM-V medium in the presence of 500U/ml IL-4 (Peprotech) and 800U/ml GM-CSF (Peprotech). Next, iDCs were cocultured with T-cells at a ratio of 1:1 for 48 hours and expression of CD40, CD80, CD83, CD86 and HLADR was measured by ow cytometry. Where indicated, CD1c-blocking antibody (clone L161, Biolegend), TNF-blocking antibody (clone MAb1, eBioscience), or control antibody was added to cultures at a concentration of 20g/ml. Secretion of TNF and IL12p70 was measured by ELISA (eBioscience).
6.6.6 Functional T-cell assays. IFN-ELISPOT was performed by coculturing 15, 000 T-cells and 50, 000 target cells (ratio 0.3:1) for 24 hours in nitrocellulosebottomed 96-well plates (Millipore) precoated with anti-IFN antibody 1-D1K (Mabtech). Plates were washed and incubated with biotinylated antibody 7-B61 (Mabtech) followed by streptavidin-HRP (Mabtech). IFN spots were subsequently visualized with TMB substrate (Sanquin) and spots were quantied using ELISPOT Analysis Software (Aelvis). With regard to T-cell clones, reactivity to CMV-infected cells and cancer cells was generally determined in the same experiment. Where indicated, blocking of CD8 was performed using 10g/ml anti-CD8 antibody clone OKT8 (eBioscience), blocking of CD8 with 10g/ml anti-CD8 clone 2ST8.5H7 (Abcam), and NKG2D-blocking with 10g/ml anti-NKG2D clone 149810 (R&D Systems).
51 Chromium-release
assays was performed as described [125, 149]. Target cells

51 Cr-release
were labeled overnight with 150Cu 51 Cr and subsequently incubated with Tcells in four effector-to-target ratios (E:T) between 30:1 and 1:1. supernatant was measured 4 6hr later. in
108
6.6.7 Cloning of TCR genes and retroviral transduction of T-cells. mRNA of T-cell clones was isolated using the Nucleospin RNA-II kit (Macherey-Nagel) and reverse-transcribed using SuperScript-II reverse transcriptase (Invitrogen). TCR- and TCR-chains were amplied by PCR using V1 (5-gatcaagtgtggcccagaag-3), V2-5 (5-ctgccagtcagaaatcttcc3), V8 (5-gctgttggctctagctctg-3) and V9 (5-tccttggggctctgtgtgt3) sense primers, and C (5-ttcaccagacaagcgaca-3) and C (5ggggaaacatctgcatca-3) antisense primers. PCR products were sequenced by Baseclear c (Leiden, the Netherlands). Codon-optimized sequences of clone TCRs were subsequently synthesized by Geneart R (Regensburg, Germany) and subcloned into pBullet. Packaging cells (Phoenix-Ampho) were transfected with gag-pol (pHIT60), env (pCOLT-GALV) [209] and pBullet constructs containing TCR-chain-IRESneomycine or TCR-chain-IRES-puromycin, using Fugene6 (Promega). PBMCs preactivated with CD3 (30ng/ml) (clone OKT3, Janssen-Cilag) and IL-2 (50U/ml) were transduced twice with viral supernatant within 48 hours in the presence of 50U/ml IL-2 and 4g/ml polybrene (Sigma-Aldrich). Transduced T-cells were expanded by stimulation with CD3/CD28 Dynabeads (0.5 106 beads/106 cells) (Invitrogen) and IL-2 (50U/ml) and selected with 800g/ml geneticin (Gibco) and 5g/ml puromycin (Sigma) for one week. Where indicated, CD4+ and CD8+ TCR-transduced T cells were separated by MACS-sorting using CD4- and CD8-microbeads (Miltenyi Biotec). Following selection, TCRtransduced T-cells were stimulated biweekly using the REP-protocol [183].
109
Patient groups CMV-reactivation Conventional graft cohort N Median age (range) Sex M/F (%) Donor/recipient relation RD MUD Diagnosis AML CLL CML MM NHL Conditioning NMA MA ATG GVHD CMV+ Patient CMV+ Donor OS at 2 years Cordblood graft cohort N Median age (range) Sex M/F (%) Diagnosis AML ALL JMML NMID NMMD Conditioning NMA MA ATG GVHD CMV+ Patient CMV+ Donor OS at 2 years 0 (0) 6 (100) 6 (100) 2 (33) 6 (100) 0 (0) 5 (83) 0 (0) 4 (100) 4 (100) 1 (25) 4 (100) 0 (0) 3 (75) 2 (33) 2 (33) 0 (0) 1 (17) 1 (17) 0 (0) 3 (75) 1 (25) 0 (0) 0 (0) 6 2 (1 10) 67/33 4 2 (1 15) 0/100 9 (100) 0 (0) 5 (56) 8 (89) 8 (89) 5 (56) 5 (56) 9 (100) 0 (0) 3 (43) 3 (43) 4 (57) 1 (14) 5 (71) 1 (11) 1 (11) 1 (11) 5 (56) 1 (11) 4 (57) 1 (14) 0 (0) 2 (29) 0 (0) 5 (56) 4 (44) 4 (57) 3 (43) 9 56 (33-62) 89/11 7 49 (35-68) 57/43 No CMV-reactivation
Table 6.1:
Patient characteristics. ALL, acute lymphocytic leukemia; AML, acute my-
110
eloid leukemia; ATG, antithymocyte globuline; CLL, chronic lymfocytic leukemia; CML, chronic myeloid leukemia; CMV, cytomegalovirus; F, female; GVHD, graft-versus-host disease; JMML, juvenile myelomonocytic leukemia; M, male; MA, myeloablative; MM, multiple myeloma; NHL, non-Hodgkins lymphoma; NMA, non-myeloablative; NMID, non-malignant immunodeciency; NMMD, non-malignant metabolic disease; OS, overall survival; RD, related donor; MUD, matched unrelated donor.
Patient groups EBV-reactivation Conventional graft cohort N % T-cells / lymphocytes % T-cells / lymphocytes % CD56pos CD16pos cells / CD3neg lymphocytes 6 30.4 1.2 34.1 10 50 1.2 74.6 0.66 0.81 0.01 No EBV-reactivation p-value
Table 6.2: Comparison of T cells, T cells and NK cells between patients with and without EBV-reactivation. EBV, Eppstein-Barr virus. p-values: Mann-Whitney U test.
Figure 6.7: Nave T cells and total T cells after allo-SCT. (A) PBMCs of patients with conventional adult stem cell donors were collected weekly after allo-SCT, and the percentage of nave CD27pos CD45ROneg T cells was analyzed by ow cytometry. (B) Absolute counts of T cells after allo-SCT with conventional donors was measured by ow cytometry. A Mann Whitney U test was performed at all time points and signicant differences are indicated ( p < 0.05).
111
Figure 6.8: TCR clonality analysis of T cells from CMV-reactivating patients. Representative spectratype analyses of V1, V2 and V3 TCR clonality in blood samples of CMV-reactivating patients that received stem cells from conventional adult donors (A) or cordblood donors (B). All patients were analyzed during CMV-reactivation. The CDR31 size of 11 amino acids, corresponding with the CDR31 size of the public V8V1 TCR [229], is indicated with arrows.
112
Figure 6.9: Phenotyping of V2neg T-cell clones. V2neg T-cell clones were generated by limiting dilution and surface expression of indicated receptors was measured by ow cytometry. Gating was established based on appropriate isotype controls.
113
Figure 6.10: Efcient retroviral expression of 1TCRs in CD4+ and CD8+ T cells. (A) Isolated 1TCRs were retrovirally transduced into T cells and surface expression of endogenous TCR and introduced TCR was determined by ow cytometry. Indicated in plots are percentages of quadrants and MFIs of TCR and TCR stainings. (B) Transduced T cells were costained for CD4 and expression levels (MFI) of TCRs on CD4 (i.e. CD8+ ) and CD4+ T cells is indicated in plots.
114
Figure 6.11: Upregulation of DC maturation markers by TCR-transduced T cells involves TNF and CD1c. (A) Immature DCs (iDCs) were cultured alone, with mocktransduced T cells, or with T cells expressing clone-derived TCRs for 48 hours and TNF levels in culture supernatants were measured by ELISA (one-way ANOVA: p < 0.05, p < 0.01). (B) iDCs were cultured as in (A) but now in the presence of control antibody or blocking antibodies against CD1c or TNF. After 48 hours CD83 expression on DCs was measured as a representative marker of DC maturation.
115
Figure 6.12: CD8 expression is associated with a differentiated effector phenotype (CD27neg/low CD28neg ) of T cells in CMV-infected newborns. Association of CD8 expression with CD27neg/low T cells (left panel) and CD28neg T cells (right panel). Stainings are representative for 11 CMV-infected newborns. Plots represent lymphocytes gated on CD3+ TCR+ phenotype.
Figure 6.13: Expression of CD11a on original clones, TCR-transduced T cells and Jurkat cells. Expression of CD11a is shown as a fold increase of MFI of the specic staining over MFI of the staining with control antibody.
116
TCR chains
N 5J-REGION J name
Clone V name
3V-REGION
B11
TRGV4*02 TGTGCCACCTGGGATGG.
CCAGGAAGG ....TATTATAAGAAACTCTTT TRGJ1*01 .AATTATTATAAGAAACTCTTT TRGJ2*01
...TTATTATAAGAAACTCTTT TRGJ1*01
D3
TRGV8*01 TGTGCCACCTGG...... TCCAGGGGGG ...CCACTGGTTGGTTCAAGATATTT TRGJP1*01
E1
TRGV9*01 TGTGCCTTGTGGGAG... ACTTCCTACCTC
E2
TRGV9*01 TGTGCCTTGTGGGAG...
CCC
117
N1 P D1-REGION N2 D2-REGION N3 P
TCR chains
5J-REGION J name D1 name D2 name
Clone V name
3V-REGION
B11
TRDV1*01 TGTGCTCTTGGGGAAC.
AGGTCG AAAAGTGGCA AATT ACA .....CTAC ...GGGGGAT... CACCA
..TTCCTA. TTGATCT ....GGGGAT...
TCC
GT
ACACCGATAAACTCATCTTT TRDJ1*01 TRDD2*01 TRDD3*01 ......ATAAACTCATCTTT TRDJ1*01 TRDD3*01 ..ACCGATAAACTCATCTTT TRDJ1*01 TRDD3*01
D3
TRDV1*01 TGTGCTCTTGGG.....
E1
TRDV1*01 TGTGCTCTTGGGGAACT CGGACGGGGAGGGA T ACTGGGGGA....
E2
TRDV1*01 TGTGCTCTTGGGGAACT
..TGGGGGATAC. AGCCTT
CTTTGACAGCACAACTCTTCTTT TRDJ2*01 TRDD2*01 TRDD3*01
Table 6.3: CDR3 sequences of V2neg T-cell clones ImMunoGeneTics (IMGT R ) JunctionAnalysis output (www.imgt.org)
118
Chapter 7
General discussion
Allogeneic stem cell transplantation is used as a curative treatment for haematological malignancies, as shown in Part I for multiple myeloma. Treatment effect is due to both the conditioning regimen before transplantation and the graft-versus-tumor effect after transplantation. However, relapses often occur and also non-relapse mortality (NRM) and morbidity is very high in both myeloablative (1-year NRM 32%) and nonmyeloablative (1-year NRM 20%) allo-SCT [59, 204]. The main goals of future allo-SCT strategies remain therefore to improve GVL and prevent or develop options to effectively treat GVHD. In Part II and III of this thesis different approaches to reach these goals were presented. First, the role of B cells and B-cell depletion in treatment and prevention of chronic GVHD, as described in Part II, will be discussed. Secondly, T cells and options to use them as an additional tool for immunotherapy for the effective treatment and therapy of leukemia and infectious diseases, as investigated in Part III, will be considered.
7.1
The role of B cells in GVHD
Several, mainly retrospective, studies have shown clinical efcacy of B-cell depletion with the monoclonal antibody rituximab, as a treatment for steroidrefractory chronic GVHD [36, 51, 172, 179, 242], thereby suggesting a signicant role of B cells in the development of chronic GVHD. However the exact mechanism behind these promising results has yet to be elucidated. In theory both donor and host B cells could mediate allo-reactivity in chronic GVHD [197]. 119
In both scenarios B cells could induce allo-reactivity directly, via production of auto-antibodies. Increase in several auto-antibodies is seen in various studies [36, 73, 212], however there is no consensus on which antibodies are important in the pathogenesis of chronic GVHD. Also several B cell subsets and activation markers are proposed to be of importance, though mostly conicting results are published on the exact phenotype of B cells involved in the pathogenesis of chronic GVHD [83, 192]. Alternatively or additionally it is suggested that B cells can act not only as direct mediators [192], but also indirectly, by presenting allo-antigens and therefore inducing allo-reactivity in the T-cell population [197]. Also this could play a role for host (direct presentation) and donor B cells (indirect presentation). Pre-emptive B-cell depletion before allo-SCT, thus depletion of host B cells has been reported to affect mainly the frequency of acute GVHD [60, 113, 180]. In contrast, our data (Chapter 3) suggest that pre-emptive B-cell depletion can also substantially affect the incidence of chronic GVHD. These results suggest that not only donor B-cells can contribute to chronic GVHD, also host B cells can prime the new developing immunity towards allo-reactivity, and therefore B-cell depletion as part of the conditioning regimen could be considered to prevent chronic GVHD. However, in Chapter 3 we also revealed a potential threat by depletion of B cells prior to allo-SCT. Severe acute GVHD occurred faster in patients treated with rituximab prior to allo-SCT. In mice, host B cells were shown to produce IL-10 and function as regulatory cells, attenuating the course of acute GVHD [188]. However two retrospective studies showed an even reduced incidence of acute GVHD in patients treated with rituximab prior to allo-SCT [113, 180], although they did not comment on the time of onset of acute GVHD. The patient numbers used to show the fast onset of severe acute GVHD in Chapter 3 were very limited. Since in a larger cohort a benecial effect was seen on the incidence of acute GVHD and also on NRM, this threat might not be of great inuence, although prospective studies addressing this issue remain necessary to answer this question. Prophylactic depletion of donor B cells after allo-SCT could also be an alternative preventive strategy. The rst clinical trial using rituximab as prophylactic treatment for chronic GVHD, administered 2 months post allo-SCT, showed an
120
CHAPTER 7. GENERAL DISCUSSION
incidence of total chronic GVHD of 20% [11]. This is a very promising result, since generally up to 70% of patients develop chronic GVHD [10, 130, 170, 225]. Although It has been documented that B-cell depletion in the rst 6 months after (T-cell depleted) allo-SCT caused prolonged life-threatening cytopenias [155], Arai and colleagues showed a non relapse mortality of only 3% [11], suggesting this problem could be overcome. Up to now no results for a randomized controlled trial have been published, therefore this chapter cannot be closed yet. Despite here-presented evidence for the prophylactic B-cell depletion in order to prevent GVHD, B-cell depletion is so far mainly used once chronic GVHD occurred. In Chapter 4 we demonstrate that patients with a sclerodermatous skin phenotype were shown to be more responsive to B-cell depletion than patients with an ulcerative skin phenotype. In patients with an ongoing response, the response could be correlated with a change in B-cell homeostasis. B-cell imbalances were found in patients with active chronic GVHD of the sclerodermatous skin phenotype. These imbalances were corrected or even reset by treatment with rituximab. After depletion the new B-cell compartment was comparable to patients without chronic GVHD. This suggests that chronic GVHD with a sclerodermatous skin phenotype is B-cell driven and patients with this phenotype will benet from anti-B cell therapy. To a large extent B-cell homeostasis is controlled by B cell activating factor belonging to the TNF family (BAFF) [195, 217]. In patients with chronic GVHD disturbances in B-cell homeostasis have been reected in disturbances in BAFF levels [3, 192]. In patients with active chronic GVHD a high BAFF/B-cell ratio and low numbers of nave B cells suggest a shift towards activated memory B cells, responsible for the allo-reactivity that leads to chronic GVHD [116]. In our prospectively studied cohort and in a prospective study by Kim et al, no signicant differences in BAFF levels were found [115]. Moreover, we found high numbers of nave B cells with an antigen-presenting phenotype. These controversial results could be due to the fact that patients in our cohort received higher amounts of corticosteroids then in other cohorts, which have been described to inhibit BAFF and therefore could inuence B-cell activation [105]. In steroid-refractory chronic GVHD patients, that are responsive to B-cell depletion, B cells seem to escape the inhibitory effect of steroids and an imbalanced
121
B-cell homeostasis can be maintained. The exact denition of B-cell imbalances needs further exploration, since different authors show different and even contradictory results [3, 83, 192, 226]. Regardless of the underlying mechanism one could argue that patients with a sclerodermatous skin phenotype and proven B-cell imbalances should be treated early with B-cell depletion, thus even before clinical signs of chronic GVHD. Although B cell depletion appears as promising strategy in order treat patients with chronic GVHD, none of the patients developed a complete resolution of GVHD and other did not respond at all. In Chapter 4 a more ulcerative skin phenotype of chronic GVHD was described as less responsive. In these patients lesions of the skin were more comparable to those seen in patients with acute GVHD. B-cell phenotype was not altered and B-cell depletion did not result in improvement of symptoms of chronic GVHD. In our cohort 17% of patients suffered from this ulcerative type of chronic GVHD, not responding to B-cell depletion. This is a substantial population; therefore one could argue that in times of personalized medicine, including B-cell depletion in the conditioning regimen for all patients seems rather blunt. When these patients can be identied a priori, they can also be treated differently. A suggestion worth researching is treating these patients with tools now used in the treatment of acute GVHD, such as extracorporeal photopheresis [9, 93] and mesenchymal stromal cells (MSC). The immune modulatory effect of MSCs is successfully used in the treatment of steroid-refractory acute GVHD [186, 230], as well as rst line treatment [112]. Also in chronic GVHD preliminary data show remission of both scleroderma and ulcera after administration of MSCs in steroid-refractory patients [245]. Finally also tyrosine-kinase inhibitors, such as imatinib and nilotinib, could be an alternative option to treat unresponsive patients or to reach complete resolution of symptoms in partially responding patients. Imatinib has been shown to decrease the amount of brosis seen in pulmonary brosis [1]. Also in systemic sclerosis and chronic GVHD there has been rationale to treat patients with tyrosine-kinase inhibitors [156, 169]. In these patients broblasts show an activated myobroblast phenotype as the result of stimulation by not only higher levels of pro-brotic cytokines such as transforming growth factor
122
(TGF)- [94, 154] and PDGF [211], but also due to production of stimulatory anti-PDGFR antibodies [16, 212]. By inhibiting the platelet-derived-growthfactor (PDGF)- receptor, tyrosine-kinase inhibitors can attenuate the activation of broblasts by PDGF and TGF- as well as their stimulation by anti-PDGFR antibodies. First clinical trials treating patients with steroid-refractory chronic GVHD with imatinib show overall response rates up to 79% [145, 146, 168]. To improve the treatment response of patients with steroid-refractory chronic GVHD treated with B-cell depletion, we recently extended the clinical trial, described in Chapter 4. Patients with steroid-refractory chronic GVHD with a skin involvement will be treated with rituximab and thereafter nilotinib will be given for 6 months (http://www.hovon.nl/studies/studies-perziektebeeld/sct.
html?action=showstudie&studie_id=90&categorie_id=11).
In conclusion, we propose that chronic GVHD should be treated as a heterogeneous disease. This is clinically suggested by the fact that various sites are affected with a different type of pathology [92, 201] and supported by the observation that immunological phenotype of chronic GVHD can differ [3, 116, 226]. Patients, who will benet from B-cell depletion, might be identied by their aberrant B-cell prole early and they should receive an early treatment with rituximab. Patients, who show no imbalances in B-cell phenotype, will most likely not respond to B-cell depletion and should be treated differently. However, larger randomized trials are needed for this conclusion backed up by phase I/II trials to explore new drugs for the treatment of chronic GVHD.
7.2
T cells as valuable tool for cellular immunotherapy
T cells directed against minor-histocompatibility antigens restricted to haematological stem cells or even leukemic cells have been shown to induce a powerful antitumor response, while GVHD is prevented [205, 233]. In addition, also T cells directed against dened tumor-associated antigens have been reported to control leukemia after allo-SCT [117, 222]. Thus, T cells are undoubtedly important tools in immunotherapy of both solid and haematological malignancies. However, major limitations arise through the fact that TCRs 123
recognize their target in an HLA dependent manner and usually only one or a limited amount of peptides [117]. Consequently, each individual patient requires and optimal set of TCR pMHC interactions in order to control leukemia and viral infections. In addition, immune responses against tumor-associated antigens frequently underlie mechanisms of tolerance, thus an optimal set of TCR-pMHC interactions can frequently not identied for patients [124]. Finally, frequencies of in particular minor-histocompatibility-antigen reactive T-cells is usually low. Thus an extensive amplication [183] or engineering [124] of immune cells is required for a potential clinical application. The here-discussed multiple obstacles hamper so far the broad application of an individualized immune therapy with T cells in nearly all clinical scenarios including allo-SCT and urged us to search for additional immunological tools in order to control leukemia and infections. We propose in this thesis that one cell population, namely T cells and their receptors might overcome multiple major limitations of T cells. Firstly, we demonstrate in Chapter 5 that the frequency of tumor-reactive T cells in healthy individuals is usually high. Secondly, we demonstrate as reported by others that T cells and in particular the TCR recognizes its target HLAindependently [72], thus T cells or their receptors could be used for as broader patient population. We investigated in detail the benet of TCRs as tool for adoptive immune therapies and suggest many advantages. Firstly, engineering of high afnity 92TCRs is feasible and allows the selection of a highly tumor-reactive receptor without substantial side effects [88]. Secondly, in contrast to TCRs [123], TCRs do not pair with endogenous TCR chains [149, 223], thus this strategy prevents generation of unwanted (potentially autoreactive) specicities. Thirdly, these genetically modied cells are not expected to be immunogenic as they do not harbor fusion or non-self-proteins as reported for multiple genetically modied T cells [21]. Forthly, T cells and their receptors do not display substantial GVHD properties [111]. Fifthly, expression of 92TCR in T cells helps to overcome defects in proliferation seen in 92Tcells. Finally, in contrast to most TCRs, expression of high-afnity TCRs leads to efcient reprogramming of carrier cells (CD4+ and CD8+ T-cells) and to enhanced immune
124
responses [88, 149]. Collectively, these features plus the MHC-independent fashion of antigen recognition make TCR an ideal tool for the generation of novel T-cell based cellular immune cells [88, 111, 149, 194]. Although many reasons have been suggested as to why T cells and their receptors could be valuable tools for immune therapies, up till now clinical studies exploiting the antitumor reactivity of T cells in adoptive T-cell transfer are rather disappointing [167]. In Chapter 5 we gave a possible explanation and solution for this problem. We showed that the strong anti-tumor reactivity of 92T cells was not a feature of the total population of these cells. Individual 92T-cell clones showed differences in their anti-tumor response in specicity and functional avidity. We showed that these features are regulated by the CDR3 regions of both 9 and 2 chain. Single amino acid changes in the highly variable part of the CDR3 region of the 9 chain, diverse amino acid compositions in the CDR3 region of the 2 chain and length of total CDR3 region of both 9 and 2 chain can all inuence functional avidity. We showed that a specic combination of a 9 and a 2 chain is particularly important and therefore we hypothesize that highly variable parts of the CDR3 regions of both 9 and 2 chain complement each other to form a structure that denes target recognition. We used a new strategy, combinatorial-TCR-chain-exchange (CTE), which results in the expression of newly combined 9- and 2-TCR chains on engineered T-cells. CTE allows us to design 92TCRs that mediate broad and strong anti-tumor response. These TCRs can be used for rapid engineering of tumor-reactive T-cells that are not limited by HLA restrictions and will not surprise us with unwanted specicity. CTE-engineered 92TCRs provided higher tumor reactivity against a broad panel of tumor cells, derived from both solid as well as haematological malignancies. Therefore one could argue that in theory these cells could be used for any patient with any type of cancer, making them a precious additional tool for future cellular immunotherapies. However, there still remain unsolved issues. One for example is the limited life span of engineered effector cells. Expansion of tumor specic, adoptively transferred T cells has only been observed in some patients and only after lymphodepletion [62, 162]. In attempts to improve the survival of these cells in a patient several techniques have been tried, for example stimulating transferred T cells
125
in vivo with IL-2 and pamidronate [149], transfer of central memory phenotype T cells after in vitro treatment with cytokine cocktails [240], using dual reactive TCRs against both persistent viruses and leukemia [227]. These studies all reported improvement of survival of these cells in vitro or in murine models. More data in clinical studies is required to optimize the transfer of transduced T cells. The use of a dual reactive TCR could efciently prolong survival and therefore anti-tumor reactivity of transferred tumor specic T cells. This could be due to more frequent encounter of both viral and tumor antigens by the TCR and therefore a more sustained impulse to proliferate is available [227]. This method furthermore offers a possibility to tackle two problems at once. CMV reactivations after allo-SCT cause life-threatening complications, such as CMV pneumonia or colitis [28]. A T cell showing both anti-tumor and CMV-reactivity would be a very powerful tool in creating a therapy to treat leukemia and at the same time prevent CMV-reactivation after transplantation. Again T cells seem interesting candidates to achieve this goal, since they recognize self-antigens upregulated on transformed and virally infected cells. Their role in both antitumor and anti-viral surveillance has well been established [29]. CMV infection induces in both healthy individuals [176] as well as immunocompromised patients after allo-SCT [118] as well as in utero [229] in vivo expansion of specically V2neg T-cells. In addition V2neg T-cells have been shown cytotoxicity to cancer cells [41, 45]. Interestingly, CMV reactivation after allo-SCT reduces the risk of leukemic relapse [18, 166]. This suggests an association between the immune response against CMV infection and clearance of tumor. In Chapter 6 we demonstrate that CMV-reactive V2neg T cells from both conventional and cordblood donors expanded in vivo after allo-SCT. These T cells were capable of cross-recognizing haematological cancers. This might explain the reduced risk of relapse of cancer in patients with CMV-reactivation. V2neg T cells could therefore very well be used as a tool to attack both CMV infection and leukemia. Since CMV-reactive T cells could also be obtained from the nave umbilical cordblood repertoire, our results emphasize the important role of these cells as a third-party stem cell source for allo-SCT. Recognition of CMV infected cells and cancer cells by V2neg T cells is demonstrated to be TCR dependent [90, 110, 235]. Therefore we hypothesized that
126
here-cloned TCRs could recognize antigens shared by CMV-infected and transformed cells. However, transduction of the isolated TCRs into an T cell showed that only cancer-reactivity, but not CMV-reactivity was transferred. Thus, either an alternative immune receptor is responsible for CMV-reactivity of the original clones, or the TCR depends on additional molecules that are not expressed on T cells. The exact mechanism of interaction of T cells with their target remains, despite thorough and extensive research, somewhat of a mystery. Involvement of the TCR itself in both tumor recognition [41, 86, 110] and infectious targets [17, 90, 96, 199] is undeniable. However, a role for natural killer (NK) receptors, such as NKG2D and NKp30, has also been described in target recognitions by T cells [53, 87, 102, 177, 184]. Our group demonstrated that transfer of a tumor-specic 92TCR to NKG2Dneg and NKG2Dpos T cells resulted in the same anti-tumor reactivity. However blocking NKG2D on 92TCR transduced NKG2Dpos T cells decreased reactivity to tumor cells [149]. These results taken together with the diverse package of suggested target molecules, such as phosphoantigens, F1ATPase, MHC-like molecules such as MICA and ULBPs, tempt us to hypothesize that recognition by T cells is not solely done by the TCR or solely done by NK-receptors, but is the result of an interaction between those receptors. Since many T cells also express NKG2D this feature can also be transferred to T cells and be of use in clinical applications of adoptive transfer of tumor specic TCRs. Within the here-presented thesis, we were able to also elucidate a novel role of CD8 as an alternative costimulatory molecule necessary for V2neg T cells to interact with tumor cells [194]. Blocking of CD8 on CD8- expressing distinct T-cell clones as well as transduced CD8+ T-cells resulted in a signicant inhibition of tumor recognition. However, the precise role of CD8 for tumor recognition in combination with V2neg TCRs needs further investigation.
7.3
Conclusion
Our presented papers contribute to the improvement of allo-SCT by offering several possibilities to shape the graft into a more rened tool to treat haematological malignancies. We demonstrated that altering the composition of a graft 127
by early or late B-cell depletion provides interesting options in order to reduce GVHD [225, 226]. In addition we introduced T cells and their receptors as interesting additional tools for (engineered) adaptive immune therapies [88, 194], which might one day replace the conventional allo-SCT.
128
Bibliography
[1] A Abdollahi, M Li, G Ping, C Plathow, S Domhan, F Kiessling, L B Lee, G McMahon, H J Grne, K E Lipson, and P E Huber. Inhibition of platelet-derived growth factor signaling attenuates pulmonary brosis. J Exp Med, 201(6):925935, Mar 2005. [2] E G Addison, J North, I Bakhsh, C Marden, S Haq, S Al-Sarraj, R Malayeri, R G Wickremasinghe, J K Davies, and M W Lowdell. Ligation of CD8alpha on human natural killer cells prevents activation-induced apoptosis and enhances cytolytic activity. Immunology, 116(3):354361, Nov 2005. [3] J L Allen, M S Fore, J Wooten, P A Roehrs, N S Bhuiya, T Hoffert, A Sharf, A M Deal, P Armistead, J Coghill, D A Gabriel, R Irons, A Essenmacher, T C Shea, K Richards, C Cutler, J Ritz, J Serody, A S Baldwin, and S Sarantopoulos. B cells from patients with chronic GVHD are activated and primed for survival via BAFFmediated pathways. Blood, 120(12):25292536, Sep 2012. [4] T J Allison, C C Winter, J J Fourni, M Bonneville, and D N Garboczi. Structure of a human gammadelta T-cell antigen receptor. Nature, 411(6839):820824, Jun 2001. [5] E Alyea, E Weller, R Schlossman, C Canning, P Mauch, A Ng, D Fisher, J Gribben, A Freeman, B Parikh, P Richardson, R Soiffer, J Ritz, and K C Anderson. Outcome after autologous and allogeneic stem cell transplantation for patients with multiple myeloma: impact of graft-versus-myeloma effect. Bone Marrow Transplant, 32(12):11451151, Dec 2003. [6] E P Alyea, H T Kim, V Ho, C Cutler, J Gribben, D J DeAngelo, S J Lee, S Windawi, J Ritz, R M Stone, J H Antin, and R J Soiffer. Comparative outcome of nonmyeloablative and myeloablative allogeneic hematopoietic cell transplantation for patients older than 50 years of age. Blood, 105(4):18101814, Feb 2005. [7] J H Anolik, J W Friedberg, B Zheng, J Barnard, T Owen, E Cushing, J Kelly, E C Milner, R I Fisher, and I Sanz. B cell reconstitution after rituximab treatment of lymphoma recapitulates B cell ontogeny. Clin Immunol, 122(2):139145, Feb 2007. [8] J. H. Antin. Reduced-Intensity Stem Cell Transplantation: . . . whereof a little More than a little is by much too much. King Henry IV, part 1, I, 2. Hematology Am Soc Hematol Educ Program, pages 4754, 2007. [9] N Apisarnthanarax, M Donato, M Krbling, D Couriel, J Gajewski, S Giralt, I Khouri, C Hosing, R Champlin, M Duvic, and P Anderlini. Extracorporeal photopheresis therapy in the management of steroid-refractory or steroid-dependent
129
cutaneous chronic graft-versus-host disease after allogeneic stem cell transplantation: feasibility and results. Bone Marrow Transplant, 31(6):459465, Mar 2003. [10] F R Appelbaum. Haematopoietic cell transplantation as immunotherapy. Nature, 411(6835):385389, May 2001. [11] S Arai, B Sahaf, B Narasimhan, G L Chen, C D Jones, R Lowsky, J A Shizuru, L J Johnston, G G Laport, W K Weng, J E Benjamin, J Schaenman, J Brown, J Ramirez, J L Zehnder, R S Negrin, and D B Miklos. Prophylactic rituximab after allogeneic transplantation decreases B-cell alloimmunity with low chronic GVHD incidence. Blood, 119(25):61456154, Jun 2012. [12] B Arden, S P Clark, D Kabelitz, and T W Mak. Human T-cell receptor variable gene segment families. Immunogenetics, 42(6):455500, 1995. [13] M Arora, P B McGlave, L J Burns, J S Miller, J N Barke, T E Defor, and D J Weisdorf. Results of autologous and allogeneic hematopoietic cell transplant therapy for multiple myeloma. Bone Marrow Transplant, 35(12):11331140, Jun 2005. [14] K Atkinson, M M Horowitz, R P Gale, D W van Bekkum, E Gluckman, R A Good, N Jacobsen, H J Kolb, A A Rimm, and O Ringdn. Risk factors for chronic graft-versus-host disease after HLA-identical sibling bone marrow transplantation. Blood, 75(12):24592464, Jun 1990. [15] T C Barnes, M E Anderson, and R J Moots. The many faces of interleukin-6: the role of IL-6 in inammation, vasculopathy, and brosis in systemic sclerosis. Int J Rheumatol, 2011:721608721608, 2011. [16] S S Baroni, M Santillo, F Bevilacqua, M Luchetti, T Spadoni, M Mancini, P Fraticelli, P Sambo, A Funaro, A Kazlauskas, E V Avvedimento, and A Gabrielli. Stimulatory autoantibodies to the PDGF receptor in systemic sclerosis. N Engl J Med, 354(25):26672676, Jun 2006. [17] C Behr, R Poupot, M A Peyrat, Y Poquet, P Constant, P Dubois, M Bonneville, and J J Fournie. Plasmodium falciparum stimuli for human gammadelta T cells are related to phosphorylated antigens of mycobacteria. Infect Immun, 64(8):2892 2896, Aug 1996. [18] C E Behrendt, J Rosenthal, E Bolotin, R Nakamura, J Zaia, and S J Forman. Donor and recipient CMV serostatus and outcome of pediatric allogeneic HSCT for acute leukemia in the era of CMV-preemptive therapy. Biol Blood Marrow Transplant, 15(1):5460, Jan 2009. [19] I Bellantuono, L Gao, S Parry, S Marley, F Dazzi, J Apperley, J M Goldman, and H J Stauss. Two distinct HLA-A0201-presented epitopes of the Wilms tumor antigen 1 can function as targets for leukemia-reactive CTL. Blood, 100(10):38353837, Nov 2002. [20] G M Bendle, C Linnemann, A I Hooijkaas, L Bies, M A de Witte, A Jorritsma, A D Kaiser, N Pouw, R Debets, E Kieback, W Uckert, J Y Song, J B Haanen, and T N
130
BIBLIOGRAPHY
Schumacher. Lethal graft-versus-host disease in mouse models of T cell receptor gene therapy. Nat Med, 16(5):565570, May 2010. [21] C Berger, M E Flowers, E H Warren, and S R Riddell. Analysis of transgenespecic immune responses that limit the in vivo persistence of adoptively transferred HSV-TK-modied donor T cells after allogeneic hematopoietic cell transplantation. Blood, 107(6):22942302, Mar 2006. [22] N L Berinstein, A J Grillo-Lpez, C A White, I Bence-Bruckler, D Maloney, M Czuczman, D Green, J Rosenberg, P McLaughlin, and D Shen. Association of serum Rituximab (IDEC-C2B8) concentration and anti-tumor response in the treatment of recurrent low-grade or follicular non-Hodgkins lymphoma. Ann Oncol, 9(9):9951001, Sep 1998. [23] K Besold, N Frankenberg, S Pepperl-Klindworth, J Kuball, M Theobald, G Hahn, and B Plachter. Processing and MHC class I presentation of human cytomegalovirus pp65-derived peptides persist despite gpUS2-11-mediated immune evasion. J Gen Virol, 88(Pt 5):14291439, May 2007. [24] B B Bjrkstrand, P Ljungman, H Svensson, J Hermans, A Alegre, J Apperley, J Blad, K Carlson, M Cavo, A Ferrant, A H Goldstone, A de Laurenzi, I Majolino, R Marcus, H G Prentice, K Remes, D Samson, A Sureda, L F Verdonck, L Volin, and G Gahrton. Allogeneic bone marrow transplantation versus autologous stem cell transplantation in multiple myeloma: a retrospective case-matched study from the European Group for Blood and Marrow Transplantation. Blood, 88(12):47114718, Dec 1996. [25] J Blad, D Samson, D Reece, J Apperley, B Bjrkstrand, G Gahrton, M Gertz, S Giralt, S Jagannath, and D Vesole. Criteria for evaluating disease response and progression in patients with multiple myeloma treated by high-dose therapy and haemopoietic stem cell transplantation. Myeloma Subcommittee of the EBMT. European Group for Blood and Marrow Transplant. Br J Haematol, 102(5):1115 1123, Sep 1998. [26] W K Bleeker, M E Munk, W J Mackus, J H van den Brakel, M Pluyter, M J Glennie, J G van de Winkel, and P W Parren. Estimation of dose requirements for sustained in vivo activity of a therapeutic human anti-CD20 antibody. Br J Haematol, 140(3):303312, Feb 2008. [27] M Boeckh and A P Geballe. Cytomegalovirus: pathogen, paradigm, and puzzle. J Clin Invest, 121(5):16731680, May 2011. [28] M Boeckh, W G Nichols, G Papanicolaou, R Rubin, J R Wingard, and J Zaia. Cytomegalovirus in hematopoietic stem cell transplant recipients: Current status, known challenges, and future strategies. Biol Blood Marrow Transplant, 9(9):543558, Sep 2003. [29] M Bonneville, R L OBrien, and W K Born. Gammadelta T cell effector functions: a blend of innate programming and acquired plasticity. Nat Rev Immunol, 10(7):467 478, Jul 2010.
131
[30] W K Born, L Zhang, M Nakayama, N Jin, J L Chain, Y Huang, M K Aydintug, and R L OBrien. Peptide antigens for gamma/delta T cells. Cell Mol Life Sci, 68(14):23352343, Jul 2011. [31] J R Brown, H T Kim, P Armand, C Cutler, D C Fisher, V Ho, J Koreth, J Ritz, C Wu, J H Antin, R J Soiffer, J G Gribben, and E P Alyea. Long-term follow-up of reduced-intensity allogeneic stem cell transplantation for chronic lymphocytic leukemia: prognostic model to predict outcome. Leukemia, 27(2):362369, Feb 2013. [32] R P Bucy, C L Chen, and M D Cooper. Tissue localization and CD8 accessory molecule expression of T gamma delta cells in humans. J Immunol, 142(9):3045 3049, May 1989. [33] J F Bukowski, C T Morita, H Band, and M B Brenner. Crucial role of TCR gamma chain junctional region in prenyl pyrophosphate antigen recognition by gamma delta T cells. J Immunol, 161(1):286293, Jul 1998. [34] J F Bukowski, C T Morita, Y Tanaka, B R Bloom, M B Brenner, and H Band. V gamma 2V delta 2 TCR-dependent recognition of non-peptide antigens and Daudi cells analyzed by TCR gene transfer. J Immunol, 154(3):9981006, Feb 1995. [35] M R Canninga-van Dijk, C J Sanders, L F Verdonck, R Fijnheer, and J G van den Tweel. Differential diagnosis of skin lesions after allogeneic haematopoietic stem cell transplantation. Histopathology, 42(4):313330, Apr 2003. [36] M R Canninga-van Dijk, H M van der Straaten, R Fijnheer, C J Sanders, J G van den Tweel, and L F Verdonck. Anti-CD20 monoclonal antibody treatment in 6 patients with therapy-refractory chronic graft-versus-host disease. Blood, 104(8):26032606, Oct 2004. [37] G Cartron, H Blasco, G Paintaud, H Watier, and C Le Guellec. Pharmacokinetics of rituximab and its clinical use: thought for the best use? Crit Rev Oncol Hematol, 62(1):4352, Apr 2007. [38] B Castella, C Vitale, M Coscia, and M Massaia. V9V2 T cell-based immunotherapy in hematological malignancies: from bench to bedside. Cell Mol Life Sci, 68(14):24192432, Jul 2011. [39] H Cheroutre. Starting at the beginning: new perspectives on the biology of mucosal T cells. Annu Rev Immunol, 22:217246, 2004. [40] H Cheroutre and F Lambolez. Doubting the TCR coreceptor function of CD8alphaalpha. Immunity, 28(2):149159, Feb 2008. [41] A Choudhary, F Davodeau, A Moreau, M A Peyrat, M Bonneville, and F Jotereau. Selective lysis of autologous tumor cells by recurrent gamma delta tumorinltrating lymphocytes from renal carcinoma. J Immunol, 154(8):39323940, Apr 1995.
132
BIBLIOGRAPHY
[42] P Constant, F Davodeau, M A Peyrat, Y Poquet, G Puzo, M Bonneville, and J J Fourni. Stimulation of human gamma delta T cells by nonpeptidic mycobacterial ligands. Science, 264(5156):267270, Apr 1994. [43] J J Cornelissen, W E Fibbe, A V Schattenberg, E J Petersen, R Willemze, T J de Witte, B Lwenberg, L F Verdonck, A vd Biezen, and R Brand. A retrospective Dutch study comparing T cell-depleted allogeneic blood stem cell transplantation vs T cell-depleted allogeneic bone marrow transplantation. Bone Marrow Transplant, 21 Suppl 3:6670, Jun 1998. [44] P Corradini, M Cavo, H Lokhorst, G Martinelli, C Terragna, I Majolino, P Valagussa, M Boccadoro, D Samson, A Bacigalupo, N Russell, V Montefusco, C Voena, G Gahrton, and Chronic Leukemia Working Party of the European Group for Blood and Marrow Transplantation (EBMT). Molecular remission after myeloablative allogeneic stem cell transplantation predicts a better relapse-free survival in patients with multiple myeloma. Blood, 102(5):19271929, Sep 2003. [45] D V Correia, M Fogli, K Hudspeth, M G da Silva, D Mavilio, and B Silva-Santos. Differentiation of human peripheral blood V1+ T cells expressing the natural cytotoxicity receptor NKp30 for recognition of lymphoid leukemia cells. Blood, 118(4):9921001, Jul 2011. [46] M Corvaisier, A Moreau-Aubry, E Diez, J Bennouna, J F Mosnier, E Scotet, M Bonneville, and F Jotereau. V gamma 9V delta 2 T cell response to colon carcinoma cells. J Immunol, 175(8):54815488, Oct 2005. [47] L Couzi, Y Levaillant, A Jamai, V Pitard, R Lassalle, K Martin, I Garrigue, O Hawchar, F Siberchicot, N Moore, J F Moreau, J Dechanet-Merville, and P Merville. Cytomegalovirus-induced gammadelta T cells associate with reduced cancer risk after kidney transplantation. J Am Soc Nephrol, 21(1):181188, Jan 2010. [48] L Couzi, V Pitard, X Sicard, I Garrigue, O Hawchar, P Merville, J F Moreau, and J Dchanet-Merville. Antibody-dependent anti-cytomegalovirus activity of human T cells expressing CD16 (FcRIIIa). Blood, 119(6):14181427, Feb 2012. [49] T Crough and R Khanna. Immunobiology of human cytomegalovirus: from bench to bedside. Clin Microbiol Rev, 22(1):7698, Jan 2009. [50] C Cutler, S Giri, S Jeyapalan, D Paniagua, A Viswanathan, and J H Antin. Acute and chronic graft-versus-host disease after allogeneic peripheral-blood stem-cell and bone marrow transplantation: a meta-analysis. J Clin Oncol, 19(16):36853691, Aug 2001. [51] C Cutler, D Miklos, H T Kim, N Treister, S B Woo, D Bienfang, L B Klickstein, J Levin, K Miller, C Reynolds, R Macdonell, M Pasek, S J Lee, V Ho, R Soiffer, J H Antin, J Ritz, and E Alyea. Rituximab for steroid-refractory chronic graft-versushost disease. Blood, 108(2):756762, Jul 2006. [52] A Dalloul. CD5: a safeguard against autoimmunity and a shield for cancer cells. Autoimmun Rev, 8(4):349353, Feb 2009.
133
[53] H Das, V Groh, C Kuijl, M Sugita, C T Morita, T Spies, and J F Bukowski. MICA engagement by human Vgamma2Vdelta2 T cells enhances their antigen-dependent effector function. Immunity, 15(1):8393, Jul 2001. [54] W de Jager, K Bourcier, G T Rijkers, B J Prakken, and V Seyfert-Margolis. Prerequisites for cytokine measurements in clinical trials with multiplex immunoassays. BMC Immunol, 10:5252, 2009. [55] J Dchanet, P Merville, F Berg, G Bone-Mane, J L Taupin, P Michel, P Joly, M Bonneville, L Potaux, and J F Moreau. Major expansion of gammadelta T lymphocytes following cytomegalovirus infection in kidney allograft recipients. J Infect Dis, 179(1):18, Jan 1999. [56] J Dchanet, P Merville, A Lim, C Retire, V Pitard, X Lafarge, S Michelson, C Mric, M M Hallet, P Kourilsky, L Potaux, M Bonneville, and J F Moreau. Implication of gammadelta T cells in the human immune response to cytomegalovirus. J Clin Invest, 103(10):14371449, May 1999. [57] J M den Haan, N E Sherman, E Blokland, E Huczko, F Koning, J W Drijfhout, J Skipper, J Shabanowitz, D F Hunt, and V H Engelhard. Identication of a graft versus host disease-associated human minor histocompatibility antigen. Science, 268(5216):14761480, Jun 1995. [58] K Deusch, F Lling, K Reich, M Classen, H Wagner, and K Pfeffer. A major fraction of human intraepithelial lymphocytes simultaneously expresses the gamma/delta T cell receptor, the CD8 accessory molecule and preferentially uses the V delta 1 gene segment. Eur J Immunol, 21(4):10531059, Apr 1991. [59] R Diaconescu, C R Flowers, B Storer, M L Sorror, M B Maris, D G Maloney, B M Sandmaier, and R Storb. Morbidity and mortality with nonmyeloablative compared with myeloablative conditioning before hematopoietic cell transplantation from HLA-matched related donors. Blood, 104(5):15501558, Sep 2004. [60] A Dominietto, E Tedone, M Soracco, B Bruno, A M Raiola, M T Van Lint, S Geroldi, T Lamparelli, B Galano, F Gualandi, F Frassoni, and A Bacigalupo. In vivo B-cell depletion with rituximab for alternative donor hemopoietic SCT. Bone Marrow Transplant, 47(1):101106, Jan 2012. [61] T Drner, A M Jacobi, and P E Lipsky. B cells in autoimmunity. Arthritis Res Ther, 11(5):247247, 2009. [62] M E Dudley, J R Wunderlich, T E Shelton, J Even, and S A Rosenberg. Generation of tumor-inltrating lymphocyte cultures for use in adoptive transfer therapy for melanoma patients. J Immunother, 26(4):332342, Jul-Aug 2003. [63] H Einsele, G Ehninger, H Hebart, P Weber, S Dette, H Link, H P Horny, V Meuter, S Wagner, and H D Waller. Incidence of local CMV infection and acute intestinal GVHD in marrow transplant recipients with severe diarrhoea. Bone Marrow Transplant, 14(6):955963, Dec 1994.
134
BIBLIOGRAPHY
[64] A H Elmaagacli, N K Steckel, M Koldehoff, Y Hegerfeldt, R Trenschel, M Ditschkowski, S Christoph, T Gromke, L Kordelas, H D Ottinger, R S Ross, P A Horn, S Schnittger, and D W Beelen. Early human cytomegalovirus replication after transplantation is associated with a decreased relapse risk: evidence for a putative virus-versus-leukemia effect in acute myeloid leukemia patients. Blood, 118(5):14021412, Aug 2011. [65] P S Evans, P J Enders, C Yin, T J Ruckwardt, M Malkovsky, and C D Pauza. In vitro stimulation with a non-peptidic alkylphosphate expands cells expressing Vgamma2-Jgamma1.2/Vdelta2 T-cell receptors. Immunology, 104(1):1927, Sep 2001. [66] F Fagioli, A Bacigalupo, F Frassoni, M T Van Lint, D Occhini, F Gualandi, T Lamparelli, M Clavio, V Vitale, and G Sogno. Allogeneic bone marrow transplantation for acute myeloid leukemia in rst complete remission: the effect of FAB classication and GVHD prophylaxis. Bone Marrow Transplant, 13(3):247252, Mar 1994. [67] J Ferrara, P Reddy, and S Paczesny. Immunotherapy through T-cell receptor gene transfer induces severe graft-versus-host disease. Immunotherapy, 2(6):791794, Nov 2010. [68] M Ferrarini, S Heltai, S M Pupa, S Mernard, and R Zocchi. Killing of laminin receptor-positive human lung cancers by tumor inltrating lymphocytes bearing gammadelta(+) t-cell receptors. J Natl Cancer Inst, 88(7):436441, Apr 1996. [69] T Feuchtinger, K Opherk, W A Bethge, M S Topp, F R Schuster, E M Weissinger, M Mohty, R Or, M Maschan, M Schumm, K Hamprecht, R Handgretinger, P Lang, and H Einsele. Adoptive transfer of pp65-specic T cells for the treatment of chemorefractory cytomegalovirus disease or reactivation after haploidentical and matched unrelated stem cell transplantation. Blood, 116(20):43604367, Nov 2010. [70] A H Filipovich. Diagnosis and manifestations of chronic graft-versus-host disease. Best Pract Res Clin Haematol, 21(2):251257, Jun 2008. [71] A H Filipovich, D Weisdorf, S Pavletic, G Socie, J R Wingard, S J Lee, P Martin, J Chien, D Przepiorka, D Couriel, E W Cowen, P Dinndorf, A Farrell, R Hartzman, J Henslee-Downey, D Jacobsohn, G McDonald, B Mittleman, J D Rizzo, M Robinson, M Schubert, K Schultz, H Shulman, M Turner, G Vogelsang, and M E Flowers. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant, 11(12):945956, Dec 2005. [72] P Fisch, M Malkovsky, E Braakman, E Sturm, R L Bolhuis, A Prieve, J A Sosman, V A Lam, and P M Sondel. Gamma/delta T cell clones and natural killer cell clones mediate distinct patterns of non-major histocompatibility complex-restricted cytolysis. J Exp Med, 171(5):15671579, May 1990. [73] H Fujii, G Cuvelier, K She, S Aslanian, H Shimizu, A Kariminia, M Krailo, Z Chen, R McMaster, A Bergman, F Goldman, S A Grupp, D A Wall, A L Gilman, and K R
135
Schultz. Biomarkers in newly diagnosed pediatric-extensive chronic graft-versushost disease: a report from the Childrens Oncology Group. Blood, 111(6):3276 3285, Mar 2008. [74] F Garban, M Attal, M Michallet, C Hulin, J H Bourhis, I Yakoub-Agha, T Lamy, G Marit, F Maloisel, C Berthou, M Dib, D Caillot, B Deprijck, N Ketterer, J L Harousseau, J J Sotto, and P Moreau. Prospective comparison of autologous stem cell transplantation followed by dose-reduced allograft (IFM99-03 trial) with tandem autologous stem cell transplantation (IFM99-04 trial) in high-risk de novo multiple myeloma. Blood, 107(9):34743480, May 2006. [75] M A Geller, S Cooley, P L Judson, R Ghebre, L F Carson, P A Argenta, A L Jonson, A Panoskaltsis-Mortari, J Curtsinger, D McKenna, K Dusenbery, R Bliss, L S Downs, and J S Miller. A phase II study of allogeneic natural killer cell therapy to treat patients with recurrent ovarian and breast cancer. Cytotherapy, 13(1):98 107, Jan 2011. [76] S Gerull, M Goerner, A Benner, U Hegenbart, U Klein, H Schaefer, H Goldschmidt, and A D Ho. Long-term outcome of nonmyeloablative allogeneic transplantation in patients with high-risk multiple myeloma. Bone Marrow Transplant, 36(11):963 969, Dec 2005. [77] H Glucksberg, R Storb, A Fefer, C D Buckner, P E Neiman, R A Clift, K G Lerner, and E D Thomas. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HL-A-matched sibling donors. Transplantation, 18(4):295 304, Oct 1974. [78] H J Gober, M Kistowska, L Angman, P Jen, L Mori, and G De Libero. Human T cell receptor gammadelta cells recognize endogenous mevalonate metabolites in tumor cells. J Exp Med, 197(2):163168, Jan 2003. [79] K T Godder, P J Henslee-Downey, J Mehta, B S Park, K Y Chiang, S Abhyankar, and L S Lamb. Long term disease-free survival in acute leukemia patients recovering with increased gammadelta T cells after partially mismatched related donor bone marrow transplantation. Bone Marrow Transplant, 39(12):751757, Jun 2007. [80] H Goker, I C Haznedaroglu, and N J Chao. Acute graft-vs-host disease: pathobiology and management. Exp Hematol, 29(3):259277, Mar 2001. [81] J M Goldman, R P Gale, M M Horowitz, J C Biggs, R E Champlin, E Gluckman, R G Hoffmann, S J Jacobsen, A M Marmont, and P B McGlave. Bone marrow transplantation for chronic myelogenous leukemia in chronic phase. Increased risk for relapse associated with T-cell depletion. Ann Intern Med, 108(6):806814, Jun 1988. [82] A Q Gomes, D S Martins, and B Silva-Santos. Targeting T lymphocytes for cancer immunotherapy: from novel mechanistic insight to clinical application. Cancer Res, 70(24):1002410027, Dec 2010.
136
BIBLIOGRAPHY
[83] H T Greinix, D Pohlreich, M Kouba, U Krmczi, I Lohmann, K Feldmann, C Zielinski, and W F Pickl. Elevated numbers of immature/transitional CD21- B lymphocytes and deciency of memory CD27+ B cells identify patients with active chronic graft-versus-host disease. Biol Blood Marrow Transplant, 14(2):208219, Feb 2008. [84] L M Grifth, S Z Pavletic, S J Lee, P J Martin, K R Schultz, and G B Vogelsang. Chronic Graft-versus-Host Diseaseimplementation of the National Institutes of Health Consensus Criteria for Clinical Trials. Biol Blood Marrow Transplant, 14(4):379384, Apr 2008. [85] V Groh, R Rhinehart, J Randolph-Habecker, M S Topp, S R Riddell, and T Spies. Costimulation of CD8alphabeta T cells by NKG2D via engagement by MIC induced on virus-infected cells. Nat Immunol, 2(3):255260, Mar 2001. [86] V Groh, R Rhinehart, H Secrist, S Bauer, K H Grabstein, and T Spies. Broad tumorassociated expression and recognition by tumor-derived gamma delta T cells of MICA and MICB. Proc Natl Acad Sci U S A, 96(12):68796884, Jun 1999. [87] V Groh, A Steinle, S Bauer, and T Spies. Recognition of stress-induced MHC molecules by intestinal epithelial gammadelta T cells. Science, 279(5357):17371740, Mar 1998. [88] C Grnder, S van Dorp, S Hol, E Drent, T Straetemans, S Heijhuurs, K Scholten, W Scheper, Z Sebestyen, A Martens, R Strong, and J Kuball. 9 and 2CDR3 domains regulate functional avidity of T cells harboring 92TCRs. Blood, 120(26):51535162, Dec 2012. [89] W Haas, P Pereira, and S Tonegawa. Gamma/delta cells. Annu Rev Immunol, 11:637685, 1993. [90] F Halary, V Pitard, D Dlubek, R Krzysiek, H de la Salle, P Merville, C Dromer, D Emilie, J F Moreau, and J Dchanet-Merville. Shared reactivity of Vdelta2(neg) gammadelta T cells against cytomegalovirus-infected cells and tumor intestinal epithelial cells. J Exp Med, 201(10):15671578, May 2005. [91] G Hale and H Waldmann. Control of graft-versus-host disease and graft rejection by T cell depletion of donor and recipient with Campath-1 antibodies. Results of matched sibling transplants for malignant diseases. Bone Marrow Transplant, 13(5):597611, May 1994. [92] P Husermann, R B Walter, J Halter, B C Biedermann, A Tichelli, P Itin, and A Gratwohl. Cutaneous graft-versus-host disease: a guide for the dermatologist. Dermatology, 216(4):287304, 2008. [93] A H Hautmann, D Wolff, J Hahn, M Edinger, N Schirmer, J Ammer, B Holler, K Landfried, M G Hautmann, N Ahrens, P Ugocsai, R Andreesen, and E Holler. Extracorporeal photopheresis in 62 patients with acute and chronic GVHD: results of treatment with the COBE Spectra System. Bone Marrow Transplant, 48(3):439445, Mar 2013.
137
[94] M Hinchcliff, C C Huang, W Ishida, F Fang, J Lee, N Jafari, M Wilkes, S Bhattacharyya, E Leof, and J Varga. Imatinib mesylate causes genome-wide transcriptional changes in systemic sclerosis broblasts in vitro. Clin Exp Rheumatol, 30(2 Suppl 71):8696, Mar-Apr 2012. [95] T Hinz, D Wesch, F Halary, S Marx, A Choudhary, B Arden, O Janssen, M Bonneville, and D Kabelitz. Identication of the complete expressed human TCR V gamma repertoire by ow cytometry. Int Immunol, 9(8):10651072, Aug 1997. [96] M Ho, P Tongtawe, J Kriangkum, T Wimonwattrawatee, K Pattanapanyasat, L Bryant, J Shaq, P Suntharsamai, S Looareesuwan, H K Webster, and J F Elliott. Polyclonal expansion of peripheral gamma delta T cells in human Plasmodium falciparum malaria. Infect Immun, 62(3):855862, Mar 1994. [97] S N Ho, H D Hunt, R M Horton, J K Pullen, and L R Pease. Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene, 77(1):5159, Apr 1989. [98] W Y Ho, H N Nguyen, M Wol, J Kuball, and P D Greenberg. In vitro methods for generating CD8+ T-cell clones for immunotherapy from the nave repertoire. J Immunol Methods, 310(1-2):4052, Mar 2006. [99] W Hobo, K Broen, W J van der Velden, A Greupink-Draaisma, N Adisty, Y Wouters, M Kester, H Fredrix, J H Jansen, B van der Reijden, J H Falkenburg, T de Witte, F Preijers, T Schattenberg, T Feuth, N M Blijlevens, N Schaap, and H Dolstra. Association of disparities in known minor histocompatibility antigens with relapse-free survival and graft-versus-host disease after allogeneic stem cell transplantation. Biol Blood Marrow Transplant, 19(2):274282, Feb 2013. [100] M C Hogenes, S van Dorp, J van Kuik, F R Monteiro, N ter Hoeve, M R van Dijk, A C Martens, and R A de Weger. Histological assessment of the sclerotic graftversus-host response in the humanized RAG2-/-c-/- mouse model. Biol Blood Marrow Transplant, 18(7):10231035, Jul 2012. [101] P D Holler, P O Holman, E V Shusta, S OHerrin, K D Wittrup, and D M Kranz. In vitro evolution of a T cell receptor with high afnity for peptide/MHC. Proc Natl Acad Sci U S A, 97(10):53875392, May 2000. [102] K Hudspeth, M Fogli, D V Correia, J Mikulak, A Roberto, S Della Bella, B SilvaSantos, and D Mavilio. Engagement of NKp30 on Vt1 T cells induces the production of CCL3, CCL4, and CCL5 and suppresses HIV-1 replication. Blood, 119(17):40134016, Apr 2012. [103] C Huisman, R A de Weger, L de Vries, M G Tilanus, and L F Verdonck. Chimerism analysis within 6 months of allogeneic stem cell transplantation predicts relapse in acute myeloid leukemia. Bone Marrow Transplant, 39(5):285291, Mar 2007. [104] H Ihn, K Yamane, M Kubo, and K Tamaki. Blockade of endogenous transforming growth factor beta signaling prevents up-regulated collagen synthesis in scleroderma broblasts: association with increased expression of transforming growth factor beta receptors. Arthritis Rheum, 44(2):474480, Feb 2001.
138
BIBLIOGRAPHY
[105] C A Jacobson and J Ritz. B-cell-directed therapy for chronic graft-versus-host disease. Haematologica, 95(11):18111813, Nov 2010. [106] B Jena, G Dotti, and L J Cooper. Redirecting T-cell specicity by introducing a tumor-specic chimeric antigen receptor. Blood, 116(7):10351044, Aug 2010. [107] L A Johnson, R A Morgan, M E Dudley, L Cassard, J C Yang, M S Hughes, U S Kammula, R E Royal, R M Sherry, J R Wunderlich, C C Lee, N P Restifo, S L Schwarz, A P Cogdill, R J Bishop, H Kim, C C Brewer, S F Rudy, C VanWaes, J L Davis, A Mathur, R T Ripley, D A Nathan, C M Laurencot, and S A Rosenberg. Gene therapy with human and mouse T-cell receptors mediates cancer regression and targets normal tissues expressing cognate antigen. Blood, 114(3):535546, Jul 2009. [108] H Jonuleit, U Khn, G Mller, K Steinbrink, L Paragnik, E Schmitt, J Knop, and A H Enk. Pro-inammatory cytokines and prostaglandins induce maturation of potent immunostimulatory dendritic cells under fetal calf serum-free conditions. Eur J Immunol, 27(12):31353142, Dec 1997. [109] D Kabelitz. Small molecules for the activation of human gammadelta T cell responses against infection. Recent Pat Antiinfect Drug Discov, 3(1):19, Jan 2008. [110] D Kabelitz, A Glatzel, and D Wesch. Antigen recognition by human gammadelta T lymphocytes. Int Arch Allergy Immunol, 122(1):17, May 2000. [111] D Kabelitz, D Wesch, and W He. Perspectives of gammadelta T cells in tumor immunology. Cancer Res, 67(1):58, Jan 2007. [112] P Kebriaei, L Isola, E Bahceci, K Holland, S Rowley, J McGuirk, M Devetten, J Jansen, R Herzig, M Schuster, R Monroy, and J Uberti. Adult human mesenchymal stem cells added to corticosteroid therapy for the treatment of acute graftversus-host disease. Biol Blood Marrow Transplant, 15(7):804811, Jul 2009. [113] P Kebriaei, R M Saliba, C Ma, C Ippoliti, D R Couriel, M de Lima, S Giralt, M H Qazilbash, J L Gajewski, C S Ha, R E Champlin, and I F Khouri. Allogeneic hematopoietic stem cell transplantation after rituximab-containing myeloablative preparative regimen for acute lymphoblastic leukemia. Bone Marrow Transplant, 38(3):203209, Aug 2006. [114] H J Kim, W S Min, B S Cho, K S Eom, Y J Kim, C K Min, S Lee, S G Cho, J Y Jin, J W Lee, and C C Kim. Successful prevention of acute graft-versus-host disease using low-dose antithymocyte globulin after mismatched, unrelated, hematopoietic stem cell transplantation for acute myelogenous leukemia. Biol Blood Marrow Transplant, 15(6):704717, Jun 2009. [115] S J Kim, J W Lee, C W Jung, C K Min, B Cho, H J Shin, J S Chung, H Kim, W S Lee, Y D Joo, D H Yang, H Kook, H J Kang, H S Ahn, S S Yoon, S K Sohn, Y H Min, W S Min, H S Park, and J H Won. Weekly rituximab followed by monthly rituximab treatment for steroid-refractory chronic graft-versus-host disease: results from a prospective, multicenter, phase II study. Haematologica, 95(11):19351942, Nov 2010.
139
[116] S J Kim and J H Won. B cell homeostasis and the development of chronic graftversus-host disease: implications for B cell-depleting therapy. Leuk Lymphoma, 53(1):1925, Jan 2012. [117] F M Kloosterboer, S A van Luxemburg-Heijs, R A van Soest, H M van Egmond, A M Barbui, M P Strijbosch, R Willemze, and J H Falkenburg. Minor histocompatibility antigen-specic T cells with multiple distinct specicities can be isolated by direct cloning of IFNgamma-secreting T cells from patients with relapsed leukemia responding to donor lymphocyte infusion. Leukemia, 19(1):8390, Jan 2005. [118] A Knight, A J Madrigal, S Grace, J Sivakumaran, P Kottaridis, S Mackinnon, P J Travers, and M W Lowdell. The role of V2-negative T cells during cytomegalovirus reactivation in recipients of allogeneic stem cell transplantation. Blood, 116(12):21642172, Sep 2010. [119] Y Kong, W Cao, X Xi, C Ma, L Cui, and W He. The NKG2D ligand ULBP4 binds to TCRgamma9/delta2 and induces cytotoxicity to tumor cells through both TCRgammadelta and NKG2D. Blood, 114(2):310317, Jul 2009. [120] P Krishnamurthy, V T Potter, L D Barber, A G Kulasekararaj, Z Y Lim, R M Pearce, H de Lavallade, M Kenyon, R M Ireland, J C Marsh, S Devereux, A Pagliuca, and G J Mufti. Outcome of Donor Lymphocyte Infusion after T Cell-depleted Allogeneic Hematopoietic Stem Cell Transplantation for Acute Myelogenous Leukemia and Myelodysplastic Syndromes. Biol Blood Marrow Transplant, 19(4):562568, Apr 2013. [121] N Krger, A Shimoni, M Zagrivnaja, F Ayuk, M Lioznov, H Schieder, H Renges, B Fehse, T Zabelina, A Nagler, and A R Zander. Low-dose thalidomide and donor lymphocyte infusion as adoptive immunotherapy after allogeneic stem cell transplantation in patients with multiple myeloma. Blood, 104(10):33613363, Nov 2004. [122] N Krger, T Zabelina, W Krger, H Renges, N Stute, J Rischewski, S Sonnenberg, F Ayuk, F Tgel, U Schade, H Fiegel, R Erttmann, C Lliger, and A R Zander. In vivo T cell depletion with pretransplant anti-thymocyte globulin reduces graftversus-host disease without increasing relapse in good risk myeloid leukemia patients after stem cell transplantation from matched related donors. Bone Marrow Transplant, 29(8):683689, Apr 2002. [123] J Kuball, M L Dossett, M Wol, W Y Ho, R H Voss, C Fowler, and P D Greenberg. Facilitating matched pairing and expression of TCR chains introduced into human T cells. Blood, 109(6):23312338, Mar 2007. [124] J Kuball, F W Schmitz, R H Voss, E A Ferreira, R Engel, P Guillaume, S Strand, P Romero, C Huber, L A Sherman, and M Theobald. Cooperation of human tumor-reactive CD4+ and CD8+ T cells after redirection of their specicity by a high-afnity p53A2.1-specic TCR. Immunity, 22(1):117129, Jan 2005. [125] J Kuball, M Theobald, E A Ferreira, G Hess, J Burg, G Maccagno, A P Barreiros, S Lth, C C Schimanski, M Schuchmann, A Schwarting, M Neurath, G Otto, P R Galle, and A W Lohse. Control of organ transplant-associated graft-versus-host
140
BIBLIOGRAPHY
disease by activated host lymphocyte infusions. Transplantation, 78(12):17741779, Dec 2004. [126] J Kuball, S F Wen, J Leissner, D Atkins, P Meinhardt, E Quijano, H Engler, B Hutchins, D C Maneval, M J Grace, M A Fritz, S Strkel, J W Throff, C Huber, and M Schuler. Successful adenovirus-mediated wild-type p53 gene transfer in patients with bladder cancer by intravesical vector instillation. J Clin Oncol, 20(4):957 965, Feb 2002. [127] M Kubo, H Ihn, K Yamane, and K Tamaki. Upregulated expression of transforming growth factor-beta receptors in dermal broblasts of skin sections from patients with systemic sclerosis. J Rheumatol, 29(12):25582564, Dec 2002. [128] X Lafarge, P Merville, M C Cazin, F Berg, L Potaux, J F Moreau, and J DchanetMerville. Cytomegalovirus infection in transplant recipients resolves when circulating gammadelta T lymphocytes expand, suggesting a protective antiviral role. J Infect Dis, 184(5):533541, Sep 2001. [129] K Le Blanc, I Rasmusson, B Sundberg, C Gtherstrm, M Hassan, M Uzunel, and O Ringdn. Treatment of severe acute graft-versus-host disease with third party haploidentical mesenchymal stem cells. Lancet, 363(9419):14391441, May 2004. [130] S J Lee. New approaches for preventing and treating chronic graft-versus-host disease. Blood, 105(11):42004206, Jun 2005. [131] S J Lee, J P Klein, A J Barrett, O Ringden, J H Antin, J Y Cahn, M H Carabasi, R P Gale, S Giralt, G A Hale, O Ilhan, P L McCarthy, G Socie, L F Verdonck, D J Weisdorf, and M M Horowitz. Severity of chronic graft-versus-host disease: association with treatment-related mortality and relapse. Blood, 100(2):406414, Jul 2002. [132] S J Lee, G Vogelsang, and M E Flowers. Chronic graft-versus-host disease. Biol Blood Marrow Transplant, 9(4):215233, Apr 2003. [133] A M Leen, G D Myers, U Sili, M H Huls, H Weiss, K S Leung, G Carrum, R A Krance, C C Chang, J J Molldrem, A P Gee, M K Brenner, H E Heslop, C M Rooney, and C M Bollard. Monoculture-derived T lymphocytes specic for multiple viruses expand and produce clinically relevant effects in immunocompromised individuals. Nat Med, 12(10):11601166, Oct 2006. [134] M. P. Lefranc. IMGT, the international ImMunoGeneTics database. Nucleic Acids Res., 31(1):307310, Jan 2003. [135] A J Leishman, O V Naidenko, A Attinger, F Koning, C J Lena, Y Xiong, H C Chang, E Reinherz, M Kronenberg, and H Cheroutre. T cell responses modulated through interaction between CD8alphaalpha and the nonclassical MHC class I molecule, TL. Science, 294(5548):19361939, Nov 2001. [136] D S Leslie, M S Vincent, F M Spada, H Das, M Sugita, C T Morita, and M B Brenner. CD1-mediated gamma/delta T cell maturation of dendritic cells. J Exp Med, 196(12):15751584, Dec 2002.
141
[137] H M Lokhorst. No RIC in high-risk myeloma? Blood, (107):34203421, 2006. [138] H M Lokhorst, A Schattenberg, J J Cornelissen, L L Thomas, and L F Verdonck. Donor leukocyte infusions are effective in relapsed multiple myeloma after allogeneic bone marrow transplantation. Blood, 90(10):42064211, Nov 1997. [139] H M Lokhorst, C M Segeren, L F Verdonck, B van der Holt, R Raymakers, M H van Oers, R M Barge, H C Schouten, P H Westveer, M M Steijaert, J J Cornelissen, P Sonneveld, and Dutch-Belgian Hemato-Oncology Cooperative Group. Partially T-cell-depleted allogeneic stem-cell transplantation for rst-line treatment of multiple myeloma: a prospective evaluation of patients treated in the phase III study HOVON 24 MM. J Clin Oncol, 21(9):17281733, May 2003. [140] H M Lokhorst, K Wu, L F Verdonck, L L Laterveer, N W van de Donk, M H van Oers, J J Cornelissen, and A V Schattenberg. The occurrence of graft-versus-host disease is the major predictive factor for response to donor lymphocyte infusions in multiple myeloma. Blood, 103(11):43624364, Jun 2004. [141] E O Long. Tumor cell recognition by natural killer cells. Semin Cancer Biol, 12(1):57 61, Feb 2002. [142] M W Lowdell, R Craston, D Samuel, M E Wood, E ONeill, V Saha, and H G Prentice. Evidence that continued remission in patients treated for acute leukaemia is dependent upon autologous natural killer cells. Br J Haematol, 117(4):821827, Jun 2002. [143] M W Lowdell, N Ray, R Craston, T Corbett, M Deane, and H G Prentice. The in vitro detection of anti-leukaemia-specic cytotoxicity after autologous bone marrow transplantation for acute leukaemia. Bone Marrow Transplant, 19(9):891897, May 1997. [144] M J Maeurer, D Martin, W Walter, K Liu, L Zitvogel, K Halusczcak, H Rabinowich, R Duquesnoy, W Storkus, and M T Lotze. Human intestinal Vdelta1+ lymphocytes recognize tumor cells of epithelial origin. J Exp Med, 183(4):16811696, Apr 1996. [145] L Magro, B Catteau, V Coiteux, B Bruno, J P Jouet, and I Yakoub-Agha. Efcacy of imatinib mesylate in the treatment of refractory sclerodermatous chronic GVHD. Bone Marrow Transplant, 42(11):757760, Dec 2008. [146] L Magro, M Mohty, B Catteau, V Coiteux, P Chevallier, L Terriou, J P Jouet, and I Yakoub-Agha. Imatinib mesylate as salvage therapy for refractory sclerotic chronic graft-versus-host disease. Blood, 114(3):719722, Jul 2009. [147] D G Maloney, A J Molina, F Sahebi, K E Stockerl-Goldstein, B M Sandmaier, W Bensinger, B Storer, U Hegenbart, G Somlo, T Chauncey, B Bruno, F R Appelbaum, K G Blume, S J Forman, P McSweeney, and R Storb. Allografting with nonmyeloablative conditioning following cytoreductive autografts for the treatment of patients with multiple myeloma. Blood, 102(9):34473454, Nov 2003.
142
BIBLIOGRAPHY
[148] D Maraninchi, E Gluckman, D Blaise, D Guyotat, B Rio, J L Pico, V Leblond, M Michallet, F Dreyfus, and N Ifrah. Impact of T-cell depletion on outcome of allogeneic bone-marrow transplantation for standard-risk leukaemias. Lancet, 2(8552):175178, Jul 1987. [149] V Marcu-Malina, S Heijhuurs, M van Buuren, L Hartkamp, S Strand, Z Sebestyen, K Scholten, A Martens, and J Kuball. Redirecting T cells against cancer cells by transfer of a broadly tumor-reactive T-cell receptor. Blood, 118(1):5059, Jul 2011. [150] V Marcu-Malina, S van Dorp, and J Kuball. Re-targeting T-cells against cancer by gene-transfer of tumor-reactive receptors. Expert Opin Biol Ther, 9(5):579591, May 2009. [151] K J Martires, K Baird, S M Steinberg, L Grkovic, G O Joe, K M Williams, S A Mitchell, M Datiles, F T Hakim, S Z Pavletic, and E W Cowen. Sclerotic-type chronic GVHD of the skin: clinical risk factors, laboratory markers, and burden of disease. Blood, 118(15):42504257, Oct 2011. [152] S R Mattarollo, T Kenna, M Nieda, and A J Nicol. Chemotherapy and zoledronate sensitize solid tumour cells to Vgamma9Vdelta2 T cell cytotoxicity. Cancer Immunol Immunother, 56(8):12851297, Aug 2007. [153] C C Matte, J Liu, J Cormier, B E Anderson, I Athanasiadis, D Jain, J McNiff, and W D Shlomchik. Donor APCs are required for maximal GVHD but not for GVL. Nat Med, 10(9):987992, Sep 2004. [154] L L McCormick, Y Zhang, E Tootell, and A C Gilliam. Anti-TGF-beta treatment prevents skin and lung brosis in murine sclerodermatous graft-versus-host disease: a model for human scleroderma. J Immunol, 163(10):56935699, Nov 1999. [155] Z McIver, N Stephens, A Grim, and A J Barrett. Rituximab administration within 6 months of T cell-depleted allogeneic SCT is associated with prolonged lifethreatening cytopenias. Biol Blood Marrow Transplant, 16(11):15491556, Nov 2010. [156] F A Mendoza and S A Jimnez. Tyrosine kinase inhibitor therapy for systemic sclerosis: Quo Vadis? Arthritis Rheum, 63(11):31993203, Nov 2011. [157] F Miyagawa, Y Tanaka, S Yamashita, B Mikami, K Danno, M Uehara, and N Minato. Essential contribution of germline-encoded lysine residues in Jgamma1.2 segment to the recognition of nonpeptide antigens by human gammadelta T cells. J Immunol, 167(12):67736779, Dec 2001. [158] K Miyazaki. MEGAWHOP cloning: a method of creating random mutagenesis libraries via megaprimer PCR of whole plasmids. Methods Enzymol, 498:399406, 2011. [159] M Mohty, O Avinens, C Faucher, P Viens, D Blaise, and J F Eliaou. Predictive factors and impact of full donor T-cell chimerism after reduced intensity conditioning allogeneic stem cell transplantation. Haematologica, 92(7):10041006, Jul 2007.
143
[160] J Molldrem, S Dermime, K Parker, Y Z Jiang, D Mavroudis, N Hensel, P Fukushima, and A J Barrett. Targeted T-cell therapy for human leukemia: cytotoxic T lymphocytes specic for a peptide derived from proteinase 3 preferentially lyse human myeloid leukemia cells. Blood, 88(7):24502457, Oct 1996. [161] A Moretta, E Ciccone, G Pantaleo, G Tambussi, C Bottino, G Melioli, M C Mingari, and L Moretta. Surface molecules involved in the activation and regulation of T or natural killer lymphocytes in humans. Immunol Rev, 111:145175, Oct 1989. [162] R A Morgan, M E Dudley, J R Wunderlich, M S Hughes, J C Yang, R M Sherry, R E Royal, S L Topalian, U S Kammula, N P Restifo, Z Zheng, A Nahvi, C R de Vries, L J Rogers-Freezer, S A Mavroukakis, and S A Rosenberg. Cancer regression in patients after transfer of genetically engineered lymphocytes. Science, 314(5796):126129, Oct 2006. [163] C T Morita, H K Lee, H Wang, H Li, R A Mariuzza, and Y Tanaka. Structural features of nonpeptide prenyl pyrophosphates that determine their antigenicity for human gamma delta T cells. J Immunol, 167(1):3641, Jul 2001. [164] S Nedellec, M Bonneville, and E Scotet. Human Vgamma9Vdelta2 T cells: from signals to functions. Semin Immunol, 22(4):199206, Aug 2010. [165] V H Nguyen, S Shashidhar, D S Chang, L Ho, N Kambham, M Bachmann, J M Brown, and R S Negrin. The impact of regulatory T cells on T-cell immunity following hematopoietic cell transplantation. Blood, 111(2):945953, Jan 2008. [166] W G Nichols, L Corey, T Gooley, C Davis, and M Boeckh. High risk of death due to bacterial and fungal infection among cytomegalovirus (CMV)-seronegative recipients of stem cell transplants from seropositive donors: evidence for indirect effects of primary CMV infection. J Infect Dis, 185(3):273282, Feb 2002. [167] A J Nicol, H Tokuyama, S R Mattarollo, T Hagi, K Suzuki, K Yokokawa, and M Nieda. Clinical evaluation of autologous gamma delta T cell-based immunotherapy for metastatic solid tumours. Br J Cancer, 105(6):778786, Sep 2011. [168] A Olivieri, F Locatelli, M Zecca, A Sanna, M Cimminiello, R Raimondi, G Gini, N Mordini, A Balduzzi, P Leoni, A Gabrielli, and A Bacigalupo. Imatinib for refractory chronic graft-versus-host disease with brotic features. Blood, 114(3):709 718, Jul 2009. [169] J Olivieri, S Coluzzi, I Attolico, and A Olivieri. Tirosin kinase inhibitors in chronic graft versus host disease: from bench to bedside. ScienticWorldJournal, 11:1908 1931, 2011. [170] S Ozawa, C Nakaseko, M Nishimura, A Maruta, R Cho, C Ohwada, H Sakamaki, H Sao, S Mori, S Okamoto, K Miyamura, S Kato, T Kawase, Y Morishima, Y Kodera, and Japan Marrow Donor Program. Chronic graft-versus-host disease after allogeneic bone marrow transplantation from an unrelated donor: incidence, risk factors and association with relapse. A report from the Japan Marrow Donor Program. Br J Haematol, 137(2):142151, Apr 2007.
144
BIBLIOGRAPHY
[171] S Park, M Labopin, I Yakoub-Agha, J Delaunay, N Dhedin, E Deconinck, M Michallet, M Robin, T De Revel, M Bernard, N Vey, B Lioure, S Lapusan, R Tabrizi, J H Bourhis, A Huynh, Y Beguin, G Soci, F Dreyfus, P Fenaux, and M Mohty. Allogeneic stem cell transplantation for chronic myelomonocytic leukemia: a report from the Societe Francaise de Greffe de Moelle et de Therapie Cellulaire. Eur J Haematol, Jan 2013. [172] F Patriarca, C Skert, A Sperotto, F Zaja, E Falleti, R Mestroni, F Kikic, E Calistri, C Fil, A Geromin, M Cerno, and R Fanin. The development of autoantibodies after allogeneic stem cell transplantation is related with chronic graft-vs-host disease and immune recovery. Exp Hematol, 34(3):389396, Mar 2006. [173] S Z Pavletic, P Martin, S J Lee, S Mitchell, D Jacobsohn, E W Cowen, M L Turner, G Akpek, A Gilman, G McDonald, M Schubert, A Berger, P Bross, J W Chien, D Couriel, J P Dunn, J Fall-Dickson, A Farrell, M E Flowers, H Greinix, S Hirschfeld, L Gerber, S Kim, R Knobler, P A Lachenbruch, F W Miller, B Mittleman, E Papadopoulos, S K Parsons, D Przepiorka, M Robinson, M Ward, B Reeve, L G Rider, H Shulman, K R Schultz, D Weisdorf, G B Vogelsang, and Response Criteria Working Group. Measuring therapeutic response in chronic graft-versushost disease: National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graft-versus-Host Disease: IV. Response Criteria Working Group report. Biol Blood Marrow Transplant, 12(3):252266, Mar 2006. [174] D Pende, G M Spaggiari, S Marcenaro, S Martini, P Rivera, A Capobianco, M Falco, E Lanino, I Pierri, R Zambello, A Bacigalupo, M C Mingari, A Moretta, and L Moretta. Analysis of the receptor-ligand interactions in the natural killer-mediated lysis of freshly isolated myeloid or lymphoblastic leukemias: evidence for the involvement of the Poliovirus receptor (CD155) and Nectin-2 (CD112). Blood, 105(5):2066 2073, Mar 2005. [175] F L Pietersma, S van Dorp, M C Minnema, J Kuball, E Meijer, R Schuurman, and D van Baarle. Inuence of donor cytomegalovirus (CMV) status on severity of viral reactivation after allogeneic stem cell transplantation in CMV-seropositive recipients. Clin Infect Dis, 52(7):144148, Apr 2011. [176] V Pitard, D Roumanes, X Lafarge, L Couzi, I Garrigue, M E Lafon, P Merville, J F Moreau, and J Dchanet-Merville. Long-term expansion of effector/memory Vdelta2-gammadelta T cells is a specic blood signature of CMV infection. Blood, 112(4):13171324, Aug 2008. [177] A Poggi, C Venturino, S Catellani, M Clavio, M Miglino, M Gobbi, A Steinle, P Ghia, S Stella, F Caligaris-Cappio, and M R Zocchi. Vdelta1 T lymphocytes from B-CLL patients recognize ULBP3 expressed on leukemic B cells and up-regulated by trans-retinoic acid. Cancer Res, 64(24):91729179, Dec 2004. [178] I Puig-Pey, F Bohne, C Bentez, M Lpez, M Martnez-Llordella, F Oppenheimer, J J Lozano, J Gonzlez-Abraldes, G Tisone, A Rimola, and A Snchez-Fueyo. Characterization of T cell subsets in organ transplantation. Transpl Int, 23(10):1045 1055, Oct 2010.
145
[179] V Ratanatharathorn, L Ayash, C Reynolds, S Silver, P Reddy, M Becker, J L Ferrara, and J P Uberti. Treatment of chronic graft-versus-host disease with anti-CD20 chimeric monoclonal antibody. Biol Blood Marrow Transplant, 9(8):505511, Aug 2003. [180] V Ratanatharathorn, B Logan, D Wang, M Horowitz, J P Uberti, O Ringden, R P Gale, H Khoury, M Arora, S Spellman, C Cutler, J Antin, M Bornhaser, G Hale, L Verdonck, M Cairo, V Gupta, S Pavletic, and Center for International Blood and Marrow Transplant Research, Milwaukee, WI, USA. Prior rituximab correlates with less acute graft-versus-host disease and better survival in B-cell lymphoma patients who received allogeneic peripheral blood stem cell transplantation. Br J Haematol, 145(6):816824, Jun 2009. [181] C Reynolds, V Ratanatharathorn, P Adams, T Braun, S Silver, L Ayash, E Carson, A Eisbruch, L A Dawson, K McDonagh, J Ferrara, and J Uberti. Allogeneic stem cell transplantation reduces disease progression compared to autologous transplantation in patients with multiple myeloma. Bone Marrow Transplant, 27(8):801 807, Apr 2001. [182] J C Ribot, A debarros, and B Silva-Santos. Searching for "signal 2": costimulation requirements of T cells. Cell Mol Life Sci, 68(14):23452355, Jul 2011. [183] S R Riddell and P D Greenberg. The use of anti-CD3 and anti-CD28 monoclonal antibodies to clone and expand human antigen-specic T cells. J Immunol Methods, 128(2):189201, Apr 1990. [184] B Rincon-Orozco, V Kunzmann, P Wrobel, D Kabelitz, A Steinle, and T Herrmann. Activation of V gamma 9V delta 2 T cells by NKG2D. J Immunol, 175(4):21442151, Aug 2005. [185] O Ringdn, M Labopin, N C Gorin, N Schmitz, U W Schaefer, H G Prentice, L Bergmann, J P Jouet, F Mandelli, D Blaise, L Fouillard, F Frassoni, and Acute Leukaemia Working Party of the European Group for Blood and Marrow Transplantation. Is there a graft-versus-leukaemia effect in the absence of graft-versushost disease in patients undergoing bone marrow transplantation for acute leukaemia? Br J Haematol, 111(4):11301137, Dec 2000. [186] O Ringdn, M Uzunel, I Rasmusson, M Remberger, B Sundberg, H Lnnies, H U Marschall, A Dlugosz, A Szakos, Z Hassan, B Omazic, J Aschan, L Barkholt, and K Le Blanc. Mesenchymal stem cells for treatment of therapy-resistant graftversus-host disease. Transplantation, 81(10):13901397, May 2006. [187] A M Rouquette-Gally, D Boyeldieu, A C Prost, and E Gluckman. Autoimmunity after allogeneic bone marrow transplantation. A study of 53 long-term-surviving patients. Transplantation, 46(2):238240, Aug 1988. [188] V Rowe, T Banovic, K P MacDonald, R Kuns, A L Don, E S Morris, A C Burman, H M Bonger, A D Clouston, and G R Hill. Host B cells produce IL-10 following TBI and attenuate acute GVHD after allogeneic bone marrow transplantation. Blood, 108(7):24852492, Oct 2006.
146
BIBLIOGRAPHY
[189] L Ruggeri, M Capanni, E Urbani, K Perruccio, W D Shlomchik, A Tosti, S Posati, D Rogaia, F Frassoni, F Aversa, M F Martelli, and A Velardi. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science, 295(5562):20972100, Mar 2002. [190] A M Russano, G Bassotti, E Agea, O Bistoni, A Mazzocchi, A Morelli, S A Porcelli, and F Spinozzi. CD1-restricted recognition of exogenous and self-lipid antigens by duodenal gammadelta+ T lymphocytes. J Immunol, 178(6):36203626, Mar 2007. [191] A Sandstrom, L Scharf, G McRae, A J Hawk, S C Meredith, and E J Adams. st T cell receptors recognize the non-classical major histocompatibility complex (MHC) molecule T22 via conserved anchor residues in a MHC peptide-like fashion. J Biol Chem, 287(8):60356043, Feb 2012. [192] S Sarantopoulos, K E Stevenson, H T Kim, C S Cutler, N S Bhuiya, M Schowalter, V T Ho, E P Alyea, J Koreth, B R Blazar, R J Soiffer, J H Antin, and J Ritz. Altered B-cell homeostasis and excess BAFF in human chronic graft-versus-host disease. Blood, 113(16):38653874, Apr 2009. [193] S Sarantopoulos, K E Stevenson, H T Kim, W S Washel, N S Bhuiya, C S Cutler, E P Alyea, V T Ho, R J Soiffer, J H Antin, and J Ritz. Recovery of B-cell homeostasis after rituximab in chronic graft-versus-host disease. Blood, 117(7):22752283, Feb 2011. [194] W Scheper, S van Dorp, S Kersting, F Pietersma, C Lindemans, S Hol, S Heijhuurs, Z Sebestyen, C Grnder, V Marcu-Malina, A Marchant, C Donner, B Plachter, D Vermijlen, D van Baarle, and J Kuball. T cells elicited by CMV reactivation after allo-SCT cross-recognize CMV and leukemia. Leukemia, Jan 2013. [195] P Schneider, F MacKay, V Steiner, K Hofmann, J L Bodmer, N Holler, C Ambrose, P Lawton, S Bixler, H Acha-Orbea, D Valmori, P Romero, C Werner-Favre, R H Zubler, J L Browning, and J Tschopp. BAFF, a novel ligand of the tumor necrosis factor family, stimulates B cell growth. J Exp Med, 189(11):17471756, Jun 1999. [196] K R Schultz, D B Miklos, D Fowler, K Cooke, J Shizuru, E Zorn, E Holler, J Ferrara, H Shulman, S J Lee, P Martin, A H Filipovich, M E Flowers, D Weisdorf, D Couriel, P A Lachenbruch, B Mittleman, G B Vogelsang, and S Z Pavletic. Toward biomarkers for chronic graft-versus-host disease: National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versushost disease: III. Biomarker Working Group Report. Biol Blood Marrow Transplant, 12(2):126137, Feb 2006. [197] K R Schultz, J Paquet, S Bader, and K T HayGlass. Requirement for B cells in T cell priming to minor histocompatibility antigens and development of graft-versushost disease. Bone Marrow Transplant, 16(2):289295, Aug 1995. [198] E Scotet, L O Martinez, E Grant, R Barbaras, P Jen, M Guiraud, B Monsarrat, X Saulquin, S Maillet, J P Estve, F Lopez, B Perret, X Collet, M Bonneville, and E Champagne. Tumor recognition following Vgamma9Vdelta2 T cell receptor
147
interactions with a surface F1-ATPase-related structure and apolipoprotein A-I. Immunity, 22(1):7180, Jan 2005. [199] Y Shen, D Zhou, L Qiu, X Lai, M Simon, L Shen, Z Kou, Q Wang, L Jiang, J Estep, R Hunt, M Clagett, P K Sehgal, Y Li, X Zeng, C T Morita, M B Brenner, N L Letvin, and Z W Chen. Adaptive immune response of Vgamma2Vdelta2+ T cells during mycobacterial infections. Science, 295(5563):22552258, Mar 2002. [200] A Shimabukuro-Vornhagen, M J Hallek, R F Storb, and M S von BergweltBaildon. The role of B cells in the pathogenesis of graft-versus-host disease. Blood, 114(24):49194927, Dec 2009. [201] H M Shulman, G E Sale, K G Lerner, E A Barker, P L Weiden, K Sullivan, B Gallucci, E D Thomas, and R Storb. Chronic cutaneous graft-versus-host disease in man. Am J Pathol, 91(3):545570, Jun 1978. [202] H M Shulman, K M Sullivan, P L Weiden, G B McDonald, G E Striker, G E Sale, R Hackman, M S Tsoi, R Storb, and E D Thomas. Chronic graft-versus-host syndrome in man. A long-term clinicopathologic study of 20 Seattle patients. Am J Med, 69(2):204217, Aug 1980. [203] G Soci. Chronic GVHD: B cells come of age. Blood, 117(7):20862087, Feb 2011. [204] M L Sorror, M B Maris, B Storer, B M Sandmaier, R Diaconescu, C Flowers, D G Maloney, and R Storb. Comparing morbidity and mortality of HLA-matched unrelated donor hematopoietic cell transplantation after nonmyeloablative and myeloablative conditioning: inuence of pretransplantation comorbidities. Blood, 104(4):961968, Aug 2004. [205] R Spaapen and T Mutis. Targeting haematopoietic-specic minor histocompatibility antigens to distinguish graft-versus-tumour effects from graft-versus-host disease. Best Pract Res Clin Haematol, 21(3):543557, Sep 2008. [206] R Spaapen, K van den Oudenalder, R Ivanov, A Bloem, H Lokhorst, and T Mutis. Rebuilding human leukocyte antigen class II-restricted minor histocompatibility antigen specicity in recall antigen-specic T cells by adoptive T cell receptor transfer: implications for adoptive immunotherapy. Clin Cancer Res, 13(13):40094015, Jul 2007. [207] S Srivastava, A Lundqvist, and R W Childs. Natural killer cell immunotherapy for cancer: a new hope. Cytotherapy, 10(8):775783, 2008. [208] E F Srour, T Leemhuis, L Jenski, R Redmond, and J Jansen. Cytolytic activity of human natural killer cell subpopulations isolated by four-color immunouorescence ow cytometric cell sorting. Cytometry, 11(3):442446, 1990. [209] T Stanislawski, R H Voss, C Lotz, E Sadovnikova, R A Willemsen, J Kuball, T Ruppert, R L Bolhuis, C J Melief, C Huber, H J Stauss, and M Theobald. Circumventing tolerance to a human MDM2-derived tumor antigen by TCR gene transfer. Nat Immunol, 2(10):962970, Oct 2001.
148
BIBLIOGRAPHY
[210] Y Sun, I Tawara, T Toubai, and P Reddy. Pathophysiology of acute graft-versushost disease: recent advances. Transl Res, 150(4):197214, Oct 2007. [211] S Svegliati, R Cancello, P Sambo, M Luchetti, P Paroncini, G Orlandini, G Discepoli, R Paterno, M Santillo, C Cuozzo, S Cassano, E V Avvedimento, and A Gabrielli. Platelet-derived growth factor and reactive oxygen species (ROS) regulate Ras protein levels in primary human broblasts via ERK1/2. Amplication of ROS and Ras in systemic sclerosis broblasts. J Biol Chem, 280(43):3647436482, Oct 2005. [212] S Svegliati, A Olivieri, N Campelli, M Luchetti, A Poloni, S Trappolini, G Moroncini, A Bacigalupo, P Leoni, E V Avvedimento, and A Gabrielli. Stimulatory autoantibodies to PDGF receptor in patients with extensive chronic graft-versushost disease. Blood, 110(1):237241, Jul 2007. [213] Y Takemoto, H Takatsuka, H Wada, A Mori, K Saheki, M Okada, S Tamura, Y Fujimori, T Okamoto, E Kakishita, and A Kanamaru. Evaluation of CMV/human herpes virus-6 positivity in bronchoalveolar lavage uids as early detection of acute GVHD following BMT: evidence of a signicant relationship. Bone Marrow Transplant, 26(1):7781, Jul 2000. [214] Y Tanaka, S Sano, E Nieves, G De Libero, D Rosa, R L Modlin, M B Brenner, B R Bloom, and C T Morita. Nonpeptide ligands for human gamma delta T cells. Proc Natl Acad Sci U S A, 91(17):81758179, Aug 1994. [215] Liane CJ te Boome, Cristina M Mansilla, Caroline A Lindemans, Lotte E Van der Wagen, Marloes Cuijpers, Ineke Slaper-Cortenbach, Henk Rozemuller, Eefke J Petersen, Eric Spierings, Marc Bierings, Jaap J Boelens, Nico Wulffraat, and Jurgen H Kuball. Treatment of Steroid Resistant Grade II to IV Acute Gvhd by Infusion of Mesenchymal Stroma Cells Expanded with Human Plasma and Platelet Lysate - a Phase I/II Study. ASH Annual Meeting Abstracts, 120(21):736, 2012. [216] T Teshima, K Nagafuji, H Henzan, K Miyamura, K Takase, M Hidaka, T Miyamoto, K Takenaka, K Akashi, and M Harada. Rituximab for the treatment of corticosteroid-refractory chronic graft-versus-host disease. Int J Hematol, 90(2):253 260, Sep 2009. [217] J S Thompson, P Schneider, S L Kalled, L Wang, E A Lefevre, T G Cachero, F MacKay, S A Bixler, M Zafari, Z Y Liu, S A Woodcock, F Qian, M Batten, C Madry, Y Richard, C D Benjamin, J L Browning, A Tsapis, J Tschopp, and C Ambrose. BAFF binds to the tumor necrosis factor receptor-like molecule B cell maturation antigen and is important for maintaining the peripheral B cell population. J Exp Med, 192(1):129135, Jul 2000. [218] W J Urba and D L Longo. Redirecting T cells. N Engl J Med, 365(8):754757, Aug 2011. [219] E M Urban, A I Chapoval, and C D Pauza. Repertoire development and the control of cytotoxic/effector function in human gammadelta T cells. Clin Dev Immunol, 2010:732893732893, 2010.
149
[220] N W van de Donk, N Krger, U Hegenbart, P Corradini, J F San Miguel, H Goldschmidt, J A Perez-Simon, M Zijlmans, R A Raymakers, V Montefusco, F A Ayuk, M H van Oers, A Nagler, L F Verdonck, and H M Lokhorst. Prognostic factors for donor lymphocyte infusions following non-myeloablative allogeneic stem cell transplantation in multiple myeloma. Bone Marrow Transplant, 37(12):11351141, Jun 2006. [221] N W van de Donk, N Krger, U Hegenbart, P Corradini, J F San Miguel, H Goldschmidt, J A Perez-Simon, M Zijlmans, R A Raymakers, V Montefusco, F A Ayuk, M H van Oers, A Nagler, L F Verdonck, and H M Lokhorst. Remarkable activity of novel agents bortezomib and thalidomide in patients not responding to donor lymphocyte infusions following nonmyeloablative allogeneic stem cell transplantation in multiple myeloma. Blood, 107(8):34153416, Apr 2006. [222] D van der Harst, E Goulmy, J H Falkenburg, Y M Kooij-Winkelaar, S A van Luxemburg-Heijs, H M Goselink, and A Brand. Recognition of minor histocompatibility antigens on lymphocytic and myeloid leukemic cells by cytotoxic T-cell clones. Blood, 83(4):10601066, Feb 1994. [223] L T van der Veken, R S Hagedoorn, M M van Loenen, R Willemze, J H Falkenburg, and M H Heemskerk. Alphabeta T-cell receptor engineered gammadelta T cells mediate effective antileukemic reactivity. Cancer Res, 66(6):33313337, Mar 2006. [224] J J van Dongen, A W Langerak, M Brggemann, P A Evans, M Hummel, F L Lavender, E Delabesse, F Davi, E Schuuring, R Garca-Sanz, J H van Krieken, J Droese, D Gonzlez, C Bastard, H E White, M Spaargaren, M Gonzlez, A Parreira, J L Smith, G J Morgan, M Kneba, and E A Macintyre. Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia, 17(12):22572317, Dec 2003. [225] S van Dorp, F Pietersma, M Wl, L F Verdonck, E J Petersen, H M Lokhorst, E Martens, M Theobald, D van Baarle, E Meijer, and J Kuball. Rituximab treatment before reduced-intensity conditioning transplantation associates with a decreased incidence of extensive chronic GVHD. Biol Blood Marrow Transplant, 15(6):671678, Jun 2009. [226] S van Dorp, H Resemann, L te Boome, F Pietersma, D van Baarle, F GmeligMeyling, R de Weger, E Petersen, M Minnema, H Lokhorst, S Ebeling, S J Beijn, E F Knol, M van Dijk, E Meijer, and J Kuball. The immunological phenotype of rituximab-sensitive chronic graft-versus-host disease: a phase II study. Haematologica, 96(9):13801384, Sep 2011. [227] M M van Loenen, R S Hagedoorn, M G Kester, M Hoogeboom, R Willemze, J H Falkenburg, and M H Heemskerk. Kinetic preservation of dual specicity of coprogrammed minor histocompatibility antigen-reactive virus-specic T cells. Cancer Res, 69(5):20342041, Mar 2009.
150
BIBLIOGRAPHY
[228] L F Verdonck, E J Petersen, H M Lokhorst, H K Nieuwenhuis, A W Dekker, M G Tilanus, and R A de Weger. Donor leukocyte infusions for recurrent hematologic malignancies after allogeneic bone marrow transplantation: impact of infused and residual donor T cells. Bone Marrow Transplant, 22(11):10571063, Dec 1998. [229] D Vermijlen, M Brouwer, C Donner, C Liesnard, M Tackoen, M Van Rysselberge, N Twit, M Goldman, A Marchant, and F Willems. Human cytomegalovirus elicits fetal gammadelta T cell responses in utero. J Exp Med, 207(4):807821, Apr 2010. [230] M von Bonin, F Stlzel, A Goedecke, K Richter, N Wuschek, K Hlig, U Platzbecker, T Illmer, M Schaich, J Schetelig, A Kiani, R Ordemann, G Ehninger, M Schmitz, and M Bornhuser. Treatment of refractory acute GVHD with thirdparty MSC expanded in platelet lysate-containing medium. Bone Marrow Transplant, 43(3):245251, Feb 2009. [231] R H Voss, J Kuball, and M Theobald. Designing TCR for cancer immunotherapy. Methods Mol Med, 109:229256, 2005. [232] H Wang, Z Fang, and C T Morita. Vgamma2Vdelta2 T Cell Receptor recognition of prenyl pyrophosphates is dependent on all CDRs. J Immunol, 184(11):62096222, Jun 2010. [233] E H Warren, N Fujii, Y Akatsuka, C N Chaney, J K Mito, K R Loeb, T A Gooley, M L Brown, K K Koo, K V Rosinski, S Ogawa, A Matsubara, F R Appelbaum, and S R Riddell. Therapy of relapsed leukemia after allogeneic hematopoietic cell transplantation with T cells specic for minor histocompatibility antigens. Blood, 115(19):38693878, May 2010. [234] M Wilhelm, V Kunzmann, S Eckstein, P Reimer, F Weissinger, T Ruediger, and H P Tony. Gammadelta T cells for immune therapy of patients with lymphoid malignancies. Blood, 102(1):200206, Jul 2003. [235] C R Willcox, V Pitard, S Netzer, L Couzi, M Salim, T Silberzahn, J F Moreau, A C Hayday, B E Willcox, and J Dchanet-Merville. Cytomegalovirus and tumor stress surveillance by binding of a human T cell antigen receptor to endothelial protein C receptor. Nat Immunol, 13(9):872879, Sep 2012. [236] L C Wu, D S Tuot, D S Lyons, K C Garcia, and M M Davis. Two-step binding mechanism for T-cell receptor recognition of peptide MHC. Nature, 418(6897):552 556, Aug 2002. [237] T A Wynn. Common and unique mechanisms regulate brosis in various broproliferative diseases. J Clin Invest, 117(3):524529, Mar 2007. [238] B Xu, J C Pizarro, M A Holmes, C McBeth, V Groh, T Spies, and R K Strong. Crystal structure of a gammadelta T-cell receptor specic for the human MHC class I homolog MICA. Proc Natl Acad Sci U S A, 108(6):24142419, Feb 2011. [239] C Xu, H Zhang, H Hu, H He, Z Wang, Y Xu, H Chen, W Cao, S Zhang, L Cui, D Ba, and W He. Gammadelta T cells recognize tumor cells via CDR3delta region. Mol Immunol, 44(4):302310, Jan 2007.
151
[240] S Yang, Y Ji, L Gattinoni, L Zhang, Z Yu, N P Restifo, S A Rosenberg, and R A Morgan. Modulating the differentiation status of ex vivo-cultured anti-tumor T cells using cytokine cocktails. Cancer Immunol Immunother, 62(4):727736, Apr 2013. [241] T Yi, D Zhao, C L Lin, C Zhang, Y Chen, I Todorov, T LeBon, F Kandeel, S Forman, and D Zeng. Absence of donor Th17 leads to augmented Th1 differentiation and exacerbated acute graft-versus-host disease. Blood, 112(5):21012110, Sep 2008. [242] F Zaja, A Bacigalupo, F Patriarca, M Stanzani, M T Van Lint, C Fil, R Scim, G Milone, M Falda, C Vener, D Laszlo, P E Alessandrino, F Narni, S Sica, A Olivieri, A Sperotto, A Bosi, F Bonifazi, R Fanin, and GITMO (Gruppo Italiano Trapianto Midollo Osseo). Treatment of refractory chronic GVHD with rituximab: a GITMO study. Bone Marrow Transplant, 40(3):273277, Aug 2007. [243] R Zeiser and J Finke. Allogeneic haematopoietic cell transplantation for multiple myeloma: reducing transplant-related mortality while harnessing the graft-versusmyeloma effect. Eur J Cancer, 42(11):16011611, Jul 2006. [244] D Zhao, C Zhang, T Yi, C L Lin, I Todorov, F Kandeel, S Forman, and D Zeng. In vivo-activated CD103+CD4+ regulatory T cells ameliorate ongoing chronic graftversus-host disease. Blood, 112(5):21292138, Sep 2008. [245] H Zhou, M Guo, C Bian, Z Sun, Z Yang, Y Zeng, H Ai, and R C Zhao. Efcacy of bone marrow-derived mesenchymal stem cells in the treatment of sclerodermatous chronic graft-versus-host disease: clinical report. Biol Blood Marrow Transplant, 16(3):403412, Mar 2010.
152
153
Hoofdstuk 8
Nederlandse samenvatting
Introductie
Allogene stamcel transplantatie (allo-SCT) wordt gebruikt in de behandeling van verschillende hematologische maligniteiten zoals het multipel myeloom (MM) en acute myeloide leukemie (AML). Het effect van de behandeling wordt voor een deel verzorgd door de conditionering die wordt gegeven voor de alloSCT. Deze bestaat uit een hoge dosis chemotherapie en bestraling, waardoor de kankercellen, maar ook het immuunsysteem van de patint vernietigd worden. Hierna volgt de transplantatie van donor stamcellen, die uit zullen groeien tot een nieuw immuunsysteem. De grootste bijdrage aan het slagen van de behandeling wordt geleverd door het zogeheten graft-versus-tumor effect. Dit houdt in dat het getransplanteerde immuunsysteem de mogelijk overgebleven kankercellen aanvalt en zorgt hierdoor dat de patint ook na de behandeling kankervrij blijft. Het graft-versus-tumor effect ontstaat doordat de patint en de donor verschillen, ondanks dat hun HLA-type gematched wordt. Hierdoor zal het immuunsysteem van de donor de kankercellen van de patint als niet eigen herkennen en deze aanvallen. Er zijn drie belangrijke oorzaken van het falen van de behandeling met allo-SCT. 1. Ondanks het graft-versus-tumor effect krijgen veel patinten (10-45%) relapse van de ziekte. 2. Er is een belangrijke sterfte door de behandeling zelf (20-32%). Deze wordt met name veroorzaakt door een bijwerking van het graft-versus-tumor effect, graft-versus-host ziekte (GVHD). Hierbij worden niet alleen kankercellen, 155
maar ook de normale cellen van het lichaam van de patint als niet eigen herkent en aangevallen door het nieuwe immuunsysteem. Er ontstaat een afstotingsreactie van het lichaam van de patint. Dit kan kort na de allo-SCT ontstaan (acute GVHD), maar ook op langere termijn zorgen voor invaliditeit en sterfte (chronische GVHD). 3. Er ontstaan na de allo-SCT, door de immunosuppressiva die gebruikt moeten worden om GVHD te voorkomen, veel infectieuze complicaties, met name reactivaties van herpesvirussen zoals het Epstein Barr virus (EBV) en cytomegalovirus (CMV). Verbetering van de allo-SCT is nodig om deze complicaties te verminderen. Deze verbetering kan bereikt worden door de complicaties beter te behandelen. Daarnaast kan men door betere kennis van de precieze mechanismen die ten grondslag liggen aan het graft-versus-tumor effect, aan GVHD en het ontstaan van infecties, nieuwe strategien ontwikkelen om de effectiviteit van de alloSCT te vergroten en de complicaties te voorkomen. Dit proefschrift bestaat uit drie delen. In deel 1 laten we het effect van allo-SCT zien in patinten met een multipel myeloom. In deel 2 gaan we in op de rol van B cellen in de ontstaanswijze, behandeling en het voorkomen van chronische GVHD. In deel 3 gebruiken we T cellen als nieuwe strategie om leukemie en virale reactivaties te controleren en GVHD te voorkomen.
(I) De therapeutische waarde van allo-SCT

Bij patinten met een multipel myeloom zijn hun plasmacellen maligne ontaard. Deze cellen houden zich met name op in het beenmerg. Er zijn verschillende (chemotherapeutische) behandelingen, die deze ziekte kunnen onderdrukken en een patint zelfs langdurig ziekte vrij kunnen laten zijn. De enige behandeling die de ziekte echt kan genezen, lijkt een allo-SCT te zijn. In deel 1 van dit proefschrift werden de uitkomsten van de behandeling met een allo-SCT van 63 patinten met een multipel myeloom in het UMC Utrecht retrospectief geanalyseerd. We hebben gekeken naar de respons op de behandeling (afname van de ziekte activiteit), de overleving na allo-SCT, de ziekte vrije overleving (als maat voor de hoeveelheid patinten met een relapse van de ziekte), naar de 156
HOOFDSTUK 8. NEDERLANDSE SAMENVATTING
sterfte door de behandeling zelf en naar de patintengroep voor wie allo-SCT nu het meest effectief is. Behandeling met allo-SCT zorgde voor een afname van de ziekte activiteit bij 90% van de patinten. Na 2 jaar was 84% van de patinten nog in leven. Helaas kreeg 42% van de patinten in de 2 jaar followup een relapse van de ziekte. Sterfte door de behandeling zelf kwam bij 12% van de patinten voor. 64% van de patinten kreeg acute GVHD en 54% van de patinten kreeg chronische GVHD. Patinten die snel na het ontstaan van het multipel myeloom behandeld werden met een allo-SCT en die de allo-SCT kregen op het moment dat er door de voorafgaande chemotherapie geen ziekte activiteit meer was, reageerden het beste op de behandeling met allo-SCT. Deze patinten bleven de gehele follow-up duur ziekte vrij. De resultaten doen ons concluderen, dat allo-SCT efcint is bij patinten met een multipel myeloom. De sterfte door de behandeling zelf in deze patintengroep was niet hoger dan in andere groepen, maar blijft een probleem. Daarnaast waren er veel patinten met een acute en chronische GVHD. Deze complicaties benvloeden de kwaliteit van leven na allo-SCT enorm.
(II) Het mechanisme van chronische GVHD

GVHD is de belangrijkste complicatie na allo-SCT. De acute vorm treedt in de eerste maanden na allo-SCT op en kan een ernstig verloop hebben. We weten dat bij acute GVHD specieke afweercellen, zogeheten T cellen (gerijpt in de Thymus), een belangrijke rol spelen in het ontstaan van de ziekte. Deze T cellen zijn gericht tegen de normale cellen van de patint en maken deze kapot. Dit resulteert in ernstige huiduitslag, met blaren en zweren, diarree en het falen van de lever. Chronische GVHD is de meest voorkomende complicatie, die op langere termijn na allo-SCT kan ontstaan en kan tot jaren na de allo-SCT voor veel problemen zorgen tot sterfte aan toe. Het is een ziekte, die de kwaliteit van leven vreselijk in de weg staat. Chronische GVHD kan alle organen aantasten en kan veel verschillende uitingsvormen hebben. Bij het grootste deel van de patinten is de huid aangedaan. Meestal zien we verlittekening van de huid. Deze verlittekening tast niet alleen de huid aan, maar ook andere organen, zoals de longen, de nieren, de (slok)darm en de lever. Door de verlittekening kunnen organen 157
minder goed functioneren. De huid kan zelfs dusdanig strak worden, dat het als een harnas om het lijf zit en dat ademhalen bemoeilijkt wordt. Bij andere patinten zien we bij chronische GVHD juist meer zweren en wonden van de huid. Het verschil tussen acute en chronisch GVHD zit niet alleen in het tijdstip van ontstaan, maar ook in de uitingsvorm. Met name de vorm van chronische GVHD, waarbij veel verlittekening gezien wordt, lijkt niet op acute GVHD. Dit verschil doet vermoeden dat ook de ontstaanswijze anders moet zijn. De verlittekening bij chronische GVHD heeft veel weg van de verlittekening die gezien wordt bij auto-immuunziekten, zoals systemische sclerose. Bij autoimmuunziekten zijn niet alleen T cellen belangrijk, maar ook een ander soort immuuncel, namelijk de B cel (Beenmerg gerijpt). De vergelijking met autoimmuunziekten heeft ervoor gezorgd, dat er studies gedaan zijn, waarbij kleine patinten aantallen behandeld werden met rituximab, een geneesmiddel dat speciek B cellen verwijdert uit het lichaam (B-cel depletie). Rituximab wordt veel gebruikt in de behandeling van lymfomen en chronische lymfatische leukemie (CLL), aangezien bij deze ziektebeelden de B cellen ontaarden in kankercellen. B-cel depletie lijkt ook bij een aantal patinten met chronische GVHD heel goed te werken als behandeling. In deel II van dit proefschrift onderzochten we als eerste de mogelijkheid om chronische GVHD te voorkomen. In een grote groep patinten, die een allo-SCT kregen voor uiteenlopende hematologische maligniteiten, hebben we onderscheid gemaakt tussen patinten die in hun behandeling voorafgaande aan de allo-SCT wel rituximab kregen en patinten die dit niet kregen. Bij patinten, die in de 6 maanden voor hun allo-SCT wel rituximab hadden gekregen en daarmee dus geen B-cellen meer hadden, zagen we dat er minder en minder ernstige chronische GVHD ontstond. De B-cel depletie zorgde ervoor dat de B cellen van de patint verwijderd werden, maar had mogelijk ook effect op het uitgroeien van de B cellen van de donor. Hier konden we geen onderscheid in maken in deze studie. Wel kunnen we concluderen, dat rituximab een waardevolle aanvulling zou kunnen zijn in de behandeling voorafgaande aan de allo-SCT om chronische GVHD te voorkomen. Daarna hebben we geprobeerd de ontstaanswijze van chronische GVHD beter 158
te begrijpen om zo beter gebruik te kunnen maken van de juiste behandeling voor de juiste patinten. Om dit te bereiken hebben we een studie uitgevoerd, waarbij we patinten met chronische GVHD, die niet goed reageerden op de standaard therapie met prednison, behandelden met rituximab. Van deze patinten werden van te voren alle kenmerken en klachten in kaart gebracht en er werd bloed afgenomen. Na de behandeling werden ze maandelijks teruggezien en werd gekeken naar de response op de behandeling met rituximab. Wij zagen dat een groep patinten wel en een groep patinten nauwelijks reageerde op de behandeling. We observeerden dat degenen die wel reageerden voornamelijk chronische GVHD hadden, die zich uitte in verlittekening van de huid en leek op de auto-immuun ziekte systemische sclerose. Bij de patinten, die goed reageerden op de behandeling met rituximab, vonden we tevoren hogere aantallen B cellen, die ook meer actief leken te zijn. Daarnaast hadden deze B cellen andere kenmerken dan bij patinten, die niet goed reageerden. De B cellen in patinten die niet goed reageerden, waren niet verhoogd aanwezig en hadden dezelfde kenmerken als bij controle patinten, die geen chronische GVHD ontwikkelden. Na de behandeling met rituximab duurde het wel een jaar voordat er weer B cellen gevonden werden in de patinten. Bij degenen, die goed reageerden op rituximab, hadden de B cellen, die na een jaar na behandeling teruggekomen waren, opvallend genoeg dezelfde kenmerken als bij controle patinten zonder GVHD. Het leek alsof de behandeling met rituximab de B cellen bij deze patinten als het ware had gereset. Uit deze studie konden we een aantal belangrijke dingen opmaken. Ten eerste is de populatie patinten met chronische GVHD geen eenheidsworst. Chronische GVHD heeft veel verschillende verschijningsvormen en niet alle patinten met de ziekte hebben dezelfde kenmerken. De uiterlijke kenmerken van de ziekte verschillen, maar blijkbaar verschillen ook de onderliggende mechanismen. Bij patinten die goed reageerden op de behandeling vonden we van te voren (dus tijdens de chronische GVHD) hogere aantal B cellen met specieke kenmerken, die na de behandeling genormaliseerd waren. Dit geeft aan, dat deze geactiveerde B cellen heel goed de verlittekening kenmerkend voor de chronische GVHD kunnen bewerkstelligen. Daarnaast zou je deze uiterlijke kenmerken van zowel patinten als van de B-cel populatie van de patinten kunnen gebruiken om te bezien bij welke patinten het wel zin heeft om B-cel depletie te gebruiken als behandeling en bij welke patinten dat geen zin heeft. Deze kennis brengt ons dichter 159
bij het precieze ontstaansmechanisme of liever gezegd mechanismen van chronische GVHD en verklaart waarom bij sommige patinten rituximab wel werkt als behandeling en bij andere niet.
(III) Nieuwe strategien om relapse en complicaties te voorkomen

Niet alleen GVHD zorgt voor problemen na allo-SCT. Bij een groot aantal patinten komt de onderliggende ziekte weer terug (relapse). Daarnaast zijn ook infecties, zoals door herpesvirussen, een belangrijke oorzaak voor invaliditeit en sterfte na de allo-SCT. Normaalgesproken kan het immuunsysteem tumorcellen en cellen, die door een virus genfecteerd zijn, herkennen en doden, zodat bijvoorbeeld leukemie en virusinfecties onder controle blijven. Belangrijke spelers in dit controlemechanisme zijn T cellen, die een speciek eiwit aan de buitenkant dragen, namelijk de T-cel receptor. Deze receptor bestaat uit twee delen, een -keten en een keten, en past precies op een molecuul aan de oppervlakte van normale cellen, het human leukocyte antigen (HLA) molecuul. Patint en donor worden gematched op hun HLA type, zodat het nieuwe immuunsysteem kan functioneren in het lichaam van de patint. Het HLA-molecuul bevat kleine eiwitdeeltjes uit de cel en laat deze zien aan de T cellen. Wanneer deze eiwitdeeltjes afkomstig zijn van een virus of wanneer ze afkomstig zijn uit een tumorcel, kan een T cel, die deze eiwitdeeltjes speciek herkent, de genfecteerde dan wel tumorcellen doden. Bij patinten met leukemie heeft het controle mechanisme gefaald. Een oplossing voor dit probleem zou kunnen zijn om grote aantallen T cellen, die speciek de tumorcellen herkennen, als behandeling aan de patint te geven. Dit kan door heel veel van deze cellen uit een donor te halen of door de T-cel receptor, met de kwaliteit om speciek tumorcellen te herkennen, op andere, eerder niet specieke, T cellen over te brengen. Er zijn een aantal nadelen aan het gebruik van T cellen. Omdat ze precies passen op een bepaald HLA-type, moeten er voor ieder HLA-type aparte T cellen gebruikt worden. Daarnaast kan het overbrengen van een T-cel receptor op een andere T cel zorgen voor zogeheten on target en off target auto-immuniteit. On tar160
get auto-immuniteit houdt in, dat de specieke T-cel receptor stukjes eiwit herkent, die met name gepresenteerd worden op tumorcellen, maar in mindere mate ook op normale cellen voorkomen. Hierdoor zal deze specieke T-cel ook normale cellen doden. Off target auto-immuniteit houdt in, dat wanneer de specieke T-cel receptor overgebracht wordt op een andere T cel, de en -keten van de specieke receptor combinaties maken met de - en -keten van de T-cel receptor, die bij de niet specieke T cel hoort. Er ontstaan dan 2 nieuwe soorten T-cel receptoren, waarvan we in het geheel niet weten wat ze herkennen. Dit kan zorgen voor nieuwe receptoren, die ook weer normale cellen herkennen en doden. Een oplossing voor al deze nadelen is het gebruik van een ander soort T-cel receptor, namelijk de T-cel receptor, die een - en een -keten heeft. T cellen met deze receptor, T cellen, maken 5% uit van alle T cellen, die in het bloed aanwezig zijn en bevinden zich in grotere hoeveelheden in de huid, long en darm. Zij herkennen ook tumorcellen en genfecteerde cellen, maar doen dit op een andere wijze dan T cellen. Zij passen niet op HLA-moleculen, maar herkennen andere eiwitten, die alleen op tumorcellen en genfecteerde cellen aanwezig zijn en zullen daarom normale cellen met rust laten, waardoor on target auto-immuniteit voorkomen wordt. Daarnaast kan n T-cel receptor voor patinten gebruikt worden met diversen HLA-typen. Wanneer een T-cel receptor overgebracht wordt op een T cel, zal deze niet combineren met de en -keten, die aanwezig is in de T cel en een T cel bevat geen - en geen -keten. Hierdoor kunnen er geen T-cel receptoren met onbekende speciciteit ontstaan en wordt off target auto-immuniteit voorkomen. In deel III onderzochten we hoe en welke T-cel receptoren het beste gebruikt kunnen worden om leukemie en virale infecties te controleren. We observeerden dat wanneer T cellen werden ingezet in de behandeling van leukemie of andere soorten kanker, dit vaak teleurstellende resultaten had. Wij vonden een mogelijke verklaring en een oplossing voor dit probleem. Een sterke reactie tegen tumorcellen is niet alle T cellen gegeven. Sommige T-cel receptoren zijn beter in het herkennen van tumorcellen dan andere. Wij hebben uit een grote groep T cellen klonen gemaakt, met verschillende T-cel receptoren. Hierdoor konden we de T-cel receptoren vergelijken, die wel goed waren en
161
niet goed waren in het herkennen van tumorcellen. Daarnaast konden we de T-cel receptoren, die heel goed waren per aminozuur veranderen, waardoor het duidelijk werd, welk deel van de T-cel receptor nu precies het verschil bepaalde. We ontdekten dat de lengte van de -keten van belang was, maar ook de combinatie van bepaalde - en -ketens. Veranderingen in zowel de - als de -keten kon het reactievermogen van de T cellen op tumorcellen veranderen. Als je de juiste - en de juiste -keten gebruikt, en deze overbrengt op een T cel, onafhankelijk van het type HLA, kun je een grote hoeveelheid T cellen genereren, die tumoren goed herkennen en doden en geen auto-immuniteit geven. Hierdoor zouden leukemie en andere tumoren beter gecontroleerd kunnen worden. Dit principe hebben we combinatorial T-cel receptor exchange (CTE) gedoopt. Een andere observatie was dat tijdens CMV infecties na allo-SCT vaak in eerste instantie T cellen zich vermenigvuldigden. Dit werd eerder ook al gezien bij gezonde vrijwilligers met CMV en bij patinten die een niertransplantatie hadden ondergaan. Dit zou betekenen dat T cellen belangrijk zijn in de controle van CMV infecties. Er bestaat een correlatie tussen CMV infectie na allo-SCT en een langere leukemie vrije overleving. Dit deed ons vermoeden dat de T cellen, die zich vermenigvuldigen tijdens CMV infectie, ook leukemie cellen kunnen herkennen en doden en daarmee leukemie na allo-SCT kunnen controleren. We vonden dat T cellen, die CMV herkenden, ook leukemie en ander tumorcellen herkenden en doodden. Echter wanneer we de T-cel receptor van deze cellen overbrachten op een T cel, herkende deze alleen tumorcellen en was de herkenning van CMV genfecteerde cellen verloren gegaan. Dit betekent ten eerste dat we T cellen kunnen inzetten tegen leukemie en CMV infecties na allo-SCT. Ten tweede hebben we opnieuw laten zien dat de T-cel receptor zeer goed te gebruiken is om grote groepen T cellen te maken, die leukemie cellen kunnen herkennen en doden. Maar dit betekent ook dat voor de herkenning van CMV genfecteerde cellen moleculen nodig zijn, die op T cellen aanwezig zijn en niet op T cellen en dat het overbrengen van alleen de T-cel receptor wel genoeg is voor de herkenning van tumoren, maar niet genoeg is voor de herkenning van CMV. Welke moleculen er precies nodig zijn om ook CMV te herkennen, hebben we nog niet kunnen achterhalen.
162
Conclusie
In dit proefschrift worden verschillende manieren beschreven om het nieuwe immuunsysteem dat gegeven wordt aan patinten met leukemie, vorm te geven. Hierdoor kan allo-SCT een verjndere behandeling worden. Door B cellen te depleteren kan GVHD voorkomen en behandeld worden. Door gebruik te maken van de unieke kenmerken van T cellen en de T-cel receptor kunnen relapse van de ziekte en complicaties zoals GVHD en infecties vermeden worden. Deze resultaten bieden mogelijkheden om in de toekomst de conventionele manier van transplanteren te vervangen door een gerichte en gendividualiseerde behandeling van leukemie.
163
164
Curriculum Vit
Suzanne van Dorp werd geboren op 8 juli 1982 te Haarlem. In 2000 behaalde zij het gymnasiumdiploma aan het Stedelijk Gymnasium te Haarlem. Tijdens haar studie Geneeskunde aan de Universiteit Utrecht deed zij onder andere coschappen in Nicaragua en Malawi. Tijdens het co-schap interne geneeskunde in het St. Antoniusziekenhuis ontstond haar interesse voor de hematologie. Haar wetenschappelijke stage in het UMC Utrecht mondde uit in een promotieonderzoek onder leiding van Dr. J.H.E. Kuball en Dr. E. Meijer bij de afdelingen Hematologie en Immunologie. De resultaten van dit onderzoek zijn beschreven in dit proefschrift. Suzanne is in opleiding tot internist in het St. Antoniusziekenhuis te Nieuwegein.
165
166
List of publications
van Dorp S, Meijer E, van de Donk NW, Dekker AW, Nieuwenhuis K, Minnema MC, Petersen E, Schutgens R, Verdonck LF, Lokhorst HM. Single-centre experience with nonmyeloablative allogeneic stem cell transplantation in patients with multiple myeloma: prolonged remissions induced. Neth J Med. 2007 May;65(5):178-84. Kersting S, van Dorp S, Theobald M, Verdonck LF. Acute renal failure after nonmyeloablative stem cell transplantation in adults. Biol Blood Marrow Transplant. 2008 Jan;14(1):125-31. Marcu-Malina V, van Dorp S, Kuball J. Re-targeting T-cells against cancer by gene-transfer of tumor-reactive receptors. Expert Opin Biol Ther. 2009 May;9(5):579-91. van Dorp S, Pietersma F, Wl M, Verdonck LF, Petersen EJ, Lokhorst HM, Martens E, Theobald M, van Baarle D, Meijer E, Kuball J. Rituximab treatment before reduced-intensity conditioning transplantation associates with a decreased incidence of extensive chronic GVHD. Biol Blood Marrow Transplant. 2009 Jun;15(6):671-8. Pietersma FL, van Dorp S, Jacobi R, Ran L, Nanlohy NM, Schuurman R, Minnema MC, Meijer E, van Baarle D. High level of perforin expression in T cells: An early prognostic marker of the severity of herpesvirus reactivation after allogeneic stem cell transplantation in adults. Clin Infect Dis. 2010 Mar 1;50(5):717-25. Westeneng AC, Hettinga Y, Lokhorst H, Verdonck L, van Dorp S, Rothova A. Ocular graft-versus-host disease after allogeneic stem cell transplantation. Cornea. 2010 Jul;29(7):758-63. Minnema MC, van Dorp S, van de Donk N, Lokhorst H. Prognostic factors and outcome in relapsed multiple myeloma after non-myeloablative allogen167
eic stem cell transplantation; a single center experience. Bone Marrow Transplant. 2011 Feb;46(2):244-9. Kuball J, de Boer K, Wagner E, Wattad M, Antunes E, Weeratna RD, Vicari AP, Lotz C, van Dorp S, Hol S, Greenberg PD, Heit W, Davis HL, Theobald M. Pitfalls of vaccinations with WT1-, Proteinase3- and MUC1-derived peptides in combination with MontanideISA51 and CpG7909. Cancer Immunol Immunother. 2011 Feb;60(2):161-71. Pietersma FL, van Dorp S, Minnema MC, Kuball J, Meijer E, Schuurman R, van Baarle D. Inuence of donor cytomegalovirus (CMV) status on severity of viral reactivation after allogeneic stem cell transplantation in CMVseropositive recipients. Clin Infect Dis. 2011 Apr 1;52(7):e144-8. van Dorp S , Resemann H , te Boome L, Pietersma F, van Baarle D, GmeligMeyling F, de Weger R, Petersen E, Minnema M, Lokhorst H, Ebeling S, Beijn SJ, Knol EF, van Dijk M, Meijer E, Kuball J. ( Both authors contributed equally) The immunological phenotype of rituximab-sensitive chronic graftversus-host disease: a phase II study. Haematologica. 2011 Sep;96(9):1380-4. Hogenes MC, van Dorp S, van Kuik J, Monteiro FR, ter Hoeve N, van Dijk MR, Martens AC, de Weger RA. Histological assessment of the sclerotic graftversus-host response in the humanized RAG2-/-c-/- mouse model. Biol Blood Marrow Transplant. 2012 Jul;18(7):1023-35 Grnder C, van Dorp S, Hol S, Drent E, Straetemans T, Heijhuurs S, Scholten K, Scheper W, Sebestyen Z, Martens A, Strong R, Kuball J. 9 and 2CDR3 domains regulate functional avidity of T-cells harboring 92T-cell receptors. Blood. 2012 Dec 20;120(26):5153-62. Scheper W , van Dorp S , Kersting S, Pietersma F, Hol S, Heijhuurs S, Lindemans C, Sebestyen Z, Grnder C, Marcu-Malina V, Marchant A, Donner C, Becke S, Plachter B, Vermijlen D, van Baarle D, Kuball J. ( Both authors contributed equally) T-cells elicited by CMV-reactivation after alloSCT cross recognize CMV and leukemia. 10.1038/leu.2012.374. [Epub ahead of print] Leukemia. 2013 Jan 1. doi:
168
Dankwoord
Lieve pap, dit boek is opgedragen aan jou. Het was mooier geweest, als ik het je had kunnen geven. Je hebt een belangrijke rol gespeeld in de keuze om te gaan promoveren. Vroeger zei je vaak, dat je eigenlijk wel gewoon kon promoveren op je afstudeerscriptie over Fockenbroch. Ik twijfelde daar toen al wat aan, en na 6 jaar promoveren denk ik dat die twijfel wel terecht was. Maar promoveren leek je wel wat, en je had gelijk. Het was zeker wat. Soms frustrerend, soms stressvol, maar meestal inspirerend en een voorrecht om te kunnen doen. Veel mensen hebben de afgelopen jaren voor die inspiratie gezorgd. Als eerste wil graag ik mijn co-promotor Dr. J.H.E. Kuball bedanken. Danke schn, Herr Doktor Kuball! Lieber Jrgen, het is ongelooijk hoe jij in vijf jaar van kortdurend mijn kamergenoot op AIO kamer 3 opgeklommen bent tot waar je nu staat. Maar het is ook weer niet geheel verbazingwekkend. Jouw intelligentie, je vindingrijkheid, je werklust en je absolute optimisme en doorzettingsvermogen zijn een inspiratie en de grootste kracht achter deze promotie. Ik ben heel erg blij, dat je mij geadopteerd hebt als AIO en het is een eer om jouw eerste promovendus te zijn. Daarnaast wil ik graag mijn co-promotor Dr. E. Meijer bedanken. Beste Ellen, na mijn wetenschappelijke stage onder jouw hoede ben ik begonnen aan dit promotie-traject. Helaas verliet je me na een jaar om je geluk elders te zoeken. Ik heb een van de belangrijkste schatten in het onderzoek van jou geleerd, het maken van een databank. Daarnaast heb ik tijdens onze hereniging in het St. Antoniusziekenhuis, gezien hoe je een goede hematoloog moet zijn! Dr. S.B. Ebeling, beste Saskia, dank je wel voor het begeleiden van mijn eerste ervaringen met lab-onderzoek, pipeteren, muisexperimenten, wetenschappelijke artikelen en berhaupt een volwassen baan hebben! Ik wil graag mijn promotor Prof. dr. H. M. Lokhorst bedanken. Beste Henk. Jij 169
bracht mij in contact met Ellen en jij bracht mij mijn eerste paper! Je staat daarmee aan de basis van mijn wetenschappelijke carrire. Dank je wel daarvoor. Jij bent voor mij ht voorbeeld voor hoe je n een geweldige klinische dokter kan zijn n tegelijkertijd heel erg belangrijk in de onderzoekswereld. De leden van de leescommissie: Prof. dr. E. Wiertz, Prof. dr. C.E. Hack, Prof. dr. A. van de Loosdrecht, Prof. dr. J. Cornelissen, veel dank voor het beoordelen van mijn manuscript. Prof. dr. P. Fisch, herzlichen Dank fr die Beurteilung meiner Dissertation. Dr. H. Dolstra en Dr. T. Mutis, dank voor het opponeren. Tuna, ook erg bedankt voor de spraakmakende discussies, de gezellige borrels en het salsaen! Dr. M. Minnema, Dr. E. Petersen, Dr. R. Schutgens, Dr. L. te Boome, Dr. N. van de Donk, Dr. E. van der Spek , dank jullie wel voor de prettige samenwerking op de afdeling Hematologie. Monique en Eefke, speciale dank voor het includeren van patinten. Liane, dank je wel voor het oppakken en voortzetten van de chronische GVHD studie en voor alle samenwerking in group Kuball! Prof Dr. M. van Dijk. Beste Marijke, dank je wel voor je hulp bij het analyseren van alle coupes. Monique Knies. Dank je wel voor je altijd supersnelle reacties, voor de jne interactie en voor het rennen naar de postkamer! Group Kuball: Vica, Zsolt, Cordula, Wouter, Sam, Sabine, Sabina, Liane, Caroline, Linda, Hakim, Julia, Carina, Edite: Thank you for all the labbesprechungen, the discussions, the experiments together, the nice dinners and celebrations. It was a pleasure and remember: Gamma delta T cells rock! Lieve Wouter, mijn Muze! Je maakte het laatste jaar op het lab tot een waar feest! Dank je wel voor al je hulp, je gezelligheid, je rust, je kunde en je heerlijke humor. Mijn ideaalbeeld voor de toekomst zou zijn om met jou een onderzoeksgroep te hebben. Ik weet niet of het uit gaat komen, want je hebt wellicht heel andere plannen, maar ik heb heel veel respect voor jou als onderzoeker en collega en niet te vergeten als vriend! Zullen we in ieder geval nog vaak samen bangen en mag ik dan alsjeblieft meezingen met lekkere hitjes? 170
DANKWOORD
Dear Vica, you have taught me so much! I admire how you take on projects from scratch and make them into successes. I wish the Dutch could have won you over, but I think that there is no place where you will be as happy as in Israel with your family! Dear Cordula, thank you for sharing your pure look on life and science. Your perfectionism led to a Blood paper! Do not forget to put stuff in perspective sometimes, since that can make things in life nd science al lot easier! Im sure that you will nd the way to do so and still be content with everything you do. Thank you for organising all those nice group events! Lieve Zsolt, toen jij kwam solliciteren, viel ik zowat van mijn stoel zo mooi vond ik je. Maar niet alleen daarom bleek het heel gelukkig dat je group Kuball kwam versterken. Je bracht een zeer hoog onderzoeksniveau, stijl, gezelligheid en humor met je mee en een heel jn Amsterdams juppengehalte samen met Jan! Lieve Sabina, Dr. Kersting. Wat was het jn om met jou onderzoek te doen en met Floor tussen ons in op de kamer te zitten! Je hebt je een ongeluk geFACSt en ik ben heel blij dat daar uiteindelijk zon mooi paper uitgekomen is! We hebben een onvergetelijk leuk congres in New Orleans gehad samen en ik hoop dat daar nog een aantal van volgen! Ik bewonder hoe je je gezinsleven met zon uitdagende baan combineert! Lieve Sabine, Sabje! Dank je wel voor alle experimenten en al het isoleren en kweken dat we samen hebben gedaan! Het was een waar genoegen om je te zien groeien in de groep tot de onmisbare kracht die je bent! Uiteraard wil ik graag de studenten bedanken, waarmee ik heb samengewerkt. Henrike: Thank you for your persistence, your efciency and your patience. It must have been hard to work with a chaotic person like me, but the result is a very nice paper in Hematologica! Nicole, ik vond het ontzettend leuk om een deel van je begeleiding te kunnen doen! Minke, Agent Rab! Ja, dat was wel fantastisch. MSCs kweken, de FACS, AMC, en heel veel lachen: een perfecte combinatie! Gelukkig werken we nu iedere dag samen in het Antonius! Op de afdeling Immunologie in het WKZ was het goed toeven! Dat kwam
171
door de gedreven, maar ontspannen sfeer en de vanzelfsprekendheid en het plezier waarmee er samengewerkt werd. Ik denk dat dat erg bijzonder is en wil iedereen daar erg graag voor bedanken! De staf van de afdeling Immunologie, zoals ik hem heb gekend: Dr. K. Tesselaar, Prof. dr. L. Meijaard, Dr. J. Leusen, Dr. D. van Baarle, Dr. J. Borghans, Dr. A. Martens, Dr. K. Denzer, Dr. L. Bont, Prof. dr. P. Coffer, Prof. dr. F. Miedema. Zeer veel dank voor de waardevolle bijdragen tijdens discussies en presentaties. Frank, ik wil jou speciaal nog bedanken voor je humorvolle en tegelijkertijd uiterst scherpe benadering van gepresenteerde data. Het was genieten met jou bij een bespreking! Petra, dank je wel voor het dynamic duo dat we waren in mijn eerste jaar. Ik vind je heel bijzonder en kundig. Yvonne en Saskia, zeer veel dank voor jullie oneindige exibiliteit en hulpvaardigheid! AIO kamer 3: Rogier, Lydia, Ellen, Floor, Kees, Kristof (Zotteke), Peter (de Hongaarse muur), Jantine, Paul, Wouter, Cordula, Annelieke, Bart, Thijs, Hilde, Vica, Sabina, Charlotte, Kirsten, Hakim. Het waren mooie tijden! Het gekkenuurtje om half 5, de onuitputtelijke bron grappen en grollen, het doorbreken van de bubbel, de dikke negermuziek, het tafeltennis, de vissen EBV en GVHD, het klaterende gelach tussen de serieuze inspanningen door, maakten het de leukste AIO kamer van alle AIO kamers ter wereld! Dank daarvoor. Verder veel dank aan Jeffrey, Dan (de Man), Jorg, Veerle, Marieke, Henk Jan, Richard, Henk, Vera, Liset, Jolanda, Ingrid, Ana, Justin, Guru, Marloes, Tessa, Robbert, Robert, Sten, Florijn, Frans, Joke, Walter, Nening, Ronald, Sanne, Gerrit, Koos, Sigrid, Maaike, Marco, Marc en Wilco voor alle gezelligheid, de jne samenwerking en de geweldige borrels!! Mijn paranimfen: Samantha, Sammie!! Jouw ongelooijke efcintie en snelheid in het proeven doen, maakten dat er veel tijd was om kofe te drinken en eindeloos te kletsen. Ik heb het zo leuk met je gehad en het was zon luxe en voorrecht om met jou te mogen werken. Dank je wel. Heel jn dat je aan mijn zijde wil staan!
172
DANKWOORD
Floor, lieve Flora! Vanaf de eerste dag dat ik kwam werken in het WKZ was je daar. Liefde op het eerste gezicht! Een onafscheidelijk duo was het gevolg en daarmee heel veel hilariteit. Dansen op het lab, samen FACSen, SPSSen, datamassage, samen uit eten met sammelebuca en birellas en ook serieuzere gesprekken, die geleid hebben tot een heel aantal publicaties (Pietersma, van Dorp en van Dorp, Pietersma et al.), waar alleen Blood nog niet zo van gecharmeerd was. Dank je wel dat je het werken op een lab zo sprankelend hebt gemaakt! Gelukkig hebben ook buiten het lab mensen direct en indirect bijgedragen aan dit boekje of aan mijn levensgeluk, zodat dit boekje makkelijker kon ontstaan. Lieve Kees, jij hebt wel een heel uitzonderlijke positie ingenomen in het ontstaan van dit proefschrift. Je maakte het al mooie leven in en buiten het lab nog mooier! Dank je wel voor alles wat je me gegeven hebt en hebt laten zien. Ook Cees (II), Nathalie, Marieke en Maarten heel erg bedankt voor de heerlijke etentjes, de discussies, het altijd warme welkom en het varen! Lieve Amsterdammers, liefste Wel, An, Ras, Soof, Maart, Daan, Shaun en Wes! Jullie zijn bevrijdend! Heel erg bedankt voor jullie. Wellie, jij begrijpt mij als geen ander, ik wil je speciaal bedanken, omdat je er onvoorwaardelijk voor me bent. Maart! Ik wil jou bedanken voor het feit, dat je altijd achter me gestaan hebt. Je kent me door en door en dat maakt dat je heel belangrijk voor me bent. Penny, dank je wel dat je me altijd een warme plek hebt gegeven, ook al heb ik er de laatste jaren wat weinig gebruik van gemaakt. Daan, door jou wilde ik professor worden (Grieks weliswaar). Dit is het begin! Lieve Niels en Jeroen! Dank voor jullie geweldigheid! Ik ben blij dat heel Nederland dat tegenwoordig inziet. Tillemakinderen en juf! Heel erg bedankt voor de vertrouwde etentjes. Antonianen, bedankt voor de bijzonder goede sfeer waarin we samenwerken! Ik besef me heel goed, dat zoveel leuke collegas eigenlijk niet kan! Dr. Geers, dank voor de gelegenheid om dingen af te maken en de jne, goed passende supervisie! Dr. de Weerdt, lieve Okke! Door jou wil ik hematoloog worden. 173
Mark, dank je wel voor je absolute optimisme, je barricades voor bijvoorbeeld de tonijn, je danskwaliteiten en Ticket to Ride! Azzie, lieverd! Ment to be vind ik ons. Ik ben echt heel erg blij dat jij me op stond te wachten op de eerste dag! Joost , mijn grote vriend! Ik kan werkelijk alles met je bespreken. Ik ben blij met je keuze voor het mooiste aandachtsgebied dat er is! Lieve Girlsch, Suzy Q, Hes en Lies. Ik ben heel blij met jullie en dat we samen groot geworden zijn! Dank jullie wel voor altijd!! Daan en Vic, jullie zijn precies goed voor de meisjes, niet veranderen! Lieve Miekie! Dank je wel voor je trouwe vriendschap, je attentheid, je organisatorisch talent en alle leuke dingen die we samen gedaan hebben (niet te vergeten jouw promotie!!). Erwin, jij ook dank je wel voor al die leuke avonden, etentjes, feestjes en voor je interesse, jnheid en Mario Cart! Lieve Daph! Samen studeren, samen wonen, samen hardlopen, samen een Duitse baas hebben. Dank je wel dat we dat allemaal konden delen! Lieve Willem. Dank je wel voor je beste vriendschap! Ik weet zeker dat mijn kinderen in de zandbak toch van jouw duifjes van kinderen gaan winnen! En ook weet ik zeker dat we er samen naar zullen gaan kijken! Lieve Debster, Inator, mijn liefste huisgenoot! Ik ben zelden iemand tegengekomen bij wie het zo gemakkelijk was allemaal en bij wie ik zoveel overeenkomsten vond. Duizend maal dank!! Ik heb Paolo en John nogmaals benaderd voor de woongroep. Ook dank voor het delen van Dit en Draak op de Koningslaan. Dit, ik ben heel blij dat je er bent!! Lieve Marie Louise, Hans, Kasper, Marlou en Jibbe. Jullie zijn geweldig! Dank voor alle vrolijkheid, openheid en liefdevolheid, waarmee jullie me meteen omringd hebben. Lieve mamma en Harry, dank jullie wel voor alle liefde, gezelligheid, heerlijke etentjes, adviezen, ondersteuning en het geweldige voorbeeld dat jullie zijn. Ik hoop ook zo van het leven te gaan genieten als jullie dat doen en ik ben heel blij dat jullie dat samen doen. Lieve Omi! Jij ook bedankt voor de weekenden ontspannen en onze telefoongesprekken, waar van alles de revue passeert. Een hele dikke iiiiiii! 174
DANKWOORD
Chris, mijn lieve broertje. Ik ben van alle mensen op de wereld het meest trots op jou. Je hebt een heleboel voor je kiezen gehad en je bent er zo ontzettend leuk en knap uitgekomen. Speciale dank voor het verzorgen van de lay-out en de kaft van dit boekje! Het is echt prachtig geworden! Arjan, lieve AJ! Ik heb hl erg veel zin in hl erg veel meer van jou. Zoveel inspiratie! Dat moet een mooi feestje worden!
175

Shaping Graft Immunity

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Shaping Graft Immunity

Încărcat de

Drepturi de autor:

Formate disponibile

Shaping graft immunity:

Suzanne van Dorp

Shaping graft immunity:

Suzanne van Dorp

Prof. dr. H.M. Lokhorst Dr. J.H.E. Kuball Dr. E. Meijer

Voor Henk van Dorp v

Therapeutic potential of an allogeneic stem cell transplantation

II Graft-versus-Host Disease and mechanisms of disease

8 Nederlandse samenvatting Curriculum Vit List of publications Dankwoord

155 165 167 169

Introduction and scope of the thesis

1.2 Graft-versus-Host Disease as major challenge after allo-SCT

CHAPTER 1. INTRODUCTION AND SCOPE OF THE THESIS

CHAPTER 1. INTRODUCTION AND SCOPE OF THE THESIS

CHAPTER 1. INTRODUCTION AND SCOPE OF THE THESIS

Therapeutic potential of an allogeneic stem cell transplantation

Netherlands Journal of Medicine, 2007

of Haematology, University Medical Centre Utrecht, Utrecht, the Netherlands

CHAPTER 2. NONMYELOABLATIVE ASCT IN MULTIPLE MYELOMA

Patients and methods

CHAPTER 2. NONMYELOABLATIVE ASCT IN MULTIPLE MYELOMA

had an elevated 2-microglobulin ( 3.0 mg/l).

< 3 mg/L > 3 mg/L

CHAPTER 2. NONMYELOABLATIVE ASCT IN MULTIPLE MYELOMA

First-line treatment n (%)

Relapse treatment n (%)

9 (25) 24 (66.7) 3 (8.3)

0 (0) 16 (69.6) 7 (30.4)

CHAPTER 2. NONMYELOABLATIVE ASCT IN MULTIPLE MYELOMA

CHAPTER 2. NONMYELOABLATIVE ASCT IN MULTIPLE MYELOMA

Figure 2.6: Nonrelapse mortality.

Graft-versus-Host Disease and mechanisms of disease

Biology of Blood and Marrow Transplantation, 2009

1 Department 2 Department 3 Childrens 4 Julius

Hospital, University of Wrzburg, Germany

CHAPTER 3. DUAL ROLE FOR HOST B CELLS IN RICT

Design and methods

No Rtx 144 (83) 56 (20 69) 28.8 (1 40) 35/65 6 29 8 5 6 62 14 6 2 6

0.662 0.011 0.417

0.925 92 (64) 9 (6) 31 (22) Continued on next page

CHAPTER 3. DUAL ROLE FOR HOST B CELLS IN RICT

CHAPTER 3. DUAL ROLE FOR HOST B CELLS IN RICT

CHAPTER 3. DUAL ROLE FOR HOST B CELLS IN RICT

CHAPTER 3. DUAL ROLE FOR HOST B CELLS IN RICT

CHAPTER 3. DUAL ROLE FOR HOST B CELLS IN RICT

CHAPTER 3. DUAL ROLE FOR HOST B CELLS IN RICT

The immunological phenotype of rituximab-sensitive chronic graft-versus-host disease: a phase II study

1 Dept 2 Dept 3 Dept 4 Dept

CHAPTER 4. RITUXIMAB-SENSITIVE CHRONIC GVHD

Design and methods

4.3 Results and Discussion

Response duration Months (range)

Skin (n = 18) 6 (3 12)

Deep sclerosis (n = 12)

CHAPTER 4. RITUXIMAB-SENSITIVE CHRONIC GVHD

CHAPTER 4. RITUXIMAB-SENSITIVE CHRONIC GVHD

HR 0.985 0.796 0.925 1.585 10.683 3.615 0.033

p-value 0.689 0.728 0.920 0.498 0.031 0.234 0.281

CHAPTER 4. RITUXIMAB-SENSITIVE CHRONIC GVHD

4.4 Supplementary Design and Methods

CHAPTER 4. RITUXIMAB-SENSITIVE CHRONIC GVHD

AML CML NHL

Table 4.4: Characteristics of healthy controls used for ow-cytometry analyses.

Median age (years; range)