Documente Academic
Documente Profesional
Documente Cultură
Marrow relapse is the major obstacle to cure for 1015% of young patients with acute lymphoblastic leukaemia (ALL).
Recent investigations into the biology of minimal residual disease indicate that many early relapses derive from residual
cells present at rst diagnosis, but some late relapses might represent new mutations in leukaemic cells not eliminated
by conventional therapy. Treatment of marrow relapse involves higher doses and more intensive schedules of the drugs
used for initial therapy with or without haemopoietic stem cell transplantation. In most reports, transplantation is better
than continuation chemotherapy in early marrow relapse, but its role in later relapse is less clear. Current therapy cures
10% of patients with early marrow relapses and 50% of those with late relapses, but outcomes have changed little in the
past two decades. Understanding the molecular biology of ALL underlies development of improved risk stratication and
new therapies. Although better drugs are needed, introduction of new agents into clinical trials in paediatric disease has
been dicult. Innovative trial designs and use of valid surrogate endpoints may expedite this process.
Introduction
50 years of large randomised clinical trials of combination
chemotherapy and CNS prophylaxis have improved the
cure rate for acute lymphoblastic leukaemia (ALL) in
childhood from less than 5% in 1950 to 85% in 2000.1
Progress in this disease has come principally from prevention of relapses (gure 1) through more eective use of
established drugs and combination regimens in riskadapted treatment strategies, rather than introduction of
new drugs. Today few recurrences occur solely in the CNS
or other extramedullary sites,27 and most are curable,
especially those that happen well after initial treatment.811
Marrow relapse is the major impediment to cure (gure 2);
assuming a recurrence rate of 1015%, roughly
1000 patients each year relapse in Europe and North
America.12 Treatment of marrow relapse has not been as
successful as primary treatment despite intensied chemotherapy and use of haemopoietic stem cell transplantation,
and better therapeutic options are needed.
We describe recent advances in the understanding of
marrow relapse and past, present, and future approaches
to treatment. Molecular techniques provide insight into
the pathogenesis of relapse and support a targeted
approach to drug discovery that has produced new classes
of drugs with promising preclinical activity in lymphoid
malignancies.13 The logistics of introducing them into
paediatric clinical trials and proving their ecacy in
marrow relapse are daunting. Innovative trial designs and
use of surrogate endpoints could facilitate this process.1416
By use of the 3-[4,5-dimethyl-thiazole-2,5-diphenyl] tetrazolium bromide (MTT) cytotoxicity assay, both casecontrol
and matched-set comparisons of initial and relapse
samples showed that relapse samples are more resistant to
glucocorticoids, L-asparaginase, thiopurines, and anthracyclines in vitrothese drugs are central components of
initial treatment. However, the samples retain sensitivity
to etoposide and cyclophosphamide, which are sparingly
used in rst-line protocols.18 This nding suggests that
patients might benet after relapse from dierent drugs or
dosing strategies but does not distinguish between
acquired and inherent drug resistance in leukaemic
blasts.
Early studies of matched initial and relapse specimens
also found shifts in morphology and in chromosome
number in a third and changes in lineage-related surface
antigens in a half of patients;19 conversion to a new lineage
with a distinct karyotype was rare and usually indicated a
new, treatment-related leukaemia.19 Developments in
multichannel ow cytometry allow phenotypic detection of
fewer than one leukaemic blast among 1000 normal cells,
making it possible to quantify minimal residual disease
(table 1).20 Comparison of matched initial and relapse
specimens from patients with B-precursor ALL showed
A
Origins of relapse
Much of our knowledge about the pathogenesis of relapse
comes from casecontrol and matched-set studies of leukaemic cells at initial diagnosis, at relapse, and during
clinical remission. Casecontrol studies access large numbers of banked specimens, but relapse and control specimens commonly come from dierent individuals with
dierent types of ALL. Matched sets of samples from one
individual control for these variables, but are dicult to
assemble.17
www.thelancet.com/oncology Vol 9 September 2008
873
Review
Extramedullary relapse
Other
Marrow relapse
100
75
50
25
0
BFM
198195
CCG
198395
DFCI
198195
Tokyo CCSG
198195
UKMRC
198097
Figure 2: Trends in outcomes for primary treatment of paediatric acute lymphoblastic leukaemia
Results from large cooperative group trials show that marrow relapse remains the largest barrier to cure despite
improvements in overall outcome. Non-relapse outcomes not reported by US Childrens Cancer Group (CCG) or
Tokyo Childrens Cancer Study Group (Tokyo CCSG)marrow relapse estimated as event-free survival minus CNS
extramedullary relapse in CCG and minus all extramedullary relapse in Tokyo CCSG. BFM=Berlin-Frankfurt-Mnster
group. DFCI=Dana-Farber Cancer Institute. UKMRC=UK Medical Research Council (Childhood Leukaemia Working
Party). Each column represents outcomes for a specic trial or set of closely related trials.
Sensitivity
Limitation
Morphology
15%
Multiparameter ow cytometry
00101%
RT-PCR
000010001%
Review
Mutation
Initiating
Transforming
Transforming
Treatment resistance
Normal lymphoid
progenitor cell
Pre-leukaemic
clones
Resistance emerges:
MRD persists
Acute leukaemia
Therapy
Selection
No resisitance:
leukaemia cured
Acute leukaemia
Mutation
Early relapse
Therapy
Late relapse
Second
mutation
Review
Early
Postremission therapy
Continuation chemotherapy
All patients who reach a second remission receive
additional chemotherapy, even if haemopoietic stem cell
transplantation is planned. To maintain control of
disease, higher dose intensity is used, and higher
regimen-related toxicity is tolerated than in rst-line
treatment. Most reports describe single-arm studies with
combinations of vincristine, glucocorticoids, methotrexate, cytarabine, etoposide, cyclophosphamide or
ifosfamide, and thiopurines, with or without maintenance
therapy for up to 2 years.36,39,41,43,44,52,54,55 CNS prophylaxis
includes high-dose methotrexate or cytarabine, intrathecal
Complete remission
Event-free survival
Overall survival
Early
Late
Earliest
Earliest
Early
Late
Early
References
Late
19722001
350
72%
76%
96%
6%
32%
71%
198194
216
72%
8%
19%
50%
37
8
198386
297
100%
83%
9%
12%
38
198389
642
66%
5%
10%
33%
11%
43%
198494
106
40%
66%
81%
13%
43%
39
198496
117
72%
16%
24%
35%
40
198790
183
48%
73%
96%
18%
44%
20%
52%
41
198990
489
47%
1%
14%
33%
42
199095
269
42%
91%
..
21%
48%
29%
26%
43
199095
121
35%
5%
35%
51%
44
Earliest relapses are less than 1824 months from rst remission, early relapses include these, as well as late relapses under 6 months from therapy discontinuation, and late relapses occurred after this time.
*Includes all relapses. Includes combined relapses. Excludes T-cell ALL relapses, and includes all combined relapses, intermediate-risk marrow relapses, and high-risk extramedullary relapses.
Table 2: Studies of outcomes after marrow relapse according to duration of rst remission
Marrow
Combined
Extramedullary
Very early
Early
Late
Very early
Early
Late
Early
Late
<18 months
18 months and
<6 months o
treatment
6 months o
treatment
<18 months
18 months and
<6 months o
treatment
6 months o
treatment
On treatment or
<6 months o
treatment
6 months o
treatment
B-precursor risk
High
High
Intermediate
High
Intermediate
Intermediate
Intermediate
Standard
B-precursor therapy
Allogeneic
HSCT
Allogeneic HSCT
MFD HSCT
Allogeneic
HSCT
MFD HSCT,
radiotherapy
Chemotherapy,
radiotherapy
Chemotherapy,
radiotherapy
Chemotherapy,
radiotherapy
T-cell risk
High
High
High
High
High
High
Intermediate
Standard
T-cell therapy
Allogeneic
HSCT
Allogeneic HSCT
Allogeneic
HSCT
Allogeneic
HSCT
Allogeneic HSCT,
radiotherapy
Allgeneic HSCT,
radiotherapy
Chemotherapy,
radiotherapy
Chemotherapy,
radiotherapy
<36 months
36 months
<36 months
36 months
<18 months
18 months
High
Intermediate
High
Intermediate
Intermediate
Low
Allogeneic HSCT
MFD HSCT
Allogeneic HSCT,
radiotherapy
MFD HSCT,
radiotherapy
MFD HSCT,
radiotherapy
Chemotherapy,
radiotherapy
Berlin-Frankfurt-Mnster
Allogeneic haemopoietic stem cell transplantation (HSCT) is transplantation of cells from any of matched familial donors (MFD), unrelated donors, or cord blood. *The Childrens Oncology Group had no very
early group. Time since diagnosis used in current studies as proxy for duration of rst remission.
Table 3: Risk stratication and treatment strategy for relapsed acute lymphoblastic leukaemia
876
Review
chemotherapy, and in more recent Berlin-FrankfurtMnster group trials, 12001800 cGy cranial irradiation.43
The Paediatric Oncology Group study 8303 reported that
the addition of four-drug reinduction pulses did not
improve upon weekly rotation of cytarabine and
teniposide with vincristine and cyclophosphamide.38
Berlin-Frankfurt-Mnster group relapse trials randomised
patients with rst remission greater than 18 months to
high-dose or long-duration infusions of intermediatedose methotrexate with blocks of multidrug chemotherapy,41,43 with equivalent outcomes. Aside from these
randomised trials, published data do not show one
chemotherapy combination to be better than others.
In children with early marrow relapse who achieve remission, leukaemia-free survival is only 1020% with
chemotherapy alone. Outcomes are better for late relapse,
with several studies predicting leukaemia-free survival of
greater than 50%.36,39,44,54 However, some of these reports
overestimate leukaemia-free survival for isolated marrow
relapse by including patients with combined or isolated
extramedullary relapse or excluding patients with T-cell
ALL.
Selection of patients
Treatment outcome
Biological and
physiological
factors
Initial treating
institution
Reinduction protocol
Referral policy
Transplanting
institution
Acceptance policy
Conditioning protocol
GVHD prophylaxis and treatment
Experience with HSCT
Socioeconomic
factors
Table 4: Potential confounding factors aecting design and interpretation of trials of haemopoietic stem
cell transplantation (HSCT)
Review
Population
Donor
Leukaemia-free survival
Follow-up
Comments
References
NMDP
Unrelated
363
36%
5 years
62
Berlin-Frankfurt-Mnster group
Matched pairs
Unrelated
81
pairs
5 years
35
Unrelated
Matched sibling
28
39
54%
39%
5 years
61
Northwestern University
ALL second
remission
Unrelated
Unrelated cord
12
16
60%
61%
63
Unrelated cord
Matched
1 antigen mismatch
2 antigen mismatches
Marrow
Matched
Mismatched
64
30
201
267
60%
3645%*
33%
116
166
38%
37%
NMDP=United States National Marrow Donor Program. *Lowest in those with low cell count, highest in those with high cell count.
Table 5: Retrospective studies comparing unrelated donor or cord-blood transplant to matched familial donor or chemotherapy for acute lymphoblastic leukaemia in second or higher remission
Review
Design
N
HSCT
Chemotherapy
Very early
Early
HSCT
HSCT
Chemotherapy
26%
Late
Chemotherapy
7%*
HSCT
References
Chemotherapy
POG-8303
Matched pair
42
192
38
IBMTR/POG
Matched pair
255
255
30%
14%
41%
7%
IBMTR/POG
Matched pair
188
40%
23%
53%
32%
46
61%
59%*
BFM REZ 87
Randomised
25
145
59%
30%
69
41
BFM REZ
Matched pair
81
81
44%
0%
49%
39%*
35
CCG-1941
Randomised
ITT/PP
29%/42%
27%/30%
58
NOPHO
Casecontrol
75
150
32%
11%
42%
29%
47
AIEOP/GITMO
Casecontrol
57
230
33%
16%
55%
40%*
74
Leiden
Casecontrol
75
150
44%
24%
75
MSKCC
Casecontrol
38
37
48%
9%
81%
37%
76
Westmead
Casecontrol
20
34
54%
10%
57%
0%*
77
Leukaemia-free survival adjusted for time to transplant; results are actuarial calculations except for two calculated from text and tables.38,69 Earliest relapses are less than
1824 months from rst remission, early relapses include these, as well as late relapses under 6 months from therapy discontinuation, and late relapses occurred after this
time. HSCT=haemopoietic stem cell transplantation. POG=Paediatric Oncology Group. IBMTR=International Bone Marrow Transplant Registry. BFM=Berlin-FrankfurtMnster group. CCG=Childrens Cancer Group. ITT=intention to treat. PP=per protocol. NOPHO=Nordic Paediatric Hematology-Oncology group. AIEOP=Associazione Italiana
Ematologia ed Oncologia Pediatrica. GITMO=Gruppo Italiano Trapianti Midollo Osseo. MSKCC=Memorial Sloan-Kettering Cancer Center. *p>005. p005. p 0005.
Table 6: Studies of leukaemia-free survival at 3 years or longer after allogeneic bone-marrow transplantation or chemotherapy for acute lymphoblastic
leukaemia in second remission
matched familial donor haemopoietic stem-cell transplantation as denitive therapy where possible.
Sequential population-based retrospective studies in
Scandinavian children treated for ALL relapse show that
transplantation of haemopoietic stem cells led to increased
long-term survival compared with chemotherapy irrespective of timing of rst remission.47,48 Over the course of
rst, second, and subsequent relapses, treatment-related
mortality was higher with chemotherapy than with
transplantation. Although these results suggest that transplantation should be used in every marrow relapse, because
of non-lethal morbidities6668 selection of patients who will
not relapse after chemotherapy alone is important.48
In the other studies, outcomes are better with transplantation of haemopoietic stem cells in patients with rst
remission lasting less than 18 months, but the advantage is
less in patients with later relapse (table 6).7477 In no
comparison is outcome after transplantation worse than
after chemotherapy alone.
Further relapse is the greatest barrier to cure after rst
relapse. Best possible allogeneic transplant oers a better
chance of cure than chemotherapy for patients who have
not had haemopoietic stem cell transplantation in the
recent past.36,48 Although some studies report occasional
long-term third remissions in patients at intermediate
and standard risk after second relapse,37,43 cure is unlikely
unless the patients have favourable features at initial
diagnosis and rst relapse.36,48
Despite the limitations of studies in marrow relapse, the
results provide reasonable support for broad principles to
guide current treatment and future clinical trials in rst
relapse. Fundamentally, both current chemotherapy and
transplantation of haemopoietic stem cells are toxic,
www.thelancet.com/oncology Vol 9 September 2008
Review
Mechanism of action
Disease subset
Rituximab
B-lineage ALL
Epratuzumab, BL22
B-lineage ALL
Alemtuzumab
Combotox
B-lineage ALL
Daclizumab
Monoclonal antibodies
Ph+ ALL
Dasatinib
Ph+ ALL
Bortezomib
Sirolimus, temsirolimus
Inhibition of mTOR
CEP-701
Inhibition of FLT3
Infant MLL
MK0572, Ly450139
T-lineage ALL
Nucleoside analogues
Clofarabine
Deoxyadenosine analogue
Nelarabine
Deoxyguanosine analogue
T-lineage ALL
Forodesine
PNP inhibitor
T-lineage ALL
Intrathecal liposomal
cytarabine
CNS ALL
Decitabine
DNA demethylation
FLT3=FMS-like tyrosine kinase. cKIT=c-kit stem cell factor (CD117). MLL=mixed lineage leukaemia. mTOR: mammalian
target of rapamycin. NFB=nuclear factor B. PDGFR=platelet-derived growth factor receptor. Ph+=Philadelphia
chromosome. PNP=purine nucleoside phosphorylase. SAHA=suberoylanilide hydroxamic acid. SRC=Src protein
tyrosine kinase.
Table 7: New drugs in development for trials in paediatric acute lymphoblastic leukaemia (ALL)
Review
Conclusion
By contrast with the orderly succession of large randomised
trials and continuous progress in ALL at initial diagnosis,
most studies in marrow relapse are single-arm studies or
retrospective reviews that have not provided substantial
improvements. Polarisation around the issue of haemopoietic stem cell transplantation has thwarted potential
collaborations that might enrol enough patients for randomised trials. Such trials in relapse are dicult because
of the heterogeneity of the population. Nonetheless, randomised trials are needed to produce credible results that
change practice. Even if randomised comparisons of
chemotherapy with stem cell transplantation are not
possible, there are many investigations that are possible,
such as trials to dene a standard chemotherapy backbone
as the foundation for subsequent transplantation or continuation chemotherapy. Randomised phase II trials in
relapse also provide a platform for assessing how a new
drug does when added to a standard therapy or how two
new treatments compare with one another. Similarly,
randomised trials in patients having haemopoietic stem
cell transplantation could help to address some of the
heterogeneity in selection of patients and conditioning
regimens, which are matters of strong local preference.
Because the clone destined for relapse is often present at
diagnosis and is inherently more resistant to induction
chemotherapy,2427 prevention of marrow relapse remains
www.thelancet.com/oncology Vol 9 September 2008
881
Review
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
882
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Review
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
883