Sunteți pe pagina 1din 7

YGYNO-976498; No.

of pages: 7; 4C:
Gynecologic Oncology xxx (2016) xxxxxx

Contents lists available at ScienceDirect

Gynecologic Oncology
journal homepage:

Review Article

Placental site trophoblastic tumors and epithelioid trophoblastic tumors:

Biology, natural history, and treatment modalities
Neil S. Horowitz, MD, Donald P. Goldstein, MD, Ross S. Berkowitz, MD
New England Trophoblastic Disease Center, Division of Gynecologic Oncology, Department of Obstetrics, Gynecology, and Reproductive Biology, Harvard Medical School, Brigham and Women's
Hospital, Dana Farber Cancer Institute, Boston, MA 02115, United States


PSTT and ETT are rare types of GTN that arise from intermediate trophoblast.
Time from antecedent pregnancy N 4 years and advanced stage are poor prognostic factors.
Hysterectomy is the primary treatment for stage I disease.
Stage I disease with high-risk features and all advanced stage disease require chemotherapy.
EMA-EP and TP/TE are the chemotherapy regimens recommended for treatment of PSTT and ETT.

a r t i c l e

i n f o

Article history:
Received 13 September 2016
Received in revised form 17 October 2016
Accepted 17 October 2016
Available online xxxx
Gestational trophoblastic neoplasia
Placental site trophoblastic tumor
Epithelioid trophoblastic tumor

a b s t r a c t
Placental site (PSTT) and epithelioid trophoblastic tumor (ETT) are rare types of gestational trophoblastic neoplasia (GTN) that arise from intermediate trophoblast. Given that this cell of origin is different from other forms of
GTN, it is not surprising that the clinical presentation, tumor marker prole, and treatment paradigm for PSTT and
ETT are quite different as well. The mainstay for therapy for stage I PSTT and ETT is hysterectomy with adjuvant
chemotherapy reserved for those presenting greater than four years from the antecedent pregnancy. Surgery is
also important for metastatic disease. There is no standardized chemotherapy regimen for advanced stage disease but often consists of a platinum-containing combination therapy, usually EMA-EP or TE/TP. Despite its rarity,
PSTT and ETT account for a disproportionate percentage of mortality from GTN likely resulting from their relative
chemotherapy resistance. Novel therapeutic modalities therefore are needed to improve the outcomes of women
with advanced stage or resistant PSTT and ETT.
2016 Elsevier Inc. All rights reserved.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Historical perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pathology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Presentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Diagnostic evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Risk factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Treatment of localized disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Treatment of metastatic disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Survival . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
International Society for the Study of Trophoblastic Disease (ISSTD) placental site and epithelioid trophoblastic tumor database
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Conict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .











Corresponding author at: Brigham and Women's Hospital, Division of Gynecologic Oncology, 75 Francis St., Boston, MA 02115, United States.
E-mail address: (R.S. Berkowitz).
0090-8258/ 2016 Elsevier Inc. All rights reserved.

Please cite this article as: N.S. Horowitz, et al., Placental site trophoblastic tumors and epithelioid trophoblastic tumors: Biology, natural history,
and treatment modalities, Gynecol Oncol (2016),


N.S. Horowitz et al. / Gynecologic Oncology xxx (2016) xxxxxx

1. Introduction
Gestational trophoblastic neoplasia (GTN) is group of malignant lesions that arise from placental villous and extravillous trophoblasts.
GTN can arise after both molar and non-molar pregnancy events and
are comprised of 4 distinct histologic subtypes; invasive mole, choriocarcinoma, placental site trophoblastic tumor (PSTT) and epithelioid trophoblastic tumor (ETT). While invasive mole and choriocarcinoma
comprise the majority of GTN cases, PSTT and ETT are rare but important
forms of GTN with unique pathology, natural history, and treatment paradigms [14]. Given the rarity of PSTT and ETT, our knowledge about
these histologies has been based generally on case reports and small, single institution case series. However, there have been two, relatively large
series from the United Kingdom and China which have further advanced
our understanding [5,6] (Table 1). Ultimately the limitations of small
sample size will be overcome when data from the International Society
for the Study of Trophoblastic Disease Placental Site and Epithelioid Trophoblastic Tumor Database ( mature. Until then,
the goal of this review will be to summarize the existing data regarding
the epidemiology, pathology, presentation, evaluation, and treatment
of PSTT and ETT and provide a framework for future investigations into
these diseases.
2. Historical perspective
Both PSTT and ETT are relatively newly recognized disease entities.
In 1976 Kurman, Scully, and Norris presented a series of 12 cases

describing a lesion, previously not well characterized, as trophoblastic pseudotumor of the uterus [7]. Previous reports of similar lesions described them as an unusual type of sarcoma associated with
pregnancy. In that initial series all but 1 patient was alive and well so
it was thought to be a benign neoplasm. After a report by Twiggs et
al. [8] revealed its malignant behavior, it was renamed placental
site trophoblastic tumor by Scully and Young who conrmed its malignant potential in an updated review [9]. Approximately 20 years
later, in 1998, ETT was described by Shih and Kurman [10]. Although
having the same cell of origin, intermediate trophoblast, ETT was distinct from PSTT and had morphologic features similar to squamous
cell carcinoma making the diagnosis more challenging [10]. Since
their initial descriptions, cumulatively the world's literature consists
of reports on b 500 cases of PSTT and approximately 110 cases of ETT.
3. Pathology
Trophoblastic stem cells develop along two lines of differentiation,
villous and extravillous. Molar pregnancies and choriocarcinoma
are derived from villous trophoblasts which are composed mostly
of cytotrophoblast (CT) and syncytiotrophoblast (ST). PSTT and ETT
are derived from extra villous trophoblast and are composed almost
exclusively of intermediate trophoblast [11]. PSTT arises from intermediate trophoblast and has a growth pattern of invasion similar to
that of normal intermediate trophoblast. In contrast, ETT develops from
chorionic-type intermediate trophoblast present in other parts of the
placenta (i.e. chorionic plate, fetal membranes, cell islands, etc.) [12,

Table 1
Summary of the largest published series on PSTT.
Authors (Reference)
Zhao et al. [6]


Reported risk factors

hCG (median)

Chemotherapy regimens




154 IU/L


Stage I
Stage III/IV


b1000 IU/L


Stage I
Stage III/IV


691 IU/L


Stage I
Stage III/IV


132 IU/L


Stage I
Stage III/IV


205 IU/L



Time from AP (N36 mo)

Schmid et al. [5]


Prognostic score
Deep invasion (N50%)
Time from AP (N48 mo)

Baergen et al. [22]


Hyman et al. [21]


Moutte et al. [30]


Mitotic index
FIGO score
Number of mets
Age (N35 yo)
Clear cytoplasm
Time from AP (N24 mo)
Deep invasion (N33%)
Mitotic rate
Term AP
FIGO Stage
Time from AP (N12 mo)
Term AP
Age (N40 yo)

Bonazzi et al. [55]



110 IU/L

Feltmate et al. [23]


Mitotic index
Tumor volume (N1 cm3)

b 500 IU/L


Stage I
Stage III


FAV oxuridine, actinomycin-D, vincristine; FAEV - oxuridine, actinomycin-D, etoposide, vincristine; EMA-CO etoposide, methotrexate, actinomycin-D, cyclophosphamide, vincristine; EMA-EP etoposide, methotrexate, actinomycin-D, cisplatin; BEP bleomycin, etoposide, cisplatin; ICE- ifosfamide, carboplatin, etoposide; MTX - methotrexate; MAC - methotrexate, actinomycin-D, cyclophosphamide; MAE - methotrexate, actinomycin-D, etoposide.
Pts = patients; hCG = human chorionic gonadotropin; AP = antecedent pregnancy; NR = not reported; mo = month; FIGO = International Federation of Gynecology and Obstetrics;
yo = years old.
Statistically signicant risk factor (p b 0.05) on multivariate analysis.

Please cite this article as: N.S. Horowitz, et al., Placental site trophoblastic tumors and epithelioid trophoblastic tumors: Biology, natural history,
and treatment modalities, Gynecol Oncol (2016),

N.S. Horowitz et al. / Gynecologic Oncology xxx (2016) xxxxxx

13]. Two other benign trophoblastic lesions may also arise from intermediate trophoblast - exaggerated placental site and placental
site nodule (PSN). PSN can be typical or atypical with up to 15% of
atypical PSN being associated with PSTT/ETT [4].
The intermediate trophoblasts that make up PSTT are varied in appearance ranging from decidual-like cells to spindle cells, to endothelium when lining vascular spaces. Morphologically PSTT intermediate
trophoblasts are predominately mononuclear with varying degrees of
nuclear atypia and cytoplasm that can be amorphic, eosinophillic, or
even clear (Fig. 1). These tumors are often associated with abundant brinoid material and have prominent vascular invasion but usually display minimal inammation and necrosis. These features are quite
different from choriocarcinoma which typically is dimorphic, has significant nuclear atypia, and marked hemorrhage and necrosis. PSTT has a
unique immunohistochemical phenotype. Mel-CAM (CD146) and
human placental lactogen (hPL) are strongly positive while placental alkaline phosphatase (PLAP) is only focally positive. The specicity of this
staining pattern for PSTT is approximately 60%. Additionally, marked
positive staining with Ki-67, alpha-inhibin, and cytokeratin 8/18 and
negative staining for smooth muscle markers help conrm the diagnosis
of PSTT [14,15]. PSTT is positive for hCG and p63 in a relatively small
proportion of cells and most are diploid though reports of tumor with
triploid have been published [16].
Although ETT arise from migratory intermediate trophoblast, it is
pathologically similar to PSTT. Grossly these are typically solid to cystic
and eshy, well dened lesions in the uterus, cervix or within the
peritoneal cavity. Histologically ETT is distinguished from PSTT by its
smaller, more monomorphic cells and by its nested, nodular, well
circumscribed growth pattern unlike the sheet-like inltrative pattern
seen with PSTT (Fig. 2). Additionally, there are often areas of necrosis
that surround islands of viable tumor cells that create a geographic
pattern of necrosis [17]. Given its frequent involvement of the lower
uterine segment and endocervix, its epithelioid histologic appearance,
and expression of p63 and cytokeratins, ETT can be confused with
squamous cell carcinoma [10,17]. Immunohistochemistry can help distinguish ETT from other GTN and epithelial carcinomas. ETT is positive
for pancytokeratin, epithelial membrane antigen, E-cadherin, and
EGFR (consistent with their epithelial origin) but all tumors are also
strongly positive for PLAP and p63 and only focally positive for hPL,
hCG, and Mel-CAM. As p63 is expressed in ETT but not PSTT and hPL
and Mel-CAM are only focally/weakly positive relative to the strong
positivity seen in PSTT, these markers (p63, hPL, and Mel-CAM) can
help distinguish the two entities [10,18].

Typically PSTT and ETT follow normal term pregnancies but can occur
after any pregnancy event including molar pregnancy and typically
present from months to many years after the antecedent pregnancy
[5,6,2124]. PSTT and ETT both exhibit slow growth rates and remain
conned to the uterus (Stage I, nonmetastatic) for extended periods of
time. As a result, there is often a paucity of symptoms other than vaginal
bleeding or amenorrhea [5,6,2124]. Other symptoms are contingent
upon sites of metastases which ultimately occur in 3050% of women
with PSTT and ETT who present with metastatic disease [25,26]. Some
women present with unique clinical situations like nephrotic syndrome
or virilization, but this is quite unusual [4,6].
Similar to other types of GTN, common sites of metastasis include
lung, vagina, and central nervous system with the lung being the most
common site, however, there is thought to be a higher propensity for
retroperitoneal node involvement [4]. Reviewing the literature and including their own data, Lan and colleagues identied 17 of 286 (5.9%)
patients with PSTT who had lymph node metastases at initial diagnosis
or at recurrence. Site of nodal metastases was known in 14 of the 17 patients; 5 (35%) were noted to have pelvic node involvement, 7 (50%)
had positive para-aortic nodes, and 2 (14%) had retroperitoneal nodes
not otherwise specied. Of the 7 women with para-aortic nodal metastases, 5 had isolated para-aortic lymph node involvement with no involved pelvic nodes suggesting direct extension to this nodal basin. In
addition the authors identied 23 patients who underwent retroperitoneal lymphatic procedures for various reasons including enlarged
nodes. In this selected population, the rate of nodal spread was 39% (9
of 23). In the 3 patients with presumed stage I disease who had nodal
metastases, all had deep myometrial invasion (N50%), suggesting this
maybe an important risk factor [25].
Because PSTT and ETT originate from intermediate trophoblast rather than syncytiotrophoblast (ST), there is a misconception that PSTT and
ETT do not secrete hCG. However, the vast majority secretes hCG (90%)
but relative to other types of GTN the levels are only mildly elevated. Despite the presence of signicant tumor volume, hCG levels were
b1000 IU/L in most series but have been reported to be as high as N
100,000 IU/L in a small percentage of patients [5,6]. Patients with high
hCG levels generally have an admixture of PSTT or ETT with choriocarcinoma tissue. These typically low levels of hCG make using hCG as a
tumor marker less reliable. While PSTT generally strongly stains for
hPL, serum hPL measurement has not been proven to be useful in monitoring the course of the disease or guiding clinical management [4].

4. Presentation

As in the case with all types of GTN, the diagnosis of PSTT or ETT has
to be considered in the differential for any reproductive age woman presenting with abnormal vaginal bleeding or amenorrhea and an elevated

PSTT and ETT are rare subtypes of GTN with an incidence of 1 in

100,000 pregnancies and representing 12% of GTN cases [19,20].

5. Diagnostic evaluation

Fig. 1. Hematoxylin and eosin (H & E) staining and Mel-CAM and Ki-67 immunohistochemistry in a representative example of PSTT. Particular note is made of the brinoid material and
sheet-like invasive growth pattern. (Photo courtesy of Bradley J. Quade, Associate Pathologist Department of Pathology, Division of Women's and Perinatal Pathology, Brigham and
Women's Hospital and Associate Professor of Pathology, Harvard Medical School, Boston, MA).

Please cite this article as: N.S. Horowitz, et al., Placental site trophoblastic tumors and epithelioid trophoblastic tumors: Biology, natural history,
and treatment modalities, Gynecol Oncol (2016),

N.S. Horowitz et al. / Gynecologic Oncology xxx (2016) xxxxxx

Fig. 2. H & E staining and p63 immunohistochemistry in a representative example of ETT. Noted is the nodular, well circumscribed growth. (Photo courtesy of Bradley J. Quade, Associate
Pathologist Department of Pathology, Division of Women's and Perinatal Pathology, Brigham and Women's Hospital and Associate Professor of Pathology, Harvard Medical School, Boston,

Ultrasonography is an excellent tool to evaluate PSTT and ETT. Although there are no absolute ultrasonographic ndings that are
characteristic of PSTT and ETT, uterine lesions can often be identied
by ultrasound. Zhou et al. described three types of sonographic presentations of PSTT: Type I - a heterogeneous solid mass in the uterine
cavity with minimal to moderate degree of vascularization on color
Doppler; Type II a heterogeneous solid mass in the myometrium
with minimal to high degree of vascularization; and Type III - cystic
lesions in the myometrium with high degree of vascularity (lacunar-type). The time from antecedent pregnancy to presentation
was lower for Type I compared to Type II or III lesions [26]. When
comparing ultrasound ndings in patients with ETT to those with
PSTT, invasive mole and choriocarcinoma, Qin and colleagues [27] did
not nd any difference in the maximal diameter or hemodynamic
parameters of uterine lesions. They did, however, identify a wellcircumscribed border with a hypoechogenic halo in all cases of ETT
(n = 12),versus in only 1 of 21 cases of PSTT and 1 of 16 cases of invasive
mole or choriocarcinoma. Additionally, they identied a peripheral pattern of Doppler signals in ETT. This pattern was noted for Doppler signal
spots around the tumor rather than within the tumor which was distinct
from invasive mole and choriocarcinoma. These sonographic ndings
were consistent with the expansive growth pattern and vascular morphology typically seen in ETT. Though the numbers were small, and larger studies are needed to validate these ndings suggest that ultrasound
could discriminate ETT from other types of GTN [27].
In addition to ultrasound, computed tomography (CT) and magnetic
resonance imaging (MRI) are commonly employed to evaluate the extent of disease in PSTT and ETT [4]. There is limited data on the role of
uorodeoxyglucose positron emission tomography (FDG-PET) in GTN.
Mapelli et al. studied the value of FDG-PET in the primary staging and
monitoring of GTN and noted that although it was not superior to conventional imaging, it could play a role in patients with high-risk disease
and to exclude false positive lesions [28]. In a more recent case report
FDG-PET was able to identify PSTT suggesting a potential role of this imaging modality in its management [29].
Given the low levels of hCG associated with vaginal bleeding and the
rarity of these malignancies, the diagnosis of PSTT and ETT is often difcult. In a report from the French Trophoblastic Disease Reference Center
approximately 30% of patients were initially misdiagnosed with ectopic
pregnancies. Additionally, the initial diagnosis was incorrect in 7 of 18
patients (40%) and there were major disagreements with the referral
pathologist in 30% of patients [30]. Since a correct diagnosis is critical
to avoid inappropriate or ineffective treatment, various markers including SALL4, percentage of free beta hCG subunit, and pregnancy associated major basic protein (pMBP) have been evaluated.
SALL4 is a zinc nger transcription factor important in embryonal
development by maintaining stem cell pluripotency. Because SALL4
has been identied as a reliable marker of germ cell tumors and non-

gestational choriocarcinoma, Strichelbout et al. evaluated its ability to

distinguish gestational choriocarcinoma from PSTT or ETT [31]. Immunohistochemistry staining of SALL4 was undertaken in 31 cases of
GTN, 19 cases of choriocarcinoma, 9 cases of PSTT, 1 case of ETT, and 2
cases with mixed histologies. All of choriocarcinomas and none of the
PSTTs or ETTs expressed SALL4. Other molecular markers (p63, hCG,
and hPL) used to differentiate various forms of GTN showed variable
and inconsistent expression in choriocarcinoma, PSTT, and ETT, suggesting that SALL4 may prove to be a better diagnostic tool [31].
Serum free beta hCG subunit level may be another pre-treatment diagnostic tool for PSTT. Assays for the free beta subunit of hCG measure
the beta chain of hCG without the alpha chain and carbohydrate moiety.
As mentioned previously, PSTT usually has low to variable concentration of hCG making it difcult to differentiate from early stage choriocarcinoma/GTN or quiescent gestational trophoblastic disease. Using
the measurement of free beta hCG as a proportion of total hCG, these authors compared the ndings in 13 cases of PSTT to choriocarcinoma and
quiescent GTD and found free beta hCG to be the predominant hCG form
in cases of PSTT; 60 19% versus 9.3 9.2% for choriocarcinoma and
5.4 7.8% for quiescent GTD. Using a cut off of N 35%, free beta subunit
had a 100% detection rate and no false-positives for PSTT [32]. A second
study from the same group reviewed free beta subunit in 128 cases referred to the USA hCG Reference Service. Forty-ve of these cases had
active invasive trophoblastic disease and 83 had persistent low levels
of hCG with or without a history of GTD. High proportions of free beta
subunit (N30% of total hCG) were identied in 18 of 128 cases. All of
these cases were suspected to have PSTT or non-trophoblastic neoplasms at the time of referral. PSTT was conrmed in 13 of 18 and
non- trophoblastic neoplasms in 5 of 18. The authors concluded that a
proportion N 30% free beta subunit could be used to diagnose PSTT in
women presenting with low hCG levels [33]. A recent study by Harvey
et al. however reported that free beta hCG was helpful but not a denitive test to discriminate PSTT from other types of GTN [34].
A third immunohistochemical marker of PSTT is pregnancy associated major basic protein (pMBP). This protein is a marker of intermediate
trophoblast and has been shown to stain positively in a high percentage
of cells in PSTT and exaggerated placental sites [4]. As a marker of intermediate trophoblast pMBP could help distinguish PSTT or ETT from
other types of GTN but its clinical utility to monitor response or in surveillance has not been investigated.
6. Risk factors
Our understanding of the optimal management, long-term outcome,
and prognostic factors for PSTT and ETT is limited by the rarity of these
malignancies and the difculties of performing appropriate statistical
analyses of risk factors identied in small series. As a result conventional
International Federation of Gynecology and Obstetrics (FIGO) staging

Please cite this article as: N.S. Horowitz, et al., Placental site trophoblastic tumors and epithelioid trophoblastic tumors: Biology, natural history,
and treatment modalities, Gynecol Oncol (2016),

N.S. Horowitz et al. / Gynecologic Oncology xxx (2016) xxxxxx

and prognostic scoring for GTN do not correlate well with outcome in
PSTT and ETT. Unfortunately, there is conicting data concerning most
of the reported potential prognostic factors in PSTT. Reviewing the
available literature however, a few risk factors generally appear to be associated with favorable or unfavorable outcomes. Advanced age
(N34 years old), deep myometrial invasion (N 50%), and tumor size
(N1 cm3) have been associated with worse outcome for women with
PSTT [5,6,2124]. A high mitotic rate has also correlated with a higher
risk of recurrence [5,23]. In one study all patients with 5 mitosis/
10 hpf recurred [23]. Not surprising, advanced stage disease, dened
as stage IIIV, uniformly has been associated with relapse and worse
survival [5,6,21]. In the largest published series (108 patients) a multivariate analysis was performed and stage IV disease was the only significant predictor of survival with an odds ratio of 3.5 (95% CI 1.111.6)
and relapse with an odds ratio of 5.7 (95%CI 1.324) [6]. Although the
exact duration from antecedent pregnancy to diagnosis of PSTT that is
most predictive of outcome is not known, there are multiple studies
that indicate that this is maybe the most important risk factor. In two series duration of approximately 2 years seemed to be important.
Though not statistically signicant, Feltmate et al. noted that the interval from antecedent pregnancy to diagnosis for those that did recur
compared to those that did not was 27 vs 10 months [23]. Similarly,
Newlands and colleagues reported that all patients who initiated treatment within 2 years of the prior pregnancy survived while 4 of 5 with an
interval from antecedent pregnancy N 2 years died [35]. In a more recent
review from the United Kingdom [5], an interval of 48 months from the
antecedent pregnancy predicted survival at 4 years with a 93% specicity, and 100% sensitivity, PPV of 100%, and NPV of 98%. Among patients
who presented within 4 years of the antecedent pregnancy, 48 of 49
(98%) survived while all 13 patients presenting N 4 years after the antecedent pregnancy died. A slightly shorter interval of 36 months between
antecedent pregnancy and diagnosis of PSTT was shown by Zhao and
colleagues to be a signicant predictor of poor survival on univariate
analysis but the signicance was lost on multivariate analysis [6]. The
one study in which interval from antecedent pregnancy (12 months)
was not associated with worse survival was the review from Hyman
et al. [21]. However, the authors discuss that their ndings should be
interpreted with caution given the small sample size.
As a result of the absence of any large single center or multi-center
clinical studies, the risk factors for ETT are even less well established.
Given the clinical similarities between PSTT and ETT it is assumed that
many of the same risk factors for PSTT would apply to ETT [3638].
7. Treatment of localized disease
Treatment decisions for PSTT and ETT are made predominately upon
stage of disease (i.e. presence or absence of metastases) with some consideration of high risk factors. Given the relative chemotherapy resistance of PSTT and ETT, hysterectomy is the treatment of choice for
localized disease conned to the uterus in women who no longer desire
fertility preservation. For premenopausal women, removal of the ovaries is not indicated in the absence of pelvic disease [21]. Based upon observational studies, patients with uterine-conned disease may be
cured with primary hysterectomy alone without adjuvant chemotherapy [37,39]. The role of lymphadenctomy in the surgical management of
PSTT is more controversial. Given the modest rate of nodal metastases
(56%) shown by Lan et al. [25], at the New England Trophoblastic Disease Center we do perform lymphadenctomy in patients with presumed
stage I PSTT, when deep myometrial invasion or bulky lymph nodes are
There does not appear to be a role for adjuvant chemotherapy after
hysterectomy for patients with low risk stage I disease [5,6]. Most reported high risk factors in PSTT are not signicantly associated with outcome when examined in a multivariate analysis except for a time
interval of N 4 years from the antecedent pregnancy and stage IV disease.
Although there is no conclusive evidence currently available to support

adjuvant chemotherapy in stage I PSTT, it is reasonable to consider adjuvant chemotherapy in patients with a long time interval (N4 years)
from the antecedent pregnancy. While some may consider adjuvant
chemotherapy in Stage I disease with deep myometrial invasion or
serosal involvement, there is currently no data to support this
For young women in whom fertility preservation is important,
there are several case series of local uterine resection with varying
degrees of success. Saso and colleagues reported on the use of a modied Strassman procedure for resection in 5 women with presumed
localized uterine PSTT. After the procedure only 1 patient (20%)
remained cancer free with her fertility intact. The other 4 women all required hysterectomy for failure to remove occult multifocal uterine disease. Despite using intraoperative frozen section, distinguishing PSTT
cells from myometrial cells was difcult [40]. Shen et al. used a
multimodality approach to preserve fertility in 6 patients with PSTT
[41]. Using both intravenous and intrauterine arterial infusion chemotherapy in addition to conservative surgery (dilation and curettage or
local tumor resection) the authors noted complete remission in all 6 patients at a mean follow up of 47 months and reported 1 successful pregnancy and normal delivery. Given these discrepant results, the role of
fertility preserving therapy is still controversial and requires extensive
8. Treatment of metastatic disease
Unlike invasive mole or choriocarcinoma, PSTT and ETT are relatively chemotherapy resistant. As such, treatment of advanced
stage, metastatic, or locally advanced disease with chemotherapy
alone is not recommended. In the United Kingdom experience with
PSTT, 60% of patients treated with chemotherapy alone had no or
an incomplete response to chemotherapy. Long-term disease control
was only achieved with surgery or surgery and chemotherapy [5].
This same treatment paradigm of multimodality therapy for PSTT
and ETT is echoed by multiple other authors with complete surgical
resection appearing to be more critical for ETT than PSTT [5,6,22
24,3539]. Depending upon sites and extent of metastases surgical
resection needs to be individualized and may necessitate thoracic
procedures, hepatic resection or other upper abdominal procedures,
bowel resection, and/or craniotomy. The role of surgery after normalization of serum hCG in patients with PSTT is uncertain. For
other forms of GTN, surgical resection of residual masses after normalization of hCG has shown no benet [42]. However, because
PSTT and ETT are less sensitive to chemotherapy and hCG is a less
sensitive tumor marker, some experts, having found viable tumor
in resected residual lesions, recommend resection of residual masses
after completion of chemotherapy [5].
Although there are several chemotherapy regimens used to treat
PSTT and ETT the optimal one is not known given the rarity of these
diseases and the lack of controlled trials. Additionally, various approaches to the sequencing of chemotherapy and surgery (postoperative, neoadjuvant, sandwich) have been employed but again the
best approach is not clear. Despite the lack of any conclusive differences between regimens such as EMA-CO, EP-EMA, or other platinum and non-platinum based regimens, the preferential rst-line
therapy is EP-EMA (Table 2). Based predominately on the experience
at Charing Cross, this regimen achieved a 100% remission in those
patients who had b2 year interval from antecedent pregnancy, a
20% remission in those with interval N 2 years, and an overall survival
of 50% for patients with metastatic PSTT [43]. Since this regimen is
associated with signicant hematologic toxicity, alternative regimens
which appear active and less toxic such as TP/TE (paclitaxel-cisplatin/
paclitaxel-, etoposide), are gaining favor [44]. Zhao et al. [6] treated 37
women who presented with stage II IV PSTT with oxuridine-based
regimens and the initial response rate was 92% with an approximate
20% relapse rate.

Please cite this article as: N.S. Horowitz, et al., Placental site trophoblastic tumors and epithelioid trophoblastic tumors: Biology, natural history,
and treatment modalities, Gynecol Oncol (2016),

N.S. Horowitz et al. / Gynecologic Oncology xxx (2016) xxxxxx

Table 2
Common chemotherapy regimens used to treat advanced stage or recurrent PSTT or ETT.
Day 1


Day 2


100 mg/m2 IV bolus over 3060 min

500 mcg IV Push
100 mg/m2 IV Push followed by
200 mg/m2 IV infusion over 12 h
15 mg PO every 12 h 4 doses
100 mg/m2 IV bolus over 3060 min
500 mcg IV Push*
150 mg/m2 IV Bolus over 60 min
75 mg/m2 IV Bolus


135 mg/m2 IV Bolus over 3 h

75 mg/m2 IV Bolus over 3060 min
135 mg/m2 IV Bolus over 3 h
150 mg/m2 IV Bolus over 1 h

Day 8
Day 1
Day 15

from antecedent pregnancy N 4 years, FIGO stage II, lymphovascular

invasion, hCG N 10,000 IU/L, deep myometrial invasion, and serosal involvement, were statistically signicant risk factors for poor outcomes
while only time from antecedent pregnancy and FIGO stage remained
statistically signicant on multivariate analysis. With a long interval
from prior pregnancy (N 48 months) the hazard ratio for recurrence
was 10.78 (95% CI 2.1254.73, p 0.004) and with stage IV disease
the hazard ratio was 18.7 (95% CI 3.01116.6, p 0.002). The database
is still prospectively collecting data on new patients. To register a patient visit http//
11. Conclusions

Some authors give etoposide and actinomycin D on day 2 with G-CSF on days 814.
Filgrastim 300 mcg SQ days 814 in the EMA-EP regimen and days 814 and days 22
28 in the TP/TE regimen or peglgrastim 6mgSQ days 2 and 16 in the TP/TE regimen h =
hours, min.= minutes.

Salvage regimens for recurrent or resistant PSTT and ETT are needed.
Current regimens are derived from those used in treating other types of
GTN. These regimens include gemcitabine, high dose chemotherapy
with stem cell rescue, combination bleomycin, etoposide, and cisplatin
(BEP), and other platinum-based and uorouracil based-regimens
[4549]. New therapeutic strategies for these diseases are needed and
as targeted biologic agents have become available and shown promise
in a variety of malignancies, these agents may be used rationally in
the refractory setting. In the small series by Lan and colleagues [25]
they demonstrated that PSTT tumor cells stained strongly positive for
EGFR and VEGF. Singh et al. [50] demonstrated strong staining in PSTT
for VEGF, angiopoietin 1 and angiopoietin 2. These targets in addition to PDGF were also identied by Cole et al. [51] If these ndings can
be conrmed by others, perhaps agents such as erlotinib, sunitinib, or
bevacizumab may prove to be effective therapies. Radiation therapy
has also been shown in limited cases to help control disease in localized
recurrent PSTT [52,53].
9. Survival
Long term survival for stage I PSTT with low risk disease after hysterectomy with or without lymphadenctomy is outstanding approaching
90% at 10 years [5]. Even for stage IIIV disease, surgical resection and
chemotherapy have resulted in approximately 50% overall survival at
10 years. Unfortunately long-term remission after recurrence was elusive with approximately 30% living N60 months [5]. These survival
data are similar to that presented by Zhao et al. in which the mean survival was 171.3 months for stage I, 43 months for stage II, 123.8 months
for stage III, and 9.5 months for stage IV [6]. There is limited data
concerning the survival for patients with ETT. While hysterectomy in
patients with stage I ETT is associated with good survival, patients
tend to be particularly resistant to chemotherapy [37,38].
10. International Society for the Study of Trophoblastic Disease
(ISSTD) placental site and epithelioid trophoblastic tumor database
In 2009, the ISSTD initiated a global effort to establish a database
to collect clinical, pathologic, treatment, and outcomes data for
women diagnosed with PSTT/ETT [54]. The multivariate analysis of
the data from 20 participating institutions on 326 patients was presented at the XVIII th ISSTD Congress in 2015. The majority of these
women had PSTT; 135 (41%) had stage I disease and were treated
with surgery alone with a mortality of 3.4%. Given the size of the
dataset traditional risk factors were subjected to univariate and multivariate analysis. By univariate analysis age N 40 years old, interval

PSTT and ETT are the least common types of GTN and disproportionately contribute to mortality due to their relative resistance to
chemotherapy. Information concerning these diseases has advanced
in recent years [56]. Patients with Stage I disease, who do not want to
preserve fertility, should be treated with hysterectomy. Women with
an interval of N4 years from the antecedent pregnancy are at highrisk and adjuvant chemotherapy should be considered. If patients
want to preserve fertility they should be carefully counseled regarding the potential risks and benets of fertility preserving surgery. All
patients with metastatic disease should be treated with intensive
combination chemotherapy and with surgical resection of all sites
of gross and resistant disease to maximize the opportunity for cure
[57]. Our growing molecular understanding of these diseases is likely
to shape future therapies.
Conict of interest statement
The authors declare that there are no conicts of interest.

We wish to acknowledge the support of the Donald P. Goldstein, MD
Trophoblastic Tumor Registry Endowment and the Dyett Family Trophoblastic Disease Research and Registry Endowment.
[1] J. Lurain, Gestational trophoblastic disease II: classication and management of gestational trophoblastic neoplasia, Am. J. Obstet. Gynecol. 204 (2011) 1118.
[2] M.J. Seckl, N.J. Sebire, R.S. Berkowitz, Gestational trophoblastic disease, Lancet 376
(2010) 717729.
[3] N.S. Horowitz, D.P. Goldstein, R.S. Berkowitz, Management of gestational trophoblastic neoplasia, Semin. Oncol. 36 (2009) 181189.
[4] B.W. Hancock, M.J. Seckl, B.W. Hancock, M.J. Seckl, R.S. Berkowitz, Placental site and
epithelioid trophoblastic tumors, Gestational Trophoblastic Disease 2015, pp. 113
( (chapter 20).
[5] P. Schmid, Y. Nagai, R. Agarwal, B. Hancock, P.M. Savafe, N.J. Sebire, et al., Prognostic
markers and long-term outcome of placental-site trophoblastic tumors: a retrospective observational study, Lancet 374 (2009) 4855.
[6] J. Zhao, W.G. Lu, F.Z. Feng, X.R. Wan, J.H. Liu, X.F. Yi, et al., Placental site trophoblastic
tumor: a review of 108 cases and their implications for prognosis and treatment,
Gynecol. Oncol. 142 (2016) 102108.
[7] R.J. Kurman, R.E. Scully, H.J. Norris, Trophoblastic pseudotumor of the uterus: an exaggerated form of syncytial endometritis simulating a malignant tumor, Cancer 38
(1976) 12141226.
[8] L.B. Twiggs, T. Okagaki, G.L. Phillips, J.R. Stroemer, L.L. Adcock, Trophoblastic
pseudotumor- evidence of malignant disease potential, Gynecol. Oncol. 12 (1981)
[9] R.E. Scully, R.H. Young, Trophoblastic pseudotumor: a reappraisal, Am. J. Surg.
Pathol. 5 (1981) 7576.
[10] I.M. Shih, R.J. Kurman, Epithelioid trophoblastic tumor: a neoplasm distinct from
choriocarcinoma and placental site trophoblastic tumor simulating carcinoma,
Am. J. Surg. Pathol. 22 (1998) 13931403.
[11] R.J. Kurman, I.M. Shih, Discovery of a cell: reections on the checkered history of intermediate trophoblast and update on its nature and pathologic manifestations, Int.
J. Gynecol. Pathol. 33 (4) (2014) 339347.
[12] K. Bernirschke, P. Kaufmann, R. Baergen, Pathology of the Human Placenta, Springer,
New York, 2006 191.
[13] R.J. Kurman, C.S. Main, H.C. Chen, Intermediate trophoblast: a distinctive form of trophoblast with specic morphological, biochemical, and functional feature, Placenta
5 (1984) 349370.
[14] I.M. Shih, R.J. Kurman, The pathology of intermediate trophoblastic tumors and
tumor-like lesions, Int. J. Gynecol. Pathol. 20 (1) (2001) 3147.

Please cite this article as: N.S. Horowitz, et al., Placental site trophoblastic tumors and epithelioid trophoblastic tumors: Biology, natural history,
and treatment modalities, Gynecol Oncol (2016),

N.S. Horowitz et al. / Gynecologic Oncology xxx (2016) xxxxxx

[15] M.A. Vardar, A. Altintas, Placental-site trophoblastic tumor. Principles of diagnosis,
clinical behavior and treatment, Eur. J. Gynaecol. Oncol. 16 (4) (1995) 290295.
[16] M. Kashimura, Y. Kashimura, K. Oikawa, C. Sakamoto, Y. Matsuura, S. Nakamura, Placental site trophoblastic tumor: immunohistochemical and nuclear DNA study,
Gynecol. Oncol. 38 (2) (1990) 262267.
[17] K. Allison, J. Love, R. Garcia, Epithelioid trophoblastic tumor: a review of a rare neoplasm of the chorionic-type intermediate trophoblast, Arch. Pathol. Lab. Med. 130
(12) (2006) 18751877.
[18] I.M. Shih, J.D. Seidman, R.J. Kurman, Placental site nodule and characterization of
distinctive types of intermediate trophoblast, Hum. Pathol. 30 (1999) 687694.
[19] C. Lybol, C.M. Thomas, J. Bulten, J.A. van Dijck, F.C. Sweep, L.F. Massuger, Increase in
the incidence of gestational trophoblastic disease in The Netherlands, Gynecol.
Oncol. 121 (2011) 334338.
[20] J. Li, Y. Shi, X. Wan, H. Qian, C. Zhou, X. Chen, Epithelioid trophoblastic tumor: a
clinico-pathological and immunohistochemical study of seven cases, Med. Oncol.
28 (1) (2011) 294299.
[21] D.M. Hyman, L. Bakios, G. Gualtiere, C. Carr, R.N. Grisham, V. Makker, et al., Placental
site trophoblastic tumor: analysis of presentation, treatment, and outcome, Gynecol.
Oncol. 129 (2013) 5862.
[22] R.N. Baergen, J.L. Rutgers, R.H. Young, K. Osann, R.E. Scully, Placental site trophoblastic tumor: a study of 55 cases and review of the literature emphasizing factors of
prognostic signicance, Gynecol. Oncol. 100 (2006) 511520.
[23] C.M. Feltmate, D.R. Genest, L. Wise, M.R. Bernstein, D.P. Goldstein, R.S. Berkowitz,
Placental site trophoblastic tumor: a 17-year experience at the New England Trophoblastic Disease Center, Gynecol. Oncol. 82 (2001) 415419.
[24] A.J. Papadopoulos, M. Foskett, M.J. Seckl, I. McNeish, F.J. Paradinas, H. Rees, et al.,
Twenty-ve years' clinical experience with placental site trophoblastic tumor, J.
Reprod. Med. 47 (2002) 460464.
[25] C. Lan, Y. Li, J. He, J. Liu, Placental site trophoblastic tumor: lymphatic spread and
possible target markers, Gynecol. Oncol. 116 (3) (2010) 430437.
[26] Y. Zhou, H. Lu, C. Yu, Q. Tian, W. Lu, Sonographic characteristics of placental site trophoblastic tumor, Ultrasound Obstet. Gynecol. 41 (6) (2013) 679684.
[27] J. Qin, W. Ying, X. Cheng, X. Wu, B. Lu, Y. Liang, et al., A well-circumscribed border
with peripheral Doppler signal in sonographic image distinguishes epithelioid trophoblastic tumor from other gestational trophoblastic neoplasms, PLoS One 9 (11)
(2014), e112618.
[28] P. Mapelli, G. Mangili, M. Picchio, C. Gentile, E. Rabaiotti, V. Giorgione, et al., Role of
18 FDG-PET in the management of gestational trophoblastic neoplasia, Eur. J. Nucl.
Med. Mol. Imaging 40 (4) (2013) 505513.
[29] S.S. Cerci, E. Erdemoglu, K.K. Bozkurt, Y. Yalcn, E. Erdemoglu, Placental site trophoblastic tumor and uorine-18-uorodeoxyglucose positron emission tomography/
computer tomography, Hell. J. Nucl. Med. 18 (3) (2015) 264267.
[30] A. Moutte, M. Doret, T. Hajri, N. Peyron, F. Chateau, J. Massardier, et al., Placental site
and epithelioid trophoblastic tumors: diagnostic pitfalls, Gynecol. Oncol. 128 (3)
(2013) 568572.
[31] M. Stichelbout, L. Devisme, H. Franquet-Ansart, J. Massardier, D. Vinatier, F. Renaud,
et al., SALL4 expression in gestational trophoblastic tumors: a useful tool to distinguish choriocarcinoma from placental site trophoblastic tumor and epithelioid trophoblastic tumor, Hum. Pathol. S00468177 (16) (2016) 3003130034.
[32] L.A. Cole, S.A. Khanilan, C.Y. Muller, A. Giddings, E. Kohorn, R. Berkowitz, Gestational
trophoblastic diseases: 3. Human chorionic gonadotropin-free beta subunit, a reliable marker of placental site trophoblastic tumors, Gynecol. Oncol. 102 (2) (2006)
[33] L.A. Cole, S.A. Khanilan, C.Y. Muller, Blood test for placental site trophoblastic tumor
and non- trophoblastic malignancy for evaluating patients with low positive human
chorionic gonadotropin results, J. Reprod. Med. 53 (2008) 457464.
[34] R.A. Harvey, H.D. Pursglove, P. Schmid, P.M. Savage, H.D.C. Mitchell, M.J. Seckl,
Human chorionic gonadotropin free beta subunit measurement as a marker of placental site trophoblastic tumors, J. Reprod. Med. 53 (2008) 643648.
[35] E.S. Newlands, M. Bower, R.A. Fisher, F.J. Paradinas, Management of placental site
trophoblastic tumors, J. Reprod. Med. 43 (1998) 5359.
[36] X. Zang, W. Lu, B. Lu, Epithelioid trophoblastic tumor: an outcome based literature
review of 78 reported cases, Int. J. Gynecol. Cancer 23 (7) (2013) 13341338.

[37] M.R. Davis, B.E. Howitt, B.J. Quade, C.P. Crum, N.S. Horowitz, D.P. Goldstein, et al., Epithelioid trophoblastic tumor: a single institution case series at the New England
Trophoblastic Disease Center, Gynecol. Oncol. 137 (2015) 456461.
[38] J.E. Palmer, M. Macdonald, M. Well, B.W. Hancock, J.A. Tidy, Epithelioid trophoblastic
tumor: a review of the literature, J. Reprod. Med. 53 (7) (2008) 465475.
[39] D.P. Goldstein, R.S. Berkowitz, Current management of gestational trophoblastic
neoplasia, Hematol. Oncol. Clin. North Am. 26 (1) (2012) 111131.
[40] S. Saso, J. Haddad, P. Ellis, I. Lindsay, N.J. Sebire, A. McIndoe, et al., Placental site trophoblastic tumours and the concept of fertility preservation, BJOG 119 (2012)
[41] X. Shen, Y. Xiang, L. Guo, F. Feng, X. Wan, Y. Xiao, et al., Fertility preserving treatment
in young patients with placental site trophoblastic tumors, Int. J. Gynecol. Cancer 22
(2012) 869874.
[42] T. Powles, P. Savage, D. Short, A. Young, C. Pappin, M.J. Seckl, Residual lung lesions
after completion of chemotherapy for gestational trophoblastic neoplasia: should
we operate? Br. J. Cancer 94 (2006) 5154.
[43] E.S. Newlands, P.J. Mulholland, L. Holden, M.J. Seckl, G.J. Rustin, Etoposide and cisplatin/etoposide, methotrexate, and actinomycin D (EMA) chemotherapy for patients with high-risk gestational trophoblastic tumors refractory to EMA/
cyclophosphamide and vincristine chemotherapy and patients presenting with
metastatic placental site trophoblastic tumors, J. Clin. Oncol. 18 (2000) 854859.
[44] J. Wang, D. Short, N.J. Sebire, I. Lindsay, E.S. Newlands, P. Schmid, et al., Salvage chemotherapy of relapsed or high risk gestational trophoblastic neoplasia (GTN) with
paclitaxel/cisplatin alternating with paclitaxel/etoposide (TP/TE), Ann. Oncol. 19 (9)
(2008) 15781583.
[45] L.M. El-Helw, M.J. Seckl, R. Haynes, L.S. Evans, P.A. Lorigan, J. Long, et al., High-dose
chemotherapy and peripheral stem cell support in refractory gestational trophoblastic neoplasia, Br. J. Cancer 93 (2005) 620621.
[46] J.R. Lurain, J.C. Schink, Importance of salvage therapy in the management of high
risk gestational trophoblastic neoplasia, J. Reprod. Med. 57 (2012) 219224.
[47] J. Yang, Y. Xiang, X. Wan, X. Yang, Recurrent gestational trophoblastic tumor: management and risk factors for recurrence, Gynecol. Oncol. 103 (2006) 587590.
[48] H.Y. Ngan, K.F. Tam, K.W. Lam, K.K. Chan, Relapsed gestational trophoblastic neoplasia: a 20 year experience, J. Reprod. Med. 51 (2006) 829834.
[49] R. Agarwal, C. Alifrangis, J. Everard, P.M. Savage, D. Short, J. Tidy, et al., Management
and survival of patients with FIGO high risk gestational trophoblastic neoplasia: the
UK experience, 19952010, J. Reprod. Med. 59 (2014) 712.
[50] M. Singh, D. Kindelberger, Z. Nagymanyoki, S.-W. Ng, C.M. Quick, H. Yamamoto,
et al., Vascular endothelial growth factors and their receptors and regulators in gestational trophoblastic diseases and normal placenta, J. Reprod. Med. 57 (2010)
[51] M.E. Cole, R. Broaddus, P. Thaker, C. Landen, R.S. Freedman, Placental-site trophoblastic tumors: a case of resistant pulmonary metastasis, Nat. Clin. Pract. Oncol. 5
(2008) 171175.
[52] N.J. Finkler, R.S. Berkowitz, S.E. Driscoll, D.P. Goldstein, M.R. Bernstein, Clinical experience with placental site trophoblastic tumors at the New England Trophoblastic
Disease Center, Obstet. Gynecol. 71 (1988) 854857.
[53] M. Bower, F.J. Paradinas, R.A. Fisher, S. Nicholson, G.J.S. Rustin, R.H. Begent, et al., Placental site trophoblastic tumour: molecular analysis and clinical experience, Clin.
Cancer Res. 2 (1996) 897902.
[54] B. Hancock, F. Froeling, R. Ramaswami, J. Horsman, L. Ellis, M. Seckl, on behalf of the
ISSTD PSTT database collaborators, The ISSTD global placental site and epithelioid
trophoblastic tumor (PSTT/ETT) database an analysis of 326 patients. Abstract #
12. ISSTD XVIII World Congress on Gestational Trophoblastic Disease. Bali, Indonesia
September 1518, 2015.
[55] C. Bonazzi, M. Urso, T. Dell'Anna, S. Sacco, A. Buda, M.G. Cantu, Placental site trophoblastic tumor: an overview, J. Reprod. Med. 49 (8) (2004) 585588.
[56] G. Santoro, A.S. Lagana, A. Micali, V. Barresi, V. Giacobbe, V. Palmara, Historical, morphological and clinical overview of placental site trophoblastic tumors: from bench
to bedside, Arch. Gynecol. Obstet. (2016) (Epub:115).
[57] F. Taylor, B.W. Hancock, Pharmacotherapy of placental site and epithelioid trophoblastic tumours, Expert Opin. Orphan Drugs 3 (2015) 7585.

Please cite this article as: N.S. Horowitz, et al., Placental site trophoblastic tumors and epithelioid trophoblastic tumors: Biology, natural history,
and treatment modalities, Gynecol Oncol (2016),