Residual lung lesions after completion of chemotherapy for gestational trophoblastic neoplasia: should we operate?

Abstract

The significance of residual lung metastasis from malignant gestational trophoblastic neoplasm (GTN) after the completion of chemotherapy is unknown. We currently do not advocate resection of these masses. Here, we investigate the outcome of these patients. Patients with residual lung abnormalities after the completion of treatment for GTN were compared to those who had a complete radiological resolution of the disease. None of the residual masses post-treatment were surgically removed. In all, 76 patients were identified. Overall 53 (70%) patients had no radiological abnormality on CXR or CT after completion of treatment. Eight (11%) patients had residual disease on CXR alone 15 patients had residual disease on CT (19%). During follow-up, two patients (2.6%) relapsed. One of these had had a complete radiological response post-treatment whereas the other had residual disease on CT. Patients with residual lung lesions after completing treatment for GTN do not appear to have an increased chance of relapse compared to those with no residual abnormality. We continue to recommend that these patients do not require pulmonary surgery for these lesions.

Full text

Page 1

Residual lung lesions after completion of chemotherapy for
gestational trophoblastic neoplasia: should we operate?
T Powles1, P Savage1, D Short1, A Young1, C Pappin1and MJ Seckl*,1
1Department of Health Charing Cross Gestational Trophoblastic Disease Centre, Hammersmith Hospitals Campus of Imperial College London,
Palace Rd, London W68RF, UK
The significance of residual lung metastasis from malignant gestational trophoblastic neoplasm (GTN) after the completion of
chemotherapy is unknown. We currently do not advocate resection of these masses. Here, we investigate the outcome of these
patients. Patients with residual lung abnormalities after the completion of treatment for GTN were compared to those who had a
complete radiological resolution of the disease. None of the residual masses post-treatment were surgically removed. In all, 76
patients were identified. Overall 53 (70%) patients had no radiological abnormality on CXR or CT after completion of treatment.
Eight (11%) patients had residual disease on CXR alone 15 patients had residual disease on CT (19%). During follow-up, two patients
(2.6%) relapsed. One of these had had a complete radiological response post-treatment whereas the other had residual disease on
CT. Patients with residual lung lesions after completing treatment for GTN do not appear to have an increased chance of relapse
compared to those with no residual abnormality. We continue to recommend that these patients do not require pulmonary surgery
for these lesions.
British Journal of Cancer (2006) 94, 51–54. doi:10.1038/sj.bjc.6602899
Published online 13 December 2005
& 2006 Cancer Research UK
www.bjcancer.com
Keywords: lung metastasis; chemotherapy; gestational trophoblastic disease
????????????????????????????????????????????
Gestational trophoblastic neoplasia (GTN) encompasses a spec-
trum of disorders from the premalignant conditions of complete
and partial hydatidiform moles (CHM or PHM) to the malignant
disorders of invasive mole, choriocarcinoma, and placental site
trophoblastic tumour (PSTT). About 16% of CHM and 0.5% of
PHM transform into any of the malignant forms of GTN.
Choriocarcinoma and PSTT can also develop after any type of
pregnancy, including miscarriages and term deliveries (Seckl et al,
2000; Palmieri et al, 2005). Malignant GTN is usually diagnosed
by a plateaued or rising human chorionic gonadotrophin (hCG)
level some weeks following uterine molar evacuation. Histological
confirmation is rarely performed because of the risk of life-
threatening haemorrhage.
Malignant GTN can metastasis, most commonly to the lungs
(Kumar et al, 1988). The vast majority of these patients are cured
with chemotherapy despite the presence of metastatic disease,
although those with liver and/or brain metastases have a poorer
prognosis (Bower et al, 1997; Newlands et al, 2000). On completion
of chemotherapy, most metastatic lesions will be completely
resolved. However, in some patients, residual lesions persist
usually in the lungs. The underlying concern is that such lesions
may still harbour nests of active cancer, so should these lung
lesions be removed?
For nongestational choriocarcinomas, which despite being
pathologically similar are genetically and clinically distinct from
GTN, the answer is yes. This is because residual masses may
contain active cancer regardless of normal tumour markers
(Thompson and Goldstein, 1969; Gels et al, 1997; Hartmann
et al, 1997a,b). However, for GTN, the situation is less clear as the
disease is more chemosensitive and has an even lower relapse rate
than that of nongestational choriocarcinomas arising from ovary
(Sarwar et al, 2004). Therefore, our centre has historically observed
residual GTN masses postchemotherapy with the exception of
those in the brain, the latter being a site where chemotherapy
might be less effective because of the blood–brain barrier
(Newlands et al, 2002).
So is it safe to avoid resection of residual GTN lesions in the
chest? Here, we attempt to address this issue by comparing the
outcome for patients with or without residual radiological
abnormalities in the chest, following chemotherapy for GTN.
PATIENTS AND METHODS
The Charing Cross GTN database was screened to identify all
women diagnosed between 1995 and 2004, with GTN and lung
metastasis as the only site of metastasis. Patients with known
metastases elsewhere including the liver and brain were excluded
as these sites are associated with a worse prognosis and could
confound the results. Only patients whose hCG levels had
normalised for 6 weeks during chemotherapy and remained
normal for 6 weeks following completion of chemotherapy, and
were therefore biochemically thought to be cured, were included
Received 5 October 2005; revised 10 November 2005; accepted 14
November 2005; published online 13 December 2005
*Correspondence: Professor MJ Seckl;
E-mail: m.seckl@imperial.ac.uk
British Journal of Cancer (2006) 94, 51–54
& 2006 Cancer Research UKAll rights reserved 0007– 0920/06$30.00
www.bjcancer.com
Clinical Studies

Page 2

in the subsequent analysis. Clinical, biochemical and radiological
details (either computer tomography (CT) or chest X-ray (CXR),
or both) at presentation and on completion of chemotherapy were
recorded. At presentation for chemotherapy, patients were staged
according to nationally and internationally accepted criteria for
GTN, stratified into low- or high-risk groups, and treated
accordingly (Bagshawe, 1976; Kohorn, 2001; McNeish et al,
2002). Individuals with low-risk disease were treated with
subcutaneous methotrexate and folinic acid, whereas those with
high-risk disease were given weekly combination chemotherapy,
comprising of etoposide, methotrexate, and actinomycin altern-
ating with cyclophospohmide and vincristine (EMA/CO). Resis-
tance to methotrexate occurred in a minority of patients while on
treatment, and therapy was changed to either actinomycin or
EMA/CO depending on the level of hCG at the time of resistance as
previously described (McNeish et al, 2002). During follow-up,
disease-free patients outcome was recorded and plotted according
to the Kaplan–Meier method. This study was approved by our
local institutional review board.
RESULTS
Between 1995 and 2004, 681 patients received chemotherapy for
malignant GTN. At presentation, 76 patients (11%) had lung
metastases as the only overt metastatic site, and following
chemotherapy, all had normal serum hCG levels and were thought
to be cured. Nevertheless, 23/76 (30%) had persistent pulmonary
radiological lesions. No pulmonary lesions were identified in the
remaining 53 patients on either CXR and/or CT scanning after
treatment.
The characteristics of patients with and without radiological
abnormality, at the end of treatment are shown in Table 1. It is
noteworthy that the median age at presentation, initial hCG,
duration of treatment and proportion of patients receiving
combination treatment, in the two groups are not significantly
different (P40.05 for each). Table 2 compares and contrasts
specific factors including number and size of lung metastasis at
presentation, the type of radiological investigation used at the end
of treatment (CXR or CT), and the type of GTN with the risk of
having persistent pulmonary lesions. Of these factors, only patients
with lung lesions 42cm at the start of the treatment or those who
were imaged by CT scan (rather than CXR) were found to have a
significantly increased chance of residual disease on completion of
treatment (Table 2). In contrast, neither the number of metastases
at presentation nor the presence of confirmed choriocarcinoma
altered the risk of detecting residual lesions.
Overall, two patients relapsed after chemotherapy. Both patients
had nonmolar choriocarcinoma that followed term delivery. One
patient had multiple metastasis at the start of the treatment, the
largest of which was 5cm, and an hCG of 102000IUl?1. She was
treated with combination chemotherapy for 21 weeks. At the end
of the treatment, she had three residual masses on CT scanning
ranging between 1 and 1.5cm. Relapse occurred 3 months after the
Table 1
metastatses
Comparability of patient groups with and without residual lung lesions following chemotherapy for malignant GTN associated with lung
All patients
No radiological abnormality after
chemotherapy
Radiologic abnormality
after chemotherapy
Number
Median age at presentation
Median follow-up
hCG at start of treatment
76
31 years (range: 17–61)
5 years (range: 1–9)
10900IUl?1(range 150–
2300000mmoll?1)
18 weeks (range 10–33)
48 (63%)
2
53
31 years (range: 17–61)
4 years (range: 1–9)
110000IUl?1(range 150–
2300000mmoll?1)
18 weeks (range12–33)
33 (62%)
1
23
32 years (range: 18–58)
5 years (range: 1–9)
106000IUl?1(range 170–
1800000mmoll?1)
16 weeks (range 10–29)
15 (65%)
1
Median duration of treatment
Proportion of high-risk patients (score48)
Number of relapses
Table 2
chemotherapy
Influence of CXR verses CT, size of lesion, number of lesions and type of malignant GTN on risk of having a persistent pulmonary lesion after
Radiological investigation at completion of
chemotherapy
No radiological abnormality after
chemotherapy
Radiological abnormality after
chemotherapy
CXR
CT
w2P¼0.018
34
19
8
15
Size of largest lung metastasis at the beginning of treatment
o2cm
42cm
w2P¼0.001
No radiological abnormality after chemotherapy
48
5
Radiological abnormality after chemotherapy
12
11
Number of lung metastasis at the beginning of treatment
Less than 10
10 or more
w2P¼0.85
No radiological abnormality after chemotherapy
38
15
Radiological abnormality after chemotherapy
16
7
Type of GTN
Post molar malignant GTN (either invasive mole or
choriocarcinoma)
Known choriocarcinoma
w2P¼0.59
No radiological abnormality after chemotherapy
34
Radiological abnormality after chemotherapy
13
1910
Residual lung lesions after completion of chemotherapy for GTN
T Powles et al
52
British Journal of Cancer (2006) 94(1), 51–54
& 2006 Cancer Research UK
Clinical Studies

Page 3

completion of treatment with a rising hCG. She went on to receive
second line chemotherapy and eventually developed chemotherapy
resistant disease and died from progressive GTN. The second
patient had numerous 1–3cm lung metastasis and an hCG of
36000IUl?1at diagnosis. She received 17 weeks of combination
chemotherapy, which resulted in complete resolution of the lung
metastasis on CXR. The disease relapsed in multiple sites,
including the brain, 6 months later. She also eventually died from
her disease.
The 5-year disease-free outcome of the whole group was 97%
(95% CI: 3–100%). For those patients with and without a residual
abnormality, the 5-year disease-free survival was 95% (95% CI:
86–100%) and 98% (95% CI: 94–100%), respectively.
There was no significant difference in the disease-free outcome
of the patients with or without persistent radiological abnormality
(Figure 1).
DISCUSSION
Little is known about the outcome of patients with residual
pulmonary lesions, following successful chemotherapy for GTN
with lung metastases at diagnosis. Additionally, there is un-
certainty regarding the management of these residual lesions with
some advocating surgical removal (Ilancheran, 1998). However,
the work presented here shows that residual lung abnormalities in
women who are in marker remission following chemotherapy for
malignant GTN do not appear to have an increased risk of relapse.
Consequently, we believe that these lesions simply represent dead
tissue. Nevertheless, the absolute number of relapses in our study
was small, so we cannot exclude the possibility that there may be a
slight increased risk of relapse in patients with residual lung
lesions when a larger cohort of patients has been studied. So is
there any other data in the literature that might confirm or refute
our findings?
There have been several smaller studies investigating the role of
surgical resection for lung metastasis in GTN (Tomoda et al, 1980;
Saitoh et al, 1983; Xu et al, 1985; Jones et al, 1993). However,
unlike our study, all these have focused on surgery in patients with
an elevated hCG associated with chemoresistant tumours and
many were carried out before the introduction of curative
combination chemotherapy for high-risk disease. The most recent
of these studies was published in 1993 (Jones et al, 1993). Seven
patients had surgical resection after chemotherapy; none had a
normal serum hCG at the time of surgery. Additionally, five of the
patients had extra pulmonary metastasis known to be associated
with a poor prognosis. Indeed, three of these patients died of
extrapulmonary disease. Interestingly, two patients with isolated
pulmonary metastasis were cured by surgery as others have
previously reported (Sink et al, 1981). Thus, a thoracotomy in the
presence of a raised hCG is only likely to be helpful for isolated
chemoresistant lesions and not when there is widespread disease
(Sink et al, 1981). However, none of these studies really provides
any additional data to that presented here concerning the
management of residual lesions when the patients are in
biochemical complete remission with a normal hCG.
In our study, not all the patients had the same radiological
investigation at the end of treatment. Indeed, a higher proportion
of patients with radiological abnormality at the end of treatment
had CT scans. This is perhaps not surprising, as CT scanning is
more sensitive at identifying lung lesions than CXR. Nevertheless,
our results suggest that the observation of persistent radiological
abnormalities at the end of the treatment has no bearing on the
outcome for these patients. Therefore, the type of radiological test
performed may be irrelevant. In view of the radiation exposure
related to CT scanning, we do not recommend this investigation at
the end of the treatment in this young patient population.
Moreover, since radiological residual abnormalities in the chest
do not appear to predict subsequent disease course, and relapses
can be easily serologically detected by a rising hCG, one might be
tempted to advocate no chest imaging at this time. Nevertheless,
we still recommend performing a follow-up CXR on completion
of therapy for women with lung metastases at presentation. This
is because such imaging provides a useful baseline for future
comparison. For example, should the patient undergo chest
imaging for other reasons years later, the finding of a lung lesion
might not necessarily prompt additional investigation if compar-
ison with the previous films postchemotherapy shows no change.
In addition, should the patient relapse and a CXR demonstrate a
lung lesion, it would be important to know whether this was new,
enlarged from before, or old and unchanged.
Perhaps unsurprisingly, this study also shows that patients with
residual disease at the end of treatment have larger lung metastasis
at diagnosis (Table 2). It may be that larger pulmonary metastasis
result in more nonviable tissue after the completion of treatment.
In summary, chemotherapy alone cures the majority of patients
with isolated lung metastasis. Radiological abnormalities at the end
of treatment are of no prognostic significance if the patient is in
hCG remission, and excision of these lesions does not therefore
seem reasonable.
ACKNOWLEDGEMENTS
This work was supported by the Department of Health and
NTRAC.
REFERENCES
Bagshawe KD (1976) Risk and prognostic factors in trophoblastic
neoplasia. Cancer 38: 1373–1385
Bower M, Newlands ES, Holden L, Short D, Brock C, Rustin GJ, Begent RH,
Bagshawe KD (1997) EMA/CO for high-risk gestational trophoblastic
tumours: results from a cohort of 272 patients. J Clin Oncol 15:
2636–2643
Gels ME, Hoekstra HJ, Sleijfer DT, Nijboer AP, Molenaar WM, Ebels T,
Schraffordt Koops H (1997) Thoracotomy for postchemotherapy
0.0
0.1
0.8
0.6
0.4
0.2
0246810 12
0.375
1
DFP-value
0.5405

Cum survival
?2
Years
A n=53
B n=23
Figure 1
A¼no residual disease post-treatment. Arm B¼residual disease post-
treatment.
Kaplan–Meier graph for cumulative relapsed free survival. Arm
Residual lung lesions after completion of chemotherapy for GTN
T Powles et al
53
British Journal of Cancer (2006) 94(1), 51–54
& 2006 Cancer Research UK
Clinical Studies

Page 4

resection of pulmonary residual tumor mass in patients with non-
seminomatous testicular germ cell tumors: aggressive surgical resection
is justified. Chest 112: 967–973
Hartmann JT, Candelaria M, Kuczyk MA, Schmoll HJ, Bokemeyer C
(1997a) Comparison of histological results from the resection of residual
masses at different sites after chemotherapy for metastatic non-
seminomatous germ cell tumours. Eur J Cancer 33: 843–847
Hartmann JT, Schmoll HJ, Kuczyk MA, Candelaria M, Bokemeyer C
(1997b)Postchemotherapy resections
metastatic non-seminomatous testicular germ cell tumors. Ann Oncol
8: 531–538
Ilancheran A (1998) Optimal treatment in gestational trophoblastic disease.
Ann Acad Med Singapore 27: 698–704
Jones WB, Romain K, Erlandson RA, Burt ME, Lewis Jr JL (1993)
Thoracotomy in the management of gestational choriocarcinoma. A
clinicopathologic study. Cancer 72: 2175–2181
Kohorn EI (2001) The new FIGO 2000 staging and risk factor scoring
system for gestational trophoblastic disease: description and critical
assessment. Int J Gynecol Cancer 11: 73–77
Kumar J, Ilancheran A, Ratnam SS (1988) Pulmonary metastases in
gestational trophoblastic disease: a review of 97 cases. Br J Obstet
Gynaecol 95: 70–74
McNeishIA, Strickland S, Holden
Seckl MJ, Newlands ES (2002) Low risk persistent gestational
trophoblastic disease: outcome
low-dose methotrexate and folinic acid, 1992–2000. J Clin Oncol 20:
1838–1844
Newlands ES, Holden L, Seckl MJ, McNeish I, Strickland S, Rustin GJ (2002)
Management of brain metastases in patients with high-risk gestational
trophoblastic tumors. J Reprod Med 47: 465–471
ofresidualmasses from
L, RustinGJS, Foskett M,
followinginitialtreatment with
Newlands ES, Mulholland PJ, Holden L, Seckl MJ, Rustin GJ (2000)
Etoposide and cisplatin/etoposide, methotrexate, and actinomycin D
(EMA) chemotherapy for patients with high-risk gestational tropho-
blastic tumors refractory to EMA/cyclophosphamide and vincristine
chemotherapy and patients presenting with metastatic placental site
trophoblastic tumors. J Clin Oncol 18: 854–859
Palmieri C, Fisher RA, Sebire NJ, Lindsay I, Smith JR, McCluggage WG,
Savage P, Seckl MJ (2005) Placental site trophoblastic tumour arising
from a partial hydatidiform mole. Lancet 366: 688
Saitoh K, Harada K, Nakayama H, Terauchi A, Nagano T, Inoue K (1983)
Role of thoracotomy in pulmonary metastases from gestational
choriocarcinoma. J Thorac Cardiovasc Surg 85: 815–820
Sarwar N, Newlands ES, Seckl MJ (2004) Gestational trophoblastic
neoplasia: the management of relapsing patients and other recent
advances. Curr Oncol Rep 6: 476–482
Seckl MJ, Fisher RA, Salerno GA, Rees H, Paradinas FJ, Foskett M,
Newlands ES (2000) Choriocarcinoma and partial hydatidiform moles.
Lancet 356: 36–39
Sink JD, Hammond CB, Young Jr WG (1981) Pulmonary resection in the
management of metastases from gestational choriocarcinoma. J Thorac
Cardiovasc Surg 81: 830–834
Thompson HT, Goldstein AM (1969) Resection in the treatment of
pulmonary metastases from testicular tumours and its influence on the
overall survival-rate. Br J Surg 56: 349–351
Tomoda Y, Arii Y, Kaseki S, Asai Y, Gotoh S, Suzuki T, Kondoh T,
Imaizumi M (1980) Surgical indications for resection in pulmonary
metastasis of choriocarcinoma. Cancer 46: 2723–2730
Xu LT, Sun CF, Wang YE, Song HZ (1985) Resection of pulmonary
metastatic choriocarcinoma in 43 drug-resistant patients. Ann Thorac
Surg 39: 257–259
Residual lung lesions after completion of chemotherapy for GTN
T Powles et al
54
British Journal of Cancer (2006) 94(1), 51–54
& 2006 Cancer Research UK
Clinical Studies