Persistent gestational trophoblastic tumor after partial hydatidiform mole

GYNECOLOGIC

ONCOLOGY

36,

358-362 (1990)

Persistent Gestational Trophoblastic Tumor after Partial Hydatidiform Mole LAUREL W. RICE, M.D.,

Ross S. BERKOWITZ,

M.D.,’

MARILYN New

England

Trophoblastic

JANICE M. LAGE, M.D., R. BERNSTEIN, BFA

Disease Center, Division of Gynecologic Brigham and Women’s Hospital, Harvard

Oncology, Departments Medical School, Boston,

DONALD

of Obstetrics Massachusetts

P. GOLDSTEIN, and Gynecology 02115

M.D.,

AND

and Pathology,

Received May 17, 1989

The current study investigatesthe clinical characteristicsof patientswith partial molar pregnancy who developedpersistent gestationaltrophoblastic tumor (GTT). Between January 1979 and January 1989, 16 of 240 (6.6%) patients, who were followed for partial mole, developedpersistentGTT. Fifteen (94%)patients were diagnosedas having a missedabortion before evacuation and only 1 patient presentedwith excessiveuterine sizeand theta lutein ovarian cystsand wasfelt to have molar disease.No patient presentedwith toxemia, hyperemesis,or hyperthyroidism. All 16 patientsdevelopednonmetastaticGTT. Fifteen patientsachieved completeremissionwith methotrexate-citrovorum factor and only 1 patient requiredcombinationchemotherapyto attain remission. None of the patientshad histologicevidenceof choriocarcinoma. Patientswith partial molewho developedpersistentGTT did not have clinical or pathological characteristicsthat distinguished them from other patients with partial mole. All patients with partial mole shouldbe followed with measurementof hCG levels to assuregonadotropinremiSSiOn. 0 1990 Academic press, h. INTRODUCTION

Hydatidiform moles are categorized as either complete or partial, based on gross morphology, histopathology, and karyotype [l-4]. Partial molar pregnancies are characterized by (1) focal trophoblastic hyperplasia; (2) two populations of chorionic villi with focal hydatidiform swelling and cavitation; (3) identifiable fetal or embryonic structures; and (4) scalloping of the villous outline and stromal trophoblastic inclusions [1,4]. The majority of partial moles are triploid (69, XXY, 69, XYY), with less than one-third being diploid [5-91. Flow cytometric analysis of DNA content can be useful in differentiating between diploid and triploid molar pregnancies [5,8,9]. When fetuses are identified with partial moles, they gen’ To whom addressed.

correspondence

and reprint

requests

should

MATERIALS

All rights of reproduction

Press, Inc. in any form reserved.

AND METHODS

Between January 1979 and January 1989, 16 of 240 (6.6%) patients with partial molar pregnancy, who were followed at the New England Trophoblastic Disease Center (NETDC), developed persistent gestational trophoblastic tumor. The diagnosis of partial molar pregnancy was based on a thorough review of all curettage specimens by one of the authors (J.M.L.) in the Department of Women’s and Perinatal Pathology at the Brigham and Women’s Hospital. Histologic sections were reviewed noting the presence or absence of fetal or embryonic trophoblastic hyperplatissue, amnion , chorion, sia/atypia, stromal trophoblastic inclusions, villous scal-

be

358 0090-8258/90$1.50 Copyright 0 1990 by Academic

erally exhibit the stigmata of triploidy, including growth retardation and congenital malformation [2,4,10,11]. The presenting signs and symptoms of partial molar pregnancy have been recently studied by several authors [ 12-151. Approximately 90% of patients with partial moles are thought to have a missed or spontaneous abortion before curettage [ 13,141. Classical features of complete molar pregnancy such as excessive uterine size, theta lutein ovarian cysts, or toxemia are usually not present in patients with partial moles. The diagnosis of partial hydatidiform mole is often considered only after histologic review of the curettage specimens. Persistent gestational trophoblastic tumor requiring chemotherapy develops in approximately 4% of patients after evacuation of a partial mole [ 12-171. Limited information has been published concerning the clinical features of patients with partial moles who develop persistent tumor. The current study, therefore, was undertaken to investigate the clinical characteristics of patients with partial moles who developed persistent gestational trophoblastic tumor (GTT).

PERSISTENT

PARTIAL

loping, villous cavitation, molar implantation site, and myometrial invasion. Each patient with a partial molar pregnancy was followed with weekly hCG values until they were normal for 3 consecutive weeks and then with monthly hCG levels until normal for 6 months [131. Persistent gestational trophoblastic tumor was diagnosed if the hCG level reelevated or plateaued for at least 3 consecutive weeks. The medical records at the time of molar evacuation were reviewed in all 16 patients to determine the patient’s age, gravidity, parity, presenting symptoms, clinical diagnosis, gestational age, uterine size, preevacuation hCG level, ultrasound findings, and the presence of hyperemesis, theta lutein ovarian cysts, hyperthyroidism, or toxemia. The hospital records at the time of treatment for persistent GTT also were studied to determine the following information: the time interval between molar evacuation and diagnosis of postmolar tumor, hCG level, extent of disease, chemotherapy treatment, and treatment outcome. All 16 patients underwent a thorough evaluation at the time of diagnosis of persistent GTT. This workup included a complete history and physical examination; assessment of hepatic, renal, and thyroid function tests; and measurement of peripheral blood and platelet counts. An endometrial curettage was performed during the first course of chemotherapy to determine the histologic type of trophoblastic tissue present in the uterus at the onset of therapy [18]. The presence of locally invasive disease was determined by physical examination, curettage, and pelvic ultrasound [ 191. Metastases were evaluated by chest roentgenogram, liver function tests, ultrasound of the abdomen and pelvis, and computed tomography of the chest and head.

359

MOLE

12 (75%) patients. The pelvic ultrasound suggested molar disease in 4 patients due to the presence of vesicular changes in the placental tissue. Histologic sections from the 16 partial moles demonstrated the following findings: fetal or embryonic tissues in 9; amnion in 12; chorion in 14; trophoblastic hyperplasia ranging from mild (10) to moderate (6) with atypia in 7; stromal trophoblastic inclusions in 13; villous scalloping 16; villous cavitation in 16; molar implantation site in 9; and invasive partial hydatidiform mole in 1 case. Table 3 reviews clinical information concerning the 16 patients at the time of their treatment for persistent gestational trophoblastic tumor. The time interval from molar evacuation to the diagnosis of persistent disease ranged from 5 to 28 weeks (mean = 12.3 weeks). All 16 patients had nonmetastatic disease and were treated with methotrexate and citrovorum factor as primary therapy [20]. Methotrexate and citrovorum factor induced complete remission with one treatment in 14 patients and with two treatments in 1 patient. One patient (D.J.) was resistant to methotrexate and citrovorum factor and achieved remission following five courses of etoposide, actinomycin D, and methotrexate-citrovorum factor. This patient (D.J.) had a previous complete mole, which did not require chemotherapy. Another patient (J.B.) also had a previous complete mole, not requiring chemotherapy. The hCG level at the time of treatment for persistent tumor ranged from 6.3 to 6000 mIU/mL (mean = 808 mIU/ml). An endometrial curettage was performed in 15 patients during their treatment for persistent disease and revealed molar tissue in 10 patients and no trophoblast in 5 patients. No patient had gestational choriocarcinoma.

RESULTS DISCUSSION Tables 1 and 2 review demographic and clinical information regarding the 16 patients at the time of evacuation of their partial mole. The patients’ ages ranged from 25 to 40 years (mean = 31.3 years) and gravidity ranged from 1 to 7 (mean = 3.1). The clinical diagnosis before evacuation was missed abortion in 15 (94%) patients and probable molar pregnancy in only 1 patient. The 1 patient, who was diagnosed as having a probable mole, presented with excessive uterine size, theta lutein ovarian cysts, and a markedly elevated preevacuation hCG level. No patient presented with hyperemesis, hyperthyroidism, or toxemia and no patient had a detectable fetal heart beat. Preevacuation hCG values were measured in 9 patients and ranged from 44 to 344,250 mIU/mL; only 3 patients had a preevacuation hCG level >70,000 mIU/ml. A preevacuation ultrasound was performed in all patients and diagnosed missed abortion in

Complete molar pregnancy results in nonmetastatic or metastatic gestational trophoblastic tumor in approximately 20% of patients [21-251. Clinical characteristics of complete mole have been identified that are associated with an increased risk of developing postmolar tumor and include excessive uterine size, theta lutein ovarian cysts, and markedly elevated hCG levels. Signs of marked trophoblastic growth in patients with complete mole are associated with the development of persistent GTT. Partial molar pregnancy has a small but finite risk of developing persistent GTT. Between January 1979 and January 1989, 16 (6.6%) of 240 patients, who were followed for partial mole at the New England Trophoblastic Disease Center, developed persistent tumor. Our experience with partial molar pregnancy is consistent with

TABLE 1 Demographic and Clinical Information at Time of Evacuation of Partial Molar Pregnancy

Patient

Age (years)

Gravidity/ parity

M.H.

25

211

J.P.

29

4/o

S.S.

28

4/l

K.S.

34

2/o

J.N.

32

l/O

S.D.

31

2/l

J.D.

26

211

E.S.

37

3/l

D.C. V.M. P.D.

30 25 31

712 110 3/2

J.B.

26

2/o

D.J.

31

610

N.T.-C.

38

512

M.P.

40

211

L.O.

38

3/l

Presenting signs and symptoms

Gestational age (weeks)/uterine size (weeks)

Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Absent fetal heart Absent fetal heart Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Vaginal bleeding Absent fetal heart Cramping Absent fetal heart Absent fetal heart

Clinical diagnosis

2218

Missed abortion

16/13

Missed abortion

20112

Missed abortion

16/12

Missed abortion

1416

Missed abortion

22110

Missed abortion

26118

Missed abortion

13/lO

Missed abortion

12/14 20124 14/13

Missed abortion Probable molar disease Missed abortion

14114

Missed abortion

717

Missed abortion

8.517

Missed abortion

l2/10

Missed abortion

14114

Missed abortion

TABLE 2 Demographicand Clinical Information at Time of Evacuation of Partial Molar Pregnancy

Patient M.H.

Preevacuation hCG level (mIU/ml) 3,140

J.P. S.S. K.S. J.N. S.D. J.D. E.S. D.C. V.M.

70,000 65,000 Not measured Not measured Not measured 440 Not measured 39,500 344,250

P.D. J.B. D.J. N.T.-C. M.P. L.O.

Not measured 259,000 340,000 Not measured Not measured 27,939

Preevacuation ultrasound findings Collapsed gestational sac, no fetus; missed abortion Missed abortion Missed abortion Missed abortion Missed abortion Missed abortion Probable mole Missed abortion Probable molar tissue Theta lutein ovarian cyst: positive fetus; negative fetal heart; probable mole Missed abortion Missed abortion Probable mole Missed abortion Missed abortion Missed abortion

360

Theta lutein ovarian cysts

Toxemia

Hyperemesis

No

No

No

No No No No No No No No Yes 20-30 cm

No No No No No No No No No

No No No No No No No No No

No No No No No No

No No No No No No

No No No No No No

PERSISTENT

PARTIAL

361

MOLE

TABLE 3 Clinical Information at Time of Treatment for Persistent Gestational Trophoblastic Tumor

Patient M.H. J.P. S.S. K.S. J.N. S.D. I.D. ES. D.C. V.M. P.D. J.B. D.J.

N.T.-C. M.P. L.O.

hCG level (mIU/ml)

Time interval from molar evacuation to diagnosis of persistent disease (weeks)

Extent of tumor

275 15 6ooo 162 45 290 1050 15 64 23.5 6.3 49 3661

12 13 6 8 18 10 5 12 5 10 28 10 28

Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic Nonmetastatic

607 20 657

II

Nonmetastatic Nonmetastatic Nonmetastatic

I5 6

prior reported data. Summarizing the data from eight published series, 20 (3.9%) of 511 patients with partial mole developed persistent GTT [ 161. While their risk for persistent tumor is less than that for patients with complete mole, patients with partial mole also require gonadotropin follow-up to assure complete remission. All of our patients with persistent GTT after partial mole had nonmetastatic disease. Furthermore, 15 (94%) of our 16 patients achieved complete sustained remission following only one or two courses of single-agent chemotherapy. Patients with persistent tumor after partial mole generally have nonmetastatic disease [161. However, Wong and Ma and Stone and Bagshawe have both reported 2 patients with metastatic tumor following partial hydatidiform mole [ 17,151. Our patients with partial mole, who developed persistent GTT, did not have clinical or pathologic characteristics that distinguished them from other patients with partial mole. Fifteen (94%) patients were diagnosed as having missed abortion prior to evacuation. Only 1 patient presented with the classic signs and symptoms of molar disease, including excessive uterine size, theta lutein ovarian cysts, and markedly elevated hCG levels. No patient presented with hyperemesis, hyperthyroidism, or toxemia. Histopathologic study also did not identify any distinctive morphologic features associated with the partial moles that developed persistent disease. None of the partial moles had marked trophoblastic hyperplasia. Investigation should continue to identify prognostic factors in partial mole that are associated with developing persistent tumor. Meanwhile, all patients with partial

Endometrial curettage findings

Treatment MTX-CF x 2 MTX-CF x I MTX-CF x I MTX-CF x I MTX-CF x I MTX-CF x 1 MTX-CF x I MTX-CF x 1 MTX-CF x I MTX-CF x I MTX-CF x I MTX-CF x I MTX-CF x 2 VP16/MTX-CF/ ActD x 5 MTX-CF x I MTX-CF x I MTX-CF x I

Molar tissue Molar tissue Molar tissue Molar tissue No trophoblast Molar tissue No trophoblast Molar tissue Invasive mole No trophoblast No second D&C No trophoblast No trophoblast

Remission Remission Remission Remission Remission Remission Remission Remission Remission Remission Remission Remission Remission

Molar Molar Molar

Remission Remission Remission

tissue tissue tissue

mole should be monitored with serial hCG measurements to ensure attainment of gonadotropin remission. REFERENCES I. Szulman, A. E., and Surti, U. The syndrome of hydatidiform moles. I. Cytogenetics and morphologic correlations, Amer. J. Obstet. Gynecol. 131, 665-771 (1978). 2. Vassilakos, P., Riotton, G., and Kajii, T. Hydatidiform mole: Two entities, Amer. J. Obstet. Gynecol. 127, 167-170 (1977). 3. Goldstein, D. P., and Berkowitz, R. S. Gestational trophoblastic neoplasms-Clinicu/ principles of diagnosis and management, Saunders, Philadelphia, pp. 157-159 (1982). 4. Szulman, A. E., and Surti, U. The syndrome of hydatidiform moles. II. Morphologic evolution of the complete and partial mole, Amer. J. Obstet. Gynecol. 132, 20-27 (1978). 5. Hemming, J. D., Quirke, P., Womack, C., Wells, M., Elston, C. W., and Bird, C. C. Diagnosis of molar pregnancy and persistent trophoblastic disease by flow cytometry. J. Clin. Patho/. 40, 615-620 (1987). 6. Lawler, S. D., Fisher, R. A., Pickthall, V. .I., Povey, S., and Wyn Evans, M. Genetic studies on hydatidiform moles. I. The origin of partial moles, Cancer Genet. Cytogenet. 5, 309-320 (1982). I. Szulman, A. E., Philippe, E., Boue, .I. G., and Boue, A. Human triploidy: Association with partial hydatidiform moles and nonmolar conceptuses, Hum. Pathol. 12, 1016-1021 (1981). 8. Lage, J. M., Driscoll, S. G., Yavner, D. L., Olivier, A. P., Mark, S. D., and Weinberg, D. S. Hydatidiform moles: Application of flow cytometry to diagnosis, Amer. J. C/in. Patho/. 89, 596-600 (1988). 9. Lage, J., Rice, L., Redline, R., Weinberg, D., and Berkowitz, R. Natural history of repetitive hydatidiform moles as analyzed by flow cytometry, Lab Invest. 60, 50A (1989). IO. Wertelecki, W., Graham, J. M., and Sergovich, F. R. The clinical syndrome of triploidy, Obstet. Gynecol. 47, 69-76 (1976).

362

RICE

11. Doshi, N., Surti, U., and Szulman, A. E. Morphologic anomalies in triploid liveborn fetuses, Hum. Patho/. 14, 716-723 (1983). 12. Czernobilsky, B., Barash, A., and Lancet, M. Partial moles: A clinicopathologic study of 25 cases, Obstet. Gynecol. 59, 75-77 (1982). 13. Berkowitz, R. S., Goldstein, D. P., and Bernstein, M. R. Natural history of partial molar pregnancy, Obstet. Gynecol. 66, 677-681 (1985). 14. Szulman, A. E., and Surti, U. The clinicopathologic profile of the partial hydatidiform mole, Obstet. Gynecol. 59, 597-602 (1982). 15. Stone, M., and Bagshawe, K. D. Hydatidiform mole: Two entities. Lnncet 1, 535-536 (1976). 16. Berkowitz, R. S., Goldstein, D. P., and Bernstein, M. R. Partial molar pregnancy: A separate entity, Contemp. Obstet. Gynecol. 31, 99-102

(1988).

17. Wong, L. C., and Ma, H. K. The syndrome of partial mole, Arch. Gynecol. 234, 161-166 (1984). 18. Berkowitz, R. S., Desai, U., Goldstein, D. P., Driscoll, S. G., Marean, A. R., and Bernstein, M. R. Pretreatment curettage: A

ET AL.

19.

20.

21. 22 23. 24.

25.

predictor of chemotherapy response in gestational trophoblastic neoplasia, Gynecol. Oncol. 10, 39-43 (1980). Berkowitz, R. S., Birnholz, J., Goldstein, D. P., and Bernstein, M. R. Pelvic ultrasonography and the management of gestational trophoblastic disease, Gynecol. Oncol. 15, 403-412 (1983). Berkowitz, R. S., Goldstein, D. P., and Bernstein, M. R. Methotrexate with citrovorum factor rescue as a primary therapy for gestational trophoblastic disease, Cancer 50, 2024-2027 (1982). Curry, S. L., Hammond, C. B., Tyrey, L., Creasman, W. T., and Parker, R. T. Hydatidiform mole-Diagnosis, management, and long-term followup of 347 patients, Obstet. Gynecol. 45, l-8 (1975). Berkowitz, R. S., and Goldstein, D. P. Pathogenesis of gestational trophoblastic neoplasms, Parhobiol. Ann. 11, 391-411 (1981). Morrow, C. P. Postmolar trophoblastic disease: Diagnosis, management and prognosis, Clin. Obstet Gynecol. 27, 21 l-220 (1984). Zongfu, X., Hongshao, S., and Minyi, T. Risk of malignancy and prognosis using a provisional scoring system in hydatidiform mole, Chin. Med. J. 93, 605-612 (1980). Kohorn, E. I. Hydatidiform mole and gestational trophoblastic disease in southern Connecticut. Obstet. Gynecol. 59, 78-84 (1982).