MASSAD, L. STEWART MD; ABU-RUSTUM, NADEEM R. MD; LEE, SUZETTE S. MMS; RENTA, VIVIAN MD
Gestational trophoblastic disease after molar pregnancy, especially its nonmetastatic subgroup, is among the most curable malignancies.1–3 However, achieving excellent outcomes requires the early institution of chemotherapy after biochemical detection of disease, before symptom onset. This, in turn, requires strict patient compliance with protocols for measurements of serum hCG. Subsequent treatment then requires delivery of cytotoxic chemotherapy under prescribed schedules.
Indigent women, especially those from minority groups, often participate in mammography and cervical cytologic screening programs at suboptimal rates.4,5 Women with abnormal Papanicolaou smears or known cervical intraepithelial neoplasia (CIN) have been lost from posttreatment surveillance programs at rates above 60%.6–9
Rates of compliance with postmolar surveillance and treatment protocols have not been studied in similar populations. High rates of noncompliance with postmolar surveillance might lead to delayed diagnosis and lower likelihood of cure. Noncompliance with treatment might lead to the development of disease that is resistant to chemotherapy. Based on anecdotal observations of noncompliance with postmolar surveillance, we initiated a registry to track compliance with surveillance and treatment protocols and to document the impact of compliance problems on women followed up after evacuation of a molar pregnancy by the gynecologic oncology service of an urban public hospital.
Materials and Methods
Patients who had evacuation for molar pregnancy at Cook County Hospital between May 1, 1994, and December 31, 1999, were referred to the Gynecologic Oncology Service and entered prospectively to a registry for tracking, as were uninsured women evacuated elsewhere and referred for care. Cook County Hospital is an urban, public hospital serving predominantly indigent patients from the metropolitan Chicago area. All molar gestations were confirmed histologically. Patients were informed of the potential for malignant transformation after molar pregnancy, the possibly lethal nature of postmolar gestational trophoblastic disease, and the high likelihood of cure with early diagnosis through biochemical monitoring. All were cautioned that noncompliance might lead to incurability and death. Protocols for surveillance were outlined. All patients were advised to initiate contraception, and use of hormonal contraception, especially injectable medroxyprogesterone acetate, was strongly encouraged.
After intake assessment, patients were asked to return weekly for a brief inquiry about symptoms and for hCG measurement. Assays for hCG were done by enzyme-linked immunosorbent assay (ELISA), with a threshold for detection of 5 mIU/mL. Once hCG levels decreased below threshold on at least 2 consecutive weeks, patients were asked to return monthly for 6 months, after which they were released from surveillance and allowed to attempt conception if desired. Pregnancies identified during follow-up were confirmed using ultrasound.
Noncompliance was defined as at least one missed appointment. To enhance compliance by minimizing inconvenience and delays during surveillance, patients were not examined unless a decision to institute chemotherapy was made, and patients were asked to present early in the morning for phlebotomy to minimize laboratory waiting times. Attempts were made by one attending physician (LSM) to contact by telephone women who failed to keep appointments. In successful attempts, women were counselled again in English or Spanish by one investigator (LSM) about the malignant potential of molar pregnancy and the importance of biochemical surveillance in ensuring curability and were asked to return. When patients could not be reached by telephone or failed to return despite telephone contact, certified letters containing similar information were sent to the last known address. Use of telegrams was at the discretion of the attending gynecologic oncologist.
Gestational trophoblastic disease was diagnosed according to ACOG criteria.10 All women beginning chemotherapy were counselled about the potentially lethal nature of gestational trophoblastic disease, its curability with adherence to chemotherapy protocols, and the problem of chemoresistance that might occur with noncompliance. Patients who had chemotherapy underwent clinical staging using criteria of the International Federation of Gynecology and Obstetrics (FIGO) and began chemotherapy as appropriate. Remission was considered complete after 2 to 3 weekly hCG measurements below 5 mIU/mL, with surveillance continuing at 2- to 5-week intervals during the subsequent year, and continued follow-up at 3- to 12-month intervals thereafter. Hormonal contraception was recommended for at least 1 year after completion of chemotherapy. Intensive efforts to contact patients who missed treatments or follow-up appointments were conducted as described above.
Comparison of women who did or did not miss appointments before achieving biochemical remission was done using χ2 tests. Significance was assumed when P < .05.
During the study period, 62 women were identified in the registry. Eleven women were excluded from study, including two who failed to return after evacuation of their pregnancies, one who was incarcerated, two who were referred to other providers, three with choriocarcinoma in products of conception, two who presented with metastatic disease, and one with unexplained persistently elevated hCG after a nonmolar gestation.
Complete molar pregnancies were found in 37 (73%) of the 51 women remaining after exclusions, with partial moles in ten (20%); histologic subtype was not specified for four (7%) women whose slides could not be retrieved after evacuation at other institutions. The median hCG level at diagnosis among the 48 women with available results was 185,835 mIU/mL (range 3558–7,200,000 mIU/mL). The mean age was 25 years (median 24 years, range 16–49 years). There were 36 residents of the city of Chicago, with 15 from the surrounding metropolitan area. Twenty-nine (57%) women were Hispanic, 19 (37%) black, two (4%) Asian, and one (2%) white.
Gestational trophoblastic disease developed in 11 women (22%), with nine (82%) having stage I and two (18%) stage III disease. The median time to diagnosis was 5 weeks (mean 6.7, range 3–19 weeks). The median hCG level at diagnosis was 1474 mIU/mL (range 8–63,834 mIU/mL). Treatment included methotrexate alone in six and methotrexate followed by dactinomycin in two. A combination of etoposide, methotrexate, dactinomycin, cyclophosphamide, and vincristine was required at diagnosis or after failure of single-agent therapy in three women. All but one of the women treated with chemotherapy agreed to hormonal contraception, with three using oral contraceptives (OCs) and seven injectable medroxyprogesterone acetate.
No patient with gestational trophoblastic disease was fully compliant with protocols for both treatment and follow-up. Five (45%) of the 11 patients with gestational trophoblastic disease missed at least one treatment, and three missed two treatments. Nevertheless, all women had complete biochemical remission. Six women are still under surveillance less than a year after achieving remission, but three of them have already missed surveillance visits, and the other three had missed appointments during treatment. Of the remaining five women, three were followed up at least 12 months without recurrence, but four missed at least one postremission visit and two conceived during surveillance, one at 4 months while taking OCs and one at 10 months after remission while using condoms with spermicide.
Five (13%) of the remaining 40 women were lost to follow-up before documented remission or diagnosis of gestational trophoblastic disease (Table 1). Spontaneous remission was documented for 35 women after a median of 8 weeks (range 8–13 weeks). However, 21 (60%) of these 35 women missed a median of two appointments (range 1–6) before remission. Four (10%) patients without gestational trophoblastic disease conceived, and 16 (40%) failed to complete 6 months of biochemical surveillance, leaving only 15 (38%) of the 40 patients without gestational trophoblastic disease who had documented negative hCG levels 6 months after remission. Five (33%) of these 15 had at least one appointment during the 6 months of follow-up, and three (20%) had missed appointments before remission. Thus, only seven (18%) of the 40 women without gestational trophoblastic disease complied strictly with postmolar surveillance protocols.
Pregnancies occurred in six (12%) of 51 women before completion of prescribed surveillance, including one woman who conceived 11 months after completing methotrexate therapy for Stage I gestational trophoblastic disease. However, none conceived before biochemical normalization. None of 25 women who received injectable medroxyprogesterone acetate throughout follow-up conceived, whereas six (23%) of 26 women who used other methods or stopped medroxyprogesterone acetate before completing surveillance conceived (P < .01).
We found a high rate of noncompliance with surveillance protocols for postmolar pregnancy among indigent, predominantly minority women. This finding requires confirmation by others, and risk factors for noncompliance should be explored at centers where greater patient volume yields sufficient statistical power to identify differences in compliance associated with such variables as age, ethnicity, type of mole, initial hCG level, and distance travelled. If confirmed, interventional trials could study ways to enhance compliance. Centers that treat trophoblastic disease should explore the feasibility of simplified surveillance strategies, both to enhance compliance among indigent women and to limit cost.
Noncompliance was described but not quantified in a review of gestational trophoblastic disease management in Asia and in a study of prophylactic chemotherapy after molar pregnancy in Korea.11,12 A MEDLINE search using the terms “mole” and “compliance” found no previous reports of noncompliance with postmolar surveillance protocols in United States women. Our finding is not surprising because indigent women have been shown to adhere poorly to other cancer prevention strategies, including breast and cervical cancer screening programs.4–9
There are many factors that contribute to the lack of reports of noncompliance with postmolar surveillance. Most studies of the outcome of postmolar surveillance have come from cancer centers, the results of which might not be generalizable to indigent women. Referral bias could result in selection at those centers for women who are motivated to comply with recommended follow-up. Cancer centers are commonly part of university hospitals, where indigent care might be a lesser component of practice, and compliance might be better among insured women. Cancer centers may have tracking and recall components not available to clinicians who care for large numbers of indigent women. Studies comparing indigent, minority women with others are needed to elucidate the relative influence of those factors.
Reasons for noncompliance with surveillance protocols among indigent women are likely complex. Indigent women face a variety of barriers to compliance with cancer prevention programs. Financial barriers to care were not likely to have been influential in this study, because Cook County Hospital provides care without copayment and without regard to insurance status. However, poverty can pose other barriers to care. Transportation problems, including lack of public transportation fare and lack of access to private vehicles were cited informally during telephone conversations with several noncompliant patients. Transportation assistance has been shown to enhance compliance with prenatal care13 and after abnormal Papanicolaou smear.6,13 Since completing this study, we have begun referring all women who have molar pregnancies to social workers who can arrange for transportation assistance. Although patients report that transportation service facilitates compliance, the problem of noncompliance has not resolved.
Although the potential for malignant transformation after molar pregnancy was emphasized repeatedly, many women might have lacked experience with cancer and placed low priority on its prevention. Indigent women might give less priority to cancer prevention than to more immediate concerns, including family care.14 By omitting examination at weekly surveillance visits and by giving postmolar patients priority in registration, we attempted to minimize, but did not eliminate, the inconvenience of compliance. Homelessness and drug abuse can interfere with care, and concern about the impact of absenteeism on employment can hinder compliance as formerly unemployed women enter the workforce. Fear also has been cited as a barrier to adherence with cancer prevention protocols and might have contributed to noncompliance in our study.15 We emphasize the curability of trophoblastic malignancy when it is diagnosed early. Lack of medical sophistication among indigent women might limit their understanding of the risk of cancer after molar pregnancy. Lack of knowledge has been suggested as a risk factor for noncompliance with cancer prevention among such women,15 and education and counseling have been shown to improve compliance after abnormal Papanicolaou smears.8,16,17 We found that it might be necessary to repeat educational efforts before patients understand the risk of malignancy after molar pregnancy. Finally, the credibility of counseling by clinicians of different cultural backgrounds might be limited, as peer counseling improved follow-up after abnormal Papanicolaou smears.17,18 In our clinics minority nurses reinforce the importance of compliance. Further studies should explore reasons for noncompliance.
Seven of 40 (18%) untreated women complied fully with protocols, whereas no treated women did so. This finding might be an artifact of the greater complexity of management protocols for treated women. If confirmed in studies with greater statistical power to determine true differences, this finding suggests that the early administration of chemotherapy might not enhance compliance with surveillance.
The rate of pregnancy during surveillance in our series was 12%, which represents a serious barrier to biochemical monitoring. Pregnancies occurred despite clear explanation to all patients that pregnancy would complicate surveillance for gestational trophoblastic disease and despite strong encouragement for hormonal contraception. We found that women using injectable medroxyprogesterone acetate were significantly less likely than others to conceive during follow-up. Although the resulting amenorrhea and spotting were disturbing to some patients in our series, injectable medroxyprogesterone acetate should be considered the contraceptive method of choice for indigent women after evacuation of molar pregnancy. All pregnancies occurred after biochemical normalization, but before the completion of surveillance, which suggests that indigent women might become complacent about contraception after the conclusion of weekly phlebotomy. Clinicians should counsel patients that the risk of recurrence persists despite biochemical normalization.
1. Smith EB, Weed JC, Tyrey L, Hammond CB. Treatment of non-metastatic gestational trophoblastic disease: Results of methotrexate alone versus methotrexate-folinic acid. Am J Obstet Gynecol 1982;144:88–92.
2. Hammond CB, Borchert LG, Tyrey L, Creasman WT, Parker RT. Treatment of metastatic trophoblastic disease: Good and poor prognosis. Am J Obstet Gynecol 1973;115:451–7.
3. Lurain JR, Casanova LA, Miller DS, Rademaker AW. Prognostic factors in gestational trophoblastic tumors: A proposed new scoring system based on multivariate analysis. Am J Obstet Gynecol 1991;164:611–6.
4. Katz SJ, Hofer TP. Socioeconomic disparities in preventive care persist despite universal coverage: Breast and cervical cancer screening in Ontario and the United States. JAMA 1994;272:530–4.
5. Hayward RA, Shapiro MF, Freeman HE, Corey CR. Who gets screened for cervical and breast cancer? Results from a new national survey. Arch Intern Med 1988;148:1177–81.
6. Marcus AC, Crane LA, Kaplan CP, Reading AE, Savage E, Gunning J, et al. Improving adherence to screening follow-up among women with abnormal Pap smears. Med Care 1992;30:216–30.
7. Paskett ED, Phillips KC, Miller ME. Improving compliance among women with abnormal Papanicolaou smears. Obstet Gynecol 1995;86:353–9.
8. Miller SM, Siejak KK, Schroeder CM, Lerman C, Hernandez E, Helm CW. Enhancing adherence following abnormal Pap smears among low-income minority women: A preventive telephone counselling strategy. J Natl Cancer Inst 1997;89:703–8.
9. Massad LS, Meyer PM. Predicting compliance with follow-up recommendations after colposcopy among indigent urban women. Obstet Gynecol 1999;94:371–6.
10. American College of Obstetricians and Gynecologists. Management of gestational trophoblastic disease. ACOG technical bulletin no. 178. Washington, DC: American College of Obstetricians and Gynecologists, 1993.
11. Teoh ES. Asian approaches in the treatment of trophoblastic disease. Obstet Gynecol Clin North Am 1988;15:545–64.
12. Kim DS, Moon H, Kim KT, Moon YJ, Hwang YY. Effects of prophylactic chemotherapy for persistent trophoblastic disease in patients with complete hydatidiform mole. Obstet Gynecol 1986; 67:690–4.
13. Melnikow J, Paliescheskey M, Stewart GK. Effect of a transportation incentive on compliance with the first prenatal appointment: A randomized trial. Obstet Gynecol 1997;89:1023–7.
14. Lacey L. Cancer prevention and early detection strategies for reaching underserved urban, low-income black women. Cancer 1993;72:1078–83.
15. Lerman C, Hanjani P, Caputo C, Miller S, Delmoor E, Nolte S, et al. Telephone counseling improves adherence to colposcopy among lower-income minority women. J Clin Oncol 1992;10:330–3.
16. Stewart DE, Buchegger PM, Lickrish GM, Sierra S. The effect of educational brochures on follow-up compliance in women with abnormal Papanicolaou smears. Obstet Gynecol 1994;83:583–5.
17. Suarez L, Nichols DC, Brady CA. Use of peer role models to increase Pap smear and mammogram screening in Mexican-American and black women. Am J Prev Med 1993;9:290–6.
18. Sung JFC, Blumenthal DS, Coates RJ, Williams JE, Alema-Mensah E, Liff JM. Effect of a cancer screening intervention conducted by lay health workers among inner-city women. Am J Prev Med 1997;13:51–7.