Gestational trophoblastic neoplasia after human chorionic gonadotropin normalization in a retrospective cohort of 7761 patients in France

BACKGROUND: The risk of malignant transformation of molar pregnancies after human chorionic gonadotropin levels return to normal is low, roughly 0.4%, but may justify an adaptation of monitoring strategies for certain patients. OBJECTIVE: This study aimed to determine the risk of gestational trophoblastic neoplasia after human chorionic gonadotropin normalization in women with molar pregnancy and identify risk factors for this type of malignant transformation to optimize follow-up protocols after human chorionic gonadotropin normalization. STUDY DESIGN: This was a retrospective observational national cohort study based at the French National Center for Trophoblastic Diseases of 7761 patients, treated between 1999 and 2020 for gestational trophoblastic disease, whose human chorionic gonadotropin levels returned spontaneously to normal. RESULTS: Among 7761 patients whose human chorionic gonadotropin levels returned to normal, 20 (0.26%) developed gestational trophoblastic neoplasia. The risk of malignant transformation varied with the type of mole, from 0% (0 of 2592 cases) for histologically proven partial mole to 0.36% for complete mole (18 of 5045) and 2.1% (2 of 95) for twin molar pregnancy. The median time to diagnosis of malignant transformation after human chorionic gonadotropin normalization was 11.4 months (range, 1-34 months). At diagnosis, 16 of 20 patients (80%) had the International Federation of Gynecology and Obstetrics stage I tumor, and 10 of 20 patients (50%) had a tumor classified as low risk in terms of the International Federation of Gynecology and Obstetrics score. In 9 of 20 patients (45%), the most common first-line treatment was combination chemotherapy. A quarter of these tumors (5 of 20) were histologically proven placental site or epithelioid trophoblastic tumors. In univariate analysis, the factors significantly associated with a higher risk of developing gestational trophoblastic neoplasia after the end of the normal human chorionic gonadotropin monitoring period were age of >= 45 years (odds ratio, 8.3; 95% confidence interval, 2.0-32.7; P=.004) and time to human chorionic gonadotropin normalization of >= 8 weeks (odds ratio, 7.7; 95% confidence interval, 1.1-335; P=.03). The risk was even higher for human chorionic gonadotropin normalization times of >= 17 weeks (odds ratio, 19.5; 95% confidence interval, 3.3-206; P<.001). CONCLUSION: In this group of patients with gestational trophoblastic disease, none of the those with pathologically verified partial mole had malignant transformation, supporting the current recommendation of stopping human chorionic gonadotropin monitoring after 3 successive negative tests. In cases of complete mole or twin molar pregnancy, we proposed to extend the monitoring period with quarterly human chorionic gonadotropin measurements for an additional 30 months in patients with the identified risk factors for late malignant transformation (age, >= 45 years; time to human chorionic gonadotropin normalization, >= 8 weeks).