Ovarian sex cord-stromal cell tumors and the risk of sex hormone-sensitive cancers.

Abstract

Background: Sex cord-stromal cell tumors are rare ovarian tumors. Some sex cord-stromal cell tumors secrete hormones that originate from the ovarian sex or stromal cells. Previous studies have indicated an increased risk for breast and endometrial cancers. However, these studies focused only on the sex cord-stromal cell tumor-subtype of adult granulosa cell tumors, and the estimates stemmed from selected cohorts and lacked a well-described cohort, making it difficult to adjust for possible confounders.

Objective: We examined the incidence of other primary hormone-sensitive cancers and compared their risk with that in a matched cohort of women without sex cord-stromal cell tumors.

Study design: This nested cohort study evaluated women diagnosed with sex cord-stromal cell tumors. We established a cohort of women diagnosed with sex cord-stromal cell tumors using Systematized Medical Nomenclature for Medicine codes from the Danish Pathology Register. Sex cord-stromal cell tumors diagnoses were considered valid if the primary diagnosis was established at a tertiary referral center or confirmed by at least 2 pathologists. A 1:10 matched control group matched by birth year was selected from national registries. Variables for the follow-up of cases and controls were drawn from national registries and included assessment of the rates of breast, ovarian, and endometrial cancer, hormone use, Charlson Comorbidity Index, and sociodemographics. Hazard ratios for cancer rates were calculated using multivariate Cox proportional hazards models with sex cord-stromal cell tumor exposure estimated in early and delayed models to capture cancer rates over time. The standardized incidence ratios for very rare sex cord-stromal cell tumors were determined using a log-linear Poisson model.

Results: Among the 1516 tumors with Systematized Medical Nomenclature for Medicine codes for sex cord-stromal cell tumors, 1387 met the inclusion criteria after pathologic chart review. The majority had benign tumors, primarily thecoma or fibrothecomas (66%), and 26% had adult granulosa cell tumors. Increased rates of breast cancer were found among patients with thecomas (hazard ratio, 1.2; 95% confidence interval, 1.0-1.4). In the analysis of all sex cord-stromal cell tumors combined, an increased rate of synchronous endometrial cancer was found (hazard ratio, 3.3; 95% confidence interval, 2.7-4.1). In the subgroup analysis, malignant and benign sex cord-stromal cell tumors showed significantly higher hazard ratios for synchronous endometrial cancer, notably adult granulosa cell tumors (hazard ratio, 10.7; 95% confidence interval, 5.7-20.1). In the model that assessed the rates of endometrial cancer 2 months after surgical removal of the sex cord-stromal cell tumor, no increased rates were found. Sertoli cell tumors were associated with an increased incidence of breast cancer (standardized incidence ratio, 18.9; 95% confidence interval, 2.7-134). Both Sertoli and Leydig cell tumors were associated with a higher incidence of synchronous endometrial cancer with standardized incidence ratios of 41.4 (95% confidence interval, 10.4-166) and 44.9 (95% confidence interval, 18.7-108), respectively.

Conclusion: Women with sex cord-stromal cell tumors have an increased rate of synchronous endometrial cancer, and women with benign sex cord-stromal cell tumors have an increased rate of synchronous ovarian cancer. A marginally increased rate of breast cancer has been observed in women with thecomas. There was no increase in the rate of breast cancer among women with adult granulosa cell tumors. It is recommended that women diagnosed with hormone-secreting sex cord-stromal cell tumors and additional risk factors for endometrial cancer (abnormal uterine bleeding and/or abnormal endometrial thickness) should undergo an endometrial biopsy to assess for the presence of cancer.