P–355 Cancer diagnosis among patients with recurrent pregnancy loss: a cohort study

Abstract

Abstract Study question Is unexplained recurrent pregnancy loss (RPL) related to long term cancer morbidity? Summary answer Recurrent unexplained pregnancy loss patients showed lower cancer morbidity. This trend was significant in the secondary aborters and in a sub-analysis for gynecological cancers. What is known already The association between infertility and cancer was studied, but has scarcely been studied in RPL; One study reported a higher incidence of breast and uterine cancers, while another found no association. Immune dysfunction is a possible cause of ‘unexplained RPL’; RPL patients have an increased number of toxic natural killer cells (NKs) in both peripheral blood and decidua. The immune system is also involved in the recognition of cancer cells, potentially leading to effective killing. It is possible that the NK populations in RPL are capable of a better response towards cancer cells in the uterine environment and related organs. Study design, size, duration A retrospective cohort study comparing RPL patients and patients with normal deliveries presenting between 1990 –2010 and followed up until 2018. Participants/materials, setting, methods The RPL (exposed) group consisted of patients with 3 or more losses between 5–24 weeks. The comparison (unexposed) group included women who gave birth, and were not listed in the registry of RPL patients. Matching was based on maternal age and year of delivery, which was matched to the date of admission to the RPL clinic. Patients’ data were cross-linked to the national cancer registry. Kaplan-Meier survival curves were used to compare cancer incidence. Main results and the role of chance The RPL group comprised of 937 RPL patients, compared to 4685 patients with a live birth. The mean follow up time was 16.3 ±5.3 years for RPL cases and 15.9 ± 4.9 for the comparison group. Groups were compared in terms of lifetime risk, post-admission risk and according to cancer type. In a Univariate analysis, the life time risk for cancer was 5.3% (49/937) among RPL patients and 6.8% (317/4685) in the comparison group (p = 0.08). Survival analysis showed the same trend - a lower cancer morbidity in RPL patients (p = 0.06). The low cancer morbidity was more prominent, reaching statistical significance in secondary RPL patients (p = 0.05) , but not in primary RPL (p = 0.4). Breast cancer was the most common tumor, but was neither more nor less common in RPL than in the comparison group. Gynecological cancers, however, were significantly less common in RPL patients: 0.3% (3/937) compared to 1.3% (60/4685) in the comparison group (p = 0.01). After adjustment for maternal age the odds ratio for gynecological cancer was 0.247 (p = 0.018, 95% CI 0.077–0.791) and significantly represented in the survival analysis (p = 0.01). Limitations, reasons for caution There was no access to BMI and smoking status. Patients were followed for a mean period of 16 years; cancer may present later than 16 years. Wider implications of the findings: Unexplained RPL is assumed to have an immunological basis. Our study may provide an indirect support for hyper-responsive immunological mechanisms in RPL patients. Further research is needed to deepen our understanding of the underlying mechanisms and possibly to facilitate treatment options. Trial registration number Not applicable