Propensity score matching analysis of ovarian cancer patients with multiple primary malignant neoplasms using multicenter real-world data.

Abstract

e17501 Background: Multiple primary malignant neoplasms (MPMNs) in patients with ovarian cancer is rare and has not attracted enough attention. It is unclear how the MPMNs affect the prognosis of ovarian cancer (OC) patients. Methods: This is a multicenter retrospective analysis of 5, 268 ovarian cancer patients from six centers who was diagnosed with ovarian cancer from January 1, 1989 to August 21, 2020. Propensity score matching was used to balance the baseline characteristics between patients with and without MPMNs. Cox regression analysis was utilized to analyze the influence of MPMNs on overall survival (OS). Results: After excluding unqualified medical record, totally 4, 848 patients were analyzed and 240 were concurrent at least one MPMNs other than OC. Ten patients had two MPMNs and one patient had three. The most common concurrent cancer was breast cancer (111/240, 46.25%), followed by endometrial cancer (37/240, 15.42%), and cervical cancer (30/240, 12.50%). Patients with MPMNs were elder than those without MPMNs (52 vs. 51, P = 0.03) when ovarian cancer was diagnosed. And the proportion of early-stage cases was lower in patients with MPMNs (25.8% vs. 27.2%, P < 0.001). Patients with breast cancer had a higher proportion of high-grade serous ovarian cancer (HGSOC) than those without MPMNs. After using the propensity score matching method adjusting age, pathological type, grade, and stage, concurrent MPMNs, including breast cancer, had no effect on OS of ovarian cancer patients. Among 240 patients with MPMNs, patients with breast cancer shared similar age and stage compared with the rest patients, while their proportion of HGSOC was higher than patients with other cancer (68.4% vs. 51.1%, P = 0.028). However, the median OS of those two groups were similar (27.3 m vs.27.1 m, P = 0.744). In addition, 94 patients were diagnosed with breast cancer prior to ovarian cancer, seven diagnosed posteriorly to ovarian cancer, four diagnosed simultaneously, and six had no precise diagnosed dates. There was no remarkable difference in clinical characteristics between the prior and posterior groups, however, the median OS of those seven patients was significantly longer than the prior group (76.0 m vs. 25.4 m, P = 0.002). Conclusions: The MPMNs showed no influence on the overall survival of ovarian cancer patients. The order of diagnosis of ovarian cancer and breast cancer might affect the prognosis.