Abstract
ObjectiveTo assess the distribution of molecular classes and their impact on the risk of recurrence in endometrial cancer patients with lymph node metastasis at the time of primary surgery.MethodsEndometrial cancer patients with lymph node micrometastasis or macrometastasis (International Federation of Gynecology and Obstetrics (FIGO) 2009 stage IIIC) after surgical staging at five referral centers worldwide from October 2013 to September 2022 who underwent molecular classification were identified. Endometrial cancers were categorized into four molecular classes: POLE mutated, mismatch repair deficient, p53 abnormal, and no specific molecular profile. Survival analyses using Kaplan–Meier and Cox models (univariate and multivariate) were conducted to evaluate the relationship between molecular class and 5-year recurrence free survival.Results131 patients were included: 55 (42.0%) no specific molecular profile, 46 (35.1%) mismatch repair deficient, 1 (0.8%) POLE mutated, and 29 (22.1%) p53 abnormal. During a 5 year follow-up period, 50 (38.2%) patients experienced a recurrence with a median time of 1.2 years (interquartile range (IQR) 0.5–1.8). Median follow-up for the remaining 81 patients was 3.1 years (IQR 1.3–4.5). Survival analysis revealed a significant difference in recurrence-free survival between no specific molecular profile, mismatch repair deficient, and p53 abnormal classes (log rank p<0.01). In a model adjusted for type of lymph node metastasis and tumor grade, the molecular class did not retain significance (p=0.13), while in a model adjusted for type of lymph node metastasis and adjuvant therapy, the molecular class retained significance (p<0.01).ConclusionAmong patients with stage IIIC endometrial cancer, POLE mutated tumors exhibited an extremely low prevalence, with no specific molecular profile emerging as the largest molecular subgroup. Despite the significant difference in recurrence-free survival between molecular classes, conventional histopathologic parameters retained crucial prognostic value. Our findings highlight the necessity of integrating molecular classes with pathological characteristics, rather than considering them in isolation as crucial prognostic factors in stage IIIC endometrial cancer.