Using Immunohistochemical Markers and Clinicopathological Factors to Predict the Prognostic Survival of Different Types of Endometrial Cancer Recurrence

  Objective  To probe for factors that can be used effectively to predict the prognostic survival of patients with endometrial cancer recurrence.

  Methods  The clinicopathological data of 473 patients with stage Ⅰ to Ⅲ endometrial cancer who underwent standard surgical treatment from October 2013 to May 2019 were retrospectively collected, and post-operative recurrence of the patients were followed up. Overall recurrence includes local recurrence and poor prognosis recurrence. The endpoint indicators of this study are the recurrence-free survival (RFS) and overall survival (OS) of patients with overall recurrence, local recurrence, and poor prognosis recurrence (PPR). The Kaplan-Meier survival curve was used to evaluate the OS and RFS of patients. Cox proportional-hazards model was used to identify factors affecting the prognostic survival of patients with endometrial cancer recurrence.

  Results  Among the 473 patients, 406 did not experience recurrence. A total of 67 patients, accounting for 14.2%, had recurrence. Among them, 27 had local recurrence, accounting for 5.7%, while 40 had poor prognosis recurrence, accounting for 8.5%. The median follow-up time of patients with recurrence was 38 months. The survival curve showed that the RFS and OS of the patients in the recurrence-free group remained unchanged, while the patients in the recurrence group, regardless of whether they had overall recurrence, local recurrence or PPR, experienced a decrease in RFS and OS(P<0.001). The overall 3-year OS rate of patients with recurrence was 44.8%, the median survival time was 29 months, and the median recurrence time was 17 months. The 3-year OS rate of patients in the recurrence-free group was 98.8%, and the median survival time was 40 months; the 3-year OS rate of patients with local recurrence was 59.3%, the median survival time was 27 months, and the median recurrence time was 15 months. The 3-year OS rate of patients with PPR was only 35.0%, the median survival time was 22 months, and the median recurrence time was 10 months. The results of multivariate Cox regression analysis showed that, for overall recurrence patients, FIGO stage Ⅲ (hazard ratio (HR)=3.432, P=0.005), increased expression of K-i67 (HR=1.015, P=0.025), and decreased expression of estrogen receptor (ER) (HR=0.985, P=0.005) are independent factors for the decline in RFS, FIGO stage Ⅲ (HR=4.918, P=0.005) and the decreased expression of progesterone receptor (PR) (HR=0.977, P=0.003) are independent factors for the decrease in OS. For patients with local recurrence, special pathological types (HR=2.545, P=0.049) and increased expression of Ki-67 (HR=1.024, P=0.033) are independent factors influencing the decrease in RFS, while decreased expression of PR (HR=0.973, P=0.009) is an independent risk factor for decreased OS. For patients with PPR, FIGO stage Ⅲ (HR=5.977, P=0.002) and decreased ER expression (HR=0.984, P=0.023) are independent risk factors for the decline in RFS, while FIGO stage Ⅲ (HR=10.098, P=0.001) is an independent factor influencing the decline of OS.

  Conclusion  FIGO stage Ⅲ, increased Ki-67 expression, and decreased ER expression can increase patients' risk of postoperative recurrence, and FIGO stage Ⅲ and decreased expression of PR can increase the risk of death in patients with recurrence.