Prognostic and risk factors
In univariable analysis, we found that open surgery, uterus invasion, and parametrium invasion negatively affected the overall survival (OS). The mixed pathological subtype might positively influence survival. Hence, we adjusted important tumor characteristics in multivariable analysis. The results showed that mixed histological subtypes had a positive impact on OS (HR,0.02; 95% CI, 0.00~0.45,P =0.014). Uterus invasion was a risk factor for OS (HR, 24.5; 95%CI, 1.37~439.31, P =0.03). However, the tumor size, age, adjuvant therapy, NACT, parametrium invasion, LVSI, surgical margins, perineural invasion, lymph node metastasis, and depth of stromal invasion did not show a clear correlation with prognosis.
With respect to the treatment strategy, the EP regimen showed better outcomes compared to the non-EP regimen in postoperative chemotherapy (HR, 0.04; 95% CI, 0~0.89, P =0.041). Furthermore, treatment without chemotherapy showed better outcomes than the non-EP regimen (HR, 0.03; 95% CI, 0~0.86,P =0.04). Neoadjuvant chemotherapy and postoperative radiotherapy had no effect on the prognosis.