Impact of surgical compliance on survival prognosis of ovarian cancer patients and associated influencing factors: A propensity score matching analysis of the SEER database

Abstract

Abstract Purpose To analyze the impact of surgical compliance on ovarian cancer patients’survival and explore the factors affecting surgical compliance. Materials and methods Ovarian cancer patients from 2004-2015 in the SEER database were selected. Clinical, demographic, and treatment characteristics of patients in two groups with good and poor surgical compliance were compared. Kaplan-Meier curves and Cox regression methods were used to analyze the effect of surgical compliance on overall survival (OS) and cancer-specific survival (CSS). Surgical compliance and other independent risk factors were included to construct OS and CSS column-line plots, and the predictive power of the models was assessed using the Harell consistency index (C-index), decision-curve analysis (DCA), subject work characteristics (ROC) curves, and calibration curves. Binary logistic regression was used to identify significant factors affecting surgical compliance. After balancing confounders using propensity score matching (PSM), the effect of surgical compliance on OS and specific survival (CSS) was again assessed. Results A total of 41859 ovarian cancer patients were enrolled in the study, of which 783 (1.87%) belonged to the group of those who were advised to have surgery but the patient refused and had poor surgical compliance, of which 41076 (98.13%) belonged to those who had good compliance in complying with surgical orders. Cox multifactorial analysis showed that surgical compliance was an independent prognostic factor for patients with ovarian cancer. Kaplan - Meier analysis also showed that OS and CSS were significantly better in patients with good compliance than in those with poor compliance ( P < 0.0001). The study data were randomly divided into a training set (n = 29301) and a validation set (n = 12558) according to 7:3, and the column-line plots of OS and CSS were constructed using surgical compliance and independent prognostic factors in the training set and internally validated by the validation set, which determined the superior performance of the nomogram by ROC and DCA curves in comparison with the Grade staging and the AJCC stage system. Time roc, C index, and calibration curves demonstrate the discrimination and calibration of the predictive model. Poor surgical compliance was associated with older age, pathological staging of germ cell type tumors, late staging, presence of local and distant metastases, higher CA-125 values, and lower household income. 712 pairs of participants were enrolled in the study after 1:1 propensity score matching (PSM) to balance confounders, and a Kaplan-Meier analysis also demonstrated that patients with good surgical compliance had significantly better OS and CSS than patients with poor compliance. superior to patients with poor compliance ( P < 0.0001). Conclusion Surgical compliance is an independent prognostic factor predicting OS and CSS in patients with ovarian cancer and is significantly associated with survival. Poor surgical compliance was associated with older age, pathological staging of germ cell type tumors, late staging, presence of local and distant metastases, higher CA-125 values, and lower family income.