This study aimed to investigate how serum lipid levels affect epithelial ovarian cancer (EOC) patients receiving bevacizumab treatment and to develop a model for predicting the patients’ prognosis.
A total of 139 EOC patients receiving bevacizumab treatment were involved in this study. Statistical analysis was used to compare the median and average values of serum lipid level variables between the baseline and final follow-up. Additionally, a method based on machine learning was proposed to identify independent risk factors for estimating progression-free survival (PFS) in EOC patients receiving bevacizumab treatment. A PFS nomogram dividing the patients into low- and high-risk categories was created based on these independent prognostic variables. Finally, Kaplan–Meier curves and log-rank tests were utilized to perform survival analysis.
Among EOC patients involved in this study, statistical analysis of serum lipid level variables revealed a substantial increase in total cholesterol, triglycerides, apolipoprotein A1, and free fatty acids, and a significant decrease in apolipoprotein B from baseline to final follow-up. Our method identified FIGO stage, combined chemotherapy regimen, activated partial thromboplastin time, globulin, direct bilirubin, free fatty acids, blood urea nitrogen, high-density lipoprotein cholesterol, and triglycerides as risk factors. These risk factors were then included in our nomogram as independent predictors for EOC patients. PFS was substantially different between the low-risk group (total score < 298) and the high-risk group (total score ≥ 298) according to Kaplan–Meier curves (P < 0.05).
Serum lipid levels changed variously in EOC patients receiving bevacizumab treatment. A prediction model for PFS of EOC patients receiving bevacizumab treatment was constructed, and it can be beneficial in determining the prognosis, selecting a treatment plan, and monitoring these patients’ long-term care.