Construction and Validation of a Predictive Model for Chemotherapy-Induced Peripheral Neuropathy in Breast Cancer Survivors Receiving Taxane-Based Chemotherapy

Background Breast cancer survivors (BCSs) undergoing taxane-based chemotherapy frequently experience chemotherapy-induced peripheral neuropathy (CIPN), potentially affecting their quality of life. Early identification of high-risk survivors can help mitigate the severity of CIPN. Objective To construct and validate a predictive model for CIPN in BCSs receiving taxane-based chemotherapy. Methods In this multicenter cross-sectional study conducted across 10 hospitals in China between April 2022 and March 2023, 569 BCSs were randomly assigned to development (n = 401) or validation (n = 168) sets (ratio, 7:3). Predictive factors were identified by multiple logistic regression, and a nomogram was constructed. Model discrimination was evaluated using receiver operating characteristic curves and area under the curve values, whereas calibration was assessed with the Hosmer-Lemeshow test and calibration curves. Decision curve analysis was performed to evaluate clinical utility. Results CIPN was observed in 82.8% of the survivors. The nomogram included 5 factors: treatment with paclitaxel liposome, treatment with albumin-bound paclitaxel, number of chemotherapy cycles, vitamin D deficiency, and fatigue levels. The area under the curve values for the development and validation sets were 0.866 (95% confidence interval, 0.817-0.914) and 0.848 (95% confidence interval, 0.761-0.935), respectively, indicating good performance. The Hosmer-Lemeshow test and decision curve analysis confirmed good calibration and clinical utility. Conclusions The nomogram model demonstrates good discrimination and calibration, offering a practical and visual tool for identifying high-risk survivors for CIPN. Implications for Practice This predictive model can assist clinicians in the early identification of BCSs at high risk for CIPN and in promptly implementing preventive measures.