Dynamic Immune-Nutritional Indices as Powerful Predictors of Pathological Complete Response in Patients with Breast...

Background/Objectives: Pathological complete response (pCR) is an established surrogate marker of neoadjuvant chemotherapy (NACT) efficacy in breast cancer; however, reliable predictors of pCR remain limited. Immune-inflammation- and nutrition-based biomarkers derived from routine blood tests may offer accessible tools for early assessments of treatment response. This study aimed to evaluate both baseline values and dynamic (Δ) changes in multiple immune-nutritional indices to determine their predictive performance with regard topCR. Methods: A retrospective analysis was conducted on 236 early breast cancer patients who received neoadjuvant chemotherapy. Pre-treatment (B), post-treatment (A), and Δ values were calculated for the prognostic nutritional index (PNI), advanced lung cancer inflammation index (ALI), hemoglobin-albumin-lymphocyte-platelet (HALP) score, systemic inflammation response index (SIRI), pan-immune-inflammation value (PIIV), global immune-nutrition-information index (GINI), nutritional risk index (NRI), and related biomarkers. Associations with pCR were examined using chi-square testing and univariate logistic regression, and diagnostic performance was assessed through receiver operating characteristic (ROC) analysis. Results: pCR was achieved in 116 patients (49.2%). Logistic regression identified the NRI (OR = 2.336), ΔGINI (OR = 2.323), ALI (OR = 1.318), PNI (OR = 1.365), HALP score (OR = 1.217), ΔSIRI (OR = 2.207), and ΔPIIV (OR = 2.001) as significant predictors. ROC analysis showed that the NRI (AUC = 0.840) and ΔGINI (AUC = 0.807) were the strongest discriminators of pCR. In aLASSO (Least Absolute Shrinkage and Selection Operator)-penalized logistic regression with 10-fold cross-validation, the NRI and ΔGINI emerged as independent predictors of pCR (OR = 1.28 and OR = 1.23, respectively), showing acceptable calibration particularly in the moderate-to-high probability range. Conclusions: Both baseline and Δ immune-nutritional biomarkers predict pCR following NACT in breast cancer. The NRI and ΔGINI demonstrated the best diagnostic performance, whereas ΔSIRI and ΔPIIV also showed meaningful associations. Easily obtainable, low-cost indices-particularly Δ markers-may support the early identification of responders and facilitate more personalized therapeutic decision-making in breast cancer management.