Dynamic Changes in NPAR and the CALLY Score for Predicting Pathological Complete Response in Breast Cancer: A Simple...

Pathological complete response (pCR) after neoadjuvant therapy is closely associated with improved survival in breast cancer, particularly in aggressive subtypes. In daily practice, treatment response is mainly evaluated using imaging modalities such as magnetic resonance imaging and ultrasonography; however, these methods may not always accurately reflect residual disease. Easily accessible and reliable laboratory-based markers are still lacking. Because systemic inflammation and nutritional status are closely linked to tumor biology, dynamic changes in composite indices may better reflect treatment response than baseline values alone. Therefore, evaluating treatment-related changes in such markers may provide a practical approach for early response assessment.

In this retrospective study, patients with histopathologically confirmed breast cancer who received neoadjuvant systemic therapy followed by surgery were included. The cohort consisted predominantly of stage II (43%) and stage III (57%) disease, with molecular subtypes representing a high-risk population. Changes in the neutrophil-to-platelet-albumin ratio (ΔNPAR) and creactive protein-albumin-lymphocyte (CALLY) score (ΔCALLY) between pretreatment and post-treatment periods were evaluated. Receiver operating characteristic curve analyses were performed, and optimal cutoff values were determined using the Youden index (-0.516 for ΔNPAR and 0.20 for ΔCALLY). For practical use, a simple combined score based on the direction of change was constructed (ΔNPAR < 0 and ΔCALLY > 0). Logistic regression analysis was performed to identify independent predictors of pCR.

The mean age of the patients was 52 ± 12 years. ΔNPAR showed moderate ability to predict pCR (area under the curve [AUC] 0.679; P < .001), with a sensitivity of 60.6% and specificity of 69.2%. ΔCALLY demonstrated similar performance (AUC 0.685; P < .001), with a sensitivity of 79.8% and specificity of 61.0%. The combined score improved predictive performance (AUC 0.755). Patients with the highest score had a pCR rate of 70.5%, with balanced sensitivity (78.7%) and specificity (78.8%). In multivariable analysis, the combined score was the only independent predictor of pCR (OR 23.5; P < .001).

Changes in ΔNPAR and ΔCALLY, particularly when combined in a simple scoring system, were associated with pCR and may help identify patients with a higher likelihood of achieving pCR. These findings suggest that routinely available laboratory parameters may represent a cost-effective and clinically applicable tool for improving response assessment and supporting treatment decision-making in the neoadjuvant setting.