Early Change in the Pan-Immune-Inflammation Value During Neoadjuvant Chemotherapy Is Associated with Pathological Complete Response in HER2-Positive and Triple-Negative Breast Cancer: A Single-Centre Study with Temporal Validation

Background/Objectives: Pathological complete response (pCR) after neoadjuvant chemotherapy (NAC) is a validated surrogate endpoint in HER2-positive and triple-negative breast cancer. Most work on blood-count-derived inflammatory indices has examined a single pretreatment value rather than change during treatment. We tested whether an early reduction in the pan-immune-inflammation value (PIV) predicts pCR. Methods: We studied 272 patients with HER2-positive or triple-negative stage II–III breast cancer treated at one centre: a discovery cohort treated in 2018–2021 (n = 191) and a temporally independent validation cohort treated in 2022–2023 (n = 81). Early ΔPIV was the proportional change between the baseline complete blood count and the count before the second or third cycle. The 30% reduction threshold was fixed before the analysis for clinical interpretability. Reporting followed STROBE and TRIPOD. Results: pCR occurred in 157 patients (57.7%). The median early ΔPIV was −0.44 in patients with pCR and −0.11 in those without (p < 0.001). The area under the curve was 0.80 (95% CI 0.73–0.86) in the discovery cohort and 0.84 (0.75–0.93) in validation. pCR was reached by 78.6% of patients with a ≥30% early reduction versus 37.6% without it in the discovery cohort, and by 77.8% versus 27.8% in the validation cohort. In the primary three-variable model, the adjusted odds ratio was 6.45 (2.90–14.32; p < 0.001) in the discovery cohort and 27.85 (3.93–197.36; p < 0.001) in the validation cohort. Adding early PIV reduction to clinicopathologic factors raised the area under the curve from 0.69 to 0.80 (difference 0.111, 95% CI 0.047–0.175; p < 0.001). Conclusions: Early PIV reduction is associated with pCR and adds information beyond established factors. At 69% sensitivity and 73% specificity it is a candidate adjunct rather than a decision rule, and prospective multicentre validation is required.