A nomogram to predict pathological complete response of neoadjuvant systemic therapy in HER2-positive breast cancer based on clinicopathologic characteristics and pretreatment systemic inflammation response index: A dual-center study.

Abstract

e12590 Background: Patients with human epidermal growth factor receptor 2 (HER2) positive breast cancer (BC) who achieve a pathologic complete response (pCR) after neoadjuvant systemic therapy (NST) have favorable prognoses. Multiple pretreatment systemic inflammatory markers (SIMs), based on peripheral blood cell counts, have been reported as predictors of pCR in BC patients. The markers include neutrophil-lymphocyte ratio, platelet-lymphocyte ratio, lymphocyte-monocyte ratio, and systemic inflammation response index (SIRI= neutrophil × monocyte / lymphocyte count), etc. However, the most significant SIM remains controversial among studies and the variations among different molecular subtypes remain unknown. Methods: We retrospectively reviewed 233 patients with stage I–III HER2-positive BC who received NST and subsequently underwent surgery in Kaohsiung and Taichung Veterans General Hospital from 2011 to 2021. Data on multiple pretreatment SIMs were collected. The most significant SIM were analyzed for predictive significance with other clinicopathologic factors by using stepwise logistic regression with forward selection. The data was conducted to illustrate a nomogram plot for determining pCR probability. Results: Among the pretreatment SIMs, only SIRI was significantly related to pCR by optimal cut-off value of 1.7 ×109/L. Stepwise logistic analyses indicated that the clinical T and N stage, HER2 immunohistochemistry (IHC) score, PR expression, and targeted therapy regimen were independent predictors of pCR, with an area under the curve (AUC) of 0.74. Sub-analysis was performed after stratifying into T stage subgroups (T1-2 vs. T3-4). Significant predictors In the T1-2 subgroup were clinical N stage, HER2 IHC score, PR, and targeted therapy regimen (AUC=0.722). Significant predictors in the T3-4 subgroup were SIRI and ER (AUC=0.824). The Hosmer–Lemeshow test revealed the predictive ability was a good fit to actual observation. A T-stage-stratified nomogram based on the logistic model was built. Conclusions: In HER2 positive BC patients receiving NST, low pretreatment SIRI is predictive of pCR in T3-4 subgroup. Our nomogram served as an efficient tool to predict pCR and to assist clinical decision support before neoadjuvant treatment.