Tumor Clusterin Expression at Diagnosis Is Associated with Residual Disease Burden After Neoadjuvant Chemotherapy in Breast Cancer

Abstract

Objectives: Clusterin (CLU) has been implicated in cell survival and treatment resistance, but its relationship with pathological response to neoadjuvant chemotherapy in breast cancer remains unclear. We evaluated CLU expression in routine diagnostic specimens as a biomarker of pathological response and long-term outcomes. Methods: This multicenter study included 152 patients with breast cancer treated with neoadjuvant chemotherapy. CLU expression was assessed by immunohistochemistry and analyzed across four predefined expression categories. Associations with pathological complete response (pCR), residual cancer burden (RCB), and long-term outcomes were evaluated using logistic and ordinal regression and time-to-event analyses. Multivariable ordinal regression assessed the independent association between CLU expression and RCB. Results: CLU was evaluable in 131 tumors, with high interobserver reproducibility (97.5% agreement; weighted κ = 0.932). Increasing CLU expression was inversely associated with histological grade (ρ = −0.29; p = 0.003). Among 73 patients with RCB assessment, increasing CLU expression was associated with greater residual cancer burden (OR per category increase = 1.53, 95% CI 1.01–2.31; p = 0.044), whereas low (<50%) CLU expression was associated with higher odds of achieving RCB 0 (Firth OR = 5.23, 95% CI 1.53–17.96; p = 0.008). Among 105 patients evaluable for pCR, pCR was more frequent in low- than high-CLU tumors (28.6% vs. 12.5%; OR = 2.80, 95% CI 1.02–7.65; Fisher’s exact p = 0.051). CLU expression was not associated with disease-free survival, distant recurrence-free survival or overall survival. Conclusions: CLU expression at diagnosis shows an ordered association with residual disease burden after neoadjuvant chemotherapy. This association was not independent of clinical stage and subtype, and CLU therefore warrants further validation as a response-associated tissue biomarker rather than as an independent predictor.