Combined COX-2 and HER-2 Biomarker Profiling to Predict Neoadjuvant Chemoradiotherapy Response in Locally Advanced Rectal Cancer

Abstract

BACKGROUND: Locally advanced rectal cancer (LARC) is commonly treated with neoadjuvant chemoradiotherapy (nCRT), but highly variable response limits outcomes and highlights the need for predictive biomarkers. Cyclooxygenase-2 (COX‑2) and human epidermal growth factor receptor 2 (HER‑2) are overexpressed in a subset of colorectal cancers and are mechanistically linked to radioresistance. Both pathways are therapeutically targetable and exhibit molecular crosstalk, suggesting that combined assessment may improve prediction of nCRT response, but their combined predictive value in LARC remains unexplored. Therefore, this study was conducted to evaluate the association between COX‑2 and HER‑2 expression and radiotherapy response in patients with LARC.

METHODS: This observational retrospective cohort study included 59 patients with stage II–III rectal adenocarcinoma treated with standardized nCRT. COX-2 and HER-2 expressions on pretreatment biopsies were assessed by immunohistochemistry, and radiologic response 4–8 weeks post nCRT dichotomized into good and poor responses using RECIST 1.1.

RESULTS: High COX-2 expression was present in 67.8% of tumors and was associated with poor response (p<0.001; OR=10.08; 95% CI: 2.92–34.78). HER-2 positivity (32.2% of cases) was also associated with poor response (p=0.039; OR=4.28; 95% CI: 1.16–15.79). In multivariate analysis, high COX-2 (adjusted OR=0.110; p=0.002) and HER-2 positivity (adjusted OR=0.197; p=0.049) remained independent predictors of poor response. Tumors with combined COX-2 low/HER-2 negative and COX-2 high/HER 2 positive profiles showed good response rates of 86.7% and 13.3%, respectively, representing a 73.4% absolute difference.

CONCLUSION: Since Low COX-2 expression and HER-2 negativity is mostly associated with good radiotherapy response, hence COX-2 and HER-2 might be independent molecular predictors of radiotherapy response in LARC, and combined biomarker profiling provides robust risk stratification that may guide treatment intensification or de escalation strategies.

KEYWORDS: COX-2, HER-2, rectal cancer, radiotherapy response, biomarker, personalized medicine