Mapping the anatomical landscape of colorectal tumours: Location-specific efficacy of anti-epidermal growth factor receptor antibodies: Pooled analysis of randomised trials.

BACKGROUND

The selection of biologic therapies in metastatic colorectal cancer (mCRC) is traditionally guided by right-left classification; however, this binary classification may not fully capture heterogeneity in responsiveness to anti-epidermal growth factor receptor (EGFR) therapy. We evaluated whether treatment efficacy varies across detailed anatomical tumour segments.

METHODS

Individual patient-level data from 12 randomised controlled trials in the ARCAD database were analysed. Patients with RAS wild-type (RAS-WT) mCRC receiving first-line doublet chemotherapy plus anti-EGFR therapy or bevacizumab were included. The primary endpoint was overall survival (OS); secondary endpoints were progression-free survival (PFS) and objective response rate (ORR).

RESULTS

A total of 2867 patients were included (815 right-sided colon cancer [RSC]; 2052 left-sided colorectal cancer [LSC]). In RAS-WT RSCs, anti-EGFR therapy was not associated with improved OS or PFS versus bevacizumab. In RAS/BRAF-WT RSCs, anti-EGFR therapy was associated with inferior PFS. In contrast, in RAS-WT LSCs, anti-EGFR therapy significantly improved OS and ORR, without a significant difference in PFS. Segment-level analyses (N = 1491) showed no significant heterogeneity within RSCs, although median OS numerically varied across subsites. Within LSCs, the OS benefit was most pronounced in the rectum.

CONCLUSIONS

The efficacy of anti-EGFR therapy in mCRC appears to exhibit additional intraregional heterogeneity beyond the conventional right-left classification. These findings suggest that anatomical tumour location may reflect underlying biological differences not fully captured by this binary classification.