Screening for Intrinsic and Acquired Resistance Mutations to Anti-EGFR in Real-World MSS Colorectal Cancer Data

Background                                                                                                                                                                                    Anti-EGFR (cetuximab, panitumumab) plus doublet chemotherapy is the first line standard of care for metastatic MSS left-sided RAS/RAF WT colorectal cancer (CRC). However, anti-EGFR clinical efficacy is limited by intrinsic and acquired resistance. Defining anti-EGFR intrinsic resistance mutations could improve patien selection for anti-EGFR treatment while defining anti-EGFR acquired resistance mutations could enable early detection of resistance via liquid biopsy.

Methods                                                                                                                                                                                                  The intrinsic resistance cohort consisted of 163 metastatic MSS RAS/RAF WT colon adenocarcinoma patients treated in the first line with anti-EGFR with aDNA-seq sample collected before treatment (Tempus xT or xF). We screened mutations for anti-EGFR intrinsic resistance defined as HR > 2 for PFS via univariate CoxPH models. The anti-EGFR acquired resistance cohort consisted of 228 MSS CRC patients with a DNA-seq sample (xT or xF) within 12 months prior to and 0.5-24 months post treatment. We defined acquired resistance mutations as those significantly enriched post treatment vs pre-treatment via McNemar test. An anti-VEGF acquired resistance cohort of 748 MSS CRC patients was used as a control. Enrichment of targetable gene fusions post-treatment was assessed via Wilcoxon test. FDR < 0.05 was considered significant.

Results                                                                                                                                                                                                The screen for intrinsic resistance mutations returned a single significant hit of PIK3CA non-exon 9 mutations (~3%, p < 0.01), while PTEN deletions (~2%), which are functionally similar, were also associated with shorter PFS but did not reach significance (p=0.08). PIK3CA non-exon 9 mutation or PTEN deletion were detected in ~5% of patients (n=9) and were associated with shorter PFS (HR = 3.6, 95% CI: 1.7-7.6, p < 0.001) and OS (HR = 2.6, 95% CI: 1.03-6.5, p < 0.05). Median PFS and OS were 6 and 14 months for patients with these resistance alterations, versus 11 and 29 months for non-altered patients. In a CoxPH model including age, sex, sidedness, and collection procedure, these alterations maintained a significant association with PFS (HR =5.2, 95% CI: 2.2-12.4, p < 0.01) but not OS (HR=2.7, 95% CI: 0.9-8.9, p=0.08). Analysis of acquired resistance to anti-EGFR treatment yielded significant mutation hits – KRAS, NRAS, EGFR, and MAP2K1 – as well significant enrichment of targetable fusions (BRAF, FGFR, NTRK, ALK, ROS1, RET, NRG1). No mutations or fusions were observed to be associated with anti-VEGF acquired resistance.

Conclusion                                                                                                                                                                                    Patients harboring PIK3CA non-exon 9 mutation or PTEN deletion displayed intrinsic resistance to anti-EGFR therapy despite being RAS/RAF WT and may be offered anti-EGFR therapy per oncology guidelines. This small underrepresented subset of patients may not benefit from anti-EGFR therapy. Prospective clinical validation will be needed to determine if this subset of PI3K pathway alterations have utility in optimizing first line therapy selection.