Her3 As A Biomarker Of Prognosis And Panitumumab (Pan) Benefit In Ras-Wt Advanced Colorectal Cancer (Acrc)

Jenny F. Seligmann,Faye Elliott,Bart Jacobs,Ace Joseph Hatch,Alexander Sibley,Susan Richman,Chen Jiang,Jenny Barrett,Kouros Owzar,Philip Quirke,Herbert Hurwitz,Matthew T. Seymour,Andrew B. Nixon
DOI: https://doi.org/10.1200/jco.2015.33.15_suppl.3583
IF: 45.3
2015-01-01
Journal of Clinical Oncology
Abstract:3583 Background: EGFR pathway activation is important in aCRC, but EGFR expression is not predictive for anti-EGFR drug efficacy. Signalling occurs across ErbB receptors and EGFR may mediate oncogenesis by dimerization with other receptors, including HER3. Therefore, we here explore HER3 mRNA expression as a biomarker of prognosis and predictor of Pan benefit in a randomized trial in aCRC. Methods: HER3 expression, KRAS, NRAS and BRAF mutations were assessed in tumor from 308 patients (pts) randomized to 2nd-line irinotecan (Ir) or IrPan (PICCOLO, Lancet Onc 14:749-59). Prognostic analysis was in all Ir alone pts. Predictive analysis, in the 208 RAS-wt pts, compared baseline values with outcomes using Cox proportional hazards models. HER3 was treated first as a continuous variable; an exploratory binary model of high vs low expression was also tested. Results: Higher HER3 was significantly prognostic for OS (HR [per 2-fold change] = 0.91 [0.83-0.99], p = 0.04), but not PFS (HR = 0.93 [0.83-1.05], p = 0.25). Higher HER3 was associated with PFS benefit in IrPan-treated pts (HR = 0.71 [0.61-0.82], p < 0.001), but not Ir-treated pts (HR = 0.96 [0.82-1.13], p = 0.65). There was a strong biomarker/treatment interaction (p = 0.001), which remained after adjustment for BRAF status and primary tumor location; this was also seen for OS (interaction p = 0.004). HER3 status (above/below 66th centile) was not prognostic for OS (p = 0.35) or PFS (p = 0.86). However, it was predictive of Pan benefit: in RAS-wt pts with high HER3, median PFS was 8.2 mo (IrPan) vs 4.4 mo (Ir) (HR = 0.33 [0.19-0.58], p < 0.001); but pts with low HER3 had no benefit: 3.3 mo (IrPan) vs 4.3 mo (Ir) (HR = 0.96 [0.67-1.38], p = 0.84); interaction p = 0.002. A predictive effect was also seen for OS (interaction p = 0.01). Conclusions: Among RAS-wt pts, HER3 hold promise as a predictive biomarker for Pan benefit. High HER3 identified 1/3rd who gained markedly from Pan, vs 2/3rds who did not benefit, with statistically significant biomarker/treatment interactions for PFS and OS. This finding is of potential clinical utility, and deserves further study to confirm.
What problem does this paper attempt to address?