P3.02b-003 Second-Line Afatinib versus Erlotinib for Patients with Squamous Cell Carcinoma of the Lung (LUX-Lung 8): Analysis of Tumor and Serum Biomarkers: Topic: EGFR Biomarkers

LUX-Lung 8 compared second-line afatinib (40 mg/day; n=398) and erlotinib (150 mg/day; n=397) in patients with stage IIIB/IV squamous cell carcinoma (SCC) of the lung. PFS (median 2.6 vs 1.9 months, HR=0.81 [95% CI, 0.69–0.96], p=0.010) and OS (median 7.9 vs 6.8 months, HR=0.81 [0.69–0.95], p=0.008) were both significantly improved with afatinib versus erlotinib. Here we report exploratory molecular (n=245) and immunohistochemical (n=288) analyses of tumor samples to assess the frequency of short variants (SVs) and copy number alterations (CNAs) in cancer-related genes and whether these tumor genomic alterations, or EGFR expression levels, have clinical utility as prognostic/predictive biomarkers in patients with SCC of the lung. We also assessed the predictive utility of the prospectively validated VeriStrat®, a serum protein test (n=675). Archived tumor samples were retrospectively analyzed using next-generation sequencing (FoundationOne™). Tumor EGFR expression was assessed by immunohistochemistry; EGFR positivity was defined as staining in ≥10% of cells. Pretreatment serum samples were assigned as VeriStrat-Good or VeriStrat-Poor according to a mass spectrometry signature. Cox regression analysis was used to correlate OS/PFS with genomic alterations (individual or grouped into gene families e.g. ErbB family), EGFR expression levels and VeriStrat status. The frequency of ErbB family alterations was low (SVs: EGFR 6.5%, HER2 4.9%, HER3 6.1%, HER4 5.7%; CNAs: EGFR 6.9%, HER2 3.7%). No individual genetic alterations, or grouped ErbB family aberrations, were prognostic of OS/PFS. Treatment benefit from afatinib versus erlotinib was consistent in all molecular subgroups. Most tumors were EGFR-positive by immunohistochemistry (afatinib: 82%; erlotinib: 86%). EGFR expression was not predictive of OS or PFS benefit (EGFR-positive PFS: HR=0.76 [0.57‒1.02]; OS: HR=0.84 [0.63‒1.12]; EGFR-negative PFS: HR=0.87 [0.45‒1.68]; OS: HR=0.77 [0.40‒1.51]). In afatinib-treated patients, both PFS (HR=0.56 [0.43‒0.72], p<0.0001) and OS (HR=0.40 [0.31‒0.51], p<0.0001) were improved in the VeriStrat-Good versus the VeriStrat-Poor group. VeriStrat-Good patients had significantly longer OS and PFS when treated with afatinib versus erlotinib (median OS: 11.5 vs 8.9 months, HR=0.79 [0.63‒0.98]; PFS: HR=0.73 [0.59‒0.92]). In VeriStrat-Poor patients there was no significant difference in OS between afatinib and erlotinib (HR=0.90 [0.70‒1.16]). However, there was no significant interaction between treatment arms and VeriStrat classification. Despite comprehensive, multifaceted analysis, no biomarkers were identified that predicted the benefit with afatinib over erlotinib in patients with SCC of the lung. Afatinib is a treatment option in this setting irrespective of patients’ tumor genetics or EGFR expression levels. However, patient outcome strongly depends on VeriStrat status.