PTEN loss as a predictive biomarker in head and neck squamous cell cancer (HNSCC) patients treated with cetuximab (C).

Abstract

e17520 Background: No biomarker of C resistance has been identified in HNSCC. PTEN loss is present in approximately 30% of HNSCC. Biomarker analysis of the E5397 study suggested that addition of C to cisplatin in R/M HNSCC improves PFS in PTEN high/PIK3CA wild type patients but not those with PTEN loss or PIK3CA mutation. We hypothesized that PTEN testing may aid patient (pt.) selection for C therapy in HNSCC. Methods: MCC15780 was a phase II randomized trial of C plus sorafenib or C plus placebo in R/M HNSCC. 52/56 pts. in this study received C. PTEN analysis using AQUA as previously described was performed on tumor from 38 pts. Automated image capture was performed with HistoRx PM-2000 using the AQUAsition software. AQUA PTEN cut off determined as 1177 based on the first tertile.Fisher’s exact test used to compare low and high expression group. Event-time distributions estimated by Kaplan-Meier and compared using log-rank. Stratified Cox proportional hazards models used to estimate hazard ratios (HR) and test for significance for OS and PFS. All p-values are two-sided. A level of p < 0.05 is considered statistically significant. Results: 12/37 (32%) tumors were PTEN low. There was statistically significant improvement in PFS in PTEN high tumors compared to PTEN low tumors (HR (high/low) =0.33, 95% CI= (0.14, 0.75), p=0.008); and this remains significant after adjusting for age, sex, race, and ethnicity (HR (high/low) =0.27, 95% CI= (0.11, 0.67), p=0.004). Conclusions: PFS is significantly longer in PTEN high tumors compared to PTEN low tumors (HR=0.33, 95% CI= (0.14-0.75), p=0.008) in patients with R/M HNSCC treated with C. This finding remains significant after adjusting for age, sex, race, and ethnicity (HR=0.27, 95% CI= (0.11-0.67), p=0.004).This warrants PTEN analysis of specimens from larger cetuximab-based RCT. [Table: see text]