Smoking history is more predictive of survival benefit from erlotinib for patients with non-small cell lung cancer (NSCLC) than EGFR expression

Gary M. Clark, Denni M. Zborowski, P. Santabárbara, Keyue Ding, M. Whitehead, Lesley Seymour, Frances A. Shepherd

Journal of Clinical Oncology · 2005 · 17 citations · 0 references

Concepts

Abstract

7033 Background: Erlotinib (E, [Tarceva]) is an orally available, reversible inhibitor of the HER1/EGFR tyrosine kinase that has recently been approved by the FDA for the treatment of patients (pts) with locally advanced or metastatic NSCLC after failure of at least one prior chemotherapy regimen. A survival advantage of E compared to placebo (P) was demonstrated in a randomized, double-blind study of 731 patients with NSCLC (Shepherd et al, Proc ASCO, 2004, Abstract 7022). In this study, submission of tumor samples for determination of EGFR expression status by immunohistochemistry (DAKO EGFR pharmDx™ kit) was optional and smoking status was collected retrospectively prior to unblinding. The results of the study showed an overall survival benefit from E. Subset analyses indicated that E was particularly effective in pts with EGFR+ tumors and in pts who had never smoked. However, controversy exists about which of these characteristics is most important, and how their interaction might impact on the efficacy of E. Methods: To address these questions, univariate and multivariate analyses of survival of the 311 pts with available EGFR status and known smoking history were performed using main effects and interactions with treatment. Results: The univariate hazard ratio (HR) of death for E relative to P was 0.74, p=0.020, for all pts in this subset. Smoking history demonstrated a marginally significant interaction with treatment, p=0.054. The HRs were 0.42 among never smokers and 0.87 for current or ex-smokers, indicating that E was beneficial in both subsets, but more effective in pts who had never smoked. The interaction between EGFR status and treatment was not significant in either univariate (p=0.199) or multivariate (p=0.127) analyses; the HRs were 0.65 for EGFR+ and 0.93 for EGFR- pts. Pts with EGFR+ tumors who never smoked had the best survival benefit from E relative to P, HR=0.28, p=0.0007. Conclusions: In summary, these data confirm a benefit from E within different subsets of pts. Never smokers and pts with EGFR+ tumors experience an enhanced benefit from E compared to P, however, smoking history is more predictive of survival than EGFR expression in pts with NSCLC treated with E. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration OSI OSI OSI