Biomarker analysis of the phase II JO25567 study comparing erlotinib with or without bevacizumab in first-line advanced EGFR+ non-small-cell lung cancer

Makoto Nishio,Shinji Atagi,Koichi Goto,Yukio Hosomi,Takashi Seto,Toyoaki Hida,Kazuhiko Nakagawa,Hiroshige Yoshioka,Naoyuki Nogami,Makoto Maemondo,Seisuke Nagase,Isamu Okamoto,Noboru Yamamoto,Yuriko Igawa,Kosei Tajima,Masahiro Fukuoka,Nobuyuki Yamamoto,Kazuto Nishio
DOI: https://doi.org/10.21037/tlcr-22-632
2023-06-30
Abstract:Background: Epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors (TKIs), such as erlotinib, are standard-of-care for patients with EGFR mutation-positive non-small-cell lung cancer (NSCLC), but most patients progress within 1 year. Previously, we demonstrated that erlotinib plus bevacizumab (EB) improved progression-free survival (PFS) in patients with EGFR-positive non-squamous NSCLC in the randomized JO25567 study. To understand this effect, we conducted comprehensive exploratory biomarker analyses. Methods: Using blood and tissue specimens from patients enrolled in the JO25567 study, angiogenesis-related serum factors, plasma vascular endothelial growth factor-A (pVEGFA), angiogenesis-related gene polymorphisms, and messenger RNAs (mRNAs) in tumor tissue were analyzed. Interactions between potential predictors and treatment effect on PFS were analyzed in a Cox model. Continuous variable predictors were evaluated by multivariate fractional polynomial interaction methodology and subpopulation treatment effect pattern plotting (STEPP). Results: Overall, 152 patients treated with EB or erlotinib alone (E) were included in the analysis. Among 26 factors analyzed in 134 baseline serum samples, high follistatin and low leptin were identified as potential biomarkers for worse and better outcomes with EB, with interaction P values of 0.0168 and 0.0049, respectively. Serum concentrations of 12 angiogenic factors were significantly higher in patients with high follistatin. Low pVEGFA levels related to better outcomes with EB, interaction P=0.033. VEGF-A165a was the only predictive tissue mRNA, showing a similar trend to pVEGFA. No valid results were obtained in 13 polymorphisms of eight genes. Conclusions: EB treatment showed better treatment outcomes in patients with low pVEGFA and serum leptin, and limited efficacy in patients with high serum follistatin.
What problem does this paper attempt to address?