Hypothesis generative head-to-head study comparing efficacy of afatinib and osimertinib based on immunological biomarkers in Japanese NSCLC patients with <i>EGFR</i> mutations: Heat on Beat randomized phase II study.

Abstract

8633 Background: Osimertinib (Osi) has been established as a standard of care for patients (pts) with EGFR -mutant advanced non-small-cell lung cancer (NSCLC). However, in the FLAURA study, survival superiority of Osi over first-generation EGFR-TKIs was not demonstrated especially in Japanese pts (hazard ratio [HR], 1.39; 95% confidence interval [CI], 0.83 to 2.34; P = 0.22). This is presumably due to the different impact of adverse events or tumor antigen-specific cytotoxicity of T cells on subsequent therapy between both EGFR-TKIs in Japanese pts. On the other hand, there has been no clinical trial comparing second- with third-generation EGFR-TKIs. Therefore, optimal first-line EGFR-TKI may not have been identified in Japanese pts. Methods: This was a randomized, open-label, multicenter, phase II study to compare overall survival (OS) between initial treatment with afatinib (Afa) (n = 50) and Osi (n = 50) in pts with advanced or recurrent EGFR -mutant NSCLC. Exploration of immunomonitoring through peripheral blood mononuclear cells (PBMC) was also performed, before, during, and after treatment. The co-primary endpoints were the superiority of Afa over Osi at 3-year survival rate and the exploration of immunological biomarkers for treatment outcomes. Enrollment started in May 2020 at 28 sites in Japan with a minimum follow-up of 3 years. Results: Overall, 95 eligible pts were analyzed (47 to Afa and 48 to Osi). Objective response rates were 63.8% for Afa vs. 62.5% for Osi. Median progression-free survival (PFS) was 16.7 months (mos) for Afa vs. 14.5 mos for Osi (HR, 1.17; 95% CI, 0.72 to 1.90; P = 0.52). Median OS was 38.8 mos for Afa and not reached for Osi (HR, 1.15; 95% CI, 0.64 to 2.05; P = 0.64), resulting in 54.7% (95% CI, 39.4 to 67.7) for Afa vs. 57.5% (95% CI, 42.2 to 70.1) for Osi at 3-year survival rate. Predominant adverse events with Afa or Osi were diarrhea (92% vs. 31%) and pneumonitis (11% vs. 21%; Grade 5, 0% vs. 6%). Treatment discontinuation rates due to adverse events were 21% with Afa vs. 29% with Osi. The efficacy of Osi varied significantly dependent on the immunological biomarkers, Th7R (stem cell-like CD4 T cells) and Th2. Pts with high Th7R (7.83% or more) had promising PFS (31.0 mos [n = 28] vs. 6.6 mos [n = 20]; HR, 0.38; P = 0.006) and OS (not reached vs. 35.5 mos; HR, 0.56; P = 0.18). In contrast, pts with high Th2 (7.20% or more) had poor PFS (6.6 mos [n = 27] vs. 31.0 mos [n = 21]; HR, 1.78; P = 0.11) and OS (35.5 mos vs. not reached; HR, 2.77; P = 0.03). On the other hand, no immunological biomarkers affected PFS and OS of Afa. Conclusions: Afa and Osi both demonstrated favorable clinical activity as first-line treatment in Japanese NSCLC patients with EGFR mutations. Although their outcomes are comparable, immunological biomarkers (Th7R/Th2) may refine treatment decisions and warrant further prospective validation. Clinical trial information: jRCTs031190221 .