Final Survival Outcomes With Ramucirumab Plus Erlotinib Versus Placebo Plus Erlotinib in Patients With Untreated EGFR-Mutated Metastatic NSCLC: RELAY Japanese Subset

Abstract

Introduction: Significant improvement in progression-free survival (PFS; primary end point) was reported in the phase 3 RELAY study with ramucirumab (RAM) plus erlotinib (ERL) versus placebo (PL) in untreated EGFR-mutated NSCLC (hazard ratio [HR] = 0.59, 95% confidence interval [CI]: 0.46-0.76, p < 0.0001), including in the Japanese subset. We report updated PFS and final overall survival (OS) for the Japanese subset.

Methods: Patients (no central nervous system metastases) were randomized 1:1 (stratification included EGFR leucine to arginine substitution [L858R]/exon 19 deletion [ex19del]) to ERL (150 mg/d) with RAM (10 mg/kg; n = 106) or PL (n = 105) intravenously every 2 weeks. The study was not powered for OS.

Results: At final OS data cutoff (median follow-up = 48.2 mo), PFS benefit was sustained with RAM plus ERL versus PL plus ERL (median [m] PFS: 19.4 versus 11.2 mo; HR = 0.69, 95% CI: 0.51-0.93); the mOS was 54.3 and 46.0 months (HR = 0.91, 95% CI: 0.65-1.26), respectively. In L858R (n = 110) and ex19del (n = 100) subgroups, the mOS was 54.3 versus 43.2 months (HR = 0.63, 95% CI: 0.40-0.99) and 53.9 versus 62.1 months (HR = 1.40, 95% CI: 0.86-2.28), respectively. T790M rates post-progression were 52.0% versus 51.1%, respectively. Osimertinib as subsequent therapy was received by 61.0% versus 55.2% patients (L858R: 58.2% versus 48.1%; ex19del: 65.3% versus 62.7%); the median (range) osimertinib treatment duration was 16.8 (0.7-58.3) versus 20.1 (2.1-77.2) months. Safety was consistent with known RAM and ERL profiles, with no increased toxicity over time.

Conclusions: The Japanese subset reported that RAM plus ERL improved PFS, and a mOS greater than 50 months was achieved. OS differed by EGFR mutation type, with an indication of benefit for patients with L858R.

Trial registration: NCT02411448.

Keywords: EGFR; Japan; Non-small-cell lung carcinoma; Overall survival; Ramucirumab.

Figure 1

Kaplan-Meier plot of investigator-assessed PFS at OS data cutoff (Japanese ITT population). (A) ITT population. (B) Patients with an EGFR L858R substitution. (C) Patients with an EGFR ex19del mutation. The maximum data collection time was 92 months. The figure has been cropped. CI, confidence interval; ERL, erlotinib; ex19del, exon 19 deletion; HR, hazard ratio; ITT, intent-to-treat; L858R, exon 21 L858R point mutation; OS, overall survival; m, median; PFS, progression-free survival; PL, placebo; RAM, ramucirumab.

Figure 2

Kaplan-Meier plot of final OS (Japanese ITT population). (A) ITT population. (B) Patients with an EGFR L858R substitution. (C) Patients with an EGFR ex19del mutation. The maximum data collection time was 92 months. The figure has been cropped. CI, confidence interval; ERL, erlotinib; ex19del, exon 19 deletion; HR, hazard ratio; ITT, intent-to-treat; L858R, exon 21 L858R point mutation; NE, not estimable; m, median; OS, overall survival; PL, placebo; RAM, ramucirumab.

Figure 3

Forest plot of final OS by prespecified subgroups (Japanese ITT population). The following categories with fewer than 10 events in either treatment arm are not revealed: unknown smoking history, a disease stage of “other” at diagnosis, and presence of liver metastases at baseline. The shaded area represents the 95% CI for the Japanese ITT population. CI, confidence interval; ECOG PS, Eastern Cooperative Oncology Group performance status; ERL, erlotinib; HR, hazard ratio; ITT, intent-to-treat; OS, overall survival; PL, placebo; RAM, ramucirumab.

Figure 4

Summary of most common TEAEs ≥20% in the RAM+ERL arm (Japanese safety population).^aILD events (included preferred terms ILD and pneumonitis): RAM+ERL: Grade 1 to 2 (1.0%), Grade 3 or higher (1.0%); PL+ERL: Grade 1 to 2 (2.9%), Grade 3 or higher (1.9%). ALT, alanine aminotransferase; AST, aspartate aminotransferase; ERL, erlotinib; ILD, interstitial lung disease; PL, placebo; RAM, ramucirumab; TEAE, treatment-emergent adverse event.

Figure 5

Kaplan-Meier plot of investigator-assessed PFS (A–B) and OS (C–D) in the exploratory liquid biopsy addendum at final OS data cutoff (TR population). Dichotomized baseline circulating EGFR-activating mutation (L858R or ex19del) allele count (low versus high; subgroups based on the count of EGFR-activating mutation alleles at baseline above or below the median count of 102) (A and C); cycle 4 circulating EGFR-activating mutation (L858R or ex19del) alleles (undetectable versus detectable) (B and D). CI, confidence interval; ERL, erlotinib; ex19del, exon 19 deletion; HR, hazard ratio; L858R, exon 21 L858R point mutation; OS, overall survival; PFS, progression-free survival; PL, placebo; RAM, ramucirumab; TR, translational research.