LBA8518 Background: Among common EGFR-mutant NSCLC, L858R mutation is suggested to be less sensitive to EGFR-TKIs and associated with poorer prognosis than ex 19 del. REVOL858R is the first randomized phase III trial conducted exclusively in patients with advanced/recurrent EGFR L858R–mutant NSCLC, comparing a treatment strategy of erlotinib plus ramucirumab (E+RAM) followed by osimertinib (OSI) upon detection of acquired T790M with first-line OSI monotherapy. Methods: Eligible patients were treatment-naive adults (≥20 years) with advanced/recurrent EGFR L858R–positive NSCLC and ECOG PS 0–1. Patients were randomized 1:1, stratified by clinical stage (IIIB/IIIC/IV vs postoperative recurrence), sex, and baseline brain metastasis (BM). Patients received OSI 80 mg once daily (Arm A) or E+RAM (E 150 mg once daily plus RAM 10 mg/kg every 2 weeks; Arm B). In Arm B, tumor rebiopsy or liquid biopsy at termination of E+RAM was mandated; patients with detected T790M subsequently received OSI, whereas those without T790M discontinued protocol treatment. The primary endpoint, time to failure of strategy (TFS), was defined as time from randomization to disease progression or death during OSI, or during the primary treatment when OSI was not administered (in Arm B). Results: Overall, 232 patients were randomized (Arm A n=116; Arm B n=116). Baseline characteristics were generally well balanced between the two arms. In the full analysis set, median TFS was 14.8 months in Arm A and 16.6 months in Arm B (HR 1.03; 95% CI 0.78–1.38; log-rank p=0.49). Key secondary endpoints and selected prespecified subgroup analyses of TFS are summarized in Table. The most common AEs (any grade) were diarrhea (41% vs 54%) and acneiform rash (30% vs 71%) in Arm A vs Arm B. Hypertension and proteinuria were observed in 39% and 18% of patients in Arm B, respectively. Interstitial lung disease occurred in 12 patients (10%) in Arm A, and 2 patients (2%) in Arm B. AEs leading to discontinuation of protocol treatment occurred in 21% vs 36% of patients (Arm A vs Arm B). Conclusions: A planned E+RAM-to-OSI strategy did not improve TFS compared with upfront OSI in EGFR L858R–mutated NSCLC. Safety findings were consistent with known profiles, with higher discontinuation for tolerability in the strategy arm. Trial information: jRCTs051200142. Clinical trial information: 051200142. Key secondary endpoints. Key secondary endpoint, median (months) Arm A (OSI) Arm B (E+RAM) HR (95% CI) OS 44.0 (31.1-NR) 38.4 (33.3-NR) 0.98 (0.66-1.45) 3-y OS (%) 56.1% 57.2% – ORR (%) 73.2% 66.3% – PFS1 14.8 14.9 1.08 (0.81-1.44) PFS2 23.8 25.1 0.96 (0.70-1.33) Time to treatment failure 12.6 11.2 1.34 (1.02-1.77) Key prespecified TFS subgroups <65 years / ≥65 years 18.4 / 14.8 21.2 / 16.0 0.90 (0.40-1.99) / 1.06 (0.78-1.45) With / Without baseline BM 11.8 / 19.2 14.7 / 17.8 0.81 (0.49-1.34) / 1.11 (0.79-1.57)
Haratake et al. (Wed,) studied this question.