1324MO CNS outcomes of lazertinib vs gefitinib in EGFR-mutated advanced NSCLC: A LASER301 subset analysis

Up to 50% of patients (pts) with non-small cell lung cancer (NSCLC) develop central nervous system (CNS) metastases (mets), which are a major source of mortality in NSCLC. Lazertinib (laz), a CNS-active 3rd-generation (gen) tyrosine kinase inhibitor (TKI) targeting mutant epidermal growth factor receptor (EGFR), improved progression-free survival (PFS) vs the 1st-gen TKI gefitinib (gef) in the LASER301 study. Here, we compared laz vs gef among LASER301 pts with CNS mets. In the phase 3 LASER301 study (NCT04248829), treatment (tx)-naïve pts with EGFR-mutated advanced or metastatic NSCLC were randomized 1:1 to laz (240 mg/day) or gef (250 mg/day). Pts with symptomatic/unstable CNS mets were excluded. If tx was required for pts with asymptomatic/stable CNS mets, radiation and/or surgery, and steroids were completed >2 weeks before randomization. Magnetic resonance imaging was performed at screening, every 6 weeks for 18 months (mo), then every 12 weeks relative to randomization, using the same modality at each follow-up. Endpoints assessed by blinded independent central review and RECIST v1.1 included intracranial PFS (iPFS), intracranial objective response rate (iORR), intracranial duration of response (iDoR), and depth of intracranial response. Of 393 pts in LASER301, 45 (23%) receiving laz and 41 (21%) receiving gef had measurable and/or non-measurable baseline (BL) CNS mets. BL characteristics were balanced between groups, with most pts having 1–3 CNS lesions (laz: 91%, gef: 83%). Median BL target lesion size was 20.0 mm (laz) and 16.0 mm (gef). Median iPFS in the laz group was 28.2 mo (95% confidence interval [CI]: 14.8, 28.2) vs 8.4 mo (6.7, not reached [NR]) in the gef group (hazard ratio: 0.42; 95% CI: 0.20, 0.89; P=0.02). iORR was higher for the laz (n=17, 94%) vs gef (n=11, 73%) group. At data cutoff, median iDoR in the laz group was NR (8.3, NR) vs 6.3 mo (2.8, NR) in the gef group. The median best change from BL in CNS target lesion size was -57% and -47% for laz (n=17) and gef (n=14), respectively. No new safety signals were identified from LASER301. In pts with BL CNS mets, laz significantly improved iPFS vs gef with more durable responses. Laz has the potential to improve CNS outcomes in NSCLC.