IMpower131: Primary PFS and safety analysis of a randomized phase III study of atezolizumab + carboplatin + paclitaxel or nab-paclitaxel vs carboplatin + nab-paclitaxel as 1L therapy in advanced squamous NSCLC.

LBA9000 Background: Atezolizumab (atezo; anti–PD-L1) demonstrated OS benefit vs docetaxel in 2L+ NSCLC regardless of PD-L1 status or tumor histology. Because cytotoxic agents can exhibit positive immunomodulatory effects, combining atezo with chemotherapy may further improve outcomes. IMpower131 (NCT02367794) was designed to evaluate atezo + carboplatin (carbo) + paclitaxel (pac) or nab-paclitaxel (nab-pac) in 1L stage IV squamous NSCLC. Methods: Patients (pts) were randomized 1:1:1 to Arm A (atezo 1200 mg q3w + carbo AUC 6 q3w + pac 200 mg/m 2 q3w), Arm B (atezo + carbo + nab-pac 100 mg/m 2 weekly) or Arm C (carbo + nab-pac). Pts received chemotherapy for 4 or 6 cycles and atezo until loss of clinical benefit. We present the primary analysis of investigator (INV)-assessed PFS per RECIST v1.1 in the ITT population for Arm B vs Arm C. Data cutoff: 22 January 2018. Results: 338 pts (Arm A), 343 pts (Arm B) and 340 pts (Arm C) were enrolled. Minimum follow-up was 9.8 mo. INV-assessed median PFS was 6.3 mo in Arm B vs 5.6 mo in Arm C (HR, 0.715; 95% CI: 0.603, 0.848; P = 0.0001). PFS benefit was enriched in all PD-L1–positive IHC subgroups and was most pronounced in TC3 or IC3. AEs related to any treatment (TRAEs) were 91.3% (Arm A), 94.6% (Arm B) and 90.7% (Arm C); Grade 3-4 TRAEs were 42.8% (Arm A), 68.0% (Arm B) and 56.9% (Arm C); serious TRAEs were 22.3% (Arm A), 20.4% (Arm B) and 10.5% (Arm C). Preliminary OS data will be presented. Conclusions: IMpower131 met its co-primary endpoint of INV-assessed PFS in the ITT population in Arm B vs Arm C. The safety profile of atezo + carbo + nab-pac or pac was consistent with the known risks of the individual treatment components; no new safety signals were identified. Clinical trial information: NCT02367794. Table: see text