SWOG/NRG S1914: Randomized phase III trial of induction/consolidation atezolizumab + SBRT versus SBRT alone in high risk, early-stage NSCLC.

8003 Background: Stereotactic body radiation therapy (SBRT) is the standard of care (SoC) for early stage, medically inoperable non-small cell lung cancer (NSCLC). While rates of in-field control exceed 90%, regional and distant control after SBRT remain suboptimal. A prior phase II randomized trial suggested a benefit to adding immunotherapy (PMID 37478883). SWOG/NRG S1914 (NCT#04214262) is a randomized phase III trial evaluating neoadjuvant, concurrent and adjuvant atezolizumab plus SBRT for early-stage NSCLC vs SoC. Methods: Eligible patients (pts) had T1-3N0M0 NSCLC ≤7cm, were medically inoperable or declined surgery, and had ≥1 risk factor for increased recurrence: tumor diameter ≥2 cm, ≥6.2, moderately/poorly/undifferentiated histology. Randomization was to SoC SBRT (S 3-8 fractions, biologically effective dose ≥100 Gy) or neoadjuvant, concurrent and adjuvant atezolizumab (AS 1200 mg IV Q3 week, 8 cycles) with SBRT initiated with cycle 3, stratifying by tumor location (central vs peripheral), size (<4 cm vs ≥4cm) and ECOG performance status (PS, 0-1 vs 2). The primary objective was to compare overall survival (OS) between the arms. Secondary objectives included comparisons of progression free survival (PFS), failure patterns, toxicity and quality of life (QoL). OS and PFS were compared using a 1-sided stratified log-rank test at the 2.5% level, confidence intervals (CI) are 95%. The accrual goal was 432 eligible pts. Results: From 8/13/20 - 9/6/24, 417 pts were randomized, 403 met eligibility 201 to S, 202 to AS. Accrual closed at the first interim analysis for futility based on OS and PFS per design. Median follow-up for pts still alive was 12 (range: 0.03-49) months. Median age was 73 (41-91) years and 89% had PS 0-1. Median tumor diameter was 2.3 cm. No protocol treatment was received for 6 pts on S and 8 on AS. With 49 deaths, OS was not different between the arms (HR (CI): 1.15(0.65-2.01), p=0.63; 2-year OS: 82% S vs 80% AS). With 88 events, PFS was not better with AS (HR (CI): 1.35(0.89-2.06), p=0.16); 2-year PFS was 71% on S vs 60% on AS. Regional (2% vs 3%) and distant (4% vs 5%) failures were not different; there were more local failures with AS (13% vs 7%). Among former (53%)/never (3%) smokers, AS had worse OS and PFS than S (HR(CI): 2.50 (1.11-5.59), p=0.03); HR(CI): 2.16(1.15-4.04), p=0.01), respectively. Grade (G) ≥3 adverse event (AE) rates were 12% on AS (N=21 G3, 1 G4, 1 G5 respiratory failure) vs 2% on S (N=3 G3, 1 G4). Conclusions: In the first reported phase III trial to assess immunotherapy (IO) added to SBRT in early-stage NSCLC, IO failed to improve survival. More G ≥3 adverse events were reported with AS. Central review of local recurrence events is ongoing. Additional investigation into subgroups, PD-L1 status, QoL and blood/tissue are pending to determine whether there are subsets who can benefit from this combination and shed further insights into these findings. Clinical trial information: NCT04214262 .