The evidence supporting this protocol is provided by a phase III multicentre, international, randomised, open-label trial (IMpower 010) involving 1280 patients comparing adjuvant atezolizumab with best supportive care (BSC) in patients with completely resected stage IB-IIIA NSCLC who received at least 1 cycle (up to 4 cycles) of investigator’s choice of cisplatin-based doublet chemotherapy (cisplatin plus either vinorelbine, gemcitabine or docetaxel; or plus pemetrexed for patients with non-squamous NSCLC).r
Between October 2015 and September 2018, 507 patients were randomised to receive atezolizumab (1200 mg every 3 weeks for 16 cycles or 1 year) and 498 patients were randomised to receive BSC.
The primary end point was investigator-assessed disease-free survival (DFS) tested hierarchically in 3 pre-specified populations:
- stage II-IIIA with PD-L1 expression of ≥1%
- all patients in the stage II-IIIA population
- intention-to-treat (ITT) population (stage IB-IIIA).
Secondary end points were overall survival (OS) in the ITT population, DFS in the stage II-IIIA population with PD-L1 ≥50%, and the 3-year and 5-year DFS in all 3 populations.
Efficacy
At the interim analysis after a median follow up of 32.2 months in the stage II–IIIA population, atezolizumab improved DFS compared with BSC with a stratified hazard ratio (HR) of 0.66 (95% CI 0.50-0.88, p=0.0039) in PD-L1 ≥1% population and 0.79 (95% CI, 0.64-0.96, p=0.020) in all stage II-IIIA population. DFS benefit for atezolizumab was not statistically significant in the ITT population (HR=0.81; 95% CI 0.67-0.99, p=0.040).r
The 3-year DFS in the stage II-IIIA with PD-L1 expression ≥1%, all stage II-IIIA, and ITT populations for atezolizumab were 60%, 56%, and 58% vs 48%, 49% and 53% for BSC, respectively.r
In the stage II-IIIA with PD-L1 ≥50% population, atezolizumab improved DFS compared to BSC (unstratified HR=0.43, 95% CI, 0.27-0.68). However, for patients with PD-L1 1-49% and PD-L1 <1% in this population, atezolizumab did not show a benefit as DFS was similar in both groups (unstratified HR=0.87, 95% CI, 0.60-1.26 and unstratified HR=0.97, 95% CI, 0.72-1.31; respectively).r
Atezolizumab did not show a DFS benefit in the EGFR and ALK-positive subgroups.r
At subsequent analysis in April 2022, with a median follow up of 45.3 months in the ITT population, OS data remains immature. While OS cannot be formally tested in the ITT population, exploratory analysis showed an OS improvement in favour of atezolizumab vs BSC in the stage II-IIIA PD-L1 ≥1% population [stratified HR 0.71 (95% CI, 0.49-1.03)]. While not powered for subgroup analysis, the OS benefit for atezolizumab vs BSC was strongest in the stage II-IIA PD-L1 >50% population [unstratified HR 0.43 (95% CI, 0.24-0.78)] which remains when patients with known ALK/EGFR alterations were excluded. No OS improvement was seen in ITT, stage II-IIIA populations or in the stage II-IIIA PD-L1 <1% population. Overall, the data appears to reflect the DFS data.r
Quality of life data was not collected.
Kaplan-Meier curve of OS in stage II-IIIA PD-L1 >50% (excluding patients with known EGFR/ALK alterations)r

©Annal Oncol 2023
5 year follow up data (presented in poster form) shows similar results, with median OS remaining immature in the ITT group, as seen in the below table.r
5 year follow up data (abstract)r

©J Clin Oncol 2024
Toxicity
Grade 3/4 adverse events occurred in 22% of patients receiving atezolizumab and 11.5% of patients receiving BSC. Treatment-related adverse events occurred in 68% of patients in atezolizumab group (11% with grade 3/4 and 1% with grade 5 severity). Treatment-related adverse events led to atezolizumab discontinuation in 18% of patients, most frequently due to pneumonitis, hypothyroidism and increased aspartate aminotransferase (1% each).r
Summary of adverse eventsr

© Lancet 2021