Association of Immune-Related Adverse Events With Efficacy of Atezolizumab in Patients With Non–Small Cell Lung Cancer

<jats:sec><jats:title>Importance</jats:title><jats:p>Immune-related adverse events (irAEs) arising from immune checkpoint inhibitor (ICI) cancer therapy may potentially predict improved outcomes.</jats:p></jats:sec><jats:sec><jats:title>Objective</jats:title><jats:p>To evaluate the association between irAEs and atezolizumab efficacy in patients with advanced non–small cell lung cancer (NSCLC) using pooled data from 3 phase 3 ICI studies.</jats:p></jats:sec><jats:sec><jats:title>Design, Setting, and Participants</jats:title><jats:p>IMpower130, IMpower132, and IMpower150 were phase 3, multicenter, open-label, randomized clinical trials to evaluate the efficacy and safety of chemoimmunotherapy combinations involving atezolizumab. Participants were chemotherapy-naive adults with stage IV nonsquamous NSCLC. These post hoc analyses were conducted during February 2022.</jats:p></jats:sec><jats:sec><jats:title>Interventions</jats:title><jats:p>Eligible patients were randomly assigned 2:1 to receive atezolizumab with carboplatin plus nab-paclitaxel, or chemotherapy alone (IMpower130); 1:1 to receive atezolizumab with carboplatin or cisplatin plus pemetrexed, or chemotherapy alone (IMpower132); and 1:1:1 to receive atezolizumab plus bevacizumab plus carboplatin and paclitaxel, atezolizumab plus carboplatin and paclitaxel, or bevacizumab plus carboplatin and paclitaxel (IMpower150).</jats:p></jats:sec><jats:sec><jats:title>Main Outcomes and Measures</jats:title><jats:p>Pooled data from IMpower130 (cutoff: March 15, 2018), IMpower132 (cutoff: May 22, 2018), and IMpower150 (cutoff: September 13, 2019) were analyzed by treatment (atezolizumab-containing vs control), irAE status (with vs without), and highest irAE grade (1-2 vs 3-5). To account for immortal bias, a time-dependent Cox model and landmark analyses of irAE occurrence at 1, 3, 6, and 12 months from baseline were used to estimate the hazard ratio (HR) of overall survival (OS).</jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p>Of 2503 randomized patients, 1577 were in the atezolizumab-containing arm and 926 were in the control arm. The mean (SD) age of patients was 63.1 (9.4) years and 63.0 (9.3) years, and 950 (60.2%) and 569 (61.4%) were male, respectively, in the atezolizumab arm and the control arm. Baseline characteristics were generally balanced between patients with irAEs (atezolizumab, n = 753; control, n = 289) and without (atezolizumab, n = 824; control, n = 637). In the atezolizumab arm, OS HRs (95% CI) in patients with grade 1 to 2 irAEs and grade 3 to 5 irAEs (each vs those without irAEs) in the 1-, 3-, 6-, and 12-month subgroups were 0.78 (0.65-0.94) and 1.25 (0.90-1.72), 0.74 (0.63-0.87) and 1.23 (0.93-1.64), 0.77 (0.65-0.90) and 1.1 (0.81-1.42), and 0.72 (0.59-0.89) and 0.87 (0.61-1.25), respectively.</jats:p></jats:sec><jats:sec><jats:title>Conclusions and Relevance</jats:title><jats:p>In this pooled analysis of 3 randomized clinical trials, longer OS was observed in patients with vs without mild to moderate irAEs in both arms and across landmarks. These findings further support the use of first-line atezolizumab-containing regimens for advanced nonsquamous NSCLC.</jats:p></jats:sec><jats:sec><jats:title>Trial Registration</jats:title><jats:p>ClinicalTrials.gov Identifiers: <jats:ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="uri" xlink:href="https://www.clinicaltrials.gov/ct2/show/NCT02367781">NCT02367781</jats:ext-link>, <jats:ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="uri" xlink:href="https://clinicaltrials.gov/ct2/show/NCT02657434">NCT02657434</jats:ext-link>, and <jats:ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="uri" xlink:href="https://clinicaltrials.gov/ct2/show/NCT02366143">NCT02366143</jats:ext-link></jats:p></jats:sec>