Pooled Results From Atezolizumab Trials Show Immune-Related AEs Can Predict Longer OS

Data from IMpower 130, IMpower 132, and IMpower 150 confirm that combining atezolizumab with chemotherapy is the best first-line treatment in advanced NSCLC when anti-PD-L1 therapy is indicated.

The rise of immunotherapy in cancer care has brought with it the challenge of managing adverse events (AEs). But immune-related AEs are known to be a sign that the treatment is working, and an pooled analysis of phase 3 trials involved atezolizumab (Tecentriq), presented during the 2021 American Society of Clinical Oncology Annual Meeting, show that the presence of low-grade AEs can, in fact, predict longer overall survival (OS).

Atezolizumab is currently approved in the first- and second-line settings for advanced non-small cell lung cancer (NSCLC). In an exploratory analysis, investigators pooled data from the IMpower 130, the IMpower 132, and the IMpower 150 studies, which evaluated the anti-PD-L1 agent in various combinations in treatment-naïve patients with nonsquamous stage IV NSCLC. The combinations were:

  • IMpower 130: patients were randomized to carboplatin and nab-paclitaxel alone or with atezolizumab
  • IMpower 132: patients were randomized to carboplatin or cisplatin alone or with atezolizumab
  • IMpower 150: patients were randomized to atezolizumab plus bevacizumab plus carboplatin and paclitaxel, or atezolizumb plus carboplatin and paclitaxel, or bevacizumab plus carboplatin plus paclitaxel.

Analyses of immune-related AEs and survival took place at 1, 3, 6, and 12 months. In total, 2503 patients were included; 1577 in the atezolizumab arm and 926 in the control arm. Study protocols called for treatment interruption or discontinuation of AEs were grade 3 or higher.

Results showed:

  • 48% of patients in the atezolizumab arm had an immune-related AE, and 32% in the control arm did so.
  • 11% of those in the atezolizumab arm had immune-related AEs grade 3-5, compared with 5% of those in the control arm.
  • The most common immune-related AEs were rash (28% for atezolizumab, vs 18% for control), hepatitis (15% vs 10%), and hypothyroidism (12% vs 4%). Median time to onset was 1.7 months for the atezolizumab group vs 1.4 months for control.
  • OS hazard ratios (HRs) between patients with immune-related AEs and those without were 0.69 for the atezolizumab arm and 0.82 for the control arm; excluding rash, the OS HRs were 0.75 for the atezolizumab arm and 0.90 for the control.

The authors concluded that patients with grade 1 and 2 immune-related AEs had the longest OS and those with grade 3 or higher had the shortest, possibly due to discontinuation of therapy. They wrote that the data offer additional support for atezolizumab with chemotherapy in the first-line setting, with or without bevacizumab.

F. Hoffman-LaRoche funded the study.


Socinski MA, Jotte RM, Cappuzzo F, et al. Pooled analyses of immune-related adverse events (irAEs) and efficacy from thephase 3 trials IMpower130, IMpower132, and IMpower150. J Clin Oncol. 2021; 39(Suppl_15): abstr 9002. DOI: 10.1200/JCO.2021.39.15_suppl.9002