Adjuvant atezolizumab benefits PD-L1-selected patients with NSCLC
Click Here to Manage Email Alerts
CHICAGO — Updated survival data from the randomized phase 3 IMpower010 continued to support adjuvant atezolizumab for patients with stage II to stage IIIA PD-L1-expressing non-small cell lung cancer, according to researchers.
A previously observed DFS benefit with atezolizumab (Tecentriq, Genentech) — an anti-PD-L1 antibody — has translated into favorable OS outcomes for certain patients, findings presented at ASCO Annual Meeting showed.
Background and methods
The IMpower010 trial included patients with stage IB to stage IIIA NSCLC who had ECOG performance status of 0 or 1.
All 1,005 patients in the intention-to-treat population underwent complete resection 4 to 12 weeks prior to study enrollment, followed by one to four cycles of cisplatin plus pemetrexed, gemcitabine, docetaxel or vinorelbine.
After chemotherapy, researchers randomly assigned 507 patients to 1,200 mg atezolizumab every 3 weeks for 16 cycles. The other 498 received best supportive care.
Investigator-assessed DFS served as the study’s primary endpoint.
Researchers assessed DFS hierarchically, starting with patients with stage II to stage IIIA disease who had PD-L1 tumor expression on at least 1% of tumor cells, followed by all patients with stage II to stage IIIA disease randomly assigned to treatment, followed by the intention-to-treat population.
OS served as a secondary endpoint but only could be formally tested when the statistical significance boundary for DFS had been crossed in all three designated populations.
Previous results
An interim analysis — performed after data cutoff in January 2021 — showed improved DFS with atezolizumab among all patients with stage II to stage IIIA disease (stratified HR = 0.79; 95% CI, 0.64-0.96), those with stage II to stage IIIA disease with PD-L1 expression ( 1% of tumor cells, stratified HR = 0.66; 95% CI, 0.5-0.88; 50% of tumor cells, unstratified HR = 0.43; 95% CI, 0.27-0.68).
The first OS interim analysis — performed after data cutoff in April 2022 — showed a trend toward improved OS in favor of atezolizumab among those with stage II to stage IIIA and PD-L1 expression (( 1% of tumor cells, stratified HR = 0.71; 95% CI, 0.49-1.03; 50% of tumor cells, unstratified HR = 0.43; 95% CI, 0.24-0.78).
Updated results
At ASCO, Heather L. Wakelee, MD, professor of medicine at Stanford University, presented updated results from a final DFS analysis and a second OS interim analysis.
After minimum follow-up of 60 months, more than half of patients in the intention-to-treat population remained on study (59.4% for atezolizumab vs. 56.6% for best supportive care).
The difference in DFS and OS with atezolizumab did not reach statistical significance in the intention-to-treat population (stratified HR = 0.85; 95% CI, 0.71-1.01), or among all patients with stage II to stage III disease randomly assigned to treatment (stratified HR = 0.83; 95% CI, 0.69-1).
Atezolizumab conferred a significant DFS benefit among patients with stage II to stage IIIA disease who had PD-L1 expression on at least 1% of tumor cells (median, 68.5 months vs. 37.3 months; stratified HR = 0.7; 95% CI, 0.55-0.91). In this subgroup, a higher percentage of atezolizumab-treated patients remained disease free at 3 years (62.7% vs. 52.1%) and 5 years (53.2% vs. 42.7%).
Results showed a numerical improvement in OS in this subgroup (median, not estimable vs. 87.1 months; stratified HR = 0.77; 95% CI, 0.56-1.06), with higher rates of 3-year OS (82.1% vs. 78.9%) and 5-year OS (74.8% vs. 66.3%).
The benefit with atezolizumab appeared particularly strong among patients with stage II to stage IIIA disease who had PD-L1 expression on at least 50% of tumor cells. In this group, atezolizumab-treated patients achieved longer median DFS (not estimable vs. 41.1 months; unstratified HR = 0.48; 95% CI, 0.32-0.72). A higher percentage remained disease free at 3 years (74.9% vs. 53.2%) and 5 years (65.1% vs. 44.5%).
OS outcomes also favored atezolizumab in this subgroup (median, not estimable vs. 87.1 months; unstratified HR = 0.47; 95% CI, 0.28-0.77), with higher rates of 3-year OS (89.1% vs. 77.8%) and 5-year OS (82.7% vs. 65.3%).
Safety analyses showed nine patients (1.8%) assigned atezolizumab and three (0.6%) assigned best supportive care died due to adverse events.
Overall, atezolizumab exhibited a safety profile consistent with prior analyses. Researchers observed no new or unexpected safety signals.
Perspective
Back to Top
Robert Chun-Hao Hsu, MD
These results are a continuation of what we saw from the previous interim analysis. There clearly is a benefit for patients who have a PD-L1 tumor proportion score greater than 50%. There seems to be modest improvement for those with PD-L1 tumor proportion scores of 1% to 49%, but there does not seem to be a benefit for patients with scores less than 1%.
Now that we have data from multiple studies of neoadjuvant or perioperative immunotherapy, the focus going forward will be trying to determine which patients may benefit from adjuvant immunotherapy. I also wonder if using PD-L1 score is a sufficient marker to make that decision, or whether we can use technologic advances to identify other biomarkers.
Despite the increased emphasis on neoadjuvant or perioperative treatment, we do still have patients who undergo upfront resection, either due to upstaging of disease or patient preference. Consequently, there is still a role for adjuvant-only immunotherapy, so this is important information for us as clinicians to have to help make decisions about what to do in the adjuvant setting for patients who do not get neoadjuvant immunotherapy.
Collapse
Wakelee HA, et al. Abstract LBA8035. Presented at: ASCO Annual Meeting; May 31-June 4, 2024; Chicago.
Click Here to Manage Email Alerts