No difference in relapse pattern in patients relapsing after atezolizumab or best supportive care after adjuvant chemotherapy in stage IB-IIIA NSCLC in IMpower010

At an interim disease-free survival analysis of the IMpower010 study, relapse rates were found to be higher in the best supportive care (BSC) arm, as compared to the atezolizumab arm. However, there was no clear difference in pattern of relapse between the study arms among patients with a relapse. As expected, post-relapse immunotherapy use was higher in the BSC arm  than in the atezolizumab arm.

Up to 60% of patients with stage I-III non-small cell lung cancer (NSCLC) still experience disease relapse despite treatment with curative intent, demonstrating the unmet medical need in this setting. IMpower010 is the first phase III study of cancer immunotherapy to demonstrate a disease-free survival (DFS) improvement in the adjuvant NSCLC setting after complete surgical resection and platinum-based chemotherapy. At ESMO 2021, Dr. Felip reported on the sites of disease relapse and post-relapse treatment in IMpower010 at the time of the DFS interim analysis.

IMpower010

IMpower010 enrolled 1,280 patients of whom 1,269 received up to four 21-day cycles of adjuvant chemotherapy (cisplatin, partnered with either pemetrexed, docetaxel, gemcitabine or vinorelbine). In total, 1,005 of these patients were subsequently randomised (1:1) to sixteen cycles of atezolizumab (1,200 mg Q3W), or best supportive care (BSC). All eligible patients had completely resected (4-12 weeks prior to enrolment) stage IB (≥4 cm)-IIIA NSCLC (AJCC/UICC v7) and had an ECOG performance status of 0-1. The primary endpoint of investigator-assessed DFS and secondary endpoint of overall survival (OS) were tested hierarchically: first DFS in the PD-L1 tumour cell (TC) ≥1% (SP263) subgroup with Stage II-IIIA disease, if positive DFS in all randomised patients with stage II-IIIA disease, and finally DFS in the ITT population (stage IB-IIIA). Only if the latter analysis was positive, the OS could be tested in the ITT population.

Results

As previously reported, the DFS significance boundary was crossed in PD-L1 TC ≥1% stage II-IIIA (HR[95%CI]: 0.66[0.50-0.88], p= 0.004) and all randomised stage II-IIIA (HR[95%CI]: 0.79[0.64-0.96], p= 0.02) patients. In all randomised stage II-IIInA patients, DFS improvement (HR[95% CI]) was seen with increasing PD-L1 expression, with hazard ratios of 0.97, 0.87 and 0.43 for PD-L1 TC <1%, PD-L1 TC 1-49% and PD-L1 TC ≥50% subgroups, respectively. In a post-hoc analysis excluding patients with known EGFR/ALK positive NSCLC, DFS hazard ratios were numerically improved in most PD-L1 subgroups.

Among PD-L1 TC ≥1% stage II-IIIA patients, 29.4% of patients relapsed in the atezolizumab arm vs. 44.7% in the BSC arm. Corresponding disease relapse rates for the all randomised stage II-IIIA and ITT population were 33.3% vs. 43.0% and 30.8% vs. 40.8%, respectively. Overall, there were no clear differences in the patterns of relapse (locoregional only, distant only, locoregional and distant, central nervous system only or second primary lung) between the PD-L1 TC ≥1% stage II-IIIA, all randomised stage II-IIIA and ITT populations. Approximately 37% of patients in each treatment arm had locoregional relapse only, 42% distant relapse only and 18% locoregional and distant. For patients in the PD-L1 TC ≥1% stage II-IIIA population, there was a trend towards a delay in median time to relapse for patients in the atezolizumab arm, with a median time to any relapse of 17.6 months vs. 10.9 months. Approximately 65% of patients received post-relapse systemic anticancer therapy. As expected, more patients in the BSC arm received immunotherapy (32%) as compared to patients in the atezolizumab arm (12%). There were no major differences in the use of radiotherapy or surgery post-relapse between the two treatment arms and the complete study population.

Conclusion

IMpower010 is the first positive phase III study of adjuvant immunotherapy after surgical resection and adjuvant chemotherapy in patients with early-stage NSCLC, showing a 34% reduction in risk of disease recurrence or death with adjuvant atezolizumab in the PD-L1 TC ≥1 % stage II-IIIA population, potentially challenging the standard of care in this population. At this DFS interim analysis, similar patterns of relapse were seen between study arms. Furthermore, the time to relapse appeared to favour the atezolizumab vs. BSC arm in the PD-L1 TC ≥1 % stage II-IIIA population, with minimal differences seen in the all-randomised stage II-IIIA and ITT populations. Finally, a higher rate of post-relapse immunotherapy use was observed in the BSC arm. Longer term follow-up is warranted and may reveal differences in relapse patterns and treatment options.

Reference

Felip E, Vallieres E, Zhou C, et al. IMpower010: sites of relapse and subsequent therapy from a Phase 3 study of atezolizumab vs best supportive care after adjuvant chemotherapy in stage IB-IIIA NSCLC. Presented at ESMO 2021; Abstract LBA9.