New Data Bolster Atezolizumab as Adjuvant Tx in NSCLC

VIENNA -- Longer follow-up in the so-called IMpower010 trial further cements atezolizumab's (Tecentriq) value in treating stage II-IIIa non-small cell lung cancer (NSCLC) in patients with tumor cells expressing the drug's PD-L1 target.

A new analysis from IMpower010, presented at the World Conference on Lung Cancer, showed a hazard ratio of 0.71 (95% CI 0.49-1.03) for all-cause mortality with the PD-L1 targeted agent versus best supportive care, as reported by Enriqueta Felip, MD, of Vall d'Hebron University Hospital in Barcelona, Spain.

With a median follow-up duration of 46 months, the 5-year overall survival (OS) rate was 76.8% for atezolizumab compared with 67.5% for supportive care. Median for overall survival was not yet reached for either arm.

These data follow an earlier interim analysis reported last year, in which adjuvant atezolizumab significantly prolonged disease-free survival (DFS) for all patients with stage II-IIIa disease (HR 0.79; 95% CI 0.64-0.96) and those with PD-L1 TC ≥1% (HR 0.66; 95% CI 0.50-0.88, P=0.004). The FDA considered those data adequate to approve the drug last October as an adjuvant treatment after resection and platinum-based chemotherapy in patients with PD-L1 TC ≥1%.

Nevertheless, data at that time were insufficient for a meaningful reading on overall survival. The new analysis, with 13 more months of follow-up, was intended to fill that gap.

The greatest OS benefit was seen in patients with PD-L1 TC ≥50% (HR 0.43; 95% CI 0.24-0.78). When the data were analyzed for all comers in the trial, which also included patients with PD-L1 TC levels below 1%, no benefit from atezolizumab was seen either for those with stage II-IIIA disease (HR 0.95; 95% CI 0.74-1.24) or in the intent-to-treat (ITT) population of stage IB-IIIA disease (HR 0.995; 95% CI 0.78-1.28).

"These data support the previously reported positive benefit-risk profile of adjuvant atezolizumab in PD-L1 positive resected NSCLC," said Felip, who noted that the OS data are still not fully mature.

The international IMpower10 trial enrolled patients with operable stage Ib-IIIa NSCLC with adequate tissue for PD-L1 expression analysis. After resection, 1,005 patients were randomly assigned to 1,200 mg atezolizumab every 21 days or best supportive care. DFS was the primary outcome measure, with OS as a key secondary endpoint.

Analysis of patients with stage II-IIIa disease and PD-L1 TC ≥1% showed a trend toward improved OS in the majority of subgroups analyzed, with the exception of patients treated with cisplatin plus gemcitabine.

Overall, Felip said, even with the additional 13 months of follow-up, the safety profile of atezolizumab remained mostly unchanged with no new or unexpected safety signals.

Adverse events occurred in almost twice as many patients assigned to atezolizumab as compared with best supportive care (22% vs 11.5%) and resulted in atezolizumab discontinuation in 18.2% of patients assigned to the drug. Fatal treatment-related adverse events occurred in 0.8% of patients in the atezolizumab arm compared with no one in the best supportive care arm.

Commenting on the results, Benjamin Besse, MD, PhD, of Gustave Roussy Cancer Campus in Villejuif, France, listed several pieces of data he was still eager to see related to these results. Specifically, when looking at OS results, Besse said it is always important to look at crossover data.

"How many patients in the control arm received atezolizumab or any other immunotherapy?" Besse said. "This has not been reported."

In addition, Besse wanted to see more data on other patient subgroups. The presentation did not show a breakdown of survival benefit looking only at patients with PD-L1 TC 1%-49% or patients whose tumors were PD-L1 negative.

"I would like very much to see if these subgroups in the control arm did not over-perform," Besse said.

IMpower010 will continue to the final DFS analysis with further OS follow-up and analyses. Its Clinicaltrials.gov listing indicates that final completion is expected in 2027.

Comment