Adjuvant Durvalumab Does Not Improve DFS Across NSCLC PD-L1 Subgroups

“The outcomes of the BR.31 study suggest that the presence of primary disease and associated tumor antigens, as in the perioperative approach, may be required for optimal efficacy [in NSCLC],” according to Glenwood Goss, MB, BCh, FCPSA, FRCPC.

Disease-free survival (DFS) was not improved when adjuvant durvalumab (Imfinzi) was given after complete resection and optional chemotherapy vs placebo for patients with EGFR- or ALK-negative non–small cell lung cancer (NSCLC) across PD-L1 expression subgroups, according to results from the phase 3 BR.31 trial (NCT02273375) presented at the 2024 European Society of Medical Oncology (ESMO) Congress.

The median DFS in patients with PD-L1 expression of 25% or more in the durvalumab arm was 69.9 months (95% CI, 57.6-not reached [NR]) vs 60.2 months (95% CI, 47.7-NR) in the placebo arm (HR, 0.935; 95% CI, 0.706-1.247; P = .642). The 18-month DFS rate in the durvalumab arm was 75.1% (95% CI, 69.9%-79.6%) vs 70.5% (95% CI, 62.5%-77.1%) in the placebo arm. The 24-month rate was 71.2% (95% CI, 65.7%-75.9%) vs 68.5% (95% CI, 60.4%-75.3%), and the 36-month rate was 63.9% (95% CI, 58.2%-69.0%) vs 62.4% (95% CI, 54.1%-69.6%).

The median DFS in patients with PD-L1 expression of 1% or more was 59.9 months (95% CI, 48.4-77.9) in the durvalumab arm and 60.3 months (95% CI, 43.8-80.9) in the placebo arm (HR, 0.989; 95% CI, 0.788-1.248; P = .926). The 18-month DFS rates were 73.4% (95% CI, 69.0%-77.2%) vs 70.1% (95% CI, 63.6%-75.7%); 24-month rates were 68.6% (95% CI, 64.1%-72.7%) vs 67.0% (95% CI, 60.4%-72.7%); and 36-month rates were 60.2% (95% CI, 55.4%-64.5%) vs 60.1% (95% CI, 53.4%-66.3%), in the durvalumab and placebo arm, respectively.

For patients in the PD-L1 all-comer population, the median DFS was 60.0 months (95% CI, 49.6-74.9) in the durvalumab arm and 53.9 months (95% CI, 36.7-67.3) in the placebo arm (HR, 0.893; 95% CI, 0.752-1.065; P = .207). The DFS rates at 18 months were 72.1% (95% CI, 68.8%-74.9%) vs 66.0% (95% CI, 60.9%-70.6%); at 24-months, they were 67.4% (95% CI, 64.0%-70.6%) vs 63.3% (95% CI, 58.1%-68.0%); and at 36-months, they were 60.4% (95% CI, 56.8%-63.8%) vs 56.4% (95% CI, 51.1%-61.3%).

“The outcomes of the BR.31 study suggest that the presence of primary disease and associated tumor antigens, as in the perioperative approach, may be required for optimal efficacy [in NSCLC],” Glenwood Goss, MB, BCh, FCPSA, FRCPC, a professor of medicine in the University of Ottawa Division of Medical Oncology, a chair of the Thoracic Oncology Site Committee, and director of Clinical and Translational Research at Ottawa Hospital Cancer Centre, said in a presentation on these data.

Patients with stage IB to IIIA NSCLC who had complete resection, an ECOG performance status of 0 to 1, and EGFR-mutated/ALK-positive disease were eligible to enroll. Patients received a platinum doublet followed by surgery and randomization at 3 weeks or more. Patients were randomly assigned 2:1 to durvalumab at 20 mg/kg every 4 weeks for 12 months or matched placebo.

The primary end point was investigator-assessed DFS in patients with PD-L1 expression of 25% or more and EGFR- or ALK-negative disease. Secondary end points included DFS in patients who had PD-L1 expression of 1% or more and EGFR- or ALK-negative disease, all patients with PD-L1 expression of 25% or more, all randomly assigned patients, PD-L1 all comers with EGFR- or ALK-negative disease, and all patients with PD-L1 expression of 1% or more; overall survival; adverse effects (AEs); and quality of life.

Overall, 1827 patients registered, with randomization of 1415 taking place between February 2015 and March 2020. There was a total of 1219 patients with EGFR- or ALK-negative disease. Of the 1415 patients who were randomly assigned, 944 and 471 were assigned to the durvalumab arm and the placebo arm, respectively. At data cutoff, 67.7% of patients in the durvalumab arm and 67.5% in the placebo arm were still receiving study treatment.

In the PD-L1 expression of 25% or more group, the median age was 65 years old vs 63 years old in the durvalumab vs placebo arms, 61.1% vs 64.6% were male, 47.5% vs 41.0% were White, and 81.6% vs 79.5% were former smokers. Additionally, the most common histology type was adenocarcinoma in 63.0% vs 59.0%, 54.4% vs 50.9% had stage II disease, and 65.8% vs 64.6% had PD-L1 expression of 50% or more.

In the PD-L1 expression of 1% or more group, the median age was 65 years old vs 63 years old in the durvalumab and placebo arms, with 62.5% vs 64.6% being male, 46.1% vs 42.1% were White, and 78.3% vs 77.5% were former smokers. The most common histology type was adenocarcinoma in 60.8% vs 58.8%, and 44.3% vs 43.3% had PD-L1 expression of 50% or more.

In the PD-L1 all-comers group, the median age was 64 years vs 64 years in each arm, respectively; 64.8% vs 66.8% were male, 46.1% vs 46.5% were White, and 78.4% vs 76.5% were formers smokers. The most common histology type was adenocarcinoma in 66.3% vs 60.6%, and 42.5% vs 40.6% had PD-L1 expression of less than 1%.

The safety analysis included all patients who had received at least 1 dose of treatment. Any AEs occurred in 93.8% in the durvalumab arm and 92.3% in the placebo arm. Grade 3/4 AEs occurred in 23.5% vs 19.6%, and AEs leading to death occurred in 0.7% vs 0.2% between either arm. Serious AEs were observed in 18.8% vs 15.4%, and AEs leading to discontinuation occurred in 14.0% vs 5.1% in the durvalumab and placebo arms, respectively.

Reference

Goss G. CCTG BR.31: a double-blind placebo-controlled randomized phase 3 trial of adjuvant durvalumab in completely resected non-small cell lung cancer. Presented at the 2024 European Society of Medical Oncology (ESMO) Congress, Barcelona, Spain; September 13-17, 2024. LBA48.