No DFS Benefit With Adjuvant Durvalumab in Early NSCLC

Adjuvant durvalumab (Imfinzi) provided no significant improvement in disease-free survival compared with placebo following resection in patients with early-stage non-small cell lung cancer (NSCLC), according to findings from the phase 3 BR.31 trial.

In the overall population, median disease-free survival (DFS) was about 6 months longer in the durvalumab arm, but the difference was not statistically significant. Even when separating patients by programmed death-ligand 1 (PD-L1) expression, DFS outcomes were not significantly better in the durvalumab group. 

These findings, presented in a late-breaking abstract session at the 2024 European Society for Medical Oncology Congress, muddy the waters on the role of adjuvant immunotherapy in early-stage NSCLC. 

Two recent studies showed a benefit with adjuvant atezolizumab (Tecentriq) and pembrolizumab (Keytruda) in a similar patient population. The IMpower010 trial, for instance, demonstrated that, in patients with stage II-IIIA NSCLC, adjuvant atezolizumab led to a significant improvement in DFS compared with best supportive care, and patients with PD-L1 expression of 50% or higher had an overall survival benefit. The KEYNOTE-091 trial found that adjuvant pembrolizumab significantly improved DFS vs placebo in patients with stage IB, II, or IIIA NSCLC who had surgery.

In the current BR.31 trial, it’s unclear why the adjuvant durvalumab and placebo groups had similar DFS outcomes after 36 months, noted Glenwood Goss, MD, of the University of Ottawa, Canada, who presented the findings at ESMO.

Overall, the impetus for this study comes from a need to improve survival outcomes in early-stage NSCLC. 

Studies show that adjuvant chemotherapy improves 5-year survival by 5% in resected NSCLC, but 60% of patients relapse and die, explained Goss. 

Given that durvalumab has been approved for metastatic, unresectable stage III NSCLC and in the perioperative setting in resectable NSCLC, the BR.31 investigators wanted to evaluate this agent in the adjuvant setting in early-stage disease.

The study included 1219 adults (815 in the durvalumab group, 404 placebo) with histologically confirmed, resected stage IB, II, or IIIA NSCLC and no EGFR or ALK mutations. Patients were from 269 sites in 19 countries and were randomly assigned in a 2:1 ratio to durvalumab every 4 weeks for 12 months or placebo. 

Patients’ median age was 64 years, 66% were men, and 62% had adenocarcinoma. Most (84%) received adjuvant chemotherapy. The two arms were well balanced, except that slightly more patients in the placebo arm had squamous cell carcinoma, compared with the durvalumab arm.

The primary patient population evaluated in the study had PD-L1 expression of 25% or higher, and the secondary analysis included patients with lower or no PD-L1 expression.

In the overall population, durvalumab did not lead to a significant improvement in median DFS, regardless of PD-L1 expression. Overall, median DFS was about 60 months in patients receiving durvalumab compared with 53.9 months in those on placebo (hazard ratio [HR], 0.89; P = .207).

Median DFS in patients with PD-L1 expression of 25% or higher was 70 months in the durvalumab group vs 60 months in the placebo group, but the difference was also not statistically significant (HR, 0.94; 95% CI, 0.71 – 1.25; P = .64). 

And in those with PD-L1 expression of at least 1%, median DFS was the same in both groups — 60 months (HR, 0.99; P =. 93).

The safety profile of durvalumab in this trial was similar to that observed in other large, randomized phase 3 trials, Goss noted.

In the BR.31 trial, adverse events of grade 3 or 4 occurred in 23.5% of patients on durvalumab and 19.6% in the placebo group. Serious adverse events possibly related to treatment occurred in about 10% of patients in the durvalumab group and 3.6% in the placebo arm. Treatment discontinuation possibly related to adverse events occurred in 12% of those on durvalumab and 2.8% in the placebo group.

In the current trial, it’s notable that the control group had much better responses compared with the control groups from KEYNOTE-091 and IMpower-010, said Goss. 

Cross-trial comparisons show similar DFS trends at 12, 24, and 36 months, but after about 36 months, the placebo arm in the BR.31 trial continued to do well, compared with the placebo arm in the other trials, which separated from the experimental arms.

“I think that probably accounts for the negative trial,” said Goss. As to why the placebo arm outcome was better in this trial, Goss said, “It may be the patient population, the surgical technique — which we haven’t looked into — or a number of possibilities.”

Regardless, the outcomes of this trial suggest that PD-L1 tumor score is a poor biomarker for DFS, and that the presence of primary disease and associated tumor antigen, as with the perioperative treatment approach, may be required for optimal efficacy in NSCLC, he said.

Given that KEYNOTE-091 and Impower-010 showed a DFS benefit with adjuvant immunotherapy, but now BR.31 shows no benefit, “things are getting confusing,” a meeting attendee said. 

So, how should clinicians move forward in practice? he asked Goss.

“We have to do what we always do,” Goss replied. “We have to explain the result to our patients, and we have to take into consideration the clinical picture and what their desires are when they are fully informed.”

The negative results also underscore the need for predictive biomarkers in that setting, said study discussant Pilar Garrido Lopez, MD, PhD, a medical oncologist and department head at Ramon y Cajal University Hospital, Madrid, Spain. 

On the plus side, she highlighted the increasing number of treatment options and advances.

“This is an exciting time with many options for our patients,” Lopez said. “We are entering a new era, not only because we have many options — adjuvant, neoadjuvant, perioperative — but also because vaccines are entering this scenario.”

Goss reported financial relationships with AstraZeneca and Merck. Lopez reported personal financial relationships with AbbVie, Amgen, AstraZeneca, Bayer, BeiGene, BMS, Boehringer Ingelheim, Daiichi Sankyo, GlaxoSmithKline, J&J, Lilly, MSD, Novartis, Pfizer, Roches, Regeneron, Sanofi, and Takeda. She also reported direct funding from Medscape and Touch Medical, as well as institutional research funding from multiple companies. 

Sharon Worcester, MA, is an award-winning medical journalist based in Birmingham, Alabama, writing for Medscape, MDedge, and other affiliate sites. She currently covers oncology, but she has also written on a variety of other medical specialties and healthcare topics. She can be reached at sworcester@mdedge.com or on Twitter: @SW_MedReporter.