Results now published in the New England Journal of Medicine show that nivolumab (Opdivo) was not superior for progression-free survival compared with chemotherapy as a first-line treatment for patients with NSCLC.
David P. Carbone, MD, PhD
Results from CheckMate-026, as previously reported and now published in the New England Journal of Medicine (NEJM), showed that nivolumab (Opdivo) was not superior for progression-free survival (PFS) compared with chemotherapy as a first-line treatment for patients with non—small cell lung cancer (NSCLC).1
With their published NEJM findings, lead study author David P. Carbone, MD, PhD, the Barbara J. Bonner Chair in Lung Cancer Research and director of the James Thoracic Center at The Ohio State University Comprehensive Cancer Center, and coauthors offered potential explanations for the PD-1 inhibitor falling short in frontline NSCLC and highlighted a subpopulation in which the treatment appeared to be more effective.
In the primary efficacy analysis population of patients with a PD-L1 expression level of ≥5% (n = 423), researchers observed no significant difference in PFS between treatment groups. Median PFS was 4.2 months (95% CI, 3.0-5.6) in the nivolumab group versus 5.9 months (95% CI, 5.4-6.9) in the chemotherapy group. The hazard ratio (HR) for PFS was 1.15 (95% CI, 0.91-1.45; P = .25).
The median overall survival (OS) was 14.4 months (95% CI, 11.7-17.4) in the nivolumab group and 13.2 months (95% CI, 10.7-17.1) in the chemotherapy group (HR for OS, 1.02; 95% CI, 0.80-1.30).
The PD-1 inhibitor appeared to produce better outcomes among patients with a high tumor mutation burden (≥243 mutations). The response rate was 47% versus 28% and the median PFS was 9.7 months versus 5.8 months (HR, 0.62; 95% CI, 0.38-1.00), among patients treated with nivolumab versus chemotherapy, respectively.
“There was no significant association between tumor-mutation burden and PD-L1 expression level,” wrote Carbone et al. “However, in the nivolumab group, patients with both a high tumor-mutation burden and a PD-L1 expression level of 50% or more had a higher response rate (75%) than those with only 1 of these factors (32% among patients with a high tumor-mutation burden only and 34% among those with a PD-L1 expression level of ≥50% only) or neither factor (16%).”
A total of 1325 patients enrolled in the phase III CheckMate-026 trial from March 2014 through April 2015. Roughly 40% of patients were randomly assigned to 3 mg/kg of nivolumab every 2 weeks (n = 271) or the investigator’s choice of platinum doublet chemotherapy every 3 weeks for 4 to 6 cycles (n = 270).
The primary efficacy analysis population made up 78% of all patients who went to randomization. Nearly all patients who went to randomization received treatment (98%).
Median time from diagnosis to randomization was 1.9 months (range, 0.3-214.9) in the nivolumab group and 2.0 months (range, 0.5-107.3) in the chemotherapy group. Roughly three-quarters of patients in the nivolumab arm and 72% of the chemotherapy arm were assigned to a treatment group within 3 months of diagnosis.
A total of 784 patients (59%) did not undergo randomization because their PD-L1 samples could not be evaluated (6% of patients), because the PD-L1 expression level was less than 1% (23%), or because they did not meet other trial criteria (30%). During screening, 71% of patients who had PD-L1 results that could be evaluated had a PD-L1 expression ≥1%.
Overall, 39% of patients had previously received radiotherapy.
Patients with nonsquamous NSCLC who had stable disease or a response after cycle 4 were allowed maintenance therapy with pemetrexed (Alimta). Treatment with nivolumab beyond progression was permitted if protocol-defined criteria were met, including investigator-assessed clinical benefit, no rapid disease progression, no unacceptable level of adverse events (AEs) related to nivolumab, a stable performance status, and if there was no interference with imminent intervention to prevent serious complications of disease progression. Concomitant systemic glucocorticoid treatment was allowed for non-autoimmune conditions, including treatment-related AEs with a potential immunologic cause.
The tumor-mutation burden, which was defined as the total number of somatic missense mutations present in a baseline tumor sample, was determined in patients with tumor and blood samples sufficient for whole-exome sequencing. Patients were classified as low (tumor-mutation burden of <0-100 mutations), medium burden (100-242 mutations), or high burden (≥243 mutations).
Baseline characteristics were generally balanced between the treatment groups, but there were more women in the chemotherapy arm (45% vs 32%). The percentage of patients with a PD-L1 expression level of 50% or more was also higher in the chemotherapy group (47% vs 32%). However, the percentage of patients with liver metastases was higher in the nivolumab arm (20% vs 13%). Patients in the nivolumab group also had a greater tumor burden.
The researchers suggest that demographics may have played a role in the results, arguing that disease characteristics associated with a better prognosis—including fewer liver metastases, smaller tumor burden, and a higher proportion of women—favored the chemotherapy group. Additionally, the high incidence of subsequent nivolumab treatment may have contributed to the favorable overall survival in the chemotherapy group.
Writing in an accompanying editorial, Edward B. Garon, MD, associate professor of hematology and oncology at UCLA’s David Geffen School of Medicine, said it was unlikely that patient selection fully accounted for the survival results.2 He noted that findings from an independent study showed that PFS was significantly longer for patients with advanced NSCLC who had been selected on the basis of PD-L1 expression level assigned first-line therapy with pembrolizumab as first-line therapy compared with chemotherapy.3
“The discrepant trial outcomes between the PD-1 inhibitors nivolumab and pembrolizumab must reflect differences in either the evaluated therapy or the studied population,” Garon wrote. “Could efficacy greatly differ between these similar drugs? In many tumor types, including previously treated NSCLC in an unselected population of patients, the 2 agents show remarkably similar results. Strong differences in efficacy that are limited to first-line therapy for NSCLC are highly improbable.”
Garon said that these results do suggest that researchers and physicians must take care to select patients who will derive the most benefit from PD-L1 inhibitors. He pointed out that the superiority of pembrolizumab over chemotherapy in previously treated patients with a PD-L1 expression level of at least 1% was based primarily on the group of patients with a PD-L1 expression level of at least 50%.
“The primary analysis in the trial conducted by Carbone et al evaluated patients with a PD-L1 expression level of at least 5%, a level seen in approximately half the screened patients,” he wrote. “The use of a cutoff point that leads to the inclusion of more patients expands the studied population at the risk of including patients who are less likely to benefit from PD-1 inhibition. Could the lower cutoff point explain the conflicting results? Not entirely, because nivolumab was not superior to chemotherapy even in patients with a PD-L1 expression level of at least 50%.”
“In light of these data, the presence of PD-L1 expression in at least 50% of tumor cells versus in less than 50% of tumor cells should be determined in patients with newly diagnosed, advanced NSCLC with the use of the assay associated with pembrolizumab efficacy until a prospectively evaluated alternative biomarker shows similar predictive value,” he said in conclusion. “The exploratory biomarker of the tumor-mutation burden that is proposed by Carbone et al is intriguing, but it is akin to an algorithm developed today that predicts last season’s World Series victory by the Cubs. Although potentially meritorious, its ability to pick this season’s champion is unclear.”