Nivolumab Misses PFS Endpoint in NSCLC Compared With Chemotherapy

Mark Socinski, MD

Nivolumab-treated patients had a median PFS of 4.2 months compared with 5.9 months for patients treated with a chemotherapy doublet chosen by the treating physician. Overall survival (OS) and response rate also did not differ significantly between treatment groups.

More than 60.4% of patients in the chemotherapy arm crossed over to nivolumab at progression, as compared with 43.6% of patients in the nivolumab arm who crossed over to chemotherapy. Whether that difference influenced the results has yet to be determined, as reported at the 2016 ESMO Congress.

“Nivolumab did not meet the primary endpoint of superior PFS compared with chemotherapy,” said Mark Socinski, MD, executive medical director of Florida Hospital Cancer Institute in Orlando. “Safety results were consistent with the known safety profile of nivolumab: there were fewer treatment-related grade 3/4 adverse events in the nivolumab versus chemotherapy arm. Exploratory subgroup analyses were consistent with the overall study.”

The randomized CheckMate-026 trial failed to live up to expectations created by the phase I CheckMate-012 trial of first-line nivolumab (ASCO 2016, Abstract 3001). The CheckMate-012 results suggested first-line nivolumab had clinical activity with increasing levels of PD-L1 expression, although investigators also observed activity in patients with low-level or no PD-L1 expression, said Socinski.

Already approved for use in patients with previously treated advanced NSCLC, nivolumab underwent evaluation of its first-line potential in CheckMate-026. The international, multicenter, open-label trial examined the safety and efficacy of the PD-1 inhibitor as front-line treatment of patients with advanced NSCLC.

Eligible patients had stage IV or recurrent NSCLC and had received no prior systemic therapy for advanced disease and had noEGFRorALKmutations sensitive to available targeted therapies. The trial was limited to patients whose tumors had ≥1% PD-L1 expression.

Investigators randomized 541 patients to receive nivolumab 3 mg/kg every 2 weeks or a platinum-based chemotherapy doublet determined by the treating physician’s preference. Treatment continued until disease progression. Assessment of tumor status occurred every 6 weeks for 48 weeks and then every 12 weeks. At progression, patients could cross over to treatment with nivolumab.

The primary endpoint was PFS, as determined by independent review, in the subgroup of patients who had tumors with ≥5% PD-L1 expression (N = 415). Secondary endpoints included PFS in the entire study population, OS, and objective response rate.

The chemotherapy arm had more women (45% vs 32%). Otherwise, the 2 treatment groups had no significant differences in key baseline characteristics. About 40% of patients in both arms had received prior radiation therapy.

More than half of the patients in each group remained on assigned therapy for more than 3 months. Thereafter, substantially more patients remained on nivolumab, including 20.6% who remained on therapy for more than 12 months, as compared with 10.6% in the chemotherapy arm.

The two most commonly used chemotherapy doublets were pemetrexed/carboplatin (43.7%) and pemetrexed/cisplatin (32.7%). About 40% of patients received maintenance pemetrexed.

The primary analysis showed no evidence of nivolumab superiority. In addition to the median PFS, the 1-year PFS also did not differ between groups, 23.6% with nivolumab and 23.2% with chemotherapy.

Median OS in patients with ≥5% PD-L1 expression was 14.4 months with nivolumab and 13.2 months with chemotherapy. The 1-year survival was 56.3% and 53.6% with nivolumab and chemotherapy, respectively.

Objective responses occurred in 26.1% of nivolumab patients and 33.5% of chemotherapy-treated patients. Substantially more patients in the nivolumab arm had progressive disease as best response: 27.5% compared with 9.9%.

Outcomes remained similar in subgroup analyses. In those with non-squamous histology (n = 412), the unstratified HRs for OS and PFS were 1.17 and 1.29, respectively, for nivolumab versus chemotherapy. Additionally, a benefit was not observed for nivolumab versus chemotherapy in those with PD-L1 expression on ≥50% of cells, with a HR for PFS of 1.07 and an HR of 0.90 for OS. Nivolumab was superior compared with chemotherapy in those with squamous histology (n = 129), with an HR of 0.83 for PFS and 0.82 for OS.

Safety and adverse events in both treatment groups were consistent with known effects of the therapies. The most common adverse events (any grade) in the nivolumab arm were fatigue (21.0% vs 35.4% with chemotherapy), diarrhea (13.9% vs 12.9%), decreased appetite (12.0% vs 27.8%), and nausea (11.6% vs 48.3%). Grade 3/4 adverse events were uncommon with nivolumab. The most common grade 3/4 adverse events in the chemotherapy arm were anemia (17.5%), neutropenia (11.0%), thrombocytopenia (8.4%), and fatigue (5.3%).

The total rate of crossover at progression was 43.6% in the nivolumab arm and 64.2% in the chemotherapy arm. In the nivolumab arm, 1.4% of patients crossed over to other immunotherapy, as compared with 60.4%. At the end of the trial, 18.7% of treated patients remained on randomized therapy in the nivolumab group versus 5.3% of the chemotherapy arm.

Despite the negative results, evaluation of nivolumab in the first-line setting of advanced NSCLC will continue, said Socinski. The CheckMate-227 trial will compare nivolumab alone, nivolumab plus ipilimumab, and chemotherapy. Additionally, the trial will enroll patients into separate groups according to PD-L1 expression (&ge;1% or <1%, and both groups will be randomized to all three treatments.

Results of a recent randomized trial showed first-line therapy with nivolumab failed to improve progression-free survival (PFS) in PD-L1—positive non-small cell lung cancer (NSCLC) compared with standard chemotherapy.