Phase III Data Show Nivolumab Improves Survival Across Lung Cancer Subtypes

Nivolumab has now been shown to have an overall survival benefit versus docetaxel in both nonsquamous and squamous non–small cell lung cancer.

Luis Paz-Ares, MD

Nivolumab (Opdivo) has now been shown to have an overall survival (OS) benefit versus docetaxel in both nonsquamous and squamous non—small cell lung cancer (NSCLC), according to results from two phase III trials presented at the 2015 ASCO Annual Meeting. The PD-1 inhibitor was also shown to be less toxic than the chemotherapy agent.

In the phase III CheckMate-017 trial,1 there was a 41% OS improvement with nivolumab versus docetaxel in the squamous setting, and in the phase III CheckMate-057 trial,2 the OS benefit with nivolumab was 27% in patients with nonsquamous NSCLC. Both studies enrolled patients who had progressed following platinum-based doublet chemotherapy.

PD-L1 status was not linked to survival in CheckMate-017; however, in CheckMate-057, stronger OS outcomes were observed in PD-L1—positive patients, including a 60% reduction in the risk of death for those with the highest PD-L1 levels.


Nivolumab is approved by the FDA in squamous NSCLC based on data from CheckMate-017. Bristol-Myers Squibb (BMS), the manufacturer of nivolumab, has applied for an additional indication in the nonsquamous setting based on the results of Checkmate-057.The open-label CheckMate-057 trial randomized 582 patients with advanced nonsquamous NSCLC after the failure of platinum-based doublet chemotherapy to nivolumab at 3 mg/kg IV every 2 weeks (n = 292) or docetaxel at 75 mg/m2 intravenously every 3 weeks (n = 290). The treatments were administered until disease progression or unacceptable toxicity. Patients received a median of 6 and 4 doses in the nivolumab and docetaxel arms, respectively.

Patients had an ECOG performance status of 0 or 1. The median patient age was 61 years in the nivolumab arm and 64 years in the docetaxel cohort. Prior maintenance with bevacizumab, pemetrexed, or erlotinib was allowed, as was TKI therapy for known EGFR mutations or ALK translocation. Forty-percent and 38% of patients in the nivolumab and docetaxel arms, respectively, had received prior maintenance therapy. In the nivolumab arm, 15% of patients were EGFR-positive and 4% were ALK-positive, with comparable rates of 13% and 3%, respectively, in the docetaxel group.

OS was the primary endpoint, with secondary objectives focused on progression-free survival (PFS), objective response rate (ORR) per RECIST v1.1, efficacy by PD-L1 expression, and safety.

The study was stopped early after an independent monitoring panel determined the primary endpoint of improved OS had been reached. Eligible patients were allowed to continue treatment or cross over to the nivolumab arm in an open-label extension of the study.

The median OS was 12.2 months with nivolumab versus 9.4 months with docetaxel (HR = 0.73; 96% CI, 0.59-0.89; P = .00155), with a 1-year OS of 50.5% versus 39.0%, respectively.

“Nivolumab is the first PD-1 inhibitor to significantly improve OS as compared to docetaxel in [NSCLC] patients with prior treatment and nonsquamous histology,” said lead author Luis Paz-Ares, MD, Hospital Universitario Doce de Octubre, Madrid, Spain, in a press conference at the ASCO meeting.

ORR was 19% with the PD-1 inhibitor compared with 12% with chemotherapy (Odds Ratio = 1.72; 95% CI, 1.1-2.6; P = .0246). Complete and partial response rates were 1% and 18% in the nivolumab arm and <1% and 12% in the docetaxel group, respectively. The stable disease rate was 25% and 42% with PD-1 inhibition and chemotherapy, respectively.

Median time to response was 2.1 months with nivolumab versus 2.6 months with docetaxel. Median duration of response was 17.2 months versus 5.6 months in the nivolumab and control arms, respectively. Fifty-two percent of the nivolumab responses are still ongoing compared with 14% of the docetaxel responses.

Median PFS was comparable between the cohorts at 2.3 months in the nivolumab arm compared with 4.2 months in the docetaxel group (HR = 0.92; 95% CI, 0.77-1.11; P = .393). One-year PFS favored nivolumab at 18.5% versus 8.1% for the control arm.

The researchers measured PD-L1 levels in pretreatment tumor biopsies with the Dako automated IHC assay. Higher PD-L1 expression was associated with improved survival outcomes among the 78% of patients for whom PD-L1 status was detectable. “PD-L1 emerged as a clear predictive factor for the benefit from nivolumab treatment,” said Paz-Ares.

In PD-L1—positive patients (PD-L1 expression on ≥1% of tumor cells), median OS was improved by 41% among 123 individuals treated with nivolumab versus 123 patients who received docetaxel (median OS = 17.2 months vs 9.0 months; HR = 0.59).

The OS benefit continued to rise as PD-L1 levels increased. The reduction in the risk of death was 57% (median OS = 18.2 months) and 60% (median OS = 19.4 months) for patients expressing PD-L1 on ≥5% and ≥10% of their tumor cells, respectively.

The researchers did not observe a similar OS benefit among patients with low or undetectable PD-L1 levels. Median OS was 10.4, 9.7, and 9.9 months among patients with PD-L1 expression levels <1%, <5%, and <10%, respectively.

Nivolumab was well tolerated and had a better safety profile than docetaxel. Among patients evaluable for safety, all-grade adverse event (AE) rates were 69% versus 88% in the nivolumab versus docetaxel arms, respectively. The most common all-grade AEs with nivolumab versus docetaxel were fatigue (16% vs 29%), nausea (12% vs 26%), decreased appetite (11 vs 16%), asthenia (10 vs 18), and diarrhea (8% vs 23%).

Grade 3-5 adverse events were reported in 10.5% of the nivolumab arm compared with 53.7% of the docetaxel cohort. The most common grade 3/4 AEs with nivolumab were fatigue, nausea, and diarrhea, at 1% each. Twenty-seven percent of patients in the docetaxel arm had grade 3/4 neutropenia versus 0 in the nivolumab arm.

Toxicity-related discontinuations occurred in 4.9% of patients receiving nivolumab compared with 14.9% of those treated with chemotherapy. Systemic therapy was administered to 42.1% and 49.7% of patients who discontinued nivolumab and docetaxel, respectively. No treatment-related deaths occurred in the nivolumab group compared with 1 in the docetaxel arm.

Checkmate 017

“The CheckMate-057 results represent progress toward establishing a new standard of care that may replace docetaxel in PD-L1 expressers,” said Paz-Ares, in a press release issued today by BMS.The open-label CheckMate-017 trial randomized 272 previously treated patients with advanced or metastatic squamous cell NSCLC to nivolumab at 3 mg/kg IV every 2 weeks (n = 135) or docetaxel at 75 mg/m2 IV (n = 137) every 3 weeks.

The primary outcome measure of the trial was OS. Secondary endpoints included ORR (RECIST v1.1), PFS, outcomes by PD-L1 expression, and safety.

Median OS was 9.2 months with nivolumab versus 6.0 months with chemotherapy (HR = 0.59; 95% CI, 0.44-0.79; P = .00025), with a 1-year OS of 42% versus 24%, respectively. Nivolumab also improved median PFS by 38% versus docetaxel (3.5 vs 2.8 months; HR = 0.62; 95% CI, 0.47-0.81; P = .0004). One-year PFS was 21% versus 6% in the nivolumab and control arms, respectively.

ORR with nivolumab was 20% (n = 27) compared with 9% (n = 12) for chemotherapy (P = .0083). The median duration of response was 8.4 months in the docetaxel arm and had not yet been reached in patients receiving nivolumab.

Unlike CheckMate-057, PD-L1 status was not linked with efficacy outcomes in CheckMate-017. Regardless of tumor PD-L1 levels, OS HRs favored nivolumab. In PD-L1—positive patients (≥1% of tumor cells express PD-L1), median OS was improved by 31% (HR = 0.69) with nivolumab versus a 42% benefit with the PD-1 inhibitor (HR = 0.58) in the PD-L1–negative group.

As was shown in the CheckMate-057 results, nivolumab was less toxic than docetaxel. Grade 3/4 adverse events were reported in 7% (9/131) of evaluable patients in the nivolumab arm versus 55% (71/129) of evaluable patients who received chemotherapy. There were no treatment-related deaths in the nivolumab cohort compared with 3 deaths linked to docetaxel.

In the BMS press release, Michael Giordano, senior vice president, head of development, Oncology at BMS, expressed his optimism with the progress of the nivolumab NSCLC program. “The survival results from [CheckMate-057], as well as from CheckMate-017…validate the Bristol-Myers Squibb development strategy for Opdivo to improve survival expectations for patients with lung cancer.”


1. Spigel DR, Reckamp KL, Rizvi NA, et al. A phase III study (CheckMate 017) of nivolumab (NIVO; anti-programmed death-1 [PD-1]) vs docetaxel (DOC) in previously treated advanced or metastatic squamous (SQ) cell non-small cell lung cancer (NSCLC). J Clin Oncol. 2015;33 (suppl; abstr 8009).

2. Paz-Ares LG, Horn L, Borghaei H, et al. Phase III, randomized trial (CheckMate 057) of nivolumab (NIVO) versus docetaxel (DOC) in advanced non-squamous cell (non-SQ) non-small cell lung cancer (NSCLC). J Clin Oncol. 2015;33 (suppl; abstr LBA109).


View more from the 2015 ASCO Annual Meeting