New Analysis of CheckMate 057: Search Continues for Predictors of Outcome With Nivolumab

Prognostic factors and tumor expression of programmed cell death ligand 1 (PD-L1) predict early mortality among patients with previously treated nonsquamous advanced non–small cell lung cancer (NSCLC) who receive nivolumab instead of docetaxel. But these features are not reliable for excluding patients from nivolumab therapy.

These were among the key findings of a post hoc analysis of the randomized phase III CheckMate 057 trial reported by first author Solange Peters, MD, Professor and Chief of Medical Oncology at the University of Lausanne, Switzerland.1

Nivolumab (Opdivo), an immune checkpoint inhibitor that targets programmed cell death protein 1 (PD-1), is currently approved by the U.S. Food and Drug Administration (FDA) as second-line therapy for both squamous and nonsquamous advanced NSCLC.

The important conclusion of this subanalysis is that no single baseline factor reliably characterizes the overall survival subgroups defined as early deaths in less than 3 months.
— Solange Peters, MD

Tweet this quote

The main results from CheckMate 057 showed a significant long-term survival benefit of nivolumab over docetaxel in patients with previously treated nonsquamous disease.2,3 After a minimum follow-up of 2 years, median overall survival was 12.2 months with the immune checkpoint inhibitor and 9.5 months with chemotherapy (hazard ratio = 0.75). However, survival during the first 3 months of treatment was actually poorer for the nivolu-mab group, with 15 more deaths in that group, primarily due to disease progression, according to Dr. Peters.

Any of the three factors associated with poorer prognosis or more aggressive disease—fewer than 3 months out since last treatment, progressive disease as the best response to the prior treatment, and Eastern Cooperative Oncology Group (ECOG) performance status of 1—in conjunction with no or low tumor PD-L1 expression was a marker for early mortality with nivolumab therapy but did not reliably identify patients who would die early.

“Our exploratory post hoc multivariate analysis has shown that patients with poorer prognostic factors and more aggressive disease combined with lower or no PD-L1 expression appear to be at higher risk of death on nivolumab vs docetaxel in CheckMate 057,” she elaborated. “However, the majority of patients with these features were still alive after 3 months in the nivolumab arm. The simple, straightforward exclusion of these patients based on any of these characteristics is probably not possible.”

“Across different pharma companies and collaborative groups, we still need to have better predictability of response to immunotherapy. There is a huge effort to identify tumor biomarkers or microenvironment biomarkers,” Dr. Peters commented.

Study Details

CheckMate 057 enrolled 582 patients with advanced nonsquamous NSCLC who had received at least 1 prior platinum-based chemotherapy. They were randomized to receive nivolumab every 2 weeks or docetaxel every 3 weeks.

In the new analysis, Dr. Peters and colleagues stratified patients according to survival status at 3 months and then tested a variety of clinical and disease factors to see whether they differed significantly for patients with early mortality, comparing both between and within treatment arms.

“The post hoc, retrospective, unplanned nature of this analysis makes it highly probable that these various parameters are imbalanced between the two treatment arms,” she acknowledged. “But the important conclusion of this subanalysis is that no single baseline factor reliably characterizes the overall survival subgroups defined as early deaths in less than 3 months.”

Multivariate analysis identified patients treated with nivolumab at an elevated risk for early mortality: those with any of the three previously mentioned adverse factors plus no or low PD-L1 expression. But for each adverse factor, more than 50% of the patients with that factor were still alive at 3 months.


Initial trial results established that response rate increased with baseline tumor PD-L1 expression, although some patients with tumors having < 1% positivity still had a response. In the new analysis, “the level of PD-L1 is continuously correlated with the level of response in this clinical trial regarding nivolumab but absolutely not regarding docetaxel,” Dr. Peters reported.

Analyses of the quality of responses according to tumor PD-L1 status showed that deep and durable responses to nivolumab were seen regardless of the level of expression. The median duration of response was 18.3 months among patients with < 1% expression and 17.2 months among patients with ≥ 1% expression. Of four complete responders, one had < 1% expression, two had ≥ 1% expression, and one had nonquantifiable levels of the ligand.

The more favorable safety profile of nivolumab compared with docetaxel in the trial population as a whole also held up regardless of PD-L1 status, according to Dr. Peters. Additionally, the same pattern of excess early mortality with nivolumab was evident in each PD-L1 expression subset.

Finally, in a landmark analysis excluding patients who died in the first 3 months, nivolumab was consistently associated with better overall survival than docetaxel in the entire intent-to-treat population (hazard ratio = 0.59) and even among the subset having tumors with < 1% PD-L1 expression (hazard ratio = 0.66).

“While enhanced clinical activity with nivolumab correlates with increasing tumor PD-L1 expression in this patient population, nivolumab-treated patients with no or low PD-L1 expression have a comparable probability of response, more durable responses, comparable overall survival, and fewer treatment-related adverse events vs those treated with docetaxel,” Dr. Peters concluded. ■

Disclosure: Solange Peters, MD, reported no potential conflicts of interest.


1. Peters S, Cappuzzo F, Horn L, et al: Analysis of early survival in patients with advanced non-squamous NSCLC treated with nivolumab vs docetaxel in CheckMate 057. 2016 World Conference on Lung Cancer. Abstract OA03.05. Presented December 5, 2016. 

2. Borghaei H, Paz-Ares L, Horn L, et al: Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer. N Engl J Med 373:1627-1639, 2015

3. Borghaei H, Brahmer JR, Horn L, et al: Nivolumab vs docetaxel in patients with advanced NSCLC: CheckMate 017/057 2-y update and exploratory cytokine profile analyses. 2016 ASCO Annual Meeting. Abstract 9025.

Related Articles

Expert Point of View: Paul Mitchell, MD

“The progression-free and overall survival curves of CheckMate 057 suggest the presence of two patient populations with respect to nivolumab (Opdivo): a relatively sensitive one and a less sensitive, possibly even resistant, one, according to invited discussant Paul Mitchell, MD, Associate...