Back Back
Lung Cancer at AACR 2018
First-line Therapy in NSCLC: What You Need to Know From AACR 2018

Released: June 01, 2018

Matthew Gubens
Matthew Gubens, MD, MS
Activity Information

Activity

Progress
1
Course Completed

At the 2018 AACR annual meeting, we heard some exciting findings in lung cancer care, particularly focusing on how we can improve our first‑line immunotherapy treatment approach. In this commentary, I want to share a couple of my key takeaways.

KEYNOTE-189
First were the findings from the phase III KEYNOTE-189 study. In May 2017, the combination of carboplatin/pemetrexed plus pembrolizumab was FDA approved as first-line therapy for patients with advanced nonsquamous NSCLC regardless of tumor PD-L1 expression based on the phase II KEYNOTE-021 study that enrolled 123 patients. However, because this study included only 123 patients, many of us in the lung cancer community were reserved about the FDA approval that followed and were waiting to see if these results would be confirmed in the phase III KEYNOTE‑189 trial. The KEYNOTE-189 study evaluated the regimen of carboplatin/pemetrexed plus pembrolizumab vs carboplatin/pemetrexed plus placebo as first-line therapy in a much larger population of patients with nonsquamous NSCLC (N = 616) with primary endpoints of both OS and PFS (by blinded independent central review).

KEYNOTE-189 enrolled patients regardless of tumor PD-L1 expression level. In the overall patient population, the risk of death for patients receiving chemotherapy plus pembrolizumab was one half that of the patients who received chemotherapy plus placebo (HR: 0.49) with 1-year OS rates of 69.2% vs 49.4%, respectively. Notably, the OS benefit was seen not only in those patients with high PD-L1 expression (≥ 50% n = 202; HR: 0.42) but also in patients with PD-L1 expression between 1% to 49% (n = 186; HR: 0.55) and < 1% (n = 190; 0.59). Similar improvements in median PFS were also reported across the PD-L1 expression level subgroups. 

In terms of other findings, approximately one half of the patients on the chemotherapy plus placebo arm received an approved immune checkpoint inhibitor upon disease progression. This is consistent with previous findings and speaks to the fact that not all patients receive second-line therapy in the real world. The incidence of both any grade (99.8 % vs 99.0%) and grade ≥ 3 (67.2% and 65.8%) adverse events was similar with chemotherapy plus pembrolizumab and chemotherapy plus placebo. The toxicity profile was generally what would be expected from chemotherapy with little additional toxicity from pembrolizumab with the exception of a slightly higher than expected incidence of grade ≥ 3 acute kidney injury (2%) and nephritis (1.5%).

Based on the KEYNOTE‑189 data, I now consider the carboplatin/pemetrexed plus pembrolizumab regimen as a standard of care that we can offer to our patients with advanced nonsquamous NSCLC, especially those patients with PD-L1 expression < 50%. But how should we treat our patients who have tumor PD-L1 expression ≥ 50% given the data that we now have? Should we recommend pembrolizumab alone or in combination with carboplatin/pemetrexed? The KEYNOTE‑024 trial showed approximately a 40% reduction in the risk of death with pembrolizumab monotherapy in this group of patients compared with an approximately 50% reduction in the risk of death with pembrolizumab in combination with carboplatin/pemetrexed in the KEYNOTE-189 trial. I certainly present both datasets to my patients, but I think that the OS benefit is fairly similar with these 2 treatment options in these patients based on the reported HRs. This leads me to recommend pembrolizumab monotherapy in most of my patients with PD-L1 expression ≥ 50%.

CheckMate‑227
Another interesting trial presented at AACR 2018 was CheckMate 227, which showed some provocative results with nivolumab plus ipilimumab that may have some bearing in the first‑line treatment setting for patients with advanced NSCLC in the future. Nivolumab and ipilimumab are already an FDA-approved combination in melanoma, and there has been a lot of interest both in NSCLC and small-cell lung cancer to see if there would be a similar benefit of this combination over single-agent immune checkpoint inhibitor therapy.

CheckMate 227 is a large, open-label phase III study that enrolled 1739 patients with metastatic or recurrent NSCLC and no previous systemic therapy regardless of histology, unlike KEYNOTE-189. The patients who had PD-L1 expression ≥ 1% (n = 1189) were randomized to nivolumab plus ipilimumab vs nivolumab monotherapy vs standard chemotherapy (1:1:1), and the patients who had PD-L1 expression < 1% (n = 550) were randomized to nivolumab plus ipilimumab vs nivolumab plus chemotherapy vs standard chemotherapy (1:1:1).

The presentation at AACR focused on a comparison of nivolumab plus ipilimumab vs standard chemotherapy in a subset of 299 patients where the investigators were able to assess tumor mutation burden (TMB) for the coprimary endpoint of PFS according to TMB with a breakpoint of ≥ 10 mutations/Mb. Of interest, in those patients with high TMB (≥ 10 mutations/Mb), PFS was significantly better with nivolumab plus ipilimumab vs chemotherapy (median PFS: 7.2 vs 5.5 months, respectively; HR: 0.58), and the ORR was also higher with the immunotherapy combination vs chemotherapy (45.3% vs 26.9%, respectively). The incidence of serious adverse events (24.0% vs 13.9%) and discontinuations due to adverse events (17.4% vs 8.9%) were higher with the combination of nivolumab plus ipilimumab than with chemotherapy, respectively, but the incidence of treatment-related deaths was similar and low for both treatments.

These results suggest that TMB, which is noted on some commercially available biomarker reports that we can use in our practices, may eventually be useful as a predictive biomarker to identify a group of patients who may benefit from a dual immunotherapy combination as first-line therapy, saving chemotherapy for later in their treatment course. It is important to note that high PD-L1 expression and high TMB are not overlapping patient populations, and so it may well be that having both of these biomarkers may enrich our ability to do precision immuno‑oncology from the beginning of treatment. However, the nivolumab plus ipilimumab combination is not yet approved in NSCLC, and I will wait to see the OS data from this study before considering dual immunotherapy over our other available first-line treatment regimens that have been shown to extend OS. It is also important to realize that TMB breakpoints that define high and low TMB levels are not standardized at this time, and therefore, the different reports are not necessarily comparable or reflective of the findings from CheckMate 227.

Looking Forward
The management of advanced NSCLC is changing rapidly right now. In addition to the 2 studies from AACR 2018 that I highlighted here, at the ASCO 2018 annual meeting, we will see new data that may also have an impact on how we treat our patients, including results from 3 important phase III trials: KEYNOTE-042 comparing first-line pembrolizumab with standard chemotherapy in patients with ≥ 1% PD-L1 expression and both IMpower131 and IMpower150, which are comparing different standard first-line chemotherapy regimens with or without atezolizumab in patients with squamous or nonsquamous histology, respectively, regardless of PD-L1 expression level. So stay tuned.

Your Thoughts
Please share your thoughts on the evolving therapeutic landscape for our patients with advanced NSCLC by joining the conversation in the comments box below and responding to the polling question at the right of your screen.