The randomized phase II POPLAR trial—reported by Fehrenbacher and colleagues and reviewed in this issue of The ASCO Post—is another key piece of information for the medical community regarding the value of immune checkpoint blockers in second/third-line treatment of patients with non–small cell lung cancer (NSCLC).1 This trial demonstrated the superiority of programmed cell death ligand 1 (PD-L1) blockade vs docetaxel in patients who had experienced disease progression after platinum therapy. Such superiority was already demonstrated in randomized phase III trials for programmed cell death protein 1 (PD-1) blockade with nivolumab (Opdivo)2,3 and pembrolizumab (Keytruda).4
Sophisticated Analysis
POPLAR provided a sophisticated analysis of PD-L1 expression as a predictive biomarker of the benefit of atezolizumab, an investigational agent. Baseline PD-L1 expression was scored by immunohistochemistry in tumor cells and in tumor-infiltrating immune cells. Increasing improvement in overall survival was associated with increasing PD-L1 expression.
As with nivolumab in nonsquamous cell carcinoma patients (CheckMate 057 trial),3 patients with the lowest PD-L1 expression assigned to the immune checkpoint blocker arm experienced overall survival similar to those assigned to docetaxel. The benefit in epidermal growth factor receptor (EGFR)-mutant patients is to be clarified in POPLAR (data not provided yet); no benefit was observed in EGFR-mutant patients with nivolumab (Checkmate 057)3 or with pembrolizumab (KEYNOTE-010).4
Interestingly, an eight-gene signature (named the Teff/IFNγ signature) was explored in the POPLAR trial and appeared to predict overall survival benefit in patients treated with atezolizumab. However, this signature did not predict a progression-free survival or objective response rate benefit.
Noteworthy POPLAR Features
It is remarkable to note that the recruitment of 287 patients (with an evaluable tumor for PD-L1 status) was achieved in less than 8 months. This clearly demonstrates the strong expectation by patients and academic physicians regarding access to immune checkpoint blockers.
Like the KEYNOTE-010 trial, POPLAR included patients who received more than one line of previous treatment (35% of patients in the POPLAR trial were third-line patients). This is in contrast with the two randomized trials of nivolumab, which were strictly limited to the second-line setting. Keeping that in mind, the objective response rate with atezolizumab was 19%, as compared to 20% for nivolumab2,3 and 18% for pembrolizumab.4
The overall toxicity profile (incidence of all adverse events and adverse events of interest) observed with atezolizumab mostly mirrors the profile already reported with anti–PD-1 agents. PD-L1 blockade with atezolizumab in the POPLAR trial did not appear to reduce lung toxicity, as originally speculated.
The proportion of patients who received chemotherapy post-atezolizumab was only 37%. This is important information, since it also has been established that only 36% of squamous cell carcinoma patients in CheckMate 017 received chemotherapy post-nivolumab.2 Thus, only a minority of patients are able to receive chemotherapy after disease progression on PD-1/PD-L1 blockade. ■
Disclosure: Dr. Soria has received consultancy fees from AstraZeneca, Astex, Covagen, Clovis, GSK, Gammamabs, Lilly, MSD, Mission Therapeutics, Merus, Pfizer, Pierre Fabre, Roche-Genentech, Sanofi, Servier, and Takeda.
References
