The caveat is that this was a highly selected cohort of patients.— Leora Horn, MD, MSc
Tweet this quote
The KEYNOTE-028 results confirm that pembrolizumab (Keytruda) is active in extensive-stage small cell lung cancer (SCLC) and lay to rest safety concerns stemming from the higher risk of paraneoplastic syndromes in this disease, according to invited discussant, Leora Horn, MD, MSc, Clinical Director of the Thoracic Oncology Program at the Vanderbilt-Ingram Cancer Center, Nashville, Tennessee.
“The caveat is that this was a highly selected cohort of patients: 147 patients were evaluated, 42 were programmed cell death ligand 1 (PD-L1)–positive, 24 were actually enrolled in the study, and there was a 33% response rate,” she explained. “So there was an 8% benefit in the overall patient population.”
The pembrolizumab dosing regimen used entailed higher and more frequent doses than those typically used in non–small cell lung cancer (NSCLC), added Dr. Horn. And of note, a sizable share of patients experienced early disease progression on pembrolizumab. Thus, a key question is “Do we have time to allow for a response in SCLC when we see a significant number of patients progressing in their first cycle of therapy?”
The results of this study are similar to what was noted in an unselected population of patients with SCLC treated with immunotherapy consisting of nivolumab (Opdivo) alone or in combination with ipilimumab (Yervoy).1 Of relevance going forward, the findings of that trial did not show a clear relationship between PD-L1 positivity and benefit.
Questions Remain
Therefore, it remains to be determined whether PD-L1 level of expression predicts pembrolizumab response in SCLC, as it does in NSCLC. Similarly, it is unknown whether chemotherapy sensitivity affects the response.
Finally, there may be better predictors of efficacy yet to be identified, according to Dr. Horn. For example, a study among patients with NSCLC has shown interplay between PD-L1 expression and mutation load when it comes to response and has identified a predictive smoking signature.2 “It would be nice to see whether these data were repeated in this group of patients with SCLC,” she commented. ■
Disclosure: Dr. Horn reported no potential conflicts of interest.
References