Insights From: Roy S. Herbst, MD, PhD, Yale School of Medicine; Corey J. Langer, MD, University of Pennsylvania; Vassiliki A. Papadimitrakopoulou, MD, University of Texas MD Anderson Cancer Center
Roy S. Herbst, MD, PhD: If someone has not had immunotherapy in the front-line setting and they have PD-L1 that is greater than 1%, I will use pembrolizumab. That’s based on KEYNOTE-010, which I actually led. There’s no doubt that that’s an improvement over standard chemotherapy. So, we’re using that. You could also use the other drugs in that setting—nivolumab or atezolizumab. Nivolumab is another PD-1 inhibitor and is given every 2 weeks, as opposed to the pembrolizumab, which is given every 3 weeks. Atezolizumab is given every 3 weeks as a PD-L1 inhibitor. All 3 would be reasonable choices in that setting.
Corey J. Langer, MD: The therapeutic landscape in the second-line setting was irrevocably altered in 2015, when nivolumab was approved based on the results of CheckMate-017, which showed a clear-cut, statistically significant, highly clinically relevant survival advantage for nivolumab given every 2 weeks, compared with our former standard, docetaxel. The median survival increased by 50% in the nivolumab arm, and that was later accompanied by a separate trial in non-squamous, non–small cell disease with, essentially, the same study design—single-agent nivolumab versus docetaxel. Again, this resulted in a statistically significant, clinically relevant survival advantage. Nivolumab was approved on that basis, and it was really the first checkpoint inhibitor to get an approval in the United States.
From March of 2015 until roughly a year later—March or April of 2016—at our institution, we treated 175 discrete individuals with nivolumab. We went back and surveyed our results, and our response rates essentially matched what had been observed in the clinical trials—about 19%. Another very interesting observation was that we saw absolutely no responses in those with oncogenic drivers, either EGFR mutations or ALK translocation. We observed a pseudo progression rate, and it was a bit lower than folks have assumed, certainly lower than what we’ve seen in melanoma. This occurred in 5 patients out of that 175—about 2.8%. It occurs, but it’s rare, and you need to be sensitized to the fact that it truly is rare.
Since June of 2016, we’ve begun to alter our approach. Pembrolizumab had received initial approval in individuals with 50% or higher expression, again, second-line and beyond. So, we started using pembrolizumab off study. And then, as of the fall of 2016, when pembrolizumab’s approval was expanded to 1% or higher in the second-line setting and atezolizumab received approval independent of PD-L1 status, all of a sudden we had 3 separate agents available in this setting. We really had to grapple with which one to use. What would be our clinical pathway?
Atezolizumab and pembrolizumab have 1 relatively mundane advantage over nivolumab: They’re both given every 3 weeks. Nivolumab is given every 2 weeks. In this era of clinical pathways, at least as of the beginning of 2017, we have decided as a group, at my institution, if individuals have PD-L1 levels that are 1% or higher, they will get pembrolizumab in the second-line setting. If the PD-L1 level is 0 or unknown, they get atezolizumab. They can still get nivolumab, but it is not strictly found in our clinical pathway. So, at this point, the 2 Q3-week agents, the PD-1 inhibitor (pembrolizumab) and the PD-L1 inhibitor (atezolizumab) are the agents of choice. But that could change, and the clinical pathways are meant to be flexible over time.
Vassiliki A. Papadimitrakopoulou, MD: In the setting of second-line therapy, I think we are seeing significant changes in the landscape because of the approvals of front-line pembrolizumab in patients with more than 50% expression and the recent accelerated approval of pembrolizumab in combination with chemotherapy for nonsquamous non–small cell lung cancer. What used to be second-line or refractory immunotherapy has transitioned to the frontline setting. However, for the patients who were treated with the old standard of care, chemotherapy, in the frontline setting, we have the option between 3 immunotherapy agents in the second-line setting.
One of the 3, pembrolizumab, requires testing for PD-L1. In a majority of the cases, this testing has really been performed in the first-line setting and can be used for selection of patients for use of pembrolizumab in the second-line setting. Frequently, patients are also being treated with atezolizumab and nivolumab without the need for PD-L1 testing upfront. There are some considerations regarding the practicality of using immunotherapy every 2 weeks with nivolumab versus every 3 weeks with atezolizumab. That may lead to different choices by different physicians, but we don’t have any data that any of these therapies is superior to the others.