Solange Peters, MD, PhD
Findings from the phase III IMpower150 study have placed renewed focus on the optimal role for immunotherapy in the frontline setting for patients with advanced non–small cell lung cancer (NSCLC), according to a talk by Solange Peters, MD, PhD, at the ESMO Immuno Oncology Congress.
Checkpoint inhibition has been a frontline option for NSCLC, since the 2016 approval of pembrolizumab (Keytruda) for patients with PD-L1–expressing NSCLC. In the United States, first-line pembrolizumab was approved by the FDA in October 2016, which was followed by a European Commission approval in January 2017. "This is now standard of care in Europe and the US since late 2016," said Peters, head of the Thoracic Malignancies Program in the Department of Oncology at the University of Lausanne in Switzerland.
While frontline immunotherapy has shown promise for some patients with NSCLC, there is a need to further improve outcomes through combinations. Several trials are currently looking at approved checkpoint inhibitors with targeted therapies, other immunotherapies, or chemotherapy. Phase II findings have shown promise for pembrolizumab with chemotherapy, resulting in an FDA approval in May 2017. Now, findings from IMpower150 offer insight into the addition of atezolizumab (Tecentriq) to VEGF-targeted therapy.1
"We’ve known the rationale since the 90's—combining antiangiogenic drugs and immunotherapy improves the function of the immune system and the trafficking of the T cells to the tumor. And, both aspects for the first time are being addressed in a clinical trial," said Peters. "PD-1/PD-L1 blockade can sensitize tumors to antiangiogenic therapy and increase its efficacy, and vice versa."
Findings from the seminal phase III KEYNOTE-024 trial were responsible for setting the baseline for future immunotherapy development in patients with NSCLC. In this trial,2
frontline pembrolizumab demonstrated a median overall survival (OS) of 30.2 months compared with 14.2 months with chemotherapy for patients with >50% PD-L1–expressing advanced NSCLC (HR, 0.63; 95% CI, 0.47-0.86; P
= .002). The 24-month OS rate was 51.5% for pembrolizumab versus 34.5% with chemotherapy.
The median PFS with pembrolizumab was 10.3 months compared with 6.0 months for chemotherapy (HR, 0.50; 95% CI, 0.37-0.68; P <.001). The 12-month PFS rate was 62% in the pembrolizumab arm and 50% for chemotherapy.
"What is interesting with immunotherapy is what we call the non-proportional hazard ratio. The benefit is delayed—the curves tend to separate after a certain time point," said Peters. "Biomarkers, which we still need to identify, help you identify a population with a very strong benefit, so the benefit is increasing over time. You need magnitude and you need time, which is important when we look at the very early data from IMpower150."
The IMpower150 study enrolled 1202 patients with stage IV non-squamous NSCLC. Patients were randomized evenly to receive atezolizumab plus carboplatin and paclitaxel (arm A), atezolizumab with bevacizumab plus carboplatin and paclitaxel (arm B), or bevacizumab plus carboplatin and paclitaxel (arm C). Those with known EGFR or ALK alterations were excluded from the study.
The atezolizumab regimen demonstrated a median PFS of 8.3 months (95% CI, 6.0-7.1) compared with 6.8 months (95% CI, 7.7-9.8) with bevacizumab and chemotherapy alone (HR, 0.617; 95% CI, 0.517-0.737; P
<.0001). The 12-month PFS rate was 37% with the atezolizumab-containing regimen and 18% with the bevacizumab/chemotherapy regimen.
"The curves start to separate when you finish the chemotherapy, so there must be something with the bevacizumab maintenance," said Peters. "There is a non-proportional hazard ratio. The difference in median PFS is very frustrating, it is a small one, but the hazard ratio is very significant. Meaning the curve starts to separate later with benefits growing over time. This means we'd like to identify subgroups who benefit."
PD-L1 expression on immune and tumor cells did not appear to impact efficacy, with benefits seen across subgroups. In those with high PD-L1 expression (>50%; TC/IC3), the median PFS with the atezolizumab regimen was 12.6 months compared with 6.8 months for the control arm (HR, 0.39).
"You can enrich with very high PD-L1, like a KEYNOTE-024, but in this patient population pembrolizumab might still be a valid option, since it is the same PFS," said Peters. "If you eliminate these high expressers from the group, it is very important to see that this strategy still works very well. This strategy is not driven by the high PD-L1, it’s even observed in patients who do not have high PD-L1."