Checkpoint Inhibition Continues Rapid Advance in Non-Small Cell Lung Cancer

Naiyer A. Rizvi, MD, discusses the future outlook for PD-1/PD-L1 checkpoint agents in non-small cell lung cancer and the potential for immunotherapy combinations.

Naiyer A. Rizvi, MD

The PD-1/PD-L1 pathway continues to make headlines as a treatment target in non—small cell lung cancer (NSCLC), with the FDA recently granting an accelerated approval to the checkpoint inhibitor pembrolizumab (Keytruda) for patients with pretreated advanced NSCLC across all histologies whose tumors express PD-L1.

The approval was based on the phase I KEYNOTE-001 trial, which found that the overall response rate (ORR) with pembrolizumab was 41% among a subgroup of 61 patients with pretreated PD-L1—positive advanced NSCLC as determined by the PD-L1 IHC 22C3 pharmDx diagnostic test, which was approved as a companion diagnostic with the drug.

Response duration ranged from 2.1 to 9.1 months. A survival improvement has yet to be demonstrated in a clinical trial, and the accelerated approval is contingent upon the eventual outcomes of confirmatory studies.

The PD-1 inhibitor nivolumab (Opdivo) is approved to treat patients with squamous cell NSCLC and is being reviewed for a nonsquamous indication, with a decision scheduled to come from the FDA by January 2, 2016. Updated phase III data from the CheckMate-057 trial presented at the 2015 European Cancer Congress (ECC) showed that nivolumab reduced the risk of death by 28% compared with docetaxel for patients with previously treated nonsquamous NSCLC.

Also at the 2015 ECC, data were released showing a potential benefit for the treatment of NSCLC with the PD-L1 inhibitor atezolizumab (MPDL3280A). When atezolizumab was added to chemotherapy in a phase Ib study, patients with newly diagnosed metastatic NSCLC had double the expected response.

Almost two-thirds of 41 evaluable patients had objective responses with atezolizumab plus different platinum-based chemotherapy doublets. The highest response rate occurred with the combination of atezolizumab, carboplatin, and pemetrexed, as 13 of 17 patients (76.5%) had partial responses. Four patients treated with the anti—PD-L1 inhibitor, carboplatin, and nab-paclitaxel had complete responses. The clinical trial is still ongoing.

Atezolizumab has also shown promise as a monotherapy in patients with PD-L1—positive advanced NSCLC in the phase II BIRCH and POPLAR trials.

OncLive: What did we learn from the KEYNOTE-001 trial, which recently led to the approval of pembrolizumab for patients with PD-L1—positive lung cancer?

Atezolizumab has also demonstrated efficacy in advanced NSCLC, specifically in phase II studies. Can you discuss the outcomes with this drug?

For a better understanding of the future outlook for PD-1/PD-L1 checkpoint agents in NSCLC and to learn more about the potential for immunotherapy combinations, OncLive spoke with Naiyer A. Rizvi, MD, director of both Thoracic Oncology and Immunotherapeutics for the Division of Hematology and Oncology at Columbia University Medical Center.Dr Rizvi: When they had a development strategy that was based on PD-L1 expression, those patients who had high-positive PD-L1 expression with more than 50% expression seemed to have a much higher response rate and durable benefit than those patients who are negative. Those patients who had PD-L1 expression that was less than 50% did not seem to have as great a benefit. Therefore, I think that it demonstrated an enrichment strategy to enhance the likelihood of benefit to pembrolizumab.Data on this were updated at ECC in Vienna. The most mature, or the most robust, data is the POPLAR trial, where patients were randomized to atezolizumab or docetaxel in the pretreated patient population. Initially, the hazard ratio that was presented at ASCO for the overall patient population was not significant, but the long-term data does show a separation of the curve, and so the hazard ratio was actually 0.73. It was a significant result with long-term follow-up, with 12.6-months median survival with atezolizumab versus 9.7 months with docetaxel.

What did we learn from the CheckMate-057 trial, which looked at nivolumab versus docetaxel in NSCLC?

It also showed enrichment in response based on PD-L1 expressions, so the high-positive patients had a hazard ratio of about 0.49 and that represented 16% of the patients with lung cancer. We are seeing similar trends when we see overall benefit to the overall patient population, but there is a subset of patients with higher levels of PD-L1 expression, which seem to do better.The study did meet its primary endpoint in terms of OS benefit in the overall patient population. I think that the nonsquamous patient population is somewhat more heterogeneous than the squamous patient population.

We did see the relative importance of PD-L1 in this study, where the patients who were PD-L1—positive seemed to have a more substantive benefit compared to those who were PD-L1–negative. We also saw that the PD-L1–negative patients seemed to do comparable to those who underwent chemotherapy, but they did not seem to do significantly better.

What are your overall thoughts on PD-L1 status as a biomarker for efficacy with anti—PD-1/PD-L1 agents in NSCLC?

If all three of these checkpoint inhibitors were eventually approved, how would you decide which of the drugs to recommend to your patients?

What potential do you see for the combination of nivolumab and ipilimumab?

We saw similar trends with the never-smokers or EGFR-mutant populations, so I think that one of the differences, in terms of the nonsquamous versus squamous patients, is that about a quarter of nonsquamous patients, who are never-smokers or EGFR-mutated patients, are less likely to benefit from immunotherapy.I think that it is clearly enriching for response, and I think it is of value. However, I do think that just because tumors are PD-L1—negative, that does not mean that they do not have the potential to benefit from immunotherapy. They may benefit from immunotherapy combinations, and that is something we need to understand better. We have the opportunity to try to look at other combinations and do better for the PD-L1–negative subset.There is not one obvious way to make that decision. Ultimately, what are going to distinguish the different companies and the different drugs are their development strategy, their pipeline agents, and how they are combining it with other therapies. It is really going to come down more to individual development strategies and where approvals are gotten and what their portfolio is, in terms of combination therapies. Beyond that, I think it is hard to do anything different than that.The future for immunotherapy is combination immunotherapy, and I think both of them showed a higher level of response rate and activity in both PD-L1—positive and PD-L1–negative patients. Both trials have shown that we have been able to give these drugs with a better toxicity profile than our initial experience with the combinations. Those are both in phase III studies in the first-line setting, so I think they are both going to be important combinations as we go forward.