Expert on PD-L1/CTLA-4 Blockade and Future of Immunotherapy in Lung Cancer

Scott J. Antonia, MD

The combination of the PD-L1 inhibitor durvalumab and the CTLA-4 inhibitor tremelimumab, showed antitumor activity in patients with locally advanced or metastatic non—small cell lung cancer (NSCLC), irrespective of PD-L1 status, according to a recent phase Ib study published in The Lancet Oncology.

In a cohort of 26 patients treated with durvalumab plus tremelimumab and followed for at least 24 weeks, the confirmed objective response rate (ORR) was 23% (95% CI, 0.09-0.44). Comparable ORRs were seen in patients from this cohort with PD-L1—positive and PD-L1–negative tumors.

Durvalumab was administered intravenously every 4 weeks for 13 doses or every 2 weeks for 26 doses, and tremelimumab was administered every 4 weeks for 6 doses followed by every 12 weeks for 3 doses.

The combination had a manageable safety profile, with 30% of patients experiencing more than 1 treatment-related grade 3/4 adverse event and 16% discontinuing treatment due to a treatment-related adverse event.

OncLive: What are the most significant findings from the trial thus far?

To learn more about the trial and benefits of dual PD-L1/CTLA-4 blockade in NSCLC, OncLive spoke with lead study author Scott J. Antonia, MD, chair of the Department of Thoracic Oncology at Moffitt Cancer Center.Antonia: There were several. First, the combination is tolerable, and the degree of toxicity is similar to what we have seen with other anti—PD-1 and anti¬¬–CTLA-4 combinations. The most significant clinical finding, which is not proven but is strongly indicated, is that the combination has significant activity in PD-L1–negative patients.

Is there an understanding at this point of why that could be the case?

Again, just to reinforce, this is certainly not proven because of the small study size. However, there was a nice objective tumor response rate in PD-L1—negative patients.There is no definite proof or any studies that have showed why that might be the case. There are theoretical possibilities, and the one that I favor is that anti—CTLA-4 targets the lymphoid compartment, whereas anti–PD-1 or anti–PD-L1 targets the tumor microenvironment.

There are likely a significant number of patients who do not make enough antitumor T cells in their lymphoid compartment. Therefore, if you don’t make enough T-cells, even if they enter into the tumor microenvironment or if they are inhibited in the tumor microenvironment, there still aren’t enough T cells to work. This is even if you blunt the immunosuppressive impact of PD-L1 by putting in these blocking drugs.

How does durvalumab compare with other PD-L1 agents? Is it possible that the combination of PD-L1 and CTLA-4 agents would have this effect with other agents?

How can oncologists decide which of these agents is best for their patients at which point in their treatment? Is there any understanding of optimal sequencing?

What do you think are the biggest questions that still need to be answered regarding the use of immunotherapies in lung cancer?

Some of the patients respond because anti—CTLA-4 is probably expanding both the numbers, as well as the repertoire of the anti-tumor T cells in those patients. Anti–CTLA-4 by itself probably does not have much activity in lung cancer because, even if you expand the T cells, you still need to block the immunosuppressive effect of PD-L1 in the tumor microenvironment. That is why the combination is necessary.It would be conjecture because it hasn’t been looked at systematically with other combinations. However, other combinations are being studied actively in lung cancer, so we will have to wait for those results. My bias is that they will behave similarly.All of those scenarios are being actively tested in clinical trials. It is the results of well-designed clinical trials that will ultimately drive how these agents are going to be used in various clinical settings.With these immunotherapeutics that we’ve tested so far—anti-PD-1/PD-L1 with or without anti—CTLA-4—the major breakthrough was that it not only produced significant improvements in clinical outcomes, but a greater impact on the field is that it gave us proof of principle that lung cancer is an immunotherapeutically responsive disease.

The challenge is that we’re probably only helping about half of patients, and only about one-quarter of patients get major tumor responses and stable disease. Of course, that means that there are still half of patients who we are not helping.

We need to understand what additional immunosuppressive mechanisms are operational in those patients, and to treat them with either other immunotherapeutics or combinations of immunotherapeutics.

Would identifying a biomarker help solve this challenge?

What do you see on the horizon for immunotherapy in lung cancer? Are there any studies that you’re excited about?

I don’t think there is any reason to think that the people who are not responding to anti—PD-1 and anti–CTLA-4 are not immunotherapeutically responsive. We probably just have not identified the proper immunosuppressive mechanisms that are operational in those particular patients.Yes. That’s absolutely required for us to move the field forward because there is such a myriad of ways that tumors can potentially evade rejection by the immune system and there is likely considerable heterogeneity among patients. To be able to identify the proper drugs and combinations of drugs for each individual, we will need biomarkers for that to be clinically efficacious.Studies that are designed to look at the strategy of increasing the number of antitumor T cells are very interesting. What will likely have an impact on the field is resurrecting tumor vaccines, developing chimeric antigen receptor adoptive T-cell therapy, or even tumor-infiltrating lymphocyte adoptive T-cell therapy.

Another big challenge we have is to come up with strategies that will drive more T cells from the blood into the tumor parenchyma. In development, there are a large number of agents that target a large variety of different and potentially immunosuppressive mechanisms in tumors.

We will just have to wait and see the results of clinical trials to know which of these are going to be relevant in lung cancer.

Antonia S, Goldberg SB, Balmanoukian A, et al. Safety and antitumor activity of durvalumab plus tremelimumab in non-small cell lung cancer: a multicenter, phase 1b study. Lancet Oncol. 2016;17(3):299-308.