Durvalumab/Tremelimumab Combo Shows Promise in NSCLC, Regardless of PD-L1 Status

Naiyer A. Rizvi, MD

The combination of the anti—PD-L1 immune checkpoint inhibitor durvalumab (MEDI4736) with the anti–CTLA-4 monoclonal antibody tremelimumab improved tumor response regardless of PD-L1 status in patients with advanced non–small cell lung cancer (NSCLC), according to a study recently published in the Journal for ImmunoTherapy of Cancer.

The phase I, open-label, dose-escalation/expansion study had 84 evaluable patients, 48 of whom had more than two prior lines of therapy. The overall response rate was 25%. Thirty-five percent of PD-L1—positive patients experienced a response, 22% of PD-L1–negative patients saw a response (<25% tumor cell staining), and 33% of PD-L1–negative patients saw a response (0% tumor cell staining).

Higher response rates were observed in those with 1 prior therapy versus 2 or more.

Across all dose level cohorts, 80% of patients experienced a greater than grade 1 treatment-related adverse event and 42% of patients had grade 3/4 treatment-related adverse event (AE). Twenty-eight percent of patients discontinued treatment due to a related AE. A greater frequency of AEs, without a corresponding increase in tumor response, was seen with increasing doses of tremelimumab.

OncLive: What were the goals of this study?

To better understand the impact of these findings and to learn more about what is on the horizon for this combination in NSCLC, OncLive spoke to lead study author Naiyer A. Rizvi, MD, an associate attending physician at Memorial Sloan Kettering Cancer Center. Rizvi: The rationale to do this study is that we know that single agent PD-1/PD-L1 is very active and, clearly, these drugs have a role in lung cancer as a second-line treatment with the recent approvals. To try and improve response and durability of response for more patients, we explored this immunotherapy combination. We hoped that there would be more potent T-cell activation and that we could improve beyond single-agent therapy.

What were the most significant findings in this study?

Were there any concerning toxicities seen with the combination?

What are the next steps in this research?

Are we any closer to understanding which patients may benefit the most from durvalumab either as a monotherapy or in combination?

We are seeing that lung cancer is an immunogenic tumor but, for a lot of different reasons, the immune system can be very suppressive. To overcome that immune-suppression, a more potent immune combination is required for a subset of patients, similar to the observation we saw in melanoma. I think we just need to drive the immune response harder. What we did in the study was examine different doses and schedules of the combination. We did find that escalating the dose of tremelimumab was associated with greater toxicities that were problematic. However, when we looked at the lowest dose level that we tested of tremelimumab, which was 1 mg/kg every 4 weeks, we actually saw that that was enough to activate T cells and induce a greater tumor response. The key takeaways were that we saw a high level of activity in both PD-L1¬—positive and ¬–negative patients in the 30% to 40% range, which is clearly superior to single-agent therapy in both PD-L1–positive and –negative patients. The toxicities were really what one expected with tremelimumab. We did see more toxicities with the combination but, for the most part, they were predictable. The toxicities we observed were gastrointestinal-related, skin-related, liver enzyme elevation, and pancreatic enzyme elevation. Particularly with the lower dose of tremelimumab, these toxicities were treatable. Most patients did not have to discontinue therapy due to side effects; we were able to work around that. I think the data are compelling enough that we are actually setting this combination, administered every 4 weeks for 1 year, in the first-line setting. Most of the patients that we treated in this trial were pretreated and had a number of lines of chemotherapy. We now have an opportunity to give this as a first-line therapy in lung cancer, and that is where we would like to go next. There are two phase III trials with the combination that are ongoing, comparing the immunotherapy combination versus chemotherapy to see if we can position this combination as an effective first-line treatment that is better than chemotherapy. I think that using PD-L1 testing to determine who is more likely to respond to single-agent therapy is useful. One of the limitations is that we still see some of the patients respond. Several researchers are trying to explore other ways to define treatments for patients better. Options include looking at mutational load, which is the total number of mutations that could be potentially immunogenic within the tumor, as well as examining certain RNA signatures or interferon signatures. I think we are getting a better handle on what the components are that may predict response to single-agent therapy. As we envision the future, we will be able to more accurately define those patients who can receive single-agent therapy versus those who should receive combination therapy.

Rizvi N, Chaft J Balmanoukian, A, et al. Tumor response from durvalumab (MEDI4736) + tremelimumab treatment in patients with advanced non-small cell lung cancer (NSCLC) is observed regardless of PD-L1 status. [published online November 4, 2015]. J Immunother Cancer. doi:10.1186/2051-1426-3-S2-P193.