James Allison on the Future of Checkpoint Inhibitors in Melanoma

Partner | Cancer Centers | <b>MD Anderson</b>

James Allison, PhD, discusses the rapidly expanding field of immunotherapy in melanoma.

James P. Allison, PhD

Immune checkpoint blockade has become a pillar of cancer treatment, says immunotherapy pioneer James Allison, PhD.

“It has become clear, in the past few years, that immune checkpoint blockade has really taken its place along with radiation, surgery, and chemotherapy,” says Allison, chair of Immunology at the University of Texas MD Anderson Cancer Center. “However, it differs from the other pillars somewhat, in that it can really be curative in certain cancers.”

Melanoma is one such cancer where checkpoint inhibitors have proven to be extremely beneficial.

The FDA recently expanded its approval for the PD-1 immune checkpoint inhibitor nivolumab (Opdivo) to include it as frontline treatment of patients with BRAF wild-type advanced melanoma. The expanded approval was based on the CheckMate-066 trial, which found that the median overall survival was not yet reached with nivolumab compared to 10.8 months for dacarbazine (HR, 0.42; 95% CI, 0.30-0.60; P <.0001). Median progression-free survival (PFS) with nivolumab was 5.1 versus 2.2 months for dacarbazine (HR, 0.43; 95% CI, 0.34-0.56; P <.0001).

Checkpoint inhibitors have also shown success in combination regimens. Pembrolizumab (Keytruda) plus low-dose ipilimumab (Yervoy) demonstrated an overall response rate (ORR) of 56% in the phase Ib KEYNOTE-029 trial, presented at the recent 2015 Society for Melanoma Research Congress. Among treatment-naïve patients (n = 63), the ORR was 57% and among those who had been previously treated, the ORR was 44%. The disease control rate (ORR plus stable disease) was 79%.

Despite these successes, challenges still remain regarding the use of checkpoint inhibition. PD-L1 testing remains controversial, with some experts stating that it does not accurately identify all patients who may respond to PD-1 immune checkpoint inhibitors. Further research is also still needed to better understand how to use checkpoint inhibitors in earlier settings, as well as in different combination treatments.

OncLive: Is there an understanding as to why certain cancers, such as melanoma, respond better than others to checkpoint inhibitors?

Within melanoma, are there clinical trials you are excited about regarding the use of checkpoint inhibitors?

Are there any biomarkers on the horizon that could make it easier for oncologists to select the best treatments for each patient?

To learn more about the rapidly expanding field of immunotherapy in melanoma, OncLive spoke with Allison, a 2014 Giant of Cancer Care and the 2015 winner of the Lasker-DeBakey Clinical Medical Research Award. Allison: The process itself is not directed toward and particular kind of cancer; it has the potential for helping enhance immune response for virtually any kind of cancer. The reality is that some respond more than others, which seems to be loosely associated with the mutational load carried by the tumors. This can be the result of their causation; in melanoma it may be because of ultraviolet radiation induced mutations from sunlight. This cancer responds really easily to checkpoint blockage, as the tumors are already infiltrated with T cells, so it is mainly these inhibitors that are really keeping the immune system from dealing with them well. In melanoma, the combination of blocking CTLA-4 and PD-1 is effective in at least half of patients with melanoma. It is really going to give durable benefit to more than half of the patients who were treated. There is a trial looking at the combination of PD-L1 inhibitors with a BRAF-targeted agent; in one setting, there was a nearly 100% response rate. There are also a few studies examining MEK inhibitors, which you might expect to be very terrible for T-cell responses since T cells need that pathway. However, it has been determined that they are not as destructive to immune responses as previously thought. It may be possible to work out proper dosing levels and really take advantage of that in treating melanoma effectively. There is no single biomarker that is really useful in my opinion. It is really controversial; there is a lot of emphasis being placed on PD-L1 testing. The problem, however, is it’s not necessarily as predictive as one would want it to be. The induction of PD-L1 on the tumors largely happens as a result of gamma interference, which is produced by T cells. That is a very dynamic process, so a particular lesion may not have T cells today, but it very well could next week. It is very dangerous to extrapolate from a baseline biopsy and also to generate an assumption about all the tumors of the body based on the expression of the legion in one tumor. It is predicative in a way, as it can enrich the population of patients who respond to about 40% or so, but if you do not treat patients who do not express it, you are missing several patients who may actually benefit.

What is on the horizon for immunotherapy in melanoma?

Do you see the use of immunotherapy moving into an earlier setting in melanoma?

The biomarker can be used, however, to predict those patients who might benefit from monotherapy with anti—PD-1 therapy, as opposed to those who may really benefit from a combination of anti–PD-1 plus anti–CTLA-4. Therefore, we are in a situation where you cannot really predict who will do better overall, but you can predict who may better respond with a combination versus monotherapy. With the variety of combinations that have come to bear, the response rates are going to continue to go up. I think we are going to get to a point very soon where the majority of patients with late-stage melanoma are going to be able to look toward decades of response. To use the “C” word, I think we are heading toward a cure. I do not think it will go down to the level where it will necessarily be used for people who were just diagnosed and have a very low tumor load. As an immunologist, my sense is that a tumor has to be of a certain size to really generate enough antigens to provoke the immune system to respond optimally. We do not know exactly what point that is yet, but there is going to be a certain cutoff point where we have to wait for the tumor to get bigger to use these types of agents; at least that has been my experience with some mouse models thus far. That is going to surely move earlier than the widely metastatic disease setting, where it is mainly used now.