Finding the Mechanisms of Immunotherapy Resistance

Jesse Zaretsky (right) at the research laboratory of Dr. Antoni Ribas (left) at the University of California, Los Angeles.

Anti–programmed death 1 (PD-1) antibodies are a form of immunotherapy that are now approved by the US Food and Drug Administration for several tumor types, including metastatic melanoma, non–small-cell lung cancer, and renal cell carcinoma. Treatment with these antibodies has resulted in durable, robust antitumor responses in patients with extensive metastatic disease, some of whom have no other treatment options. Yet, disease progression occurs, at least in patients with metastatic melanoma, which has been documented after an initial objective response. In a recent New England Journal of Medicine article, researchers at the University of California, Los Angeles (UCLA) have identified genetic mechanisms of immuno-resistance to these antibodies. Today, we are speaking with one of the authors of the study, Jesse Zaretsky, BS, a medical scientist in training and research in the laboratory of Antoni Ribas, MD, PhD, at UCLA.

- Interviewed by Anna Azvolinsky

Cancer Network:First, can you describe the type of relapse that’s been seen in patients treated with an anti–PD-1 antibody, who have initially responded?

Jesse Zaretsky: Sure. Thank you for having me on today. So, if you look at the clinical trial results for the anti–PD-1 therapy in melanoma, the typical pattern that’s been emerging is that two-thirds of patients have no response at all-they are upfront or intrinsically resistant. The other one-third of patients do have an objective response, meaning tumor regression, and it’s typically long term. From the early phase I data from a clinical trial with the longest follow-up-almost 2 years out now-75% of the patients that had a good response are still responding. That means that 25% have relapsed. In our experience at UCLA, out of 42 patients that had a good response to therapy, 15 of those had some form of relapse. Sometimes that means a single new lesion or nodule that comes back. Sometimes it means a new metastasis and progression of their disease. Of those 15 we saw at UCLA, we had available tumor to study in about 4 of those patients.

Cancer Network:Can you talk about how you addressed the potential mechanisms of resistance in this new work?

Jesse Zaretsky: Anti–PD-1 therapy and immunotherapy in general is a bit different from traditional cancer therapy, in that the drug is modulating the activity of the immune system, but it’s the immune cells, the T cells, that are actually doing the work of killing the tumor cells. In thinking about resistance, we had come up with a few hypotheses. It could have been that the immune cells, the T cells, weren’t even around in the tumor area to do their job or they could have been around but dysfunctional, either not recognizing the cancer, not getting properly activated, or perhaps being actively suppressed by other immune cell types or the tumor itself. Or it could have been that the tumor had just become resistant to the way the immune system kills the cancer.

In the specific cases we looked at, the immune cells were still there, they were still present at relapse, and that was an easy thing to look at. But in these cases, the tumor had shrunk down, almost disappeared, and took a year or almost 2 years to come back, and so this looked more to us like natural selection, akin to antibiotic resistance where the few cells that survive had escaped and then came back and were resistant. We tried to look at the tumors at the genetic level with whole-exome sequencing. That was an unbiased method where we looked at every gene for possible mutations and then at the tumors. We had paired biopsies from these individual patients, one that was taken before the patient went on therapy and then another after relapse, so that we could compare and see what had changed. We then narrowed down the list of candidates that could cause resistance.

Cancer Network:Can you describe the results of the study?

Jesse Zaretsky: In two cases, we found that 90+ percent of the mutations that were present before they went on therapy were still present at the end of therapy. But we did find new mutations in JAK1 and JAK2; those genes are critical for interferon pathway signaling, and signaling through that pathway leads to immune sensitization and growth arrest. The term “interferon” comes from the ability of these molecules to interfere with the growth of, originally, virally infected cells. Interferon, and interferon gamma in particular, is one of the main factors secreted by activated T cells. Those mutations were inactivating, the tumor cells had lost the other normal wild-type copy and they had become completely insensitive to this molecule that the immune system was using to try and slow them down and kill them.

In another case, we found a mutation in B2M, or beta-2-microglobulin, which is necessary for presenting molecules on the surface of cancer cells that the immune system uses to recognize the cancer in the first place. The tumor cells were sort of hiding out by deactivating that gene. And in the fourth case we were not able to find a mechanism, but we are still looking.

Cancer Network: Have other mechanisms of immuno-resistance with either other anti–PD-1 antibodies or other immune-based therapies been documented previously?

Jesse Zaretsky: The two mechanisms that we found were the first demonstration of them in actual patients and in a real clinical scenario where they were lost or selected against during the course of treatment. The mechanisms themselves have been described. The importance of interferon gamma in immune rejection was first studied by Robert Schreiber’s group back in the 1990s. Nicholas Restifo’s and Steven Rosenberg’s work at the National Cancer Institute had identified B2M loss in patients with an earlier immunotherapy. There are also lots of other hypotheses about immune resistance-other checkpoint inhibitor molecules that are suppressive factors for the immune system, other things in the microenvironment, but none of these have been demonstrated in patients yet.

Cancer Network: What are the implications of these new findings?

Jesse Zaretsky: We hope that if we begin to catalog the main escape routes that the tumors use to begin avoiding the immune system, we can design drugs, combinations of drugs, or sequencing of therapy to begin to address that. This is the first step in understanding the way that the tumor gets around these drugs to inform our decisions of how to move forward in the future.

Cancer Network: Thank you so much for joining us today, Jesse.

Jesse Zaretsky: Thank you.