Immunotherapy Increases Melanoma Survival, Yet Toxicity Gaps Persist: Igor Puzanov, MD, MSCI, FACP

Immunotherapy has dramatically improved outcomes in melanoma, helping it become the single most improved solid tumor in terms of survival. The next frontier, however, is understanding and reversing the epidermal-mesenchymal transformation that renders some tumors invisible to the immune system, according to Igor Puzanov, MD, MSCI, FACP, senior vice president of clinical investigation at Roswell Park Comprehensive Cancer Center.

In a conversation with The American Journal of Managed Care^® (AJMC^®) ahead of the 2026 annual meeting of the American Society of Clinical Oncology (ASCO), Puzanov also discussed the gap between immune-related adverse event guidelines and real-world practice, along with the need for better long-term toxicity management infrastructure.

Specifically, he noted that immune-related adverse events remain widely mismanaged in community and emergency settings despite published guidelines. Because of this, Puzanov called for a chronic disease-style management infrastructure to address both acute toxicities and long-term survivorship risks associated with immune checkpoint inhibitors, such as cardiovascular disease.

This transcript has been lightly edited for clarity.

AJMC: Melanoma has arguably been the poster child for modern immunotherapy. For a managed care audience, how would you characterize where melanoma immunotherapy stands today, and what is the most important unanswered question heading into ASCO 2026?

Puzanov: I have been involved in research and development, especially phase 1 of new drugs for patients with melanoma, since 2005. So, I have been fortunate to see all this change for the better happening. At this point, when we talk about stage 4 or stage 3 unresectable patients, we truly cure about half of the patients. It used to be that in the US alone, about 15,000 people died every year from melanoma. Now, that number is 7700, according to the American Cancer Society. Melanoma is actually the single most improved solid tumor in terms of survival and treatment, so we are happy about that. But we still have unanswered questions, like what about the other half of the patients?

Melanoma has been at the forefront of drug development: precision medicine, checkpoint inhibitors, immunotherapy in general, and cytokines in the past, as well as maybe in the future. I want it to continue, and I think the single most unanswered question to me is the question of how we approach the tumors, which are undergoing this epidermal-mesenchymal transformation, becoming more primitive-looking, more characteristic of stem cells.

Those cells, which kind of look like neural cells, are melanoma, but they undergo this transformation. I think there are lessons to be learned, not just for melanoma but for other tumors, because these are the hard tumors to treat, hard situations, and those tumors are not visible to the immune system, probably at least partially because they resemble stem cells, and stem cells do not want to be visible to the immune system. If we can crack that, if we can learn what makes these tumors transform to that phenotype, and you see it in some of the patients treated with checkpoint inhibitors, they transform.

If we learn what actually drove that and learn to reverse engineer it, we can actually deploy these treatments, or these manipulations, in other tumor types, which are very hard to treat, like all the gastrointestinal tumors; they are not really sensitive to checkpoint therapy except for some subgroups. So, I would love to see that approach. There are actually other unanswered questions we can discuss later, but this, to me, is the fundamental huge question, which can have implications for patients with all different types of solid tumors.

AJMC: More patients with melanoma are achieving long-term remission on immunotherapy than ever before, but these treatments can also cause immune-related adverse effects across multiple organ systems. From a health system perspective, when these toxicities are not recognized and managed early, what are the consequences in terms of infrastructure, hospitalizations, provider time, etc.?

Puzanov: To answer your question about the cost, I do not know. It certainly is costly, but we have ways to go, like managing toxicities, developing systems and processes for recognizing it, detecting it, ideally early, and managing it properly. It requires a multidisciplinary approach, probably the deployment of things like long-distance monitoring, and then you touch on survivorship. People who actually had the immune checkpoints and the long-term sequelae, the chronic inflammation, the increased risk of coronary artery disease and vascular diseases were relatively 2.34 times higher compared with people who were not exposed to the checkpoints.

We have actually published a white paper in The Journal for ImmunoTherapy of Cancer recently, with first author Tessa Flores, MD, and others. It was basically our attempt to put together a group of experts to talk about survivorship issues and what we need to do to approach that, like capturing the data, including economic data, because that actually will tell us how to pay for it. I think it will be paid by savings for the health care system; that's probably the best payment. We also discussed what needs to be done, what kind of follow-up is needed, and what needs to be watched.

I recommend that my patients really follow strict guidelines for somebody with coronary artery disease risk, ie, blood pressure of 120 over 70 or better and low-density lipoprotein below 70. I tell them to try not to be too overweight. That's always the struggle, but now it's a little bit easier with GLP-1s, paired with exercise and a diet with fiber and fermented foods, kind of feeding the microbiome to the best we know how.

Obviously, these are just kind of opinions; it's not data-driven advice. For that, we need to create databases of survivors and look at these databases, probably with the help of large datasets and large kinds of data analytic groups, maybe like Flatiron or similar, look into that in more detail, and then deploy systems and measures to ameliorate these long-term toxicities.

AJMC: Immune-related adverse events are now a recognized category of harm that primary care physicians, emergency departments, and non-oncology specialists are increasingly encountering. What is the gap between guideline recommendations for toxicity management, and what is actually happening in community settings?

Puzanov: It's a large gap. That is something I would like to improve before I retire. Though people don't recognize it, I'm actually surprised. You probably know that we created the first US-based system for toxicity management for checkpoint inhibitors with the Society of Immunotherapy in Cancer group and then the American Society of Clinical Oncology, National Comprehensive Cancer Network, and European Society of Medical Oncology. All of the groups worked together, coordinated it, and created them. They are accessible online, and there are books.

Marc S. Ernstoff, MD, and I coauthored one of them, and I gave it to everybody at all the different emergency departments I work with. To my surprise, people still do not recognize it. When they do recognize it, they do not adjudicate the severity based on the Common Terminology Criteria for Adverse Events (CTCAE) grading system, which is critical because you need to know if it's grade 1, 2, 3, or 4. The gradient is not opinion. Diarrhea, for example, is not bad or moderate; there are numbers. People don't do it.

When I was in training 20-plus years ago, we actually carried books in our pockets, and I think we actually used them more than people do now, when everything is electronic.

AJMC: You co-authored the original Society for Immunotherapy Cancer consensus recommendations on managing immune checkpoint toxicity, now in a second edition. What has changed most significantly in those guidelines since the first version, and what drove those updates?

Puzanov: We also worked on CTCAE criteria because they were developed for chemotherapy drugs. The old ones could be very well used for targeted agents and toxicities, but immunotherapy toxicities are different, so one of these adjudications of grades 1, 2, 3, and 4. Some of the toxicities were recognized recently. Their importance and their management, I would point out some changes to the cardiotoxicity.

It's still not completely non-controversial, because for these rare toxicities, but potentially lethal, like its cardiotoxicity, for example, it’s rare, less than 1%. But for the cases developed, unless managed immediately and aggressively, the mortality is between 30% and 50%. It's hard to do prospective trials because of the low numbers in multiple centers dispersed around the country.

So, we tried to actually develop some recommendations. These are expert opinions based on some retrospective data, like monitoring of troponins weekly, considering that strongly for the first 6 weeks, because if cardiotoxicity becomes clinically symptomatic, it usually causes severe damage to the heart, to the myocardium, and to other organs. If you just find a troponin increase, like weak troponin, it may be falsely elevated. Yes, you may end up doing some things that may not be completely necessary, but in the end, you will actually cherish that for that patient, where it's real and where you actually found out about cardiomyocytes, when the damage is minimal, or just biochemical markers are elevated, vs a patient in third-degree AV block with depressed left ventricular systolic function and potentially dying.

A lot of these patients may at the same time have a nice response to their cancer, so their cancer is well managed, yet the patient is not here with us anymore. That’s not what we want from our treatments; we want to do no harm, and in the same way we manage the cancer, we want to manage the adverse effects. I think it's so critical to manage the adverse effects, toxicities, skin toxicities, chronic toxicities, and some of the endocrinopathies.

Also, I will point out that we recognize that things like type 1 diabetes may develop. We also recognize, speaking of cardiotoxicity, not just the classic cardiomyocytes, inflammation, and destruction of myocardium by lymphocytes, but also things like conduction abnormalities and arrhythmias. Even with coronary artery disease exacerbation, I always had a suspicion that it would come to it, because it's all inflammation; it's a rupture of the plaque; that’s how heart attacks happen. I said, “I would not be shocked if we find out that, with these drugs, we can cause rupture of the plaque,” and it can happen. So, for somebody with coronary artery disease, it's of utmost importance to watch them closely.

In my practice, I do the weekly troponin measurements, and sometimes it's kind of complicated, but the alternative is to do nothing and then hear about something too late, and I don't want that for my patients at all. I'd rather spend some time doing unnecessary work than miss that one case where it's all necessary, but we need better systems. These large organizations can actually use that organizational and logistical knowledge and ability and use that to manage this in the same way we learned how to manage congestive heart failure patients after discharge: call centers, calling them, managing their weight, the set of questions, taking more diuretics, and preventing readmissions.

I think we can actually create these systems for our patients on immunotherapy, which I think is the way forward. Something like the health ways are doing for these, like blood pressure management, diabetes management, and kind of a chronic disease management. We just need to develop those systems, and I definitely want to be part of it. I'm highly motivated to do that in the latter part of my career.