AACR 2026 | Advancing research in immune-related adverse events in solid tumors

Yeah, so in terms of the critical research bottlenecks, you know, I think the most pressing is that with immune-related adverse events, it’s really difficult in patients to get a pre-treatment sample. So in the context of cancer, getting a pre-checkpoint therapy sample followed by a post-checkpoint therapy sample has been really informative because what many have been seeing in the field is that it’s not what the immune system is doing at baseline that really is the best indicator of how a patient will respond, but that delta from before they got the immunotherapy agent to shortly after they started the immunotherapy agent and how the immune system changes in response to the early initiation of checkpoint blockade is really, really, really informative. And, you know, in cancer, we can very easily say, okay, the tumor is here. We’re going to biopsy it before we start treatment. And then we’re going to do either another biopsy or a surgical resection, you know, at maybe one or two doses after we’ve given checkpoint immunotherapy. We can really understand how the immune system is changing, that dynamic movement. In the context of immune-related adverse events, we really have no idea where they’re going to hit or when they’re going to hit. And so there’s this huge variability in terms of who’s going to develop them. So not everybody actually gets an immune-related adverse event. And then of the subset of patients that does, it can happen anywhere in the body. It can be, you know, the brain or the joints or the heart or the pancreas or the gut. Like it can literally happen in any organ in the body. And even within an organ, you can, you know, for example, biopsy the cardiac tissue if there’s suspected myocarditis. And within six biopsies, some of them can have immune infiltrate and some of them cannot. So even if you know that there’s a suspected toxicity, kind of getting that snapshot in humans can be challenging. And because of that, it’s really hard, if not impossible, to get a pre-treatment biopsy in a location that’s actually going to be relevant. And so we can study the reaction after it pops up, but kind of understanding the ontogeny in humans is really hard. So that’s one of the research bottlenecks. I think another major research bottleneck is that right now, as a field, we tend to study the cancer response separately from the immune-related adverse event. And, you know, even though these look like two very different reactions, they are happening in the same person. And so trying to understand how they’re related and how they’re different and how if you use therapeutics against one, it’ll impact the other is becoming progressively more important. So those are two of the major bottlenecks. I think understanding the relationship between the adverse event and the tumor response is an easier bottleneck to overcome because, you know, banding together, doing team science, really leveraging the expertise of multiple investigators can help overcome that problem. You can get, you know, oncologists, internal medicine specialists, and immunologists together to really do a deep dive into each individual patient to understand that relationship between the tumor response, the toxicity, how they’re similar, how they’re different. You know, a harder bottleneck to overcome is this idea that, you know, we’re always going to be limited in terms of the pre-treatment biopsies that we can get in these, you know, non-malignant tissues because we just don’t know where they’re going to pop up. And to overcome some of those limitations, I think developing better preclinical models has really kind of advanced the field tremendously. I think a major issue with developing preclinical models that are relevant is that there’s still so many unknowns in humans. There’s still so many unknowns for why some and not others. And when they do develop, why they develop where they develop. And so, you know, we proposed a framework that was originally devised by Pam Sharma on the oncology side, where you take observations directly from patients back to mouse models in a process called reverse translation to be able to say, okay, this is what we’re seeing in patients, and then bring it back to the mouse and dive into deep mechanism and then take those insights back to patients again. So that kind of dynamic back and forth. And so I think in the immune-related adverse events space, what we really need to be doing is figuring out the most important drivers of immune-related adverse events in people first, and then taking those observations from our patients back into preclinical models and say, okay, what are the factors that we need to actually model in mice so that they can be translationally relevant and useful? And then we can start probing into deeper mechanism, things like, are cells actually resident or are they coming in? How is it different if they’re resident versus migratory? Is it different if they’re resident versus migratory? And then we can take those lessons back to patients. So I think if we could develop better preclinical models that were more translationally relevant, it would advance the field quite a bit. But I think that we need to start with the human example and really try to learn as much as we can from our patients.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.