Ira Mellman Receives 2022 AACR-CRI Lloyd J. Old Award in Cancer Immunology

Ira Mellman, PhD, a longtime member of the Cancer Research Institute (CRI) Scientific Advisory Council and current vice president of cancer immunology at Genentech, has been named the recipient of the 2022 Lloyd J. Old Award in Cancer Immunology, presented on behalf of the American Association for Cancer Research (AACR) and CRI.

Mellman’s career started with basic research, where he first achieved recognition for discovering endosomes, which help immune cells ingest cancer cell fragments and launch a response against them. Then, in collaboration with CRI scientist and Nobel laureate Ralph Steinman, MD, he defined the development and activity of dendritic cells, crucial immune cells that act as the conductors of immune responses. But Mellman soon felt a stronger pull and left academia for industry where he believed he could make a bigger impact in the lives of those with cancer.

At Genentech, he made perhaps his most “popular” contribution to cancer immunology: the cancer-immunity cycle, which he co-authored with Daniel S. Chen, PhD Describing a series of seven steps, the diagram details in a clear and simple fashion the processes that must occur for successful anti-cancer immune responses. Importantly, the framework also emphasizes the continuous, cyclical aspect that’s required of immune responses if they are to keep cancer in check long-term.

This cancer immunity cycle work, which has already been cited over 4,000 times and become a fixture at cancer immunology conferences, has redefined the way researchers and doctors think about how to effectively employ cancer immunotherapy in the clinic. In addition, it clarified the distinct mechanisms at play in cancer immunity, prompting many to explore these avenues of investigation in deeper detail.

“Dr. Mellman’s success at both ends of the spectrum, from basic research to clinical discovery and application, has been nothing short of remarkable,” according to Jill O’Donnell-Tormey, PhD, CRI’s chief executive officer and director of scientific affairs who first met Mellman while they were postdoctoral fellows at Rockefeller University during the same time. “The future of immunotherapy is on a more prosperous course thanks to his numerous discoveries and paradigm-shifting insights.”

To learn more about Dr. Mellman’s incredible journey that led him to this well-deserved award as well as his thoughts and hopes for the future of the field, we spoke with him in advance of his upcoming Lloyd J. Old Award speech to be given at the AACR 2022 annual meeting.

VIDEO TRANSCRIPT

Arthur Brodsky, PhD

Hi, I'm Dr. Arthur Brodsky, assistant director of scientific content at the Cancer Research Institute, and today I'm grateful to be joined by Dr. Ira Mellman, VP of cancer Immunology at Genentech. Dr. Mellman is receiving the 2022 Lloyd J. Old Award in Cancer Immunology presented on behalf of the Cancer Research Institute and the American Association for Cancer Research. Welcome Dr. Mellman!

Ira Mellman, PhD

Morning.

Arthur Brodsky, PhD

Like many of the great cancer immunologists who have won this award, you didn't start out in cancer. Your early work looked at endosomes, which, for our listeners, are basically kind of envelopes that cells use to uptake materials from the outside and then transport them around inside the cell. We now know that these endosomes are really vital to adaptive immunity, including against cancer. But I'm curious, you started as a basic biologist, what first led you in the immunology direction in general, and later into cancer immunology specifically?

Ira Mellman, PhD

I think the immunology direction was more or less accidental. When I finished my graduate work at Yale Medical School, I was attracted by the work of Ralph Steinman and Zanvil Cohn at Rockefeller University, who were studying endocytosis at that time, and I got there and found out that they were studying endocytosis for the purpose of understanding how the immune system processes antigen so that T cells can be stimulated to generate adaptive responses. So since I was there already, I figured well it just seems like a nice place and these are terrific people, but I knew really very little immunology at that time, so I set upon reading, talking, writing, learning about the field, and what that interface actually was and why it was important. And it didn't take very long, but I was completely converted that taking a cellular mechanistic approach to understanding key problems in understanding the immune response was not only an immense challenge, and scientifically very exciting, but was something that really hadn't been done before. In general, immunologists were not so interested in mechanism, understanding how the cells they studied actually worked in those days. And as I also like to say, cell biologists on the other hand have the problem of losing interest in problems, in scientific problems, as a function of the square of the number of cells involved. Obviously, the immune system consists of a bunch of cells that are doing very bespoke, very complex, very interesting things, but there are a lot of them. So, there was a gaping hole, I think, to try and unite both types of approaches really, for the benefit of both ends of the field and specifically in immunology.

Arthur Brodsky, PhD

Great. Now I want to fast forward to one of your more well-known contributions to the cancer immunology field: the cancer immunity cycle. I think it's really a beautiful checklist, as I kind of think of it. It provides a great overview of what's required for successful anti-cancer immunity, what needs to go right, as well as what can go wrong. And I think in this way, it provides a great overview for doctors to think about what needs to, what might need to be done to a particular patient to get their immune biology on a track that it can then go eliminate their cancer.

Now almost a decade out, how do you feel that this framework developed by you and Dan Chen has helped advance the field? And more importantly, what have you learned since that might expand its scope?

Ira Mellman, PhD

The framework, I think, has turned out to be useful, at least to initially just to us, just to think systematically about the various steps that must occur. And I would be the first to say that none of those steps is a surprise. Anybody who knows anything at all about the immune system will be able to understand that each of these things has to take place. But I think the insight as it relates to cancer immunotherapy was a bit unexpected, which is that if you frame those events in a cyclical series, then you can look at it as you would look at any set of reactions in a closed biological system, meaning that by definition, at any one point in time, there has to be a rate-limiting step. And where that rate limiting step resides–in any one patient or any one indication, but I prefer to think of it in terms of any one patient–defines the site at which therapeutic intervention has to take place. That has a number of very important implications that I think are still not entirely accepted or understood by most of us who are working in this area. For example, checkpoint inhibitors, if you improve the activity of checkpoint inhibitors to the point where they are no longer rate-limiting, it doesn't make any sense to try and improve them some more. You can do it, and maybe you will be able to make more T cells, but if it turns out that penetration through the tumor stroma is, in fact, in a class of patients, the rate-limiting step, you're not going to achieve very much therapeutic benefit, just by providing more and more and more T cells. You have to realize that there's a secondary problem, which is actually the primary problem downstream from T cell generation, so that's where you should be focusing your attention, to targeting either the development or application of new therapeutics or, when they're available, treatment plans for individual patients.

Arthur Brodsky, PhD

Another important part of this obviously is, as you alluded to, the immune system is super complex, and each of these steps, while presented in a simple diagram, and the concept might be simple, the mechanisms and the cells and factors at play are still very complex.

Arthur Brodsky, PhD

One of the kind of big things in the field has been trying to find biomarkers that can let doctors know what's going on inside a patient's body with respect to these different steps, as well as the overall cancer and immune biology. And I know something that you're a really big proponent of is called reverse translation. Typically, in the past, it's been unidirectional, traditionally, from the lab, and then we'll try something in patients, hopefully it works. But then if not kind of go back to the drawing board. But again, you're one of the proponents of reverse translation, where it's a bidirectional where what you learn in the what you learn in people, then you use to do the next round of experiments in the lab, back into the clinic. So, especially with all these biomarkers that we're looking for, not to mention all the different, the number of options with respect to immunotherapy agents these days, how is this reverse translation paradigm, how might it help us discover more cures and biomarkers more efficiently?

Ira Mellman, PhD

I think I can start answering that by really referring to two concepts or two things that Lloyd Old would routinely say, certainly to me that are just exceptionally insightful and influential. First is I learned from him to hate the term translation. So, in fact, I don't like the phrase reverse translation, I like much more his phrase for it, which was clinical discovery, which really relates to precisely what you said, just observe what happens in human patients after a particular intervention. And then worry about analyzing that and not, you know, "translating" it between mouse and human, which, you know, often is fraught with all sorts of misdirections and inaccuracies. So, that's one thing, so the concept I prefer is one, clinical discovery. But try as I might, and he prior to me, no one could get anyone away from this whole translation aspect. So, you can't beat them, join them. Second, that really goes very closely with that is another one of his bon mots, and those of you who knew him, know that Lloyd was filled with them. Many of them actually were, however, quite good. And this one is that the only model for human cancer is human cancer. I think that is true, even if you're studying cancer biology, per se, just to understand how cancer cells work, what gets them to do what they do. And it's particularly true in the case of cancer immunity or cancer immunology because the accretion of slight differences that occur between how a tumor actually, the life history of a tumor as it develops in an individual combined with the slight but nevertheless additively very substantial differences in the immune response that one sees in humans versus mice means that you can get basic mechanisms from looking at mice. You can understand that checkpoints exists, you can understand contributions that the stroma is important, that the tumor microenvironment is important for guiding or sculpting a particular immune response. But to understand that with sufficient precision and accuracy that you can actually do something about it in patients is an entirely different matter. I think the real challenge is to use patients as the source material, as the experimental material. We refer to these as looking for biomarkers. But I prefer to, although one doesn't want to engage, obviously, in human experimentation, that's certainly a bad phrase. But learning, you can kind of say you have the the privilege of learning from patients who are sick, who are engaged in therapies, both experimental and approved, and who are willing to sacrifice of themselves to allow you to study them. So, in some sense, as cancer immunologists, we have a moral obligation to do that well, and to gain as much information as we possibly can, from the sacrifice of our patients, while we are trying to help them knowing full well that not everybody at this point in time is going to be helped. But nevertheless, their struggle, their sacrifice can help us and hopefully, if we're smart enough, we can help the next cadre of patients who undoubtedly are waiting closely behind.

Arthur Brodsky, PhD

Absolutely. Another thing too that Jill [O'Donnell-Tormey] has mentioned – unfortunately, I never got to meet Lloyd – but another thing that Jill has mentioned that he said, was the importance of learning something from every patient. I mean, as you reference, obviously, these are, these are people, they're important, they should never be treated as an experiment. And we don't want their, as you mentioned, their sacrifice to be for naught. In another way, going back to your cancer immunity cycle, when everything is working smoothly, not that you can't learn anything from that, but it's interesting that when something doesn't go right almost offers another opportunity for learning how you can, it shows you then what the cancers trick might be, and you can address it. And it's kind of crazy to you know, Lloyd died in 2010, I believe, right before ipilimumab (the first checkpoint inhibitor) was approved. And it's obviously been a great decade for immunotherapy. It's shown the promise, and I think it's really validated a lot of the past few decades of work. But at the same time that, it still doesn't work for the majority of patients, and we still clearly need better options and better strategies.

Arthur Brodsky, PhD

Looking forward, it has been an amazing, we've gotten so far, but it seems like there's still so much untapped potential. In your opinion, what are the biggest gaps in the field, and what are the most pressing challenges that we're facing right now in order to move the field forward and help even more patients?

Ira Mellman, PhD

I think the biggest challenge, well, obviously there could be one biggest challenge, but there are multiple that are important. One is I think developing the tools and computation ability to be able to understand what's actually happening in patients. At the moment, we rely almost entirely on biopsies and resections to gain material to study. I don't see that changing anytime soon, although what is happening are very exciting advances in terms of what it is you can do with these patient specimens. But you're limited in terms of the numbers of them you can take, so the advantage of mice is of course you can do this set at any time course you like, prior to and following treatment. Humans doing longitudinal studies, doing multi sampling studies, is obviously much more invasive and something that ethically you don't want to do necessarily to another human being. So, it's important to get as much information out of those things as possible. I think there is more activity on that front at the moment, which is great. But one place where there is not a sufficient amount of activity, I think, is to it the development of non-invasive approaches or approaches that will allow you to assess what's going on in the blood. One non-invasive approach would be imaging. I think there's a lot of potential in being able to use biologically active imaging agents that can be developed to understand in a more basic sense what patient status is, what's going on immunologically in the tumor, what's going on metabolically in the tumor, and then gauging the number and different types of cell types that are coming in and out, or not, as the case may be. That's I think another very important way of going about it.

Second, I think we have to take much more stock of those things that are extrinsic to cancer, extrinsic to the immune system, that nevertheless may be playing a very important role. We have our own data now to show that certain drugs that are given even just as palliative measures to cancer patients can have decidedly negative outcomes for patients who are on immunotherapy. And it would be important to understand what those are, and to see what happens if you avoid them. Do you get better outcomes for people if, prospectively, you avoid those particular types of agents?

Microbiome is another one that's created a lot of interest, but I think it's fair to say we understand very little about how that works. Nevertheless, it's certainly something that has to be taken into account. Patient genetics, tumor genetics, the amalgamation of all of these things, I think really needs to be accreted and computed and taken into consideration. So, I guess, the question you asked is what's the biggest challenge and it's all rolled up in that. I think the general problem of how it is you study biology and pathophysiology in actual living human beings. That I think is the best thing. I usually use, although it's a bit nerdy is the Star Trek analogy, which is that anything you can do with a tricorder, you should be able to do in reality. I've watched Star Trek episodes where they're sequencing insertional viruses just by pointing like a mini mass-spec thing and they can get the DNA sequence. Yeah, I mean, they obviously had advice enough from somebody that they could say well this isn't completely nuts. But that's the level that obviously, we will not get there anytime soon, but we need more non-invasive approaches, even things that we can use in the blood. Just looking at peripheral blood is something that's non-invasive. How can we track the progress of an immune response, for example, just by monitoring the dynamics of the T cell receptors that are made? I think to do that, in a fashion that's really meaningful, it's not just a question of looking at how clonotypes vary – I'm sorry, getting a little bit technical here – but rather, in addition, how does the functional activity that's specified by those T cell receptors change, if at all? What do they see, how well do they see it, and what the functional consequences are of a T cell receptor actually seeing an antigen, which may be a tumor associated antigen? And I think I'll end by saying, I think we as immunologists, and maybe it's easy for me to say that starting more on the cell biology genetics side of the equation is immunology is often a very T cell centric field. It seems like there's more T cell immunologists than anything else. That's good and fine, and I love T cells. But, you know, we have to realize that the immune system is a system and that there are other components associated with it that, in complex yet wonderful ways, control what it is the T cells do. The dendritic cells initially tell them what to do, and then they go out, but I think it's becoming more and more obvious that T cells cannot be left to their own devices, even subsequently, because tissues react, tumors react, and we have to understand what those reactions are and how they have to be modified in order to prevent them from creating the next rate-limiting step that again, to use that analogy, has to be addressed in order to really convert a good idea into a good therapy.

Arthur Brodsky, PhD

Absolutely agree. And I think your cancer immunity cycle does a great job of illustrating that the T cells don't come in until after a few steps have already gotten underway. What you said about the less invasive methods really rang home for me. On the one hand, obviously, it helps patients, it's less invasive, and it makes it easier for them, which is a good thing. But also, and this is further down the line, certainly, but I get really excited about the possibility of being able to like detect cancer in advance. As with all new treatments, we're detecting them, we try all new treatments, including immunotherapy, usually in the most advanced cases first, but we've seen a lot of good signs over the past few years that immunotherapy can be more effective earlier in treatment, before cancer has developed all its tactics and strategies. Throughout the years that the cancer is developing these tricks, it's co-opting the whether it be the myeloid cells or the stroma, as you mentioned earlier, so we really have to learn kind of how to attack that ecosystem. And as you mentioned at the beginning, not only the the non-invasive methods, but having the computer power to to sort through these mountains of data and actually make sense of it, find some needles in the haystacks. So, before we go, I just wanted to ask one last question. Given the Cancer Research Institute's pioneering role in the cancer immunology field, and Lloyds Old's role specifically, what does this career award mean to you?

Ira Mellman, PhD

It means a lot because CRI and Lloyd via the Ludwig Institute were really the only beacons in the darkness at a time when neither immunologists nor cancer biologists had any interest in the interface between those two things. You know, go back to another very famous construct of cancer, the famous Weinberg hallmarks, the immune system was not even included in the first hallmarks of cancer, which is a paper that's gotten 5×10 to the-

Arthur Brodsky, PhD

That was, I think, like the early 2000s, right? The first one.

Ira Mellman, PhD

Oh yeah, already at least it should have been there. And now the revision of it does include the immune system, but they've included all sorts of other stuff as well. So it's just like one small thing. Fair enough, but that's not a way to really create a field and emphasize something that by the time over the last 10 years, it was clear this was really going to work. This is not just a fantasy of people like Lloyd Old and Jim Allison and Thierry Boone and many others. This really was something that had legs. CRI and the Ludwig, with part of Lloyd's direction and his inspiration to really lead the field, kept this going and awarded people. The CRI had multiple fellows who have gone on to fantastic careers, given out some awards that have also been extremely important for career development. The Ludwig was a little bit more contained because of its structure, but nevertheless at a time when the NIH and the NCI were really uninterested in promoting or advancing or even testing out whether the field would work. This was the last redoubt and being associated, even in this way, with that history, I think is humbling and immensely satisfying.

Arthur Brodsky, PhD

That's awesome! It still gives me goosebumps. Obviously, I came on board in 2015, so I wasn't there for any of the "dark" years, but it just makes me makes me emotional inside thinking about what all y'all went through. I hear about like all the criticism and doubt and you mentioned places like even the NIH thought these were crazy ideas. And it's got to be it's got to be a great feeling to have seen how far it's come.

Ira Mellman, PhD

To be fair, the field was not in great shape in the early days because there was really not that much to study and immunology as a field didn't really have the types of techniques that we now need, we know we need in order to unravel this. It is also my opinion, although some of my colleagues would vary in theirs, I do not believe that the immune system developed to control cancer. It is there to control infectious disease as its primary object. So, I remember also in the early days there were many really good, real mainstream immunologists who just didn't think that this was a valid course of study, just because that's not what the immune system does. So, why would you want to study basic immunology, which was something that essentially was an artifact? Well, there's a good reason to do it, which is that it's a) interesting and b) the two things do co-exist and c) there was always the chance, and that chance has now been realized, to understand you can make an absolutely fundamental impact on cancer as a disease by persuading the immune system to do its job in a slightly different way.

Arthur Brodsky, PhD

Very well put. Thank you very much Dr. Mellman for taking the time to speak with me, and congratulations on your very well-deserved award. I can't wait to see your speech at AACR 2022.

Ira Mellman, PhD

Thank you very much. It's been a pleasure.