Proteomics Predictor for Immunotherapy Benefit

Original source here.

‍

Podcast: Proteomics Predictor for Immunotherapy Benefit

• Dr. Rafeh Naqash (Host),
• Dr. David Gandara (Guest)

JCO PO author Dr. David R. Gandara at UC Davis Comprehensive Cancer Center, shares insights into his JCO PO article, “Plasma Proteome–Based Test for First-Line Treatment Selection in Metastatic Non–Small Cell Lung Cancer,” one of the Top Articles of 2024. Host Dr. Rafeh Naqash and Dr. Gandara discuss how the PROphet® blood test supports first-line immunotherapy treatment decisions for metastatic NSCLC patients.

TRANSCRIPT

‍

Dr. Rafeh Naqash: Hello and welcome to JCO Precision Oncology Conversations where we bring you engaging conversations with authors of clinically relevant and highly significant JCOPO articles. I'm your host, Dr. Rafeh Naqash, Podcast Editor for JCO Precision Oncology and Assistant Professor at the OU Health Stephenson Cancer Center at the University of Oklahoma.

Today, we are absolutely thrilled to be joined by Dr. David R. Gandara, Professor of Medicine Emeritus, Co-Director of the Center for Experimental Therapeutics and Cancer and Senior Advisor to the Director at UC Davis Comprehensive Cancer Center and also the senior author of the JCO Precision Oncology article entitled “Plasma Proteome–Based Test for First-Line Treatment Selection in Metastatic Non–Small Cell Lung Cancer.” This was one of the top performing articles of 2024, which is one of the reasons why we wanted to bring it in for a podcast discussion.

At the time of this recording, our guest’s disclosures will be linked in the transcript.

David, it is an absolute pleasure to have you today. For somebody like you who's led the field of lung cancer over the years, I'm really excited that you are going to be talking to us about this very interesting article, especially given that I think you're one of the big proponents of liquid biopsies and plasma-based testing. So, for the sake of our listeners - which comprises of academic oncologists, community oncologists, trainees - could you tell us where the biomarker landscape for non-small cell lung cancer is currently, and then we can try to take a deeper dive into this article.

Dr. David Gandar: Okay. Well, thank you, Rafeh. It's a pleasure to be with you here today. And I think the current landscape for biomarkers for immunotherapy in non-small cell lung cancer is a mess. There's no better way to describe it. That makes this paper describing a new plasma proteomic assay even more important. So I'll just give you a perspective. There are 14 trials, phase three trials, that were done in first line non-small cell lung cancer advanced stage of immunotherapy versus chemotherapy and some other aspects, although they vary tremendously. Some of them were checkpoint monotherapy, some combined with chemotherapy, some combined with CTLA-4 inhibitors and so forth. 12 out of the 14 were positive, 12 got FDA approval. So there are 12 different options that an oncologist could use. Some of them were squamous cell only, some non-squamous, some used PD-L1 as a biomarker driven part of the study. Some used TMB, tumor mutational burden, some were agnostic. So when you put all of this together, an oncologist can pick and choose among all these various regimens. And by and large, it's PD-L1 that is the therapeutic decision maker.

ASCO actually, I think, has done the very best job of making a guideline, and it's, as you well know, called a living guideline, it's dynamic. And it is much easier to interpret, for me and I think for oncologists, than some of the other guidelines. It's got a green light and a red light, it may be kind of orange. And so the green light means this is a strong recommendation by the guideline committee. The orange means it's weak. For this purpose, non-small cell lung cancer, advanced stage, only a very few of the recommendations were green. It's mainly monotherapy and patients with cancers with a PD-L1 over 50%. In our surveys, at our meetings, less than 50% of oncologists in the United States are following these guidelines. Why? Because they don't trust the biomarker. And TMB has the same sort of limitations. They're not bad biomarkers, they're incomplete. They're only looking at a part of the story. So that means we need a new biomarker. And this is one that, I think, the data are quite impressive and we'll discuss it more.

Dr. Rafeh Naqash: Absolutely. Like you said, abundance of many therapy options, but not necessarily everything works the same in different subsets of PD-L1 positivity or different subsets of patients with different levels of tumor burden. And like you said, again, difficulty in trying to identify the right biomarker. And that's a nice segue to this PROphet test that you guys ran. So can you tell us a little bit about the plasma proteomic assay? Because to the best of my knowledge, there's not a lot of validated plasma proteomic assays. A lot has been done on the tumor tissue side as far as biomarkers are concerned, but not much on the blood side, except for maybe ctDNA MRD testing. So what was the background for trying to develop a plasma-based proteomic test? And then how did this idea of testing it in the lung cancer setting come into play? And then we can go into the patient population specifics, the cohort that you guys have.

Dr. David Gandara: Okay. Well, of course there's a company behind this assay, it's called OncoHost, and I'm a consultant for them. And they came to me two years ago and they said, “We have something different from anyone else.” And they explained the science to me, as well as some other lung cancer experts here in the United States. I'm not a proteomic expert, of course, but they developed an AI machine learning platform to assess plasma proteins in normal people and in people with cancer, and specifically then in people with non-small cell lung cancer. They identified over 7,000 proteins that had cancer implications for therapy, for resistance, for prognosis, etc., and they categorized them based on the literature, TCGA data, etc., and used this machine learning process to figure out which proteins might be most specific for non-small cell lung cancer. And that's where they started. And so out of that 7,000 proteins, where they've identified which ones are angiogenic, which ones are involved with EMT or cell cycle or whatever it might be, they distilled it down to 388 proteins which they thought were worth testing in non-small cell lung cancer. And that's when I became involved.

They had a retrospective cohort of patients that had been treated with various immunotherapies. They looked at the analytic validation first, then applied it to this cohort. It looked good. Then they had a very large cohort, which they split, as you usually do with an assay, into a test set and then a validation set. For the test set, they wanted something more than a response. They wanted some indicator of long term benefit because that's where immunotherapy differentiates itself from chemotherapy and even targeted therapy. And so they picked PFS at 12 months. And I became involved at that point and it looked really good. I mean, if you look at the figures in the manuscript, the AUC is superb about their prediction and then what actually happened in the patient. And then in this paper, we applied it to a validation set of over 500 patients in a prospective trial, not randomized, it's called an observational trial. The investigator got to pick what they thought was the best therapy for that patient. And then in a blinded fashion, the proteomic assay experts did the analysis and applied it to the group.

And so what that means is some of the patients got chemotherapy alone, some got checkpoint immunotherapy monotherapy, some got in combination with chemotherapy. None of the patients in this study got a CTLA-4 inhibitor. That work is ongoing now. But what the study showed was that this assay can be used together with PD-L1 as what I would call a composite biomarker. You take the two together and it informs the oncologist about the meaning of that PD-L1. I'll give you an example. If that patient has a PD-L1 over 50% in their cancer and yet the PROphet test is negative, meaning less than 5 - it's a 0 to 10 scale - that patient for survival is better served by getting chemotherapy and immunotherapy. However, if the PROphet test is positive and the PD-L1 is over 50%, then the survival curves really look equivalent. As I said earlier, even in that group of patients, a lot of oncologists are reluctant to give them monotherapy. So if you have a test and the same sort of example is true for PD-L1 0, that you can differentiate. So this can really help inform the oncologist about what direction to go. And of course then you use your clinical judgment, you look at what you think of as the aggressiveness of the tumor or their liver metastases, etc. So again, that's how this test is being used for non-small cell lung cancer. And maybe I'll stop there and then I'll come back and add some other points.

Dr. Rafeh Naqash: I definitely like your analogy of this therapy de-escalation strategy. Like you mentioned for PD-L1 high where the PROphet test is negative, then perhaps you could just go with immunotherapy alone. In fact, interestingly enough, I was invited to a talk at SITC a couple of weeks back and this exact figure that you're referring to was one of the figures in my slide deck. And it happened by chance that I realized that we were doing a podcast on the same paper today.

So I guess from a provocative question standpoint, when you look at the PD-L1 high cohort in the subset where you didn't see a survival difference for chemo plus immunotherapy versus immunotherapy alone, do you think any element of that could have been influenced by the degree of PD-L1 positivity above 50%? Meaning could there have been a cohort that is, let's say PD-L1 75 and above, and that kind of skews the data because I know you've published on this yourself also where the higher the PD-L1 above 50%, like 90% PD-L1 positivity survival curves are much better than 50% to 89%. So could that have somehow played a role?

Dr. David Gandara: The first thing to say is that PD-L1 and the PROphet score, there's very little overlap. I know that sounds surprising, but it's also true for tumor mutational burden. There's very little overlap. They're measuring different things. The PD-L1 is measuring a specific regulatory protein that is applicable to some patients, but not all. That's why even in almost all of the studies, people with PD-L1 0 could still have some survival benefit. But in this case they're independent. And not in this paper, but in other work done by this group, the PROphet group, they've shown that the PROphet score does not seem to correlate with super high PD-L1. So it's not like the cemiplimab data where if you have a PD-L1 of greater than 90%, then of course the patient does spectacularly with monotherapy. The other thing that's important here is they had a group of around a little less than 100 patients that got chemotherapy alone. The PROphet score is agnostic to chemotherapy. And so that means that you're not just looking at some prognostic factor. It's actually clinical utility on a predictive basis.

Dr. Rafeh Naqash: I think those are very important points. I was on a podcast a couple of days back. I think there's a theme these days we're trying to do for JCO Precision Oncology, we're trying to do a few biomarker based podcasts, and the most recent one that we did was using a tissue transcriptome with ctDNA MRD and you mentioned the composite of the PD-L1 and the PROphet test and they use a composite of the tissue transcriptome. I believe they called it the VIGex test as well as MRD ctDNA. And when your ctDNA was negative at, I believe, the three month mark, those individuals had the highest inflamed VIGex test or highest infiltration of T cells, STING pathway, etc. So are there any thoughts of trying to add or correlate tissue based biomarkers or ctDNA based correlations as a further validation in this research with the company?

Dr. David Gandara: Right. So there are many things that are being looked at, various composites looking at the commutations that might affect the efficacy of immunotherapy and how they correlate with profit positivity or negativity. And I'll just give the examples of STK11 and KEAP1. As you know, there's some controversy about whether these are for immunotherapy, whether they're more prognostic or predictive. I'm one of the co-authors among many in the recently published Nature paper by Dr. Skoulidis and the group at MD Anderson which report that for KEAP1 positive especially, but also SDK11 mutated getting immunotherapy, that that's where the CTLA-4 inhibitors actually play the greatest role. So realizing that this is still controversial, there are preliminary data, not published yet, that'll be presented at an upcoming meeting, looking at many of these other aspects, P53, SCK11, KEAP1, other aspects, TMB, that's actually already published, I think in one of their papers. So yes, there's lots of opportunities.

The other cool thing is that this isn't a test, it's a platform. And so that means that the OncoHost scientists have already said, “What if we look at this test, the assay in a group of patients with small cell lung cancer?” And so I just presented this as a poster at the world conference in San Diego. And it turns out if you look at the biology of small cell, where neither PD-L1 nor TMB seem to be very important, if you look at the biology of small cell and you form an assay, it only shares 44 proteins out of the 388 with non-small cell. It's a different biology. And when we applied that to a group of patients with small cell lung cancer, again it had really pretty impressive results, although still a fairly small number of patients. So we have a big phase three study that we're doing with a pharmaceutical company developing immunotherapy where we are prospectively placing the PROphet test in a small cell trial.

The platform can also be altered for other cancer types. And at AACR, Dr. Jarushka Naidoo presented really impressive data that you can modify the proteins and you can predict immunotherapy side effects. So this is not like a company that says, “We have one test that's great for everything.” You know how some companies say, “Our test, you can use it for everything.” This company is saying we can alter the protein structures using AI machine learning assisted process to do it and we can have a very informed assay in different tumor types and different situations. So to me, it's really exciting.

Dr. Rafeh Naqash: Definitely to me, I think, combining the AI machine learning aspect with the possibility of finding or trying to find a composite biomarker using less invasive approaches such as plasma or blood, definitely checks a lot of boxes. And as you mentioned, trying to get it to prospective trials as an integral biomarker perhaps would be likely the next step. And hopefully we see some interesting, exciting results where we can try to match or stratify patients into optimal combination therapies based on this test.

So now to the next aspect of this discussion, David, which I'm really excited about. You've been a leader and a mentor to many. You've led ISLC and several other corporate group organizations, et cetera. Can you tell us, for the sake of all the listeners, junior investigators, trainees, what being a mentor has meant for you? How your career has started many years back and how it's evolved? And what are some of the things that you want to tell people for a successful and a more exciting career as you've led over the years?

Dr. David Gandara: Well, thank you for the question. Mentoring is a very important part of my own career. I didn't have an institutional mentor when I was a junior investigator, but I had a lot of senior collaborators, very famous people that kind of took me under their wing and guided me. And I thought when I basically establish myself, I want to give back by being a mentor to other people. And you wouldn't believe the number of people that I'm even mentoring today. And some of them are not medical oncologists, they're surgeons, they're radiation oncologists, they're basic scientists. Because you don't have to be an expert in that person's field to be a mentor. It helps, but in other words, you can guide somebody in what are the decision making processes in your career. When is it time to move from this institution onward because you can't grow in the institution you're in, either because it's too big or it's too small? So I established a leadership academy in the Southwest Oncology Group, SWOG. I've led many mentoring courses, for instance, for ISLC, now for International Society Liquid Biopsy, where I'm the executive committee liaison for what's called The Young Committee. So ISLB Society, totally devoted to liquid biopsy, six years old now, we have a Young Committee that has a budget. They develop projects, they publish articles on their own, they do podcasts. So what I'm saying is those are all things that I think opens up opportunities. They're not waiting behind senior people, they are leading themselves.

We just, at our International Lung Cancer Congress, reestablished a fellows program where a group of fellows are invited to that Huntington beach meeting. It's now in its 25th year and we spend a day and a half with them, mentoring them on career building. I'll just give you my first, I have the “Letterman Top 10”. So my first recommendation is if all you have is lemons, make lemonade. And what I'm meaning is find what you can do at your institution if you're a junior person, what you can claim to be your own and make the very best of it. But then as you get further along in my recommendations, one of them is learn when to say ‘no’. Because as a junior investigator the biggest threat to your career is saying ‘yes’ to everybody and then you become overwhelmed and you can't concentrate. So I'll stop there. But anyway, yes, mentoring is a big part of my life.

Dr. Rafeh Naqash: Well, thank you, David. This is definitely something that I'm going to try to apply to my career as well. And this has been an absolute pleasure, especially with all the insights that you provided, not just on the scientific side but also on the personal career side and the mentorship side. And hopefully we'll see more of this work that you and other investigators have led and collaborated on. perhaps more interesting plasma based biomarkers. And hopefully some of that work will find its home in JCO Precision Oncology. Thank you again for joining us today.

Dr. David Gandara: My pleasure.

Dr. Rafeh Naqash: And thank you for listening to JCO Precision Oncology Conversations. Don't forget to give us a rating or review and be sure to subscribe so you never miss an episode. You can find all ASCO shows at asco.org/podcasts.

The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions.

Guests on this podcast express their own opinions, experience and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service organization, activity or therapy should not be construed as an ASCO endorsement.

‍