Pedro Barata: Hi and welcome. My name is Pedro Barata, I'm a GU Medical Oncologist. And it's my pleasure to be joined by Dr. Maxime Meylan. Maxime is a research fellow at the Dana-Farber Cancer Institute and Harvard Medical School. We're going to be chatting today about his great work he presented, which has to do with the assay and analysis from the BIONIKK trial. So welcome, Maxime.

Maxime Meylan: Hi. Thank you very much for the invitation.

Pedro Barata: Absolutely. So first of all, congratulations for your great presentation at ESMO in Paris. And so I'm sure you'd love to go back to France.

Maxime Meylan: Sure.

Pedro Barata: Always a good excuse, right? So congratulations for that, and very good. And of course, we know the data well from the BIONIKK. Fantastic. To me, it's one of my favorite studies, because one of the first biomarker based studies in advanced studies.

Maxime Meylan: Thank you.

Pedro Barata: Do you mind reminding the audience simply, what the BIONIKK design was, endpoints, and then basically, we'll dive into the correlatives for this trial. But can you just remind us really quick, the design and the main findings?

Maxime Meylan: The BIONIKK trial is one of the first biomarker driven trial in the metastatic previously untreated clear-cell renal-cell carcinoma. Patients have been randomized, based on their molecular tumor, molecular profile. So their tumor were analyzed by iron sequencing, and they were assigned a molecular group, from one to four.

From previous studies, we knew that molecular group number two responded well to VEGFR TKI therapies, so patients of group two. But nothing was known about the combination nivolumab plus ipilimumab. So patients of molecular groups two were randomized 50/50 to either VEGFR TKI or nivolumab plus ipilimumab. And patients from group one and four, received nivolumab monotherapy or nivolumab plus ipilimumab.

The group three was actually very small, and was randomized with either TKI, VEGFR TKI, or nivolumab plus ipilimumab. And very briefly, this study aimed to evaluate response and progression-free survival. I evaluated by RECIST 1.1. It's a non-comparative trial, so it does not re-aim to compare arms, but evaluated efficacy of these treatments according to the randomization. And basically, we observe high response rate for patient treated by VEGFR TKI in group two, and nivolumab plus ipilimumab as well. That was interesting. And similarly, we saw high response rates of patient randomized from group one and four, that were treated by nivolumab plus ipilimumab.

So as a very brief summary, the BIONIKK trial showed that it was possible to randomize patients there, based on molecular characteristics. That's one important findings. And identified as well, that nivolumab plus ipilimumab worked on a large array of molecular profiles.

Pedro Barata: Got it. Now that's a great summary. Thank you for that. And again, it was one of the first to kind of suggest that it would be a good idea for us to select patients based on biomarker. And of course, I'm thinking after that, we saw data from IMmotion and JAVELIN, doing an RNA-Seq based signature, and predicting whether or not it would be a combination regimen based on immunotherapy would make sense. So kudus to you guys for being one of the first to do it.

And I guess, diving into your presentation this year at ESMO, has to do with this quantification of the immune populations. So first of all, kudos to you guys, because I believe, when you presented the data, you showed there was a high number of specimens collected out of BIONIKK. In my calculations, over, well over 80%. I think it was over 160 samples or so, from the 190 plus patients enrolled. Which is great, because it's not easy sometimes to get access to tissue. So that's awesome.

And then, you guys quantified T-cells, B-cells, tertiary lymphoid structures, other markers of immune activation like PD-1. I guess the question is, was this pre-planned, and why these for immune population? Why did you want to look into that?

Maxime Meylan: Exactly. So first of all, I should remind the viewers that the BIONIKK trial is a French study, and I've participated in this study while I was doing my PhD. Now I'm continuing here in the US. And the translational part of this trial was really embedded with the study at the beginning. So the goal was really to collect as comprehensive materials as possible.

And first of all, tissue, which is obviously needed for molecular characterization. So it is a part of the inclusion criteria, if I remember well, to have a tissue availability. And then, we also collected blood and other patient materials. And so, this clinical study is associated with the translational programs that's led with immunologists, Wolf and Catherine Fridman, are a big part of this project, and the main part. And so, we had in mind, to study the tumor microenvironments in depth. So we did RNA-Seq. We did multiple immunochemistry markers. And obviously, the ESMO presentation be five minutes. It's a short time to present the full spectrum of what we have prepared, and what will be published in the accompanying manuscript. So we chose to focus on the most interesting results, and the most associated with relevant clinical parameters, such as response and progression-free survival.

And recently, I mean in the 2020, the roles of B-cells and tertiary infrastructures have been discovered, and put in perspective with immunotherapy response, particularly in many cancer types such as melanoma, sarcoma and so on. Therefore, that's why we focused. So you see it's the work of Catherine and Wolf Fridman in lab in the big parts. So that's why we focused on these elements in this BIONIKK trial. And we were then happy and intrigued to find, that these TLS and the T-cells that were present inside of them, were associated with response for the nivolumab plus ipilimumab arm, and the nivolumab as well. And a longer progression-free survival for the nivolumab arm.

Pedro Barata: No, that's very, very helpful. It makes sense. And of course, as you're talking, I'm thinking, you guys have also in France, you guys have a very comprehensive way of analyzing tumors, and that's great, obviously. Now, I was trying to think, how can we apply this in clinical practice? Right? Because not all sites have these availability, and obviously, most of them would outsource it, I guess.

Maxime Meylan: Mm-hmm. Yep.

Pedro Barata: So I guess, the next question I have for you is, if you look at the world today, we've come a long way, because we now have ipi/nivo, that was included in BIONIKK, but we also have IO based combos including TKI. Right?

Maxime Meylan: Exactly.

Pedro Barata: And actually, it sounds like oxy/pembro, we had cabo/nivo, and lenv/pembro, are being used more commonly, and so, they are challenging the role of ipi/nivo as well. So ipi/nivo is right there, but a big number of patients are being treated with IO TKI. So I guess, the questions for you. One, do you see a role for these kind of predictor marker to be used with the IO TKI, or are you aware of that research being conducted with those combos?

And the second question is, how do we envision this to be available to treating physicians out there? I have to imagine, we'll have to outsource the tissue, send the tissue somewhere. Who's going to do that for us? That we might have other issues like, time to send the tissue, turn around time.

Maxime Meylan: Exactly. Yeah.

Pedro Barata: What information we're going to get. Does that depend on the front-line therapy going to use, IO/IO versus IO/TKI? So how do you put all these together? What are your thoughts?

Maxime Meylan: Two great questions. And if I go back to the first one, at the kind of therapeutic strategies we can apply to treat patients is evolving at an amazing speed, and that's a fantastic opportunity for the patients, so that's great. For researchers, sometimes things can evolve a bit too fast for us to design long studies. But we are first of all, right now, validating our markers on nivo/ipi in the independent cohort, like the first step. We are also looking at IO/TKI treated patients, because as of right now, it's a bit of a question mark, how our marker are going to stratify or not the patients. But from a biological, or maybe larger overview, we can see that in cancer, many treatment strategy revolves around immunotherapy. Whether that is immunotherapy in combo, or in combo with other type of drugs. And therefore, I think the immune markers are still relevant, to identify patients with benefit for this therapy, because we are targeting active or inactive immune sets.

So I think, there will be still markers, maybe not necessarily exactly the one we have identified, not necessarily the same thresholds, but I think there's still many thing to do, and obviously, we have to look at the new treatment approaches, whether they are combo in one way or the other.

Concerning the other question, it's also a different challenge. How can we go from an idea on interesting results, to something we can use routinely in the clinic? For this, I think prospective clinical trial, as we run on these kind of biomarkers. It has been done in sarcoma by Professor Antoine Italiano, where a patient was selected based on the presence of TLS in their tumor, and they're receiving current immunotherapy. And I was published recently in Nature Medicine, the studies for pembro sarc. And so, they are a beginning of a selection of patient based on immune biomarkers, with more specifically tertiary infrastructures.

And from a feasibility point of view, there is the threshold. Do we use zero, one, two, and so on? But more specifically, how we detect them. A pathologist, or anyone who has done, looked at many images, can tell you that, we can see them from the H&E slide. So it's not something that necessarily needs to be outsourced. Any pathologist can count, on a given area or slide, how many organized structures you can see. And so, that's one point.

Of course, there are details. How the structure mature, how many cells, and so on. That's technical details. And also, we are working towards finding maybe, proxies in the blood. Can we identify from the blood if we have a structure or not, or other parameters that we may correlate with the tissue. And this is maybe the next frontier. Right? That, that might be how we translate that into the clinic, and we stratify patient.

Pedro Barata: Yeah. No, those are very, very important points. And as you were talking about some of the things actually pathologies can do under the microscope, I guess. And that will also require a learning opportunity, and some time for education there, for us to be starting paying attention to certain factors or parameters, that we are not paying attention today, as much. Right?

Maxime Meylan: Exactly. Yeah.

Pedro Barata: The common report, from path reports, for example. So no, those are great points, Maxime. This was a great conversation. Again, congratulations for your great talk. I know it was-

Maxime Meylan: Thank you so much.

Pedro Barata:... fast, fast, very fast. Right? Because they limit the time.

Maxime Meylan: Yes.

Pedro Barata: And I think, what you did made a lot of sense, to select it into those four immunes specifically. So you showed that with TLS. And it was a very, very nice conversation today, and I'm looking forward to reading your paper.

Maxime Meylan: Thank you.

Pedro Barata: And also, on this. And obviously, I'm looking forward to seeing the validation cohort, and what's coming out of IO/TKI combos. So great, great job, great future ahead. And thank you for taking the time to sit down with us today.

Maxime Meylan: Thank you so much.