Read the Full Video Transcript

Pedro Barata: Hi, I'm Pedro Barata. I'm a GU medical oncologist at Tulane Medical School, New Orleans, Louisiana. This morning, I have the pleasure to be joined by Dr. Yann Vano. He's a medical oncologist at the European Hospital Georges-Pompidou in Paris, France. Dr. Vano, such a pleasure to have you here with us.

Yann Vano: Thank you for the invitation.

Pedro Barata: Absolutely. And the reason why we sat down, we're going to chat a little bit about your trial called BIONIKK. We know about that trial presented in important meetings like ASCO, but the paper came out in Lancet Oncology. And what a fantastic job, for those of us who are listening, this is a phase two biomarker-driven trial in patients with advanced clear cell renal cell carcinoma. And what's unique about this trial is one of the first, really to be a biomarker-driven approach. So multiple cohorts that Dr. Vano will comment more on, but basically based on a gene expression signature that I think Dr. Vano and the group have developed in the past.

So Dr. Vano, I don't know if this is a fair summary of this trial. I really love your design. So maybe I'll ask first things first, right, can you comment a little bit about, share with us the methodology behind it? Because you were sequencing the tumors from primary tumor, but also from metastatic sites, from patients with advanced RCC, and you send them out, and then you got the results and you were able to allocate patients to the right cohort in your particular trial that included nivolumab, Ipi-Nivo, but also TKI sunitinib pazopanib. Can you talk to us a little bit about, how do you put that together and how did that methodology work for you? Because the turnaround time had to be relatively short for you to be able to do this.

Yann Vano: Yeah, yeah. That was a big challenge. Yeah. In the prior study, we identified the four molecular group based on transcript in data. And we found that these four groups are distinct biological profile and distant micro environment and had not distant sensitivity to sunitinib in a prior cohort. So roughly group one and four had lower efficiency to sunitinib compared to group two and three.

And we think about that. And we design, we constricted a classifier of 35 genes to perform on qPCR, to allocate molecular group patient by patient prospectively. That's why we propose this kind of design in the binding trial, because group one and four, were not very sensitive to Sunitinib, so we decided to randomize them between nivolumab and nivolumab Adalimumab and as group two and three, were quite well, really sensitive to sunitinib we just compare, we randomize these patients to TKIs, sunitinib pazopanib or nivolumab Ipilimumab. The main challenge is that we had to group, we had to perform this qPCR on frozen tumor samples. So we used the available tumor samples that we have in your archives center. So that's why we used a mix of primary tumor in 70% of the patient. And sometimes we have archived frozen tumors from metastatic sites, so we use them. And we decided if we can avoid any biopsy, we decided to avoid it.

Pedro Barata: Yeah, I know this is very, very helpful. Can you comment briefly on the turnaround time. Were you outsourcing the sequencing or you're doing it in house? Because, I predict a lot of the readers going to think, can I do this, right, at my own institution?

Yann Vano: It's it was based on qPCR on low density. And we decided to do a run every week or every two weeks. So was a turnaround time, we fixed it at less than two weeks and we succeeded to have a turnaround time around nine to 10 days.

Pedro Barata: Wow.

Yann Vano: So, but we decide it cost a lot of one day because sometimes we have three, four samples to put it on the plate where we can put 48 samples. So sometimes we have to run the analysis before we get all the plates filled with the sample, so that's quite challenging. And we performed it in the central lab, which is a research lab. I have to say that it is a research lab and not a clinical practice. It's not a routine lab, it's research center. And we analyzed all the sample parts in other lab and we found similar findings.

Pedro Barata: Got it. That's very helpful. Thank you. So I guess I'll ask, I mean the papers out, but I'll let you of course highlight the main results of the different cohorts, if you'd like. And also I'm kind of curious, can you give us a little bit of idea about what can we see or expect with your molecular signature, if you will, in the different risk group criteria, good risk, intermediate, and poor risk?

Yann Vano: Yeah. From the prior study, we already know that group one and four were associated with poor prognosis, poor response to sunitinib of course, but prognosis with higher rates of IMDC intermediate to prior group, group one and four. The main differences between one and four, that one were quite immune desert. And four were quite very inflammatory, very highly infiltrated by immune cells.

And not surprisingly in bionic, we found that the higher rate of IMDC poor risk group was group four. But we, and the higher rate of variable risk group as expected was in group two. Particularly in group two, there were nearly 40% of IMDC variable risk group, which is quite high. And we know that this group two is highly enriched in antigen disease factors. And it has been demonstrated by the work of Bernard Bouslog that this group is very driven by angiogenesis. So it is not surprising to see that in group two, we have high rates of, high response rate with TKI, around 50%, and a very good progression-free survival median at 14 months, around 14 months, which is quite similar to nivo-ipi arm in this group with 50% also of response rate and 11 months of median PFS for this group too. So group two is really a pro angiogenic group and enrich in IMDC variable risk group. We know that these patients are, their disease are driven by angiogenisis.

And for group one and four, it is, for group four it is clear that we have the high rate of IMDC poor risk group, in this group four. This is a very poor prognosis factor to part of this group. And that in this group, the IO based arm provided the best efficacy with best overall response rate, 44% with nivo, 50% with Nio-Ipi, which is higher than in group one. Group one, it's quite a challenging group because I think nivo and nivo-Ipi do not perform very well. 29% of response rate with nivo, 39 with nivo-Ipi, median PFS lower than eight months, or for both arms. It's quite challenging in this group, I think.

Pedro Barata: Right. That's a fantastic summary. And we're thinking as we are going through the numbers again, right, we're thinking, what can we do better, right.

And I guess, that kind of brings me to the next question, which is, this is a fantastic proof of concept in my opinion, right. And we start seeing other studies being explored, whether we use this gene expression signature or we use a different one, there's probably one of the best developed out there right now, similar to yours is perhaps the one from Emotion, right. That work from Tesobath and from Dr. Reeni, Dr. Moser and others. And there's actually one trial going on as we speak, another proof of concept randomizing patients based on those clusters.

But going back to BIONIKK, I guess the question is, of course things change, right. When you enrolled these patients was a few years ago, and now we have a boom of different IO based combos available. And so I guess the question is, what do next steps look like for, for BIONIKK? Are you working on a BIONIKK 2.0? How are you going to incorporate these IO-TKI combos for instance? Are you going to apply that to the group one, where the responses to IO don't seem to be as good?

But I'm wondering your thoughts about where can we go with it, for us to really bring it to the clinical practice? What we need to prove so that practicing physicians all over the world can actually start using this information to offer a personalized approach to their patients?

Yann Vano: I think it's a first step, but we have a lot of works to do because first it was made on frozen tumor tissue. And it's very difficult in the clinical practice to have frozen tumor tissue. So I think we have to, we are looking about a lot of biomarkers in C2 in FFP tumor samples.

And I think the future relies on the availability of tumor samples. So I think we are currently working on finding some surrogate biomarkers that recapitulates these four groups, and to look at if we can do better in the prediction of response because you see that we are not reaching 80% of response rates.

Pedro Barata: Right, right.

Yann Vano: So we can do better. And we see with TKI, IO such as Pembro that we are very high in response rates. So we can, I think find a better way to send this patient. And this is currently ongoing. I think it's a very important to look and to put all the biomarkers that we are looking for in perspective, to think about the BIONIKK 2. So it's, I think it will come in the next month, but we have to look at all the institute biomarkers and probably in the blood circulating biomarkers.

We are looking at that and we are... The work that is ongoing it's to see if we can link something that happens in the past on the tumor tissue, in the primary tumor. And that I'll be reflecting in the blood at the baseline. I think it's very critical. And some work has been done in arterial tumors or other works on the repair trial of T cells and something like that.

I think as they are, the regular groups on frozen tumor tissue, it's not easily usable in routine clinical practice, frozen tissue. I think it's so difficult. It's too easy, to the cost is too high, but the transcriptome data in FFP samples, I think it's valuable. And with the lower cost, I think in the near future, we are working on that new biomarkers to find a better way to separate responders from non-responders.

Pedro Barata: You raise fantastic, fantastic points. I mean, we all want ideally, right, to have a less invasive methodology like liquid biopsy, right, that goes beyond CTDNA and explore these transcriptomic information to select our patients. That's fantastic. Dr. Vano, was such a pleasure to sit down and chat with you today. Thank you for giving us this personalized overview of BIONIKK, it's out at Lancet Oncology. Congratulations, again. It's a big effort. I didn't mention, but I think you have quite a few number of institutions helping, over 300 patients consider for this trial. I mean, what a big effort you guys did so big congratulations for the work. Fantastic work. I think it's the first of many as we discussed. And I'm looking forward to reading more of your work in the near future. Thank you.

Yann Vano: Thank you very much for invitation. Thank you.