Alicia Morgans: Hi, this is Alicia Morgans, GU medical oncologist and Associate Professor of Medicine at Northwestern University. I am so excited to have here with me today a friend and colleague, Dr. Rana McKay, who is an Assistant Professor of Medicine at the University of California San Diego, and has done some wonderful work and is presenting that at ASCO Virtual Meeting 2020. Rana, I'd love to talk to you about the work you're presenting on the OMNIVORE trial in advanced kidney cancer. I think this is really an important and very practical study that highlights a need in the community. Can you tell us a little bit about it, please?

Rana McKay: Sure. Thank you so much for having me, Alicia. It's really a pleasure to be here and discuss the OMNIVORE trial. Just to put this trial in context, the treatment landscape for advanced RCC has been rapidly evolving over the last now over a decade. We've exited the TKI era. We've entered into the IO-IO combination era, and we've seen advancements and great strides with regards to efficacy in these IO therapies. This really came around with CheckMate 025 of nivolumab in the previously treated space, followed by CheckMate 214, a landmark study of nivolumab plus ipilimumab for treatment-naïve patients with advanced RCC. But this improvement of efficacy really comes at a cost, because with the improving efficacy and combination therapy, we were also seeing worse toxicity.

There is an increased rate of grade 3/4 toxicity with nivo/ipi, higher rate of use of steroids to mitigate immune-mediated adverse events. So we're really kind of left balancing the risks and the benefits of treatment, which really were the genesis of the OMNIVORE trial and why we designed this trial from the get-go. We asked the question of, is CTLA-4 blockade required for all patients? Does everybody need upfront ipi? And people who are not responding to nivolumab, can we convert them into responders by adding ipilimumab? And then, of course, there's that question of, well, what do we do for the patients who are doing well on nivolumab, for those patients who have a complete response or have a partial response? Can we safely discontinue therapy in those patients? This was a very practically designed trial to really begin to answer some of these questions about how these IO therapies are really handled in the clinical setting.

To highlight the hypothesis for our study, we hypothesized that nivolumab discontinuation in responding patients can maintain durable responses at one-year post-therapy. And the addition of ipilimumab to nivolumab non-responders will improve response rates for patients. This led to our Phase II multicenter adaptive trial. The trial enrolled any histology RCC, both clear cell and non-clear cell. It allowed patients to enroll independent of their prior treatment status, so patients who were untreated or previously treated could enroll. They were not allowed to have received a prior checkpoint inhibitor. They needed to have measurable disease with good performance status.

All patients initiated treatment with nivolumab induction and they underwent tumor assessments at meet eight, 16 and 20. And depending on their response during the first six months of treatment, that designated what arm they were actually allocated to. So patients who had a radiologically confirmed partial response or complete response were allocated to arm A and had treatment discontinued. Patients who were non-responders, who either had a confirmed stable disease or a progressive disease, were allocated to arm B and had ipilimumab added to their regimen for a total of two doses, and those patients were continued to be followed.

The primary endpoint for arm A was the proportion of patients with a partial response or complete response at one-year post-treatment discontinuation, and the primary endpoint for part B what's the proportion of nivolumab non-responders who converted to a PR or CR. The study also had extensive correlative studies that were embedded. All patients underwent a baseline biopsy upfront and a subset of patients underwent a biopsy at progression with sequential blood sampling throughout the context of the study, really to try to understand were there any predictive biomarkers of response and looking at trying to understand resistance as well.

The trial enrolled 85 patients. 83 patients actually initiated treatment, and 12 patients were allocated to arm A, and 57 patients were allocated to arm B. The median follow up for the trial was 19.5 months at the time of last data lock. So to review the baseline characteristics, about half of the patients were treatment-naïve and half of the patients were previously treated. The majority of patients were male with good performance status, and 96% of patients had clear cell histology, 8% had sarcomatoid differentiation, 15% rhabdoid differentiation, and the majority of patients had IMDC intermediate or porous disease.

So when we break down the data for the trial, just breaking up the trial into its different components, looking at the efficacy of nivolumab for the induction nivolumab treatment within the first six months, the objective response rate was 14%. There were no CRs that were observed. The PR rate was 12%, and there were two unconfirmed PRs. So that was during the first part of the trial before any arm allocation happened. Now, for those patients that actually got allocated to arm A, the 12 patients that got allocated to arm A, five patients, 42%, remained off nivolumab at one-year post-treatment discontinuation, and one patient did achieve a CR that was obtained after about 17 months of treatment. Now, I'm looking at the arm B patients, really the PR conversion rate was low. One, there were no CRs that were observed. Two patients converted into a PR with the addition of ipilimumab. And of those two patients that converted, they were previously treated and their best response to nivolumab induction was actually PD. At the time of data lock, 54 of the 57 patients that had been allocated to arm B had discontinued treatment, and the primary reason for treatment discontinuation in the context of the arm B patients was actually radiographic or clinical progression.

With regards to the toxicity, again, breaking this up into the different components of the trial, with regards to the nivolumab induction, the rate of grade 3/4 toxicity that was treatment-related with 7%, the AEs leading to drug this continuation was 8%, and use of high dose steroids was 8%. Now, looking at arm B, this is nivolumab with two doses of ipilimumab that was added to the non-responders. Again, grade 3/4 toxicity was seen in 25% of patients, toxicity leading to discontinuation in 19% of patients, and 18% of patients actually required high dose steroids. This trial is occurring in the backdrop of two additional studies. TITAN-RCC was a study that was presented at ESMO last year, looking at sort of an adaptive design as well of adding ipilimumab boost to nivolumab non-responders. And really the big take-home from that trial as well was that the PR/CR conversion rate was low at 10%, and the CR rate was also very low with this approach. I'll also highlight Abstract 5006.

My colleague, Dr. Michael Atkins, who'll also be presenting this data at Virtual ASCO, conducting a similar trial in the frontline space with nivolumab induction. And then based on response, if there's a CR/PR, continue nivolumab, or if there's SD/PD, the addition of ipilimumab. And in that trial, they actually did four doses as opposed to the two doses that we did in our trial. The primary endpoint was PFS, so we're really excited to hear the results of that trial.

But in summary, really the take-home message for clinicians, we struggle with this combination sometimes in the community just given the toxicity, but really the strategy of nivolumab followed by ipilimumab results in both PR, low CR. Really there were no CRs that were observed in the context of the main portion of this trial, and two doses of delayed ipilimumab really are insufficient to achieve a response in patients. And so we really cannot recommend this strategy. The recommended strategy is upfront four doses of ipi given upfront with nivolumab. While a subset of patients treated with nivolumab alone can maintain durable responses off treatment at one year, again, over 50% of patients actually were not able to maintain those responses. And so early nivolumab discontinuation in the absence of toxicity also cannot be recommended.

I think we're really trying to tease out the biomarkers in this trial because it would be really cool and we can actually try to identify who that 42% was upfront, and you can feel confident in being able to discontinue treatment and have that patient remain safe and with disease stability. But that's to come in the future. I think this work is super exciting and really is the efforts of a huge collaborative team, mentor, Dr. Choueiri, the team at the Dana-Farber. Just a lot of people, well that had been involved in this work. It's really a great... Thank you for the opportunity to share this data today with you, Alicia.

Alicia Morgans: Of course. And I think we're always happy to highlight data positive or negative because I think this is actually a really informative trial that helps us understand as we practice in the community or in academic centers that it is the combination of ipi/nivo that seems to have the highest response and adding ipilimumab later is not really going to get us that bang for the buck, but certainly get some toxicity, which is important to recognize. I think as you were saying, this is a very practical or pragmatic trial design, and we've been fortunate in kidney cancer to have several of these types of trials where we think about either treatment discontinuation or we think about adding additional treatments if we don't see a response.

And I'd love to hear your thoughts on where that's to go in the future because I think that in kidney cancer, there is a real opportunity there. And we, as oncologists, always want to try to get the most benefit with the least toxicity, and it's a really incredible strategy.

Rana McKay: Yeah. Thank you so much. Yes, I think thinking about how we can escalate and deescalate therapy is a critical question in the field. And I'm going to give a shout out to Dr. Tian Zang and the PEDIGREE trial that she is conducting that's currently accruing through Alliance, that's seeking to actually answer that very question. So in this trial, patients with advanced RCC that's treatment-naïve will start on treatment with nivolumab plus ipilimumab. For those patients who have a complete response, they'll be maintained on nivolumab backbone alone. For those people that have PD that are really progressing, we'll switch to cabozantinib in the context of the trial.

But everybody that's in the middle that either had a partial response or stable disease, there's a randomization to either continue nivolumab alone or layer in cabozantinib added nivolumab to see if we can convert, again, those... To try to attain deeper responses or convert those stable disease patients into responders. So I think this is certainly very much a question that we're seeking to answer in the field. I will also highlight the COSMIC-313 study. What we're learning, it seems to be combinations earlier are better. And so with COSMIC-313, this trial is looking at the combination of nivo/ipi plus cabo in the frontline space, so the triplicate to see if that results in improved outcomes for patients over nivo/ipi alone.

Alicia Morgans: Wonderful. Wonderful work and wonderful to help us as a field understand that when we're treating patients, it's probably best to do full-court press at the beginning if that's what we choose to do, if that's the combination we want to do, ipi/nivo combination is what we need to do, and really adding ipi later does not seem to get that added benefit that we perhaps thought might exist. Thank you so much. I'd love to hear your closing thoughts. What's your overarching message from this trial?

Rana McKay: I think the overarching message is while we want to be able to personalize and tailor therapy for our patients, we have not yet validated that approach in the context of planned studies, rationally designed studies. So I think it's important to practice by the evidence that is out there. Now, of course, if you have a patient that's sitting before you that has comorbidities or has reasons that would preclude use of the combination, of course, you would adjust and modify. But if you don't have a reason not to use the combination, then follow the evidence that is out there.

Alicia Morgans: Wonderful message. We can answer these questions if we design rational clinical trials, and we design them in a way that answers this question. And doing things off study or outside of the trials that we have designed may help, but in many cases may not. And it's important for us to actually get the data to answer those questions before we implement those sorts of approaches in our clinical practice. So thank you so much for your guidance today, Dr. McKay. Congratulations on a wonderful trial that really and truly informs clinical practice. Thank you.

Rana McKay: Thank you for having me.