Andrew Allen: Yes, it depends a little on our ability to meet the FDA’s criteria around GMP raw materials. So obviously that’s a little contingent on further interactions with the agency, which is why we are cautious about specific guidance at this point.
Unidentified Analyst: Okay. Thank you so much.
Andrew Allen: Thank you.
Operator: The next question comes from Catherine Novack with JonesTrading. Please proceed.
Catherine Novack: Oh, hi, guys. Good afternoon. Thanks for taking the question. Just a couple, I guess. I wanted to ask about the prevalence of high risk disease versus low risk. And do you consider this a predictive enrichment factor or mainly saw benefit due to the fact that patients had more mature PFS curves?
Andrew Allen: Yes. Thanks, Catherine. So the way we did this analysis was to look at the, we calculated the baseline circulating tumor DNA on every subject where we had that information. And then we took the control group and we split it right down the middle and we just bifurcated that population. We then applied that same cutoff to the vaccine arm, the test arm, and it actually was slightly lower than the median for the vaccine arm. So the vaccine patients actually had slightly higher baseline ctDNA than the control arm, which obviously works against the vaccine arm, all things being equal, which is good from a conservative data analysis point of view. So that’s how we set it. So it was a 50-50 in the control arm, slightly more in the vaccine arm.
The reason we did that was simply to try and generate a more mature data set, because of course, everybody is asking the question, ourselves included, what will these data look like when we have the mature data? And so what we were trying to do was find a very fair way of identifying a population of patients who have events faster. And baseline ctDNA is remarkably efficient at selecting for the patients who have faster events. And that’s clear from the Kaplan-Meier curves that we showed. So that’s why we did it. And you could argue reasonably, that the high risk population is the population that’s going to do the worst on vaccine, because as I mentioned in my response to another question, there is a general belief, based now on data, that lower volume disease does better on immunotherapy, and perhaps particularly on vaccine immunotherapy.
The lowest volume disease is in neoadjuvant setting. And that’s where, of course, we’ve seen particular success by Moderna, who obviously did not see signal in advanced metastatic disease, but have seen signal in high risk adjuvant melanoma. So within a metastatic colorectal population, there are obviously some patients with very extensive disease and some patients with very mild or minor disease, but they all have metastatic colorectal cancer. ctDNA splits them and it gives you more data faster in that high risk group with more extensive disease. So the fact we saw such a strong signal there is reassuring that as the data mature, we will then see a signal at least as strong in the low risk group may be stronger. And therefore our intention is not to think about this high risk stratification once we’re past Q3.
And at that point, we anticipate we’ll be including everybody and treating everybody and hopefully seeking a label in everybody, because in principle, if the product works in everybody, which is what we anticipate, then of course, you don’t want to be selecting out against particular patients.
Catherine Novack: Got it. Thank you. That’s very helpful.
Andrew Allen: Yep.
Operator: The next question comes from Roy Buchanan with Citizens JMP. Please proceed.
Roy Buchanan: Hey, thanks for taking the questions. I think you just answered my GRANITE question, which is stratifying by ctDNA. So it sounds like you’re not going to do that going forward into the Phase 3.
Andrew Allen: Well, let’s be clear, Roy. We haven’t got the data yet in the low risk group, so everything we’re saying today is speculation. What we have got is data in the high risk group, which looks very good. Now we wait for data in the low risk group. So I want to be super clear about this. It’s not that we have no signal in the low risk group. We have no data in the low risk group because very few of them, as of early March, had achieved a progression or death event. So it’s not evidence of absence of effect, its absence of any evidence about anything, and therefore you just wait. But what you would expect, based on the literature and what we’ve seen from others, is that the effect of the vaccine in that low risk group will be at least as good as in the high risk group, perhaps better. And if that’s indeed what we see, then there would be no reason to worry about patient selection going forward.
Roy Buchanan: Okay, got it. Okay, thank you. I guess a few on CORAL. So this RRPV holding vehicle or whatever it is, and you’re waiting to hear back from that. Any sense on when you might hear back on that? And is that a new set of people that are going to decide whether they like this program or that program or technology versus the first decision on the GRANITE?
Andrew Allen: It is a different entity, but obviously there is relatedness, but the exact nature of those relationships is not very apparent to outsiders like us, so related but distinct.
Celia Economides: It is a consortium that’s under BARDA’s direction.
Roy Buchanan: And do you have any timelines of when they’re going to get back to you?
Celia Economides: We have not provided any guidance on that.
Roy Buchanan: Okay. And then you mentioned potentially partnering to fund the ID programs. Will that also potentially include COVID-19? Thanks.
Celia Economides: Yes, potentially.
Roy Buchanan: Okay, thank you.
Andrew Allen: Thanks, Ron.
Operator: The next question comes from Arthur He with H. C. Wainwright. Please proceed.
