EE: Hello, Happy New Year. We’re here in San Francisco, do not believe anything you hear that it’s boring, it’s a beautiful day out there. We’re actually here in GU ASCO and we’ve been through a fabulous new day of presentations in prostate cancer development that brought us some very exciting news from what was called today the brute force of clinical trialsmanship, STAMPEDE. We heard some very, to me at least, I know you were more excited about the docetaxel data, but hearing this data about celecoxib today actually intrigued me a lot. I have with me Nick James, who is the Man of the Year 2015 and it looks like 2016 as well with all this new data, who will tell us more about it. Starting from the beginning, what inspired you to create STAMPEDE?
NJ: When we set this trial up new agents were starting to appear in CRPC, so zoledronic acid had just been licensed, docetaxel was in late stage phase III, it looked active from our impressions of it. When we looked around at trials they were all happening in what was then called hormone refractory prostate cancer and we thought in most settings in oncology you get a bigger gain if you do things earlier. So we thought, right, we’ll do a trial in hormone sensitive prostate cancer and then we had the question of what agents to put in. So zoledronic acid was a no-brainer, that was already licensed for CRPC, docetaxel looked like it would be licensed, but we figured there were enough patients around to put another drug and maybe a couple of combinations in as well. Now the reason we wanted to do this is if you look at most phase III trials only one in three comes out positive so we thought if we had multiple arms we’ve got more chances to win, basically, so hence the multi-arm. But then, having set off on a multi-arm trial, you don’t want to put thousands of patients into something that looks not very exciting so we also had the multi-stage thing. So that was the genesis of the multi-arm, multi-stage thing and then we looked around and we shopped around for a third agent to put in. The data on non-steroidals actually looked quite compelling, it was quite a big protective effect in epidemiological studies, there was quite plausible biology underlying it so we thought we’d put that in as the third drug. Pfizer duly obliged and provided the drug for us and we still thought there were enough patients to look at some combinations as well. So with the combinations we were, in essence, because all the drugs were from different classes, we figured that the effects, if you saw effects, would add at the very least. We weren’t necessarily postulating synergy but we were expecting at least additivity because they were from different classes.
EE: But then you dropped it.
NJ: Yes.
EE: I didn’t know that until you said it last year in ASCO.
NJ: Yes, so the multi-stage thing is like a high jump contest, the barrier to stay in gets higher. So actually celecoxib got through two rounds of the high jump contest, if you like, it failed at the third round. So the hazard ratio on failure free survival was less than 1.0 but just not enough less than 1.0 to carry on recruiting. In fact what we’ve seen today is that the weak failure free survival effect did carry on and it was most pronounced, in fact there was a heterogeneity of effect, it carried on in the metastatic setting. The hazard ratio was quite significantly less than 1.0, it was less than 0.9. Actually for the zoledronic acid, although we formally reported that as a negative arm in the paper we just wrote, the hazard ratio for zoledronic acid is also less than 1.0. So what we think we saw today was that if you add two moderate weak effects together you actually see a significant effect. So the failure free survival in the combination of celecoxib and zoledronic acid arm was significant, a hazard ratio of 0.77, and that translated through into a survival advantage of the same amount, 0.78 in fact.
EE: But there’s probably also some biology there because we’ve tried similar agents in combination with docetaxel and we failed.
NJ: Yes, so that is true. None of the docetaxel combination trials have come up positive.
EE: Disastrous.
NJ: And interestingly we didn’t add celecoxib to docetaxel, which is now a shame in hindsight, because that might have been very interesting. So we’re busy scratching our heads about what’s happening here in the biology and also statistically we still haven’t bottomed out, we need more data as to whether there’s just an additive effect or whether there’s actually genuine synergy between the two drugs.
EE: So what are you going to do with this celecoxib? What Oliver said today is that probably it’s like a hot potato, nobody wants to pick it up in the US. I think we’re exaggerating here. There’s a signal there.
NJ: Well there are a couple of things we want to do. We want to dredge into the data a bit more precisely so, for example, the bone subset, which is about half the patients in the trial, you may postulate there’d be a bigger effect because that’s where zoledronic acid mostly works. We haven’t looked at that data yet. And then the second thing is to try… we’re busy trawling across literature to see if there are pathways that we haven’t thought about that might be affected by these drugs. I have to say, if I was running a pharmaceutical company I’d be busy going through my library of rejected compounds to see…
EE: Trying to reach the STAMPEDE investigators.
NJ: Yes, and trying to see. We’re going to pull the blocks out as well. So, for example, in the colorectal trials there’s quite a strong suggestion that the patients who over-express COX2 see a benefit from COX2 inhibitors although those trials were canned because of cardiovascular problems. So of course pulling the blocks out we’ll stain them for COX2, so that is in progress.
EE: So how are you going to prove Oliver wrong, that you’re not a brute force and that you’re actually leading precise medicine?
NJ: We’ve prospectively consented all of the patients for tissue collection from the paraffin archive and we’ve collected germline DNA from the majority of the patients in the trial. We’re recollecting it with salivary DNA because the initial collection didn’t work so well from the technical point of view. So we’re doing already some DNA studies, Ros Eeles is heading up those for us. We’ve got a big grant application in at the moment to pull out as many as we can of the 8,000 odd blocks from the patients in the trial so that obviously gives us a fantastic resource. From the data that Ros has looked at already it looks like the percentage with a BRCAness phenotype going into the trial is quite high, much higher than the general population. We're already working out the feasibility of adding a PARP inhibitor arm for the patients with the BRCAness phenotype and we’ve worked out the statistical models that we need to make that work. So we’ve done the stats theory, we’re pulling out the blocks, we’re staining them for the BRCA phenotype genes that were in Johann and others’ presentation this morning. We’re moving that way.
EE: And you’re posting job applications for investigators like me to come over, I think.
NJ: We’re more than happy to collaborate with anybody that wants to work with us. We’re delighted to work with everybody.
EE: It’s wonderful work, I’m really… not only impressed, I’m just glad that it’s happening at this time. I think within a few years you will be leading the way into precision as well.
NJ: I hope so. I hope so.
EE: I’m sure you will, not just hope.
NJ: Yes. The model is fantastic and it has been so fantastically supported by all the UK investigators but the oncologists and urologists have just bought the model. It’s just a default thing for patients with advanced prostate cancer, they go into STAMPEDE across the whole of the UK. So it’s just a very effective way of answering questions.
EE: I’ve had the opportunity to come twice to your country this year and I was impressed by almost the fanaticism by which investigators participate in the trial. It’s really a dedication to the cause to actually make prostate cancer, in this case, history.
NJ: Yes, obviously it has recruited a fantastic number of patients and we’ll be back next year with data from the abiraterone arm which is just under 2,000 men recruited out two years ago. The radiotherapy in M1 disease, irradiating the primary, we may well have mature survival data for that next year as well and we’ve got a sub-question within that around fractionation, between a conventional fractionation and a fraction a week at 6Gy per fraction hypofractionated as well. So we’ve got a lot of data coming out next year and then a couple of years after that we’ve got the abiraterone enzalutamide combination arm data maturing as well. So we’ve just got fantastic…
EE: I’m looking forward to it.
NJ: Yes.
EE: It’s like sequels.
NJ: Yes, it is, yes.
EE: Thank you very much for joining me to discuss your data. Thank you all very much for joining us.