So there are other ways for DCVax-L to go to market. What is the chance of that happening?
I certainly think there are several pathways the company can use to file a BLA. Are they all sure fire winners? No, probably not. And each pathway's chance to gain FDA approval differs. But there are 4 RAs, and each one is going to have a choice in the process. And some of these RAs are the actual payers/payors in this process, and if they see a clear benefit for their citizens, they may be apt to make different decisions then those who are not.
There are likely other pathways to approval, but below are some of the ones that I think are possible. In the end, it'll likely be just one, but that'll be dependent upon what the unblinded data indicates. Below are those possible pathways that I think will get DCVax to market, and my explanations for each one. There are likely others, but I like knowing that the company has multiple options, and I think the rework of the SAP PRIOR TO UNBLINDING will help strengthen and define those options. Stat Sig PFS (Primary) and OS (Secondary)
Obviously this is the grand slam home run pathway. As most of you know, I believe
(I do not know) they allocated the alpha for this trial unevenly, not splitting it (.025 and .025), but giving PFS .02 and OS .03. Now, if the primary (PFS) is stat sig with .02 alpha (they need about 4 months separation between the two arms to achieve this), the .02 rolls over to OS. The trial protocol had established what is known as a "Closed Testing Procedure" (Avii had asked me to look for that terminology in the protocol quite some time back and yes, it is absolutely there in the 2013/14 versions). What that basically means is that the primary must be stat sig for the alpha allocated to it to roll over to the secondary, and if the secondary is also stat sig, all the alpha rolls over to the tertiary endpoints. But in this case, since only .02 was allocated to PFS (see August 2014 Enhancements PR
), that leaves about .03 alpha to clearly put elsewhere. As I've explained several times, I think that was allocated to OS. That would allow the company to state both endpoints are independently powered (which they have stated), and they wouldn't say they split the alpha (as .02 and .03 clearly isn't split), and they haven't said that. So... this pathway would mean the data shows PFS is stat sig with about 4 months or more separation between control and treatment, so OS would be measured with a full .05 alpha then and it is stat sig, and the FDA will pretty much have to approve DCVax. With a full .05 alpha, the separation between the two arms doesn't have to be as large as it would need to be if they can't use the .02 initially allocated to PFS. And that's why I think they've spent the time and the money to adjudicate PFS rather than just abandon the endpoint. So if the unblinded data indicates stat sig PFS and OS endpoints, there will be little to nothing those who aren't invested can say to stop a pile on into this stock. I mean, they can say stuff, but no one will listen. And the pile on would likely be pretty massive. PFS is not stat sig and OS is stat sig
This is obviously also a possibility. Most longs think PFS was messed up with pseudo progression. With adjudication using the latest immunotherapy criteria, hopefully PFS can come out stat sig, but it's also possible that may turn out to not be the case. But as explained in some detail in the previous "pathway", OS still has, IMO, a back up, and that is their "independent powering" that I suspect is something like .03 alpha. So ex and Avii have both told me that if that is how the alpha is allocated, then the separation for control and treatment needs to reflect somewhere between 6 and 8 months. That is harder to achieve, and would require a nice fat long tail. I personally suspect, but obviously do not know, that this is what LP has been gearing towards achieving - in the event that the previous pathway is not hit. I think the company hopes the first pathway is a slam dunk, but in the event that it isn't, this pathway should almost absolutely garner FDA approval.
So now, those are two pathways, both of which we have a good chance of achieving, IMO. I think that because with the first one, they will have adjudicated the PFS determinations based on the correct progression criteria. And we know that in February 2017, there were approximately 80 or so patients that hadn't even progressed, and so with such a large group of patients who were progression free still - 15 months since the last patient had been enrolled (which is 18 months from surgery) - it would seem PFS still has a good chance of being found stat sig. But in the event it isn't, by letting OS play out and the long tail develop, the math will favor a stat sig OS.
Now for more pathways.... Neither endpoint is stat sig, but there is clear separation between both arms (4 to 5 months)
Okay, here's where the pathway is still clear, but the possibility of approval is lessened. The market will likely not understand what has happened, at first glance, because it will see a very clear, appreciable separation between treatment and control, but still the trial is not stat sig. The trial will not have met what are their endpoints. Of course, that doesn't sound so great in a PR. So will there be a market pile in? Certainly not the same as there would be with the first or the second pathway I've delineated above. The pathway forward will likely be described in the SAP, and could still be presented as a randomized trial that's not stat sig, but the medical and treatment benefit is still visibly obvious and demonstrated, just not through traditional means. For the FDA not to approve this, they would likely face a great deal of public pressure and/or even outrage. So I think this pathway still has lots of promise, but the market may not see that initially. PFS/OS not stat sig, but there are clear separations between EACH of three arms - control who never crossed over, control who crossed over, and treatment - and the separation obviously favors the later two
This is yet another pathway that the trial could request approval for based on the data. I'd suggest that the SAP will better define how this is done, and then the RAs will have to decide whether that is satisfactory enough for approval or not. Again, if the data shows very clear evidence of treatment benefit, then I think the odds of approval are still very good. Will the market get this? Maybe, maybe not. Will those not invested fight this in the market and on the blogs? Very possibly. In the end, and what's more important in terms of approval anyway, is what the RAs think of the data and this pathway. PFS/OS not stat sig, separation between control and treatment is not as clear (2 to 3 months), but as a blended group (treatment and control combined), they beat all historical comparisons
So this represents yet another pathway forward. Is it as ideal as the first two? Maybe not. But the CAR-Ts were approved based on single arm trials, and those treatments even kill some people. DCVAx doesn't kill its patients, and for some number, it looks to extend life for a very long time when compared with all other available treatments. As a blended group, the blended data already beats Optune's, and that makes it difficult for an RA to not approve DCVax too. Still, this one could end up being fought out with an ADCom, or even in Congress, depending. I personally think the UK will still go for all of the above as I believe the prestigious UK neurosurgeons and neurooncologists there will fight tooth and nail to make that happen. And if that happens, it will look pretty bad if the FDA doesn't do the same and approve DCVax. So this scenario will likely not represent as easy a path forward, and there will likely be some fighting over the idea both in the market, and with the regulators. However, if the UK pulls approval through, then patients from anywhere will basically be able to access it via the UK. This will bring revenue into NWBO, and will, IMO, save the company, save the treatment, while the FDA and the EU decide what to do. Neither endpoint is stat sig, the separation between the two arms is negligible, but there are 100 patients (the long tail) or about 1/3 of the patients, for whom the treatment is clearly a benefit.
Again, the pathway to approval is still possible, and likely defined, but this requires RAs to be on board with this. Again, I think the UK will be, and the FDA may have to be dragged along kicking and screaming, with some of the FDA on board with it, and others, not so much. IMO, all of the above still represent viable shots at approval because each of them still reflect a benefit to GBM patients, and that should be what's most important to the RAs.
Now this one represents yet another option that I would be remiss in not mentioning. Methylated Subgroup Approval
Depending upon what the data shows, I think via the methylated patient data in the trial alone represent an excellent shot at approval as the treatment clearly appears to benefit this subgroup. It looks to help the unmethylated patients as well, as it certainly appears to prolong their lives, but not as many unmethylated patients made it to 36 months, so the FDA could get sticky about that. IMO, the unblinding will show whether the treatment was clearly a benefit to these patients or not, because if it is, it'll be mainly or all treatment unmeth patients who made it to 36 months.
And finally... Neither end point is stat sig, the separation of the treatment and control arms is negligible, the overall data reflects little difference between this trial and historical comparisons
Well that outcome is unlikely to result in an approval. However, I think it's INCREDIBLY unlikely
(maybe impossible) that the unblinded data will reflect this because we can already see from the blinded data back in November 2018 that there are almost 100 patients that lived to 3 years (28%), and half of them reached almost 5 years (14%).
So really, my ultimate point here is that, IMO, there are substantially more pathways available to the company than what most would consider the standard and obvious ones, such as the first and second options I've listed. These multiple options also increase the odds, again IMO, that RA approval is ultimately achieved, because even if the data doesn't work for the first pathway, there's still the second one, then the third one, and so on, and so on, with the backup pathways I've cited until we reach the last one, which I don't see as viable.
Now to be clear, it would likely just be one pathway that they ultimately move forward with, and that which one it is will depend on the data. But with so many options, I pretty much think filing a BLA is inevitable. It's the approval that will either be a clear cut yes, or it'll have to be fought over with some of the RAs.
Anyhow, these are just my thoughts on the matter of RISK, for what they are worth.