Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
To screen and turn away viable main cohort enrollees en masse once filled in the main cohort at 240 just to find the remaining psPDs to fill the 72 would seem unethical to me, though I guess possible. Why not just enroll them? -- Pyrr
Nope. (But now I really will be incommunicado until Sunday night.)
Scratch that. I meant "Dramatic Enough" (not "striking")
They will even grant accelerated approval for single arm small studies.
http://support.cureduchenne.org/site/PageNavigator/FDAWebinar.html
"Dramatic enough" results can lead to approval for treatment even from a modest sized phase I study in a serious disease with only a single arm study held up against a historical control. (Start slightly before the 54 minute mark through 55 minutes 23 seconds.)
If you look at the timing for approval on recent cancer drugs, they've been getting approved in "4 months much less 6." (Start at 45 minutes 58 seconds.)
Robert J. Temple
Deputy Center Director for Clinical Science
Center for Drug Evaluation and Research
U.S. Food and Drug Administration
DMD Webinar
February 20, 2013
Absolutley, interim analysis can happen before full enrollment. AVII gives case specific examples, but he needs to site a regulation that states unequivocally that it can't. I'm on the water, so I'll be unavailable for the weekend. AVII admits this idea of full enrollment is his theory. Finally, you must remember this is an orphan designation. In some cases, the FDA will even approve orphan designated therapies in single arms studies "where the results are striking." (I think the term was striking.) This was confirmed in an FDA webcast as recently as February.
Relying upon anticipated completed price negotiations to then influence PPS resulting in enough spike to draw warrants by August/Sept is admirable and knowledgeable but demands many ifs to materialize on track.
I will say that I expect some ifs to show up as well. Namely, full enrollment on DCVAX-Direct phase I by, at most, June 30, and increasingly impressive quantitative and qualitative results as we head through July and into August. The background scientific literature on immunotherapy seems to suggest these results will build quite a bit over time.
I agree with you in saying that one would imagine its not up to the sponsors to make calls like
extending the trial beyond the minmum necessary when there is evidence of its success above and beyond the current SoC . -- Basin Street Blues
So we have two very interested commentators. AF and LS.
One commentator wants to pound the price and market cap as low as possible. The other one wants to stretch out the trial to "learn as much as possible" -- and suggests this won't hurt placebo patients even if the trial looks successful because there is a crossover.
What happens to a small company when their price is constantly hammered away at while they take too long to complete their trial and in the meantime need to get financings because their manufacturing expansion is so large?
What happens? How desperate might they become to just sell cheap at some point?
Does anybody think this is a correct statement?
I think that halting the trial for efficacy has almost no chance to occur. [Because] This is a potentially paradigm changing therapy and we want to learn as much as possible. Because, there is a cross over possibility for non-responders there is no ethical issue if the trial continues even if it looks like it is already successful. L. Smith
Saving, Seeking and Spending Alpha.
It would have made a good title for Post #12627. Maybe Pyrr can use it sometime on S.A.
Agreed. It could be easily be misconstrued as letting some air out of the balloon ahead of time so it does not pop. However, who takes 3 months to make a "little" adjustment? Knowing full well it frustrates investors, unless maybe the delay is the thing. Saving alpha spend until the 88th event is the thing.
1. February 2014. Linda stated she was waiting for the DMC recommendation just like the rest of us.
2. March 7 DMC Review "Pending" (Data in Germany?)
3. March 10 DMC Review "Outstanding." (Data back in the States?)
4. Since then, NWBO has been quiet regarding the DMC. However, NWBO has intensified their position that they are "actively considering" enrollment enlargement.
5. L. Smith opines that perhaps still no DMC analysis has taken place, and the current reason is because NWBO is now considering enlargement, and thus does not want to spend its alpha before this possibility -- meaning since at least March 27, NWBO is delaying DMC review -- by "considering enlargement." (If Smith's guess is right.)
6. AF has written articles clearly hoping NWBO is spending alpha. He wants DMC results -- another alpha spend.
7. IS AF clearly panicking (for someone?) who knows if NWBO does not waste alpha, DCVAX-L is far more likely to succeed? Something in the neighborhood of 18 articles for every NWBO PR or Webcast.
8. If NWBO is too slow, competition could catch up; Too fast and they may not get the P value they need.
9. Question, is the DMC review still outstanding?
Let's combine one of Smith's and one of Pyrr's ideas. This post is not to knock holes in either theory; it is instead meant to extrapolate them.
Smith is too wise to be wrong very often and Pyrr may very well be right about enrollment.
1. Smith states the DMC recommendation may not be available because it has not occurred, and it may not have occurred because NWBO had an agreement to run a size reevaluation.
(While Pyrr does not believe this would occur so early, let's set that aside.)
2. Now Pyrr estimates that the enrollment at the time of the 66 events means pfs may be somewhere around 15 months.
Let's combine those two hypothesis.
If Pyrr is right, the trial could be halted on the primary endpoint right now, but if Smith is right, NWBO has the authority to delay a DMC analysis in order to consider and perhaps modify trial size, albeit with only a blinded look at the situation -- because the first unblinded analysis has not occurred yet.
So, in effect, if Pyrr is right and Smith is right, then the DMC analysis is currently being held up for "active consideration" regarding trial size, and the analysis will not apparently occur until the trial size decision is made. If correct, this is an extraordinary amount of control for a company that (if Pyrr is right) already reasons their pfs numbers are probably good enough for a trial halt!
Why on earth do this? One must consider the nature of immunotherapy -- particularly that it improves over time. "So what," you may say, "under that scenario, flipper, NWBO already has its numbers." But consider, every day that goes by leads us closer to 88 events, and if we get to 88 events, then, as Smith points out, we will not have burned up the alpha spend from the 66 interim event analysis. That is powerful stuff. Imagine a p value of .0001 v. a p value of .00001, and pfs goes from 15 months to 17 months.
In other words, we will have 88 events (lowering P value) and less alpha spend (lowering p value), and more long tail events, thus creating a better chance for success -- now when the DMC reviews and ends the trial prematurely, NWBO will be in a little better position for approval.
Neither Mr. Smith nor Pyrr probably agree with my thoughts here, but when one looks at the problems AVII has shown all of us regarding premature trial termination resulting in no approval and a dead trial, it is clear that such a strategy by NWBO, in the unlikely event it is playing out, might be at once conservative, ethical (because it saves cancer treatment for the world) and pragmatic, but it would leave the rest of us in the dark.
So what of clinical and surrogate endpoints. How does that fit into this theory?
I realize how you carefully made your determination, but I watched IMUC investors do the same thing, even though I did not invest in IMUC; and until the company gives us a more accurate enrollment picture, I am loathe to give too much credit to what otherwise are very reasonable conclusions regarding enrollment. I appreciate your work.
Multiple pseudo progressions (each one ultimately telling us the trx. is working -- but easily misinterpreted as true progression) might also partially explain German reluctance to include this group in the trial.
The other being that Germans don't want to give a placebo to patients they already know benefit from the treatment. They also know this from work done in Cologne.
At each interim efficacy analysis, in the primary group, the statistician first analyzes the 2011-2013 cohort, then combines the 33 cohort for analysis; if and only if that reaches statistical significance will they then combine the tertiary group. The tertiary group in combination with the other 2 cohorts is not necessary to meet the primary endpoint. If memory still serves.
Thank You Tina. My thought and prayers go out to you and yours.
Your factual insight gives more reason to believe in the treatment. Fortunately, the event rate would suggest post pseudo progression events after initial treatment with DCVAX-L are rare in the primary group.
So if this is occurring with higher frequency in the tertiary group, they may have to modify the protocol in that group. I do not think the tertiary trial will ultimately slow down approval, but it may slow down information regarding the approval process.
But they have the choice to leave….Still, retention is probably high once enrolled -- regardless.
I think John could clear up whether it is 2 or 8 weeks. There was an applicant who blogged through his acceptance into the program, once in, he stopped blogging. He had pseudo progression. He described the steps he went through pre-enrollment.
Fair enough, perhaps that is yet another reason the expanded access trial moved forward very recently on clinical trials.gov.
Moreover, NWBO consciously or unconsciously try to make certain it does not happen again before the patient is put on DCVAX-L. How? If you demonstrate progression or pseudo progression after initial chemo radiation you are set aside, given an extra stronger dose of chemo/radiation, then if it shows 8 weeks later to resolve somewhat, you are enrolled in the pseudo group. Indirectly, this extra chemo/radiation pre treatment, likely reduces the chances that DCVAX-L will cause a similar inflammation response. It's too involved for me to go into why this is right now.
Thanks Lunatik. That was a nice combination of fact and commentary. Very accurate.
Fair enough, that was my thinking as well.
Clarified: The alpha spend which penalizes for introducing bias is not increased by review from someone like a walled off Boynton (just using him as a possibility for example purposes) also getting a chance to see what the DMC sees at the interim review -- they only take one spend hit. Moreover, adaptive design reduces the Type I error effect if the modification was an adaptation (plan B) predicted during trial design if a certain outcome occurred at an interim review.
I cannot say we had 192 enrolled by winter, but if your info is correct, I completely agree with you. (I hope the 192 does not include the additional amount enrolled in the pseudo progression group, in which case, I think there is more to think about.)
OK, now that I gave my impression of the worst case scenario, it begs the question, why wouldn't the DMC simply give an efficacy continuance in the meantime? Well, that brings us back to Accelerated Approval and Adaptive Design. If this decision has little to do with the tertiary group, then I think we can say that the trial is on a very unusual but sensible 2 track course, and the fork in the road is the 88th event.
I think you are assuming we know the enrollment numbers -- we don't yet.
JD,
Worst case scenario is they are barely eking out 4 months pfs advantage in the primary group, and they need more patients, which would take more time to get statistical significance. Because it is taking a long time to decide, under this worst case scenario, they A. are waiting for the 88 events to determine if the long tail effect is starting to creep in more firmly in the 2011-2013 cohort or B. It's really complicated to add a "little bit" more to the trial. (B is not the case)
(I know Pyrr does not think they are also looking at the unblinded interim numbers to make this decision, but a walled off, for example, Boynton could get the interim info to assist in making the sizing determination.)
That is the worse case scenario. Part A, above could be the case with or possibly w/o the combined assessment of the 33 patients. If you include the 33 patients, you should not have to worry about waiting for a long tail effect. It is possible, and I never really said this before, that since the 2nd cohort 2011-2013 is initially looked at separately for scientific purposes, the DMC may want to see minimal statistical evidence there before they will combine the 2008 carry overs. There is a concept called adaptive design. I will not go into it at the moment, but orphan status should allow them to adapt the design if its obvious that waiting to combine the 33 patients is absurd. It is also possible that 88 events will prove the second 2011 - 2013 cohort got there, or it will get there w/o addition by the 110th event, again, in either case they could combine the 33 patients for analysis. So, the primary group could have 4-5 month issues that in the end, after combining the group of 33, will greatly exceed 4-5 months. But worse case is that even with the 33 combined, they are barely looking at a 4 month advantage. I find the latter very difficult to believe.
Continued.
Let's say they add 38 patients to the group of 72.
You might think this would only likely add data points on the near end, but this also allows further maturation of earlier enrolled patients on the back end. The hope is, with a few more ("a little bit") patients you'll get more data points on the front and back end.
Meanwhile, even though the treatment group hasn't had enough patients in long enough to show amazing separation from the control group, that's because both the control and treatment group in pseudo progression respond better to the initial preDCVAX protocol, the additional patients (increasing enrollment) plus time will give a better chance to start to show separation, even as little as 4-5 months (maybe less, as long as it is increasing over time), even though if the trial went on another year it could very well show 16 to 17 months separation and 28-29 months the year after that. The key here, in this immature group, is having enough data points for analysis, and enough time for the immature group to mature and separate to demonstrate beginning minimal significance, or maybe just a solid trend -- since it is the tertiary group.
I'm not advocating the parties wait forever for this group, in fact I think they simply move this group to compare against Historically Highly Reliable SOC PFS times post approval of the Primary Group (which I hope and opine is demonstrating more than 6 months separation).
I think they would increase the size of the group of 72, not the group of 240.
For approval purposes, especially where the primary, surrogate and likely clinical endpoints are all based upon pfs, they probably would also like some confirmation in the pseudo progression group.
Either way, if they do need to increase enrollment, I hope they tell us which group(s) they are adding to, and by how much.
No, I'm speaking about the pseudo progression group.
In general, enlarging a trial = you need more data points because your end-point was to optimistic and the actual benefit of the drug is less when compared to the control group.-- Reefrad
I hope they do not blind any DCVAX-Direct trials.
1. There is no S.O.C on inoperable patients. (Except palliative)
2. A placebo would be cruel and an almost certain death sentence.
3. There is no way in my mind that a double blind trial would be more accurate under these conditions.
Thanks. Very much appreciated.
Highwayman, any word?
AVII's post history at I-Village, message 1107 and related posts. He has amazing, although very pessimistic posts on this topic. He has visual aids that are very accurate, he left no stone unturned.
(I just disagree with his pessimistic interpretations on some matters. Still, his data compilation is superb, and he is very bright.)
Thanks
but
no
thanks.
I beg to differ. Let me propose a rule, let's call it the Trump-Card Effect -- or the PYRR Rule.
When Treatment PFS exceeds HHR SOC Overall Survival.
When blinded Progression Free Survival PFS, with 60% or more events complete, nears historically highly reliable (HHR) Standard of Care Overall Survival, such that there is a 99% or better chance that the treatment arm PFS exceeds HHR Standard of Care Overall Survival, the DMC will unblind the trial after consulting with the sponsor and the FDA, and where the Treatment Arm PFS is then found to equal or exceed the HHR S.O.C Overall Survival, the trial shall be immediately halted, and and FDA application for approval shall be filed immediately and updated on a rolling basis up through the approval decision. This Rule only applies where the safety profile is excellent.
I'll bet you that the PEI, FDA, DMC the CRO and possibly one walled off manager from NWBO all know what that unblinded number is right now. If the average unblinded PFS is close to that high, they should not sleep at night until the trial is halted….
So, let's just assume the average pfs was 15 -- including the placebo. That means there is a 100% chance pfs in the treatment group exceeded OS in the historically best performing S.O.C group. If patient advocates found out that was the case, they would not stand for trial continuance. And remember, these numbers can be determined by those with true access to the data, without unblinding the trial -- with only negligible affect on alpha spend.
Correction ("December 2013")….not 2014
Note: Typo did not affect calculations.