Phase 3 trials of Custirsen is designed to work in combination with Taxol, a leading chemotherapy drug, to decrease resistance and increase its effectiveness.
Custirsen is one of two major drugs licensed to OncoGenex by Isis Pharmaceutical (NASDAQ:ISIS), considered a leader in antisense drug or nucleic acid technology which precisely targets specific genetic codes,” explained McCamant. “Teva Pharmaceutical Industries (NYSE:TEVA) is the partner and they are paying most of the costs for the Phase 3 trial. We believe that’s a nice validation for the technology and the drug candidate..
and 4) a safety concern. In TROPIC they saw serious neutropenia issues including deaths in the treatment arm possibly related to treatment. They amended the protocol to allow G-CSF and it appeared to address it. Perhaps they also wanted more safety data. (Interesting that they escaped the imposition of a clinical hold while it was investigated - no rules, just guidelines, lol)
Avii -- I would think this would be especially true if (1) you beat the interim threshold, but by a modest margin, (2) you were trying to get approval from a single registrational trial, and (3) the likely time to the final event count was relatively modest (months not years).
The one thing I've learned from getting my nose broken over and over in this space is: you DO NOT know what's going on behind the scenes. Everything in this domain that feels like a hard and fast rule ends up just being a guideline.
“However, this trial also failed its interim. This was not publicized at all. ”
Summer, I'm not trying to be argumentative but this is an important point: We don't know that TROPIC “failed the interim”, I would guess that they met their stopping criteria yet continued regardless.
The DSMB met in June of 2009 and decided to “continue”. The clinical cut-off date for the FINAL analysis was 3 months later in Sept 2009. The only reason stated for the “continue” was:
“At the second interim analysis in June 2009, since 365 deaths had occurred instead of the planned 307 deaths, the Independent Data Monitoring Committee (IDMC) recommended that the trial should continue to the final analysis.”
Now, they didn’t SPECIFICALLY say they did or did not cross the interim stopping boundary, but the above quote (from the Australian regulatory review) suggests that they COULD have stopped but decided not to because they were already close to achieving the final event target.
Non-truncated trials are more reliable/convincing. The FDA beat their PDUFA date by 3 months.
I guess my point is that a “continue” does not always mean the trial failed the statistical test at the interim. Remember the Canner quote regarding DSMBs and interims: “No single statistical decision, rule, or procedure can take the place of the well reasoned consideration of all aspects of the data by a group of concerned, competent, and experienced persons with a wide range of scientific backgrounds and points of view.”
Perhaps of more relevance is their timing: For Cabazitaxel/TROPIC the data freeze date (final event hit) was 9/25/2009. 46 days later on 11/9/2009 they requested a pre-NDA meeting w/FDA (presumably they knew they had a drug). Synergy hit target events around 2/11/14, if data scrubbing time is similar to TROPIC then unblinding may be around March 29, 2014. Considering ASCO LBA deadline is April 1 my guess is that they'll shoot for that.
So what I am trying to say that maybe the multiple markers being reached now is a good thing.
I would find it hard to believe significance wasn't achieved. CEO says the timing of the multiple markers was in line with what they thought would happened. But Co gave conservative guidance ( a later date) to avoid noise questions it longer than projected. By that statement you would think the trial passing the markers would be a matter of concern for not reaching P. All the arguments have been that its either too soon or at medium. Can someone explain why going longer would be detrimental? Only reasoning I can think of is then there are not enough deaths. And following a linear projection the control arm is doing better than would be expected. Thanks!
Thanks to Summers, Mr_ssssamsa, Dr Kuvasz, bagholderzunite, and many others for making this one the best message boards out there. I appreciate all the different perspectives and insights from everybody that contributed. I am going to be signing off for 3-4 months until final results are revealed. (I going to focus more on family, friends, and work instead).
I am comfortable in my current stake in OGXI and I am somewhat optimistic of it's chances. Good luck to all and hopeful next time I sign on we can all be cockier and richer (as opposed to poorer and wiser).
Looks like RA still holds a material amount of warrants. The strategy may have been to sell the shares in the run up to data (which hasn't yet materialized) and hold the warrants for the event.
Thanks everyone for the contributions. Looks like this will be a close one. Does anyone gave a sense as to when data would typically be available?
Thanks Summer, you seem to have an encyclopedic knowledge of clinical trials to go with your mastery of statistics.
I feel somewhat reassured that the miss on the interim, while disappointing, is not necessarily a harbinger of doom as I was afraid it might be. Based on what you and some others have posted in the last few days, the treatment group in SYNERGY may not achieve a median OS of 30.2m and the blended median may not be 27.8m, but it's also possible the controls won't achieve a median OS of 25.4m. And in any case, the trial doesn't need to achieve p less than 0.0001 or a difference in the medians of 4.8m, since you estimated earlier today that around 3m might suffice. So on the whole I feel a little better. I hope you're a bit more optimistic than you seemed to be a couple of days ago as well.
thanks Summer, I will continue to check. I am nervous, I wish i had your statistical analysis and ability to predict. I believe we are close on this, but I have faith. Stephen
The co announced that the trial has 90% power based on a hazard ratio of 0.75. That can be reversed to be ~500 events. Of course I make a lot other assumptions about the trial design.
I dont know much about the other questions you asked. They are good questions though.
I saw the sites that were selected around the globe and they are top notch. In Ontario, the cancer centre at McMaster University was one of the primary sites. That's just down the hill from me. World class cancer centre.
You ask an interesting question, and when I saw the sites, my first reaction is how positive it is that more patients can be enrolled world wide due to the emerging global economy and access to health care. Access to medical care is so important, at least in Canada it has been for a century. But also, on a commercial level, that means an expanding market.
If you don't mind me saying this, in Canada (and in Western Europe and the Commonwealth), we ask how America can have 32 million people uninsured without access to the same type of medical care that so many Americans are fortunate to have….really, the best in the world. So maybe the question is: are Americans able to go to the testing sites? Are they sicker?
I state this only because the level of conversation on this board is at such a high level, that I am sure you will be very polite in telling me to mind my own business. But as far as OGXI teasing sites go, the study will be unaffected by the location of the sites.
Without doing much math, assuming 0.02 for statistical significance level for the interim, I'll say about 6-7m. However, we do not know what "considerably more stringent" means or the shape of the KM curve.
Just to give an example of a similar trial, take a look at the cabazitaxel pivotal trial which met its study objective by showing a hazard ratio of 0.70 p less than 0.0001. That's really good.
However, this trial also failed its interim. This was not publicized at all.
Cant post link but search accessdata fda gov and select cabazitaxel.
The second interim analysis
was added after a protocol amendment to be
performed at the time of the 307 deaths (the 60%
of the 511 deaths in the final analysis of the
protocol) to assess the primary efficacy endpoint of
OS based on the O’Brien-Fleming type I error
spending function. The actual number of deaths at
the second interim analysis was 365 instead of 307.
Therefore, the type I error of the second and final
analyses were adjusted according to the O’Brien-
Fleming type I error spending function. The
corresponding statistical significance levels for the
interim analysis and the final analysis of OS were
0.0160 and 0.0452, respectively.
The median OS was 12.7 months (control) vs 15.1 months (cabazitaxel). So if we match this to SYNERGY, we will have 25.4 vs 30.2 which is just about right. Interim at 365 (or 307) of 511 to 375 of 500 are also not too different.
That's about 5m difference. Failed the interim but passed the final with a lot of room to spare (p less than 0.0001).
The provenge trial was the opposite closer call. I did not research the interim median OS numbers but I think it was at 3/4 and the final p was just 0.03.