A very logical post, as always.
We have had some discussion on these issues, but I am curious to know whether you still agree with the inferences drawn.
1. The issue of PFS. If I am not mistaken, this is the primary endpoint and in order to achieve it, 4 months delta over SOC/placebo is the endpoint. How SOC will exactly be staked is unknown at least to those on the "outside" for now. There is the issue of confoundment wrt pseudo-progression, however, that is hoped to be resolved by adjudication. This is perhaps why PFS was not addressed in the JTM article as the data was affected by the confoundment and thus not clear and unambiguous. I do not think that failure to achieve this endpoint is necessarily fatal to the overall success of the trial. This will depend upon length of survival and the delta from PFS to OS. If the delta is significant but PFS did not reach the endpoint, what is the reason for this? One possibility is that although PFS eventing, overall, is not significantly better than control, progression is dramatically slowed as the effects of DC VAX L begin to manifest themselves and take hold over the longer run since immunological therapies take time for their effects to become evident. Thus, the slowing of progression could, in and of itself, contribute to a lengthier survival. Furthermore, the PFS endpoint may be achieved for some classifications such as M+ even though, overall, the endpoint was not achieved.
2. Overall survival. We have seen from the JTM article and subsequent SNO update that the blended data as compared to historicals(from surgery as opposed to randomisation) is rather remarkable for both M- and M+, where the deltas are 7 months and 13 months from surgery(or 4 months and 10 months from randomisation). We also know that the overall blended mOS is 23.1 months from surgery or about 20.1 months from randomisation as compared to a historical SOC of about 17 months. The blended DC VAX L mOS(from randomisation) compares very favourably to the Optune treatment mOS.
We can infer that the vaccine seems to work remarkably well in the M+ categorisation(I do not know the molecular sun-group(s) making up M+), where anything at or over 8 months delta would be considered a very major advancement. The blended delta is 10 months from randomisation and likely higher for treatment. M+ also appears to be a significant percentage of the ITT population at about 44% of the total classification and almost 40% of the ITT as per the JTM article and updates.
In addition, Drs. Liau and Prins have suggested that the vaccine is very effective for the MES molecular sub-group which overall accounts for about 25%~48% of all stage IV GBM. From the February 26, 2018 abstract(which is included in the iHub preamble to the message board), it appears that MES is categorised as M-. However, I do not know the extent of overlap, if any, with M+. However, I believe it is safe to infer that DC VAX L has significant efficacy wrt to as much as 50% or more of all GBM.
Wrt M+, this appears to be characterised by the effectiveness of Temador in interfering with the cell repair mechanism and supplementation with the DC VAX L adjuvant appears to significantly enhance the effect resulting in the lengthy delta over historical controls.
Wrt to MES, Drs. Liau and Prins have stated that it is more immunogenic than other classifications which appears to mean that there are more residual T cells available to build upon by the MOA of the vaccine and that the microenvironment of the tumour is less immunosuppressive thus allowing more time for TIL, etc. infiltration of the tumour(s). MES is an extremely aggressive strain with an SOC of anywhere between 12 and 15 months. Upon recurrence it appears that perhaps 85% of these cancers morph into MES where DC VAX L appears to be most effective wrt that classification. Significant efficacy in this molecular sub-group alone, which is extremely aggressive and which appears to be a significant percentage of stage IV GBM, would warrant expeditious approval by the RAs, including the FDA, at least for that indication.
Dr. Ashkans has stated, in his capacity as a clinician and PI at Kings College, that the vaccine works to a greater or lesser degree across the spectrum and that the vaccine should be made available to ALL patients at an affordable cost. Dr. Ashkans has indicated that the vaccine should be made available as soon as possible and that he would be continually pushing for unblinding.
Cross-over is an issue that could confound results. Dr. Liau has said as much. There may not be significant amount of separation between the treatment and cross-over arms and that 90% of all patients in the trial have received either early or later vaccination. In fact, Dr. Liau has stated that this may be a trial that measures effectiveness of early versus later vaccinations. I wonder whether lack of significant separation between the cross/non-crossover and Tx arms would necessarily place approval in jeopardy given the above discussion. If there were not significant separation, this might suggest that even later vaccination may be to some extent effective and would bode well, not only for ndGBM but also for rGBM as well.
3. Accordingly, taking the foregoing into consideration, it appears that DC VAX L works as well if not better than the medical device Optune at the outset when looking at blended data and better as time goes out. DC VAX L will have at least 4 years of data in the most data rich trial for GBM in medical history as opposed to Optune. Even if DC VAX L works just as well as Optune, patients should have viable alternatives especially as between a device and a vaccine. Factors such as QOL, cost, ease of administration, relative effectiveness wrt to certain GBM classifications etc. should be given the utmost consideration in addition to safety and efficacy.
4. Further, considering recent FDA approvals, we are witnessing a more open, expeditious and flexible approach by the FDA which bodes extremely well for DC VAX L approval if unblinded results are positive as they are appearing likely to be. I would argue that the FDA in approving DC VAX L would render broad as opposed to limited approvals leaving it up to physicians and their patients to decide upon treatment options rather than imposing same through regulatory fiat. After all, GBM cancers are personal and different from individual to individual. Unless you try it, you may not know whether it works personally for you. Desperate patients should have the flexibility to try this treatment through consultation with their physicians. Furthermore, broad approval may make it easier for insurance reimbursals as opposed to off-label use.
Thus, it seems to me that the odds are much more in favour of a positive out-come rather than failure--at least I believe it to be so. I do not know this for a fact.
The foregoing summary brings me around to the following curiosity for want of a better expression. In one of your recent posts you stated that oncologists "know" that the vaccine works and that while yours and their paths rarely cross, you and they work at the same trial site medical center. This declaration is highly encouraging and is much in line with my belief but not knowledge. However, in a subsequent post you seemed to back track a bit and rather than confirming these oncologists' knowledge, you stated rather that they have a strong belief that it works. Belief and knowledge are differences in kind not degree.
Do I necessarily think that your statement that oncologists at your medical center "know" it works, even though there is officially at least no unblinding, is an exaggeration? No, I do not. Clinicians may be "blind" to the trial data/results prior to unblinding, but I would submit that clinicians, dealing with patients every day and knowing their detailed medical history, probably "know" what's going on just as Dr. Ashkans hinted. Their eyes are not blind. Is that a fair characterisation of what you meant?
Would appreciate any comments you may have on all the foregoing. Thanks for reading this and I must say that I find your posts professional, prescient and encouraging.
TIA.
