I want to throw in a comment or two about subgroups.
I share the scepticism expressed by others about the Optune results.
Novocure was able to 'demonstrate' that Optune conferred a PFS and OS benefit right across the board (somewhat suspiciously so at least in my mind.)
Whichever subgroup; methylation status, age ,gender, performance status, extent of resection, etc etc, Optune demonstrated an incremental benefit.
Now, of course, that meant that not only was Optune approved, it was approved for the entire GBM population.
Quite a result for Novocure...
The FDA, I believe, is mindful of the fact that for the last decade or so, the Stupp protocol of TMZ plus RT has been used around the world.
And that scientific opinion is coming round to the opinion that the TMZ element probably confers no clinical benefit to unmethylated patients, which, I believe, makes up a slight majority of the GBM population.
As a result, I believe the FDA may be a degree more reticent now about giving new GBM treatments approval for the global GBM population (especially if the treatment may become SOC.)
In short, predictive biomarkers are very much in vogue.
The FDA wants treatments to be targeted at sub-populations, that are most likely to receive clinical benefit (and rightly so.)
A predictive biomarker allows predicting the response of the patient to a targeted therapy and so defining subpopulations of patients that are likely going to benefit from a specific therapy.
That is pretty much the standard definition of a predictive biomarker.
Witness the whole PDL-1 expression issue, that surrounds the checkpoint inhibitor space. And the tension between BP manufacturers who want the widest possible approval, and the FDA, which, I believe, wishes to link approval to predictive biomarkers, such as PDL-1 status.
(Unless you give credence to conspiracy theory, and therefore think any drug beginning with K and ending in A is guaranteed multiple approvals..)
Now I mention all this, because it may cast a degree more light on why NWBO feels the need to continue unblinded.
Subgroup survival analysis is going to be extremely important to NWBO.
I suspect that they wish to demonstrate that the treatment benefit is 'across the board' and warrants a global GBM approval.
The degree of statistical validity that the subgroup analyses will have, is directly linked to the number of counting events in each subgroup.
Optune was basically double the trial size, so much of their data, including subgroup data, had greater statistical power than the DCVaxL trial can command.
Now the way NWBO can counteract this, is by increasing the event count, by which the final analysis will be informed.
And the only way to do this is to extend the blinded trial...
By the way, I don't believe that transcriptional subclasses (mesenchymal, proneural etc) were a pre-defined set of subgroups in this trial, and I suspect the transcriptional status of each subject can only be retrospectively established.
What NWBO cannot afford (literally!), is being required to undertake a confirmational trial to confirm subgroup benefit, unless Accelerated Approval comes beforehand.
And this is definitely an unpalatable possibility, if their data lacks adequate statistical power..
I would suggest that board members are careful not to be too readily convinced, that there is no good reason to continue a blinded trial.
LP knows why she is continuing blinded for the moment, better than anybody. And she is no fool...
So please be careful what you wish for (if it is immediate unblinding), because if that were to happen prematurely, it may just lead to a very undesired outcome to the whole lengthy journey for investors, the company, the scientific community, and ultimately for patients.
AI
