Some impressions. There are four more endpoints not yet revealed. The first two have to do with CONFIRMED PFS. One of those using ECA control, and the other using placebo control. The next endpoint concerns OS against a placebo control. The final one is tumor response against a placebo control. Seeing the excellent p values and hazard ratios today on the primary and follow up endpoint, i’m assuming the next four endpoints might be better than some crtitics might assume. Just because they weren’t shared with topline results, does nit mean they are a foregone conclusion. The last four endpoints were not chosen haphazardly, and despite media gloating that psPD confounded (previous) unconfirmed PFS, um, newsflash, no kidding, NWBO has been saying that for four years.
On another note, the five year survival we heard about Tuesday is real survival — not extrapolated. In other words, they did not take a pool strewn across two to six years to arrive at five year survival, instead, they took patient data on all patients initially enrolled through five years. Over 13% survived. It appears all lost to follow up were found. (Hint: look at the fine print under the original PFS endpoint graph, stating some treatment patients may have had psPD instead of progression)
There were only 3 idh-1 mutant patients in the treatment group (same as in the smaller placebo group) so this was not a subgroup that could have padded results to any significant extent.
UK’s NICE QALY pricing determinations should look at the two endpoints discussed on Tuesday, because the data is not extrapolated, and because this ultimately became (primarily) a single arm trial in an orphan disease, the valuation should be favorably reasonable, imo.
The surprising (I was not expecting this) statistical significance against residual disease is likely due to the same checkpoint inhibitor phenomenon had when administered before surgery. In other words, it looks as though this bodes very very well for DCVax-Direct, because it has the opportunity to work against a “living tumor” which can allow an even more natural activation and expression process. It also bodes well in DCVax-l for patients that can’t get most of the GBM tumors out.
That residual disease stat, along with the lack of idh-1 mutant patients, in addition to the elderly doing well, essentially wipes out the cherry picking argument.
I think MHRA will move quickly now. There are interim reimbursement mechanisms — pre-NICE formal price evaluation — that could allow a very abbreviated process.
It will be interesting to see the publication.
