NorthWest Biotherapeutics Inc (NWBO)

[sentiment_stocks]

There is nothing ridiculous about it. You don't understand statistics and how they work... and while I believe I understand it better than you do, I get not getting it, because I'm hardly a stat person. But I certainly understand the importance these data play in determining if something is "considered" efficacious or not. I believe that's why they've assembled a phone book size Statistical Analysis Plan (SAP).

Additionally, it becomes problematic to simply look at the standard of care (SOC) figures of GBM patients in general, or in other trials, as these patients were not subject to the exact criteria that the DCVax trial was subject to. For instance, the Rindo trial screened for patients with EGFRv3... who typically do a bit better, as that was the protein that Rindo was attempting to eliminate. Which, by the way, it did. So that would mean that their control group were primarily EGFRv3 patients. On top of that, the control patients were given an adjuvant called keyhole limpet hemocyanin (KLH) which actually slightly improved the control group's OS, which is likely why the treatment showed no increase in OS over the control patients. I doubt Celldex knew that KLH would enhance the control group's performance when the trial was set up, but that's what happened.

Then the IMUC's ICT-107 trial targeted six different types of tumor antigens, rather than the full spectrum, and so their control group would consist of patients reflecting those tumor antigens. Additionally, their control group was given a dendritic cell vaccine without any antigens loaded on them (I think that's what they used), and so, once again, the control group actually performed better because the non-loaded DCs helped those patients.

And so it becomes quite difficult to pull out a control group from the entire world's GBM population to compare to NWBO's criteria. These patients were aged 18 to 70, they'd been screened for progression for 3 months before being able to enter the trial, which was meant to eliminate all rapidly processing patients, as well as all pseudo progression patients (from chemo/rad). But as Avii is so very fond of bringing up, there could have been some pseudo progression patients from chemo/rad that made it into the trial. And as I am very fond of also saying, the trial was randomized, so that would have happened to the control arm and the treatment arm as well. If a control patient entered the trial and was a pseudo progression chemo/rad patient, they would have crossed over to DCVax, and guess what? They could have done really well with that combination. Mind you, there is an arm we often all forget about of 32 clearly pseudo progression patients in this trial (in addition to the 331 main arm trial patients) who pseudo progressed in the screening pipeline. At that time, there were about, what, 165 or so patients in the trial by May 2014 when that arm was closed? And the pseudo arm wasn't really established until the trial became a P3 trial announced on May 17, 2012, so remove another 39 or so from the 165 enrolled by May 2014, because those patients who failed screening were directed into the information arm. So during the time that the pseudo arm was established, 126 patients made it into the main arm of the trial, and 32 were put into the pseudo arm. So from 158 total patient, about 20% were weaned out as psuedoprqgressors due to chem/rad.

https://nwbio.com/northwest-bio-provides-update-on-dcvax-l-brain-cancer-trial-2/

But I digress into the minutia.

There is probably a way to compare the data from this group to historical data, but the comparison wouldn't be considered exact. Now, if they are able to show that say 15 to 20% of those patients who were treatment only lived to 5 years, that could fly, as that is obviously not what the entire GBM population is living to now. Rather, 3 to 5% live to 5 years. And one has to wonder, how many of those long survivors in the general population came out of the P1/2 and even P3 DCVax trials?

So it's unfortunately not cut and dry, even though you think it should be. Ask exwannabe how cut and dry using historical data should be.

Never mind, don't ask him.

lol