NorthWest Biotherapeutics Inc (NWBO)

[sentiment_stocks]

So the entire trial was 20,0, the patients who received DC ever where 19.3. And the patients who never recieved it?

At this time, they haven’t given us the data on the 35 patients who never received DCVax-L. And while it is an assumption that the reason these patients never crossed over is because they were too sick to do so, it’s also a likely assumption. One I will add that you have promoted for years, especially when you argued that using those who never crossed over as the sole comparator would be unfair as they were likely the sickest and died the soonest.

Instead of assuming that the 35 who never crossed over to treatment were the sickest (although they very likely were), I’d like instead to start with the mPFS of the actual placebo arm in the DCVax where the NYAS slide showed 7.6 mPFS. We need a starting point to begin with to use the KM OS chart we’ve been given from the rGBM set of 64 patients.

Of course, using this 7.6 median isn’t exactly representative of when each of those 35 might have evented, but given that it’s the median of the entire group (and not just the sickest), 7.6 months seems a fair place to begin a starting point for the OS of the rGBM arm - the 64 placebo patients who crossed over.

So assuming the mPFS of 7.6 was the starting point for the median patient from the original 99 placebo group, we can get very specific as to what the mOS might have been for the entire original 99 control patients. The median would have been between the 49th and 50th patient (of 99).

Now we don’t know when those 35 patients actually died (we haven’t been given those numbers), but we do know they entered the DCVax trial with the hope that they would receive their treatment. So it would seem logical that their intent would be to cross over upon recurrence, and so seeing as they didn’t, it’s likely they were too sick to cross over. Given that the median survival of GBM is around 14.6 months, I think it would be reasonable to suggest that these patients passed at around that same time. But to be really fair and reasonable in our assumptions, let’s pad that 14.6 with the 3 months from randomization instead of from surgery, meaning that these patients would have really lived 17.6 months from surgery, but for the purposes of comparison to the rest of the control group, we’d make that 14.6 months from randomization.

Now, why am I going to all this trouble to explain how long the 35 patients might have lived? The reason is because if we look at the 64 patient rGBM KM curve, we can then add on the additional 15 patients from that set of 64 cross overs and see where that person died. You can see from the KM chart that I’ve posted that the 15th patient on the chart passed at about 8 months. So 7.6 + 8 months = 15.6 months from randomization. To be clear, that would really be 18.6 months from surgery, an improvement of four months over the average GBM survival rate. I think that extra 4 months speaks for the healthier patients that usually are able to enroll in trials like this.



So to be clear, because I think it’s reasonable to assume that none those 35 patients who didn’t cross over went on to live past 15.6 months (which could potentially move this estimated median slightly), then it’s reasonable to point to the 15th patient on the rGBM KM curve as a likely representative of the mOS of the entire 99 control arm. I’ll add to that Linda Liau indicated this set of 35 patients fared poorly in her interview with Dr. Musella.

And so to sum up, I’m reasonably suggesting that the mOS for the entire 99 placebo arm patients was around 15.6 from randomization. When you compare that to the 232 treatment arm mOS of 19.3 months, that would theoretically represent an increase of 3.7 (almost 4) months.

If this theoretical is actually the case, that’s a decently impressive comparison between the two original trial arms.