NorthWest Biotherapeutics Inc (NWBO)

[Dan88]

It seems the short have raised some seemingly savvy analytical observation that the control arm outperforms treatment. This has confused a lot of investors or would be investor, but what on earth is the truth?

Before tipping the toe into it in more detail, let us all remember the overall picture: DCVax-L is indeed working statistically significantly for both nGBM and rGBM. In overall, the data shows:

NEWLY DIAGNOSED GBM:
mOS: 19.3 mos from randomization (22.4 mos from surgery)
vs. 16.5 mos from randomization in controls
mMGMT mOS: 30.2 mos from randomization (33 mos from surgery)
vs 21.3 mos from randomization in controls
Survival Tail: 13% vs 5.7% at 5 years

RECURRENT GBM: mOS: 13.2 mos vs. 7.8 mos from recurrence
Survival Tail: 20.7% vs. 9.6% at 24 mos after recurrence
11.1% vs. 5.1% at 30 mos after recurrence

Now some context: is unblinded data in aggregate as of October 2020 less stellar than the blended data in 2018?

First, we have not seen unblinded data in aggregate yet (not contained in May 10 NYAC presentation), but by just looking at the seemly innocent data, it does seem the treatment arm underperform the data in 2018. The reason is in part because the last batch or the last batches of patients are sicker than the previous' as it's suggested from Germany where the last patients sicker (lower KPS rating at the baseline visit for example) were enrolled. As a result, if the blended patients in 2018 have a five year survival of 28%, the five year survival at the unblinding in October 2020 would be most likely less than 28%. So it's not apple to apple if comparisons are made, and this is possibly a factor which reveals the short's argument dishonest.

Then, most importantly, does the control arm outperform the treatment arm? From the incomplete data presented at NYAC, on surface the answer: probably, but after analyzing a bit more in depth, then answer is a resounding no.

Background: the control consists of 232 nGBM patients; the control consists of 99 nGBM patients initially, and then 64 patients who crossed over to receive DCVax-L treatment plus 35 patients who have never been treated with DCVax-L.

Okay please bear with me: it's known that about 30–49% of GBM tissues have been classified as the mesenchymal subtype. Patients with this subtype, both with primary and recurrent tumours, tend to have the worst survival rates compared to other subtypes (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6485274/). [Before DCVax-L emerges as the most effective treatment for this group, see below]

The beauty with DCVax-L in treating this subtype of patients is that as Dr. Liau has persistently pointed out over the years, despite the worst prognostication of this type treated with any other treatments available before DCVax-L, it has shown stellar efficacy when treated with DCVax-L [this conclusion can trace back to when Dr. Liau conducted Phase1/2 DCVax-L trial for nGBM more than a decade ago].

So excluding other possible but less impact factors and putting everything in equal, it seems the mesenchymal subtype is the deciding factor in solving many puzzles arise by the short, including the confusing puzzle: does the control arm outperform the treatment arm?

So depending on the percentage of the mesenchymal patients in each arm, it would be easy to understand/predict the outcomes of these patients post DCVax-L treatment.

So because the 232-patient treatment arm are randomized nGBM, the percentage of its mesenchymal patients, though similar or equal to that of its comparator ECT arm, would be significantly lower than the control arm which consisting mostly the 64 cross-over patients as the patients would shift towards mesenchymal subtype upon glioma reccurrence. See below quotes within the quotation marks:
"The shift towards mesenchymal subtype upon glioma recurrence [from the same source cited above]

Interestingly, the mesenchymal subtype at tumour recurrence tended to have a worse overall survival (Wang et al., 2016). However, the frequency of mesenchymal and proneural subtypes was highest among recurrent tumours, confirming previous findings about classical being more sensitive to treatment (Wang et al., 2017b)."

So in effect, it is not the control arm which probably outperforms the treatment arm, and as a matter of effect it's because the control arm consists mostly cross-over patients with overwhelming mesenchymal subtype patients while the treatment arm is composed of much less percentage of the mesenchymal subtype patients.
If given the same percentage of the mesenchymal patients in the treatment arm as equal to that of the control arm, it would not be difficult to see the treatment arm will significantly outperform the control arm.

At this point, a new question may arise: so wouldn't it be better for all the patients to wait until progressions for DCVax-L treatment, because in this way most patients may transit to the mesenchymal subtype which would be most favorable for DCVax-L treatment?

Well, first of all, there are patients in the treatment which are already of mesenchymal subtype, so these patients should be treated with DCVax-L ASAP; second, not because DCVax-L may be most effective in treating the mesenchymal subtype, doesn't mean it is not working with other subtypes, so all the patients should be treated with DCVax-L immediately after surgery and radiochemotherapy.

In the future, all the nGBM patients treated with DCVax-L made from the initial tumors should continue with the treatment with newly made DCVax-L from the recurrent tumors. Obviously, in this way, the overall survival of these patients will be most likely increased substantially further again.

Hope this will be helpful in debunking the short's argument, and this message/opinion is not detail, complete or comprehensive enough which will take a much longer time to write; it's however may serve as a clue for those interested to start digging further.

I also believe the publication when comes out will answer many questions or much of the confusion we may currently be having, and there will be more efficacy evidences from the publication.

Lastly, this fiasco or confusion may not be as prevalent as what we have seen today, if most investors understand why the halt:

https://investorshub.advfn.com/boards/read_msg.aspx?message_id=145658266

I will be absent for a few hours from now so sorry there will be no response from me; also sorry for this rash-in commentary without much editing.