Celebration Thread! Biogen is going to the FDA with Aducan.

Alzheimer's, cardiovascular, and other chronic diseases; biomarkers, lifestyle, supplements, drugs, and health care.
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

I drew in the "U"s for the blue and red curves to show the broad trend of the data near the middle. In the two figures at the top of the previous post I then applied 20 point moving averages (medians) for the 302 high treatment arm (after removing those points which were above 4 CDR-sb change -- this was the cognitive decline level that was posted recently that was found to be a rapid progression outcome from maintaining amyloid positivity through time. A few patients in the 302 high dose exhibited this behavior and I felt it best to remove these data points because they distort the more typical pattern.)

What these figures show is a strong parabolic trend for both the average and the median. It was especially strong for the moving average with a correlation of ~0.90. I had been very unclear earlier about what this pattern could mean. The post from yesterday apparently provides an explanation. The strong non-linear trend is likely a result of the covariate differences that were noted. The patients on the ends had higher cognitive decline risk and so the parabola is exhibiting more this increased risk than any SUVR effect.
The effect is extraordinarily high and is completely unobservable by simply looking at the blob of dots.

For me the big lesson here is that data analysis is really all about showing these obscure patterns in the data. Whoever can construct the biggest jigsaw puzzle with the most order and meaning wins -- er, here, that would be $100 billion and the gratitude of tens of millions of people coping with a progressive neurodegenerative illness. Creating meaning through constructing the data into the higher order of the correlation plots is part of this meaningfulness construction. As we can see in the above figures this creation of meaning also might be possible even closer to the individual level (or at least ~ the level of 20 patient moving averages). If we could get the overall regression from the other large anti-amyloid trials to fit well into this figure, then we would have very powerful evidence on all scales of anti-amyloids.

Also wonder with this style of thinking whether a randomized trail for aducan would actually be needed. What might be done instead is to simply treat everyone with high dose. Different patients would clear different SUVR amounts and then they would fall along the regression line. In this way all the patients would contribute meaningfully to the result; with placebo we know that they will have an average of ~ 0 SUVR and and average of ~ 1.5 on CDR-sb; this does not seem overly helpful. Having more patients defining the regression line would be much more relevant.
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

Another interesting insight from today was CAA -- cerebral amyloid angiopathy. Previously I had wondered whether people might be interested in removing amyloid from their brains aside from AD. Apparently, yes. It seems that people are asking about off label aducan for this indication; it is not clear whether this would be safe or effective as CAA amyloid is of a different nature than AD amyloid. Nevertheless, 80% of AD patients actually do have CAA pathology.

It is truly truly staggering! There is such pervasive amyloid neuropathology. Where do we even start? Upwards of ~30+% of normal 80 year olds have CAA pathology; perhaps 15% in their mid 60s. CAA contributes to ICH (Intracranial hemorrhage). I am not sure though ARIA might simply be a euphemism for ICH. This is the age of amyloid medicine. It could take years to resolve all of this amyloid risk that exists in the community. The placebo ARIA events simply might be spontaneous hemorrhaging due to existing "normal" CAA deposits. This CAA risk will need to be managed through time with goal of eliminating the substantial medical risk that it poses. Anti-amyloids offer us a potential future world with greatly reduced risk for dementia, CAA/ICH (though perhaps using specifically designed anti-CAA medications) and perhaps other cognitive challenges.

I am not entirely clear whether I have this correct, though my first guess is that amyloid broadly considered is causing substantial medical issues in the community. I can certainly appreciate the perspective that caution is warranted when applying anti-amyloids in such a context; especially in novel clinically ill-defined populations.
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

SUVR 2.jpg
You do not have the required permissions to view the files attached to this post.
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

There is some very interesting reading materials concerning the background regulatory analysis available at:
https://www.accessdata.fda.gov/drugsatf ... 000TOC.cfm

I have been reading through the Clinical Pharmacology Review (and posting comments on it) which is accessible from the above url, or more directly through the url below:
https://www.accessdata.fda.gov/drugsatf ... dacted.pdf


It is highly discouraging that so much of the online discussion does not refer to any of this evidence or indeed any evidence at all. The online commentary is more based upon general opinion and philosophical outlook, though rarely if ever makes any reference to the extensive FDA aducan research. If they want to win a grown up argument, then they will need to do much better than that and actually present reasoned arguments and actual research. I have found the line of thinking presented on this thread to be highly persuasive; I suppose I am biased in that regard as I have been the most frequent contributor (perhaps excessively so), though objectively I think it is fair to say that I have built up a falsifiable framework that helps to provide some coherence to the aducan dataset and more broadly the anti-amyloid research. I do acknowledge that there are quite a few just so stories in which I create a narrative around the data after I have seen the data, though this tends to create a binding logic that then needs to form a larger structural cohesion. Much of the opposing commentary does not even attempt to build up such higher level structures; it is mostly in the category of argumentation of "No, it ain't". Yet, whoever builds the biggest jigsaw puzzle wins!


Before I forge ahead with all the new and exciting ideas from the pharma review, I want to return to a few ideas from yesterday. Firstly, CAA. I had not been aware of how frequent CAA becomes through the ageing process. The recent news reports that leaders in CAA advise against treatment with aducan as a strategy to reduce cerebral amyloid angiopathy suggests to me that this has arisen due to patient requests to go off-label. As I noted yesterday, apparently upwards of 80% of AD patients actually do have CAA pathology. It would seem obvious that patients in the aducan extension could be assessed for CAA risk (to the extent that this is diagnostically possible) and perhaps this could guide future research into the clinical application of aducan (and other anti-amyloids)
into reducing CAA risk. I would in particular be very interested to know the ARIA risk in those treated long term with aducan, also the risk of ICH. Arguably CAA should be a much easier illness to conceptualize than AD. With CAA there is amyloid obstruction in the cerebral vasculature-- presumably removing such obstruction would reduce risk. With AD even after decades of research there can be still heated debate about the role of amyloid; yet CAA does not have the potential for the same argumentation. Vasculature amyloid is without question pathological. The central question then becomes how to safely remove it more than focusing on the efficacy (cognitive, etc.). CAA could be an added win for the anti-amyloids, once the issues of safety etc. could be resolved.

Another highlight for me from yesterday is the potential ability to extrapolate the linear regression further to the left. The aducan 302 high centroid SUVR was ~-0.28 with a CDR-sb of -0.39. What were to happen if we could report a group of ~20 patients with "typical" covariates at ~ -0.55 SUVR? Would we really move down to an expected regression value of -0.78? The legitimate yelling and screaming for 302 high dose gradually loses enthusiasm after about ~-0.35 and would reasonably stop by ~ -0.45. -0.78? That would be well beyond the range of plausible disagreement. The problem as noted previously is that in the aducan p3s the normal distribution peaked ~1.4, so the average patient only had 0.3 of amyloid to lose before becoming amyloid negative. There really weren't that many patients who had 0.5 of amyloid available for removal.

The figure above shows something interesting from the pharma review. The purple annotation in the placebo figure on top shows the dona distribution of amyloid. The review noted that the dona p2 had patients with an average of 20 centiloids more amyloid than in the aducan trial. {I am not totally sure though my conversion is 0.01 SUVR = 2 Centiloid. So, 2* 10 centiloids = 0.01 * 10 SUVR
Instead of a peak of the curve at 1.4 SUVR, the peak should now be ~1.5.) What i find exciting here is that now there could be a substantial subset of patients who had amyloid to spare. Perhaps we could push back the regression line to -0.4 or -0.5 SUVR. If the linear regression were to still hold at these levels than we could move into the beyond argument region. I am still unclear on this point, though. Patients with that much amyloid to be removed really are no longer "average" patients; comparing these patients to the average placebo might then no longer be valid. [Err, somewhat frustrated that they changed scales when moving down from the placebo to the low and high dose. Good idea with a series of similar figures not to change the scale midstream. I show the placebo scale next to the low dose scale to the left. There is almost a 40% inflation of the placebo scale relative to that of the low dose.]
Last edited by J11 on Wed Jul 07, 2021 10:12 pm, edited 2 times in total.
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

"When the same scenario was analyzed (4 endpoints, high dose, and week 78), the larger
sample size and more data from earlier visits based on MMRM analysis resulted in an
even smaller false positive rate, 0.000008, compared with 0.00007 when only 300
subjects were simulated in each group only at week 78. When all positive estimates from
4 endpoints, 3 visits and 2 dose levels were considered, the overall false positive rate
was below 1/10,000,000. An accurate false positive rate could not be estimated because
only 10 million clinical trials were simulated due to computation capacity :roll: :shock: :? :D . Similar analyses
were conducted for Study 301 and Study 103. Even though CDR-SB and MMSE results
were not positive for the high dose in Study 301, 18 out of 24 combinations (4 endpoints,
3 visits and 2 dose levels) showed positive point estimates, leading to an overall false
positive rate of 0.0001 for Study 301 under the null assumption. For Study 103, there
were 5 different dose levels, 2 endpoints and 2 visits. Seventeen out of 20 combinations
showed positive point estimates despite the small sample size in each arm. As a result,
the overall false positive rate was estimated to be 0.0001 for Study 103. Given these
results, the review team believes that it is practically impossible to observe the overall
positive results in this program by chance if aducanumab is similar to placebo."

page 35 pharmacology review
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

The quote above refers to extensive computer modeling that the FDA performed on the aducan dataset. Apparently their analysis calculated that the overall false positive rate for the phase 3 aducan trials was less than 1 in ten million; the 103 trial had an estimated false positive rate of 0.0001. they concluded that it was essentially impossible for aducan to be similar to placebo given the trial results.

I highlighted "due to computation capacity" in brilliant pink and with no less than 4 emoticons. Qu'est-ce qui se passe? I am not sure whether this is scientist droll humor, though I found it hilarious! Um, they couldn't calculate a better estimate of the overall false positive rate for aducan because the FDA supercomputer did not have enough CPU? Really? Are they kidding? FDA at home initiative? If they need to go to e-10 we're here for them. I am guessing that they applied a lot more CPU to finding an honest estimate of the false positive rate than 100% of the chatbots online discussing the deficiencies of aducan.
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

"... Based on these analyses, the review team concluded that there is a clear relationship
between reduction of Aβ plaque burden in brain and preserving of clinical function in the
aducanumab program, which is consistent across all 6 other available programs of anti-
Aβ antibodies under development over the past decade. A larger reduction of Aβ plaque
level in brain is clearly associated with a better maintenance of function as measured by
CDR-SB. In contrast, compounds at the tested doses with no/minimal changes in Aβ
plaque level consistently failed to demonstrate superiority over placebo in slowing the
disease progression in clinical studies with treatment duration of 1.5 to 2 years ..."

page 29 pharmacology review
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

Above is an extended quote from the pharmacology review about the clinical efficacy of aducan. This is what the central point of discussion has been all about with aducan. Is it clinically effective? These FDA experts carefully studied the dataset for months (if not years). What is their conclusion after all of this in depth research? They unequivocally find that reducing Aβ plaque burden is related to preserved cognitive function.
Last edited by J11 on Wed Jul 07, 2021 10:02 pm, edited 1 time in total.
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

Final Model.JPG
You do not have the required permissions to view the files attached to this post.
J11
Contributor
Contributor
Posts: 3351
Joined: Sat May 17, 2014 4:04 pm

Re: Celebration Thread! Biogen is going to the FDA with Aducan.

Post by J11 »

Here it is the answer! This is the final CDR-sb model for aducan. It shows the variables that determine the expected response for AD patients treated with aducanumab. For all those paying $56,000 per year for aducanumab treatment this is what you are getting for your money. ... Does anyone understand what it means? They don't actually give the shape of the formula of the model. What they found was that there are slow/typical/ and fast progressors; MMSE at baseline matters as does race, though aside from that it is all somewhat mysterious.

Explanation of this table with typical patient examples and the error bars would be extremely helpful. It might even quite silence some of the more vocal critics if they could better appreciate the benefits that some patients would derive from treatment. My impression is that some of the patient subgroups do very well on treatment. Providing a clear description of the model would help to reveal this truth.
Post Reply