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 »

Covariates.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 »

Yeah! I was worried that I would not be able to find this one again. This shows the covariates in the 302.
First notice how there is only 8% in the >=75 age category in sub-group 4 SUVR (0, -0.16). Yet, >=75 were the patients that did much better than average: they loaded up sub-group 4 with patients that would not be expected to do well. Same with sub-group 1 (SUVR (-0.79. -0.34). They have 50% APOE-. APOE- did not show benefit in 302. These were the patients to the left in the post a few up. these patients rotated the red line clockwise and "flattened out" the linear relationship. Sub-group 3 SUVR (-0.26.-0.17) looks like the winner. It has 77% APOE+; this is where we see the bottom of the saucer and many of the patients do quite well.

I did not want to post this earlier, though when I was using parabolic regression on 302 I kept getting readouts such as:
23.0625 x^2 + 13.19 x . What I found interesting was when I differentiated this with respect to x and found the minimum; the minimum was often around 0.27 (as above). This is right near the boundary of sub-group 3.
Last edited by J11 on Mon Jul 05, 2021 8:17 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 »

Thank you flora for commenting.

Yes, it seems that much of the argument against aducan evaporates once grown-ups take a look. Go grown-ups!
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 »

This is another quote about the correlation issue (Page 130). The idea here is that you can randomly assign patients to group dose groups yet it is not as obvious how to randomly assign patients to SUVR amyloid reduction values as the trial is underway. In fact it is understood that various covariates will tend to influence amyloid clearance differently in the different demographics. I thought that one might be able to overcome this problem by appealing to the law of large numbers. Combining 301 and 302 did appear to average things out, though with 302 the strength of the covariates was able to overturn the strength of n. It is important to clearly understand this resolution of the correlation issue point because the lack of individual level correlation on the surface appeared to be a weakness in the aducan results.

Hmm, perhaps the MMRM should have included a term for placebos that became amyloid negative at week 78.

"Consequently, it is highly likely to achieve a balance of
prognostic factors (known and unknown) across dose groups (i.e., group-level), improving
the confidence in the underlying relationship.
On the contrary, if the patients were randomized at group-level, and the relationships
between endpoints were assessed at individual-level within a dose-level, such a balance
(in prognostic factors, known and unknown) can no longer be guaranteed (e.g., an
Reference ID: 48010710219
131
artificially “flat” relationship may arise because of potential imbalances in baseline
characteristics of patients). In other words, there can be multiple confounders across
different individual subjects, and when multivariate analyses are conducted, it is
extremely difficult to correctly adjust the imbalance of multiple confounders across
individual patients due to potential nonlinear relationship and complex interactions. This
challenge was well recognized in the field of pharmacometrics and FDA’s 2003 exposure response
guidance (https://www.fda.gov/media/71277/download) clearly explains these
challenges and recommends alternative methods."
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 »

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 »

This is a particularly interesting result from the FDA pharma report. It provides the model to determine the SUVR values for various covariates. I am not sure if I understand this though for APOE it appears that the formula is:

= (1 + theta * APOE e4) *BSL.

BSL is the baseline SUVR value.
theta is -0.0404 for a non-carrier
0 for a hetero zygote
0.00896 for a 44
APOE e4 is somewhat mysterious -- number of e4 alleles?

here for a 34 = (1+ 0 * APOE e4) BSL
= BSL

for a 33 not sure = ? (1+ (-0.0404 ) * BSL
= 0.9596 BSL ?
(perhaps APOE e4 is more a placeholder than an actual variable)

44 = (1+ 0.00896 ) *BSL
= 1.00896


Age seems much easier:

(AGE /71)^theta *Emax
theta = 1.94
Emax = 0.702

Perhaps these values could all be combined together to give a final estimate for SUVR.

The individual prediction for SUVR in Figure 28 appears impressive.
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.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 »

Think I'll call it with this one. This is just odd. The top figure shows the placebo SUVR for 301 and 302. Yet, it has patients with SUVR below 1.0! that is considered amyloid negative. Why did they enroll amyloid negative patients into the aducan trials? Wasn't the whole idea to exclude people who were amyloid negative? I would have thought that the cut-off would be more like 1.3; not 0.8.

The 301 high dose at baseline SUVR was skewed somewhat to the right relative to that of 302 high dose.
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 »

302 High 20 Point MA.JPG
302 High 20 Point Moving Median.JPG
CDR-sb Amyloid 4d.gif
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 »

I want to go back to the correlation plot as this has been a focus of the criticism on aducan. Yesterday's posts helped with the expert opinion on the appropriateness of using individual level correlations; bottom line there was that it is not appropriate because covariates which were so carefully apportioned at the group level are assigned without control with respect to SUVR removal.

The bottom figure returns to the individual CDR-sb vs SUVR figure. The thin red and blue wavy lines are the originally included local regression lines. As mentioned previously there is no biological interpretation that can apply to these lines as the blue line increases from ~1 at -0.1 to ~5 at ~0.5. Such volatility cannot be translated into any obviously meaningful insight. One expects that if another data set were collected substantially different curves would result. This is especially true at the ends. The ends are box in orange and as can be seen a few data points drive large changes with no self-apparent relevance to treatment outcomes.

I also drew in the green regression line from what was found when considering thousands of AD patients with various current generation anti-amyloids. This time I extrapolate this line from -0.3 to -0.8. The idea now is that we are no longer "stuck" in the group amyloid removal region of ~ -0.27. When we have a trial everything averages out and we are then more towards the middle of the data set. aducan was able to greatly increase amyloid removal as compared to other trials though it still appeared to be somewhat limited. I thought perhaps that there wasn't much further that this could go.

Yet, as we see above continues to -0.5 and beyond. The last post from yesterday was quite revealing in this respect. Apparently the reason why we seemed to be running out of amyloid was because the patients often did not have that much amyloid to start with. for the average patient -0.3 was about all that they could lose before becoming amyloid negative. What I am wondering is whether enrolling patients with more amyloid might help to actually show anti-amyloids efficacy. If the linear association holds all the way out to -0.8 than there would not be 0.4 CDR-sb benefit for these patients but another 0.8 (total 1.2). It would be of considerable interest to better establish how patients at the extreme high end of amyloid removal fared cognitively. It could be powerful rebuttal to those questioning effectiveness.
Post Reply