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

Alzheimer's, cardiovascular, and other chronic diseases; biomarkers, lifestyle, supplements, drugs, and health care.
Fc1345linville
Contributor
Contributor
Posts: 136
Joined: Tue Feb 24, 2015 1:49 pm
Location: Washington DC area

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

Post by Fc1345linville »

Here's the coverage on this from the Washington Post.

https://www.washingtonpost.com/health/2 ... -coverage/
J11
Contributor
Contributor
Posts: 2127
Joined: Sat May 17, 2014 4:04 pm

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

Post by J11 »

Thank you for your reply Fc1345linville.

I had not anticipated that the CMS would restrict coverage in the way that they did.
I will be very interested to see whether additional interventions are forthcoming.
J11
Contributor
Contributor
Posts: 2127
Joined: Sat May 17, 2014 4:04 pm

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

Post by J11 »

APOE + - 2.gif
You do not have the required permissions to view the files attached to this post.
J11
Contributor
Contributor
Posts: 2127
Joined: Sat May 17, 2014 4:04 pm

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

Post by J11 »

Sorry for posting these yet again, though they really do add some strong evidence to the anti-amyloid side. These were published early last year and might have helped the FDA decide in Aducanumab's favor.

I notice that the CMS did reference both of these studies in its statement yesterday, though did not actually appear to delve into the support that they provide for the anti-amyloid side.

The first figure above was from the Lecanemab phase 2 study. The APOE 4 subgroup with the highest dose is of particular interest. In the orange rectangle we see 99.2 (bottom number). There was a 99.2% probability that this dose arm was superior to placebo. Remember that the goal posts at some point changed: being better than placebo for an anti-amyloid is no longer felt that exceptional; the new stretch goal is 25% better than placebo. What is of especial interest here was that this dosing arm was halted partway through the trial when the ARIA risk for APOE 4s was considered too elevated at this dosage. With only 45 patients, they still achieved 99.2% probability. As seen in the pink rectangle (bottom number), the reduction in decline on ADCOMS was ~47% (1 - 9.6/18). 47% reduction in decline? This is the dosage level for all ~900 ? patients in the Clarity trial (not all APOE4)? There was even a 93.6% probability that treatment was clinically superior to placebo (25% better). The next lower dose for the APOE 4s also showed strong probabilities favoring treatment (95.6% probability superior to placebo).

I really hope that Clarity will prespecify the APOE 4s as the primary readout. We saw in 302 that the epsilon 4s did much better than the non-e4s. This also happened with the phase 2 leca trial (noted above). While the high dose arm of EMERGE did achieve statistical significance overall, the e4s clearly received a clinically significant treatment benefit; the non-e4s not so much. This is entirely unsurprising from the e4 perspective: e4s are a homogeneous genetic group that have tended to do better in previous trials, while the non-e4s are more of a mix of dementing subtypes. A strong slowing of decline of ~40% would not be open to the criticism that this was not clinically relevant: It clearly would be.

The next two figures above were from the Donanemab phase 2. The bottom figure shows a remarkably close replication of the Aducanumab 302 trial; the iADRS was a nice touch as with more test items it averaged out more of the noise while averaging in more signal while showing a consistently strong treatment benefit through time. What was even more impressive was then to go one step further and find the responder group in the low subgroup of the intermediate tercile. Basically, this subgroup appears to carry the entire treatment benefit. I found this result to be highly impressive; I still do.

I might be wrong on this, though what this seems to be telling me is that the entire multi-year anti-amyloid clinical trial paradigm is no longer a given. The figure shows ~30 highly selected patients producing a clinically significant result after a mere ~30 weeks. After ~78 weeks, (unselected by APOE 4) they still showed a 50% reduction over placebo. This was the state of Alzheimer science, early 2021. Even at that time I was pondering out loud how this could be a game changer for Alzheimer's clinical medicine. Why if there needed to be more clinical evidence then more clinical evidence could be manufactured tout de suite. Why don't people listen to J11? I am here to help!

It is not clear whether the CMS statement recognized how with more focused selection, more rapid clinical evidence could be produced. It would be especially interesting if the Clarity trial might be able to include such tau selection -- perhaps this could even be added in now (if this hasn't been done earlier).

The evidence supporting the amyloid position continued to accumulate even after the Aducanumab results had been locked. The above research powerfully supports the anti-amyloid perspective. Interestingly, the current anti-amyloid phase 3 trials also provide support to the extent that they have not been declared futile.
J11
Contributor
Contributor
Posts: 2127
Joined: Sat May 17, 2014 4:04 pm

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

Post by J11 »

103 Long Term.JPG
Phase 1 4 years a1.gif
Absolute SUVRb.jpg
You do not have the required permissions to view the files attached to this post.
J11
Contributor
Contributor
Posts: 2127
Joined: Sat May 17, 2014 4:04 pm

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

Post by J11 »

The above figures show the longer term outcomes for those on Aducanumab. The clinical trial results only give us
~78 weeks, though these trials then went into longer term extensions. What we see in the first two figures above is how the benefit accumulates through time. What might have been only ~1 point on CDR-sb after 78 weeks extends out to a massive amount by 220 weeks 5 points? The decline on MMSE of ~13 points is truly large and of course very clinically meaningful. The suggestion that this evidence should somehow be ignored is not credible.

The last figure shows how patients fared according to their SUVR level at the end of 302 (posted to this thread on June 29 2021; page 109 of the Medical Review of Aducanumab as recently posted to the FDA). As can be seen on the left, those with SUVR >1.1 often developed runaway dementia that could not be reversed. What is of interest with the figure on the right is that none of the patients with SUVR <=1.1 had this uncontrollable runaway pattern. A few patients did exhibit a lift off type pattern but then it seemed to stabilize. It would be interesting if a more complete dataset could be given for these patients. For example, what were the SUVR readings for those near the bottom of the figure? Were these patients those with the lowest SUVR levels? However, the write-up that accompanies these figures adds the caution "For these reasons and more, the explorations of the data provided above do
not meaningfully contribute to the evidence used to support the effectiveness of aducanumab.
On the other hand, there were no trends in the data which diminish the persuasiveness of the
results observed in placebo-controlled periods of the studies."

There is endless uninformed chat online about the CMS decision. Essentially none of this babble makes reference to anything factual. It is important that those experiencing cognitive issues have factually based evidence driving their decision making processes. The figures above demonstrate powerful evidence supporting anti-amyloid treatment.

These figures also provide us with insight into the post-301 and 302 trial results. These longer term views show us that anti-amyloids actually have lasting benefits for patients. The CDR-sb gains shown in these figures are actually quite substantial. It is impressive that the black line on the figure on the right only reaches ~1 CDR-sb while the black line on the left reaches 4(!) at 220 weeks! It is important that those with AD to fully appreciate how large the gains
can be on treatment especially over the longer term which the clinical trials under randomization do not explore.
J11
Contributor
Contributor
Posts: 2127
Joined: Sat May 17, 2014 4:04 pm

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

Post by J11 »

dona 1b.gif
You do not have the required permissions to view the files attached to this post.
J11
Contributor
Contributor
Posts: 2127
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 J11 exclusive.

The above figure is a reworking of the Donanemab phase 2 results.
What I noticed was that the reported endpoint for the phase 2 was at the week 76 mark.
Yet, the titration schedule for dona was different from that of aducanumab.
In order for the correlation plots to be consistent, the time points at which cognition/amyloid
are read off need to be consistent.

In the dona phase 2, a 700 mg dose was given at time points 0, 4, 8 weeks. (I am not sure about the week 0 time as this does not seem to be explicitly stated in the article, though this appears to be what is implied.) So, 52 weeks of treatment after 8 weeks is week 60. I data extracted the week 64 result as seen above. Using the week 64 iADRS ,
one has a 41% not 32% reduction in decline. {Note above that I hit the 32% result with the calculation( though after a wiggle of the result from 194 pixel to 193 pixel. 194 was at the bottom of the box on the figure; 193 more to the middle.}

When I extracted the week 60 result I hit 44.3% reduction in decline. The topline in the phase 2 was reported out as significant at 4.2% with a 32% slowing on iADRS; my best estimate is that the truer topline should have been 44.3% reduction in decline with p< 1.7%. This is more congruent with the time frame reported with Aducanumab.

I did a similar calculation for CDR-sb. This time the result was a 30% reduction in decline for the week 60 calculation versus the reported 23%.

Something very interesting than happened with the correlation plot.
J11
Contributor
Contributor
Posts: 2127
Joined: Sat May 17, 2014 4:04 pm

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

Post by J11 »

Amyloid Correlations 2a.gif
Dona Amyloid.JPG
You do not have the required permissions to view the files attached to this post.
J11
Contributor
Contributor
Posts: 2127
Joined: Sat May 17, 2014 4:04 pm

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

Post by J11 »

This is pretty exciting!

The top figure shows us where the new Donanemab result would be for the CDR-sb if we take the 30% reduction of benefit and scaled it up to the full ~78 weeks as reported for Aducanumab. The small pink rectangle on the left hand end of the purple regression line is this new result. As seen above, what I did was scale up the reported 0.36 CDR-sb by a factor of 30/23 to account for the larger percentage change seen at the week 60 result. The answer as seen to the top right is ~0.4696. The regressions that have been reported repeatedly on this thread have found a slope on the CDR-sb versus SUVR axis of around 1.27. When we take the 0.372 reported SUVR change for the dona phase 2 we find a CDR-sb expected change of ~0.472 ( quite close to the 0.4696 calculated previously).

Yeah!

The donanemab phase 2 was the one outlier result for the anti-amyloid mab. It had been off the regression line by about ~0.15. Now it is almost exactly on the line. One objection to this adjustment could be that the amyloid measurement was for the week 76 time frame and not week 60. Yet, as seen above the centiloid reading for week 52 was -82.3 versus -85.06 for week 76. Using a linear interpolation to week 60 we would have 83.22 centiloids --> 0.364 SUVR with a regression using 1.27 slope of 0.4622. (again quite close to the calculated value of .4696).

This is one possible way of explaining why the dona phase 2 result did not seem to align with the other mabs in the correlation plots. What seems to have happened is that the differences in the titration schedules created different cognitive results. For dona the better (possibly more accurate way to compare with aducanumab) is to choose the earlier cognitive readout which had a larger percentage benefit for dona than the later readout ( 30% versus 23% on CDR-sb). Admittedly one problem that lurks here is that the actual CDR-sb benefit at the earlier time benefit was somewhat less (0.33 vs 0.36). As can be seen in the first figure 2 posts up, the placebo decline at week 76 extended to around ~1.7 versus ~1.2 at week 60. While this might not be a perfect way to try and make the adjustment for the varying titration it does provide some insight into how the donanemab result could align better with the other results.
Post Reply