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

Alzheimer's, cardiovascular, and other chronic diseases; biomarkers, lifestyle, supplements, drugs, and health care.
NF52
Support Team
Support Team
Posts: 2772
Joined: Tue Oct 25, 2016 9:41 am
Location: Eastern U.S.

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

Post by NF52 »

J11 wrote:An investor Q&A with Biogen scheduled for 5 PM today.
http://investors.biogen.com/events-and- ... ing-events

Anyone know where we can access the pdf from today's CTAD presentation?
If I remember correctly from the Barcelona CTAD conference in October 2018, the PDFs were not released for several months. Can't find the source at the moment.
4/4 and still an optimist!
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, NF52.

Do they really intend to wait months?
We want to carefully read their latest update now.
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 »

aducan is such monumental news for the global dementia community.
It is startling.

Hundreds of millions of people around the world could soon live a very different life.

From what I can make out of my family's history, we have been deported, excommunicated, transported, exiled and otherwise expelled from almost everywhere on earth. Without aducan we would have had to start again at the beginning of the list and be deported, excommunicated ... from them all for a second time. A sizable proportion of the weirdness of history likely has some connection to dementing illness. I suppose that when we first arrived somewhere we could put on a fairly good show (comb our hair, polish our shoes etc.), though once dementia became obvious there would no longer be much point in pretending. The complete rejection of community is probably one of the more difficult aspects of dementia for others to understand; when you are caregiving 24/7 there is no room for a shared life.

Yet, here we are now and aducan offers us an opportunity to move into a new future.

Commentaries that I have read online downplaying the effectiveness of aducan are not sensible to me.
On an earlier page in this thread there is a figure that showed the brain imaging at baseline and treatment at 10 mg/kg aducan. How could one reasonably suggest no difference? The baseline showed a brain glowing in bright red (i.e., filled with amyloid); the picture beside it shows a brain of tranquil blue. The gradation of less amyloid with higher aducan is obvious.

The figure that showed the MMSE change in those who had a good versus poor amyloid reduction response is also notable. There was a 3 point MMSE difference in favor of high amyloid clearers. That is a large difference and was measured after 54 weeks! Also on a previous page there was long term tracking of MMSE changes out to 3 years. 10 mg/kg maintained ~ a 5 point advantage. These are large and clinically meaningful differences that are maintained through time. Current AD drugs offer zero disease modification. How can they be approved and marketed when the next wave of treatment that does offer disease modification is questioned concerning its efficacy?

Fixating merely on p-values entirely overlooks this compelling evidence. From what I can see, everyone has become so pre-occupied with p-values that they are not appreciating how much benefit aducan is offering some of the patients. If they could report that 30% (or more) of patients treated at 10 mg/kg gained 3 MMSE points over non-responders at 52 (78 etc.) weeks that would mean a great deal more to me than hitting the top-line p-value. It is only when I went back to the earlier aducan clinical studies did I clearly see how the actual evidence strongly supports the claims of efficacy. Interestingly these earlier studies were already reporting statistically significant results even with a small number of patients. With some wiggle room, I do not think that it would be reasonable for the FDA not to approve aducan.

I am greatly looking forward to the initial comparison of treatment versus placebo when the patients return to the clinical trial sites. It would be very impressive if those treated with aducan continued to show a prolonged advantage over the placebo. I am not entirely sure what would be expected to happen to the cognitive status of those who were on high dose but then did not receive treatment since March. Perhaps the advantage of the treated has continued to expand? Might they be able to immediately claim that significance had been achieved if the first visit were to demonstrate this?

Yet, what I am even more encouraged about is that merely continuing the dosing should move down amyloid levels in those treated, so that over an interval of months the gap between treated and placebo will be too obvious to dispute. We should over the nearish term expect to have the definitive answers that we have wanted for so long. My one concern is that aducan might lose its effectiveness in the now later stage patients who will return to the clinic. The trial was designed so that only early stage patients were enrolled. The risk is that the patients might no longer be as responsive to treatment as they were earlier in the trial.

Nonetheless aducan offers the global dementia community an opportunity to live a life without the cognitive impairment that has often put us at the margins of society.
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 »

After all of these years a wave of treatments with substantial evidence of efficacy are approaching the market accompanied by a wave of optimism that with effective treatment there will be a new more normal life for those in the dementia community.

Once the Engage and Emerge trials are fully restarted, over 3,000 AD patients will be on high dose aducan. Conclusive evidence for the efficacy of aducan will be determined next year. The slides for the CTAD presentation are given in the url below. They helped clarify some of the questions that I had concerning the numbers. Yet, i am still somewhat unclear why the previously reported phase 2 saw a decline in placebo of ~3 points on MMSE at 76 weeks, while 10 mg/kg declined by ~1. In the latest study this was reported as a 0.5 advantage of treatment over placebo. This might not be an apples to apples comparison, as all patients might have been included in the latest trial instead of only (10 mg/kg), at different treatment durations etc. . I would also have liked to see more disaggregated data. For example, a waterfall plot (i.e. an ordered comparison of treated versus placebo by response) would have been informative. Aggregating everything into top line numbers greatly obscures the probably large clinical benefit that a subgroup experienced under treatment. Perhaps such dramatics are considered unseemingly, though it would certainly have ended any argumentations. Clearly, our APOE4 forum would have interest in seeing the APOE4 results, though this too was kept obscure.

http://investors.biogen.com/static-file ... 557697c06f

CTAD also provided yet more clarification on the Grifolis phase 3 results. An over 60% reduction in decline in those with moderate AD continues to greatly impress me. The entire clinical evolution of AD from the earliest detection on amyloid testing to later stages into moderate (possibly even severe AD) now appears set to be amenable to intervention.

This is only the first wave of treatment, largely based on amyloidal theory. Other treatment paradigms are also approaching shore with possibly even larger treatment effects.

Happy 2020 everyone!
2020 is going to be an enormous year for all of us.

The dementia pandemic had to be resolved: apparently, it has been. Clearly life defined exclusively devoted to caregiving for our demented elders was neither financially, nor socially sustainable. When problems have to be solved, they are.

Over the next few years, it will be fascinating to watch as effective treatments reduces dementing illness in our community.
A return to vitality and to a more typical pattern of developmental psychology.
As Dickens would have said, "Recalled to Life".
roxanne
Contributor
Contributor
Posts: 124
Joined: Sat Apr 11, 2015 8:48 pm

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

Post by roxanne »

J11 I can't wait. I have a brother with MCI that would greatly benefit from this drug. He lives outside of the U.S. Hopefully we will not have to wait too long.
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 »

roxanne, thank you for commenting.

After all of the years and the decades, I have gradually mellowed and am now somewhat unfazed by the talk of cures that typically never seem to arrive. However, ... yes (I know the following should create some well deserved skepticism) this time it is different.

With over 3,000 patients on high dose aducan, a conclusive readout should be expected rapidly. One must remember that the futility analysis was done with the results locked at December 2018. With that dataset, neither of the trials reported a significant result. After the futility analysis and the stopping of the trial, 3 more months of patient data turned Emerge significant. As long as aducan is still effective in these now somewhat more advanced patients, only a few more months of treatment will decisively establish efficacy.

I do not understand why there is all of this drama? Why does everything have to be a stress-out? With what the FDA has now it could be a tough call: probably more in terms of optics than substance. Why not let the readout speak for itself? The over 3000 patients in the trials are known quantities by now- it should be clear which patients might not even have AD-- the rate of their expected decline
should be determined to 3 significant figures. It is all set-up and we can wait a few months and everything should be perfectly clear. Finally, we have a bet that is rigged in our favor. Society wins as well; with what is on the table now there is at least a $500 billion dividend to be cashed.
Obviously, the dementia strategy that has been pursued got behind the wave. Aggressive basic research should have started decades ago.

Other AD medications have also reported positive phase III results. For example, Grifolis has reported a significant slowing decline by 60% in moderate AD. Clearly, one does not want to create false expectations, however, what we are seeing now can be realistically described as the arrival of effective Alzheimer treatments.
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 Celebration is one step closer to a full party!
Biogen completed its BLA submission today for aducanumab and requested priority review from the FDA.
The FDA must respond within 60 days (accept or reject the submission).
If the FDA were to grant priority review, then they would have until March 2021 for a Yea or Nay decision.

Yeah! This is great!

Biogen has seen the deck and they know the scores for the patients once they arrived back at the clinic
in March and perhaps 3 months of additional treatment of high dose treatment for possibly ~2000+ patients.
They likely have a strong hand or they wouldn't have requested priority review. Of course, bluffing is not
allowed in this game.

I am very interested in the question of the expected cognitive levels of the patients when they arrived back for their
first visit in March. I vaguely remember reading research on this very question when I was posting last October.
What will happen to the patients after a year without treatment? Placebo is an easy one to guess: - 3 1/3 MMSE points.
Perhaps Biogen might even clean up their dataset by removing from the analysis those patients who did not appear
to exhibit typical declines (perhaps these patients did not have AD). Also wonder whether including the idea of fast
and slow decliners might be added to the data analysis.

With the treatment group decline over the year of non-treatment I am less certain. Would these patients quickly
revert back to the cognitive level of those on placebo? Might the differences in scores continue to increase or at
least stabilize? As a rough guess I will call it that non-treatment for a year will result in static absolute scores.
The amyloid has been cleared; amyloid that has taken a lifetime to accumulate has been removed and the brain
should have some room to heal. With the placebo group we saw that amyloid levels were mostly constant through
time, though all of that toxic protein in the brain drives forward the neurodegenerative process. The treatment group
had much of these toxins removed. If this thinking is correct, then the MMSE difference should expand.
Normalized to the Prime trial which saw a 1.1 MMSE expansion in the second year of treatment I will hedge
and call it a 0.5 improvement in EMERGE (Remember Prime understated the true treatment gain, so 1.1
might actually be ~1.5). If such an expansion did occur from last labs after treatment and then first labs
at restart in March, then the result should already have been pushed to higher statistical significance
even without additional treatment.

There are a few other positive features that I have noticed when examining the clinical trials with adu more
carefully. In the first clinical trial (so called PRIME {nominally Phase 1, in reality a phase 2}) there was no dose
titration. It was just straight 1 mg/kg, 3 mg/kg, 6 mg/kg, 10 mg/kg, placebo [all per month]. In contrast,
the EMERGE/ENGAGE trials used a titration scheme in which the doses 10 mg/kg etc. were not reached until week ~24.
This helped to solve the mystery that I raised earlier in this thread that the different dosing arms appeared
to all have the same amyloid reduction benefit for the first 24 weeks of the trial. I could not understand why
high and low dosing with adu had the same effect? Now we know.

This implies that the nominal 78 week trials adu phase 3s are not really 78 weeks; they are actually ~54 weeks
(in particular for the 10 mg/kg dosing arm (i.e. the effective dosing level)). The 10 mg/kg dose needed 24 weeks to
hit its dose level with titration.

Yet, here again there is an underestimate. Week 54 PRIME is not really equal to week 78 EMERGE because
Emerge patients actually did receive 78 weeks of treatment, though not full dose due to titration.

Go to https://apps.automeris.io/wpd/ and upload the MMSE response curve to extract the
data points for the placebo and 10 mg/kg dosing arms from the PRIME trial. (The website inexplicably provides 15 significant figures after the decimal point, though amusingly I was unable to correctly achieve even +- 10% accuracy with the x values as 24 weeks somehow managed to be transformed into 27 weeks during the extraction.) I uploaded the figure earlier in this thread.

Remember no titration here with PRIME. The MMSE benefit of 10 mg/kg vs. placebo increases (namely,
0.5, 2.2, 2.7, 3.3, 4.3, 4.0); for the 24, 52, 76, 108, 132, and 164 week readings, respectively.

(Note: There is confusion because the placebo in PRIME was no longer a placebo after 52 weeks. Placebo received 3 mg/kg or 6 mg/kg treatment after 52 weeks. Therefore the 2.7, 3.3, 4.3, and 4.0 are likely underestimating the MMSE benefit. If we simply use a typical placebo decline of ~3 1/3 MMSE points per year, we expect placebo decline of ~10 points at ~ year 3. Online extraction tool found the reported decline of placebo at week 164 as ~7.9. They underestimated placebo decline by 2 MMSE points? The "real" difference between 10 mg/kg and placebo is 6 points at nominal time point 164 weeks? Thesaurus please. Big? Mammoth? Gargantuan? How could you have a good argument with such decisive empirical findings?)

From extraction above, observe that the 52 week reading is a 2.2 point MMSE benefit of 10 mg/kg over placebo in PRIME. This is the equivalent time point that the EMERGE and ENGAGE trials reported for 78 weeks. What is highly notable here is that the Prime's benefit continued to increase from 2.2. In fact it almost doubled at zenith to ~4.3 (This occurred at 132 weeks). This implies that Biogen has the treatment momentum on its side. It is always helpful to have the mo with yo. The trend is your friend.

Thus, for 80 weeks from the resumption of the trials in March, 2020, the MMSE treatment difference should expand. Biogen has announced that the current Phase 3 extension will run for ~100 weeks. If the current dataset were deemed insufficient by the FDA, then Biogen would have another even more convincing dataset available within 24 weeks, and so on and so on. Basically, they can keep taking shots at net until they score. Such an advantage would end much enjoyable recreational argumentation. One might plausibly argue whether a difference of 2.2 points were in the 95% confidence interval of no effect. Yet, at 4.3 points for Prime's equivalent time point week 132 the room for legitimate debate would no longer exist.

The above makes a great deal of sense to me. However, when I look at the numbers from EMERGE and ENGAGE confusion emerges.
From PRIME, the week 52 (i.e., week 78 reported by EMERGE/ENGAGE) MMSE scores were -2.7 (placebo), and -0.4 (10mg/kg). This is an 85% reduction in decline. Yet, EMERGE reported a 23% reduction in decline for 10 mg/kg vs placebo. Are the two reductions actually referring to the same thing? Could they possibly be that different? 23%? vs. 85%? Even with differences in sample sizes?


Nevertheless, this is a highly encouraging sign that the BLA has been filed and that the winds of scientific evidence are pushing in our favor (We have the mo). The first disease modifying treatment for Alzheimer's would be a blessing for the tens of millions of those currently affected with dementing illness and the hundreds of millions of the worried well.
Last edited by J11 on Fri Jul 10, 2020 5:40 pm, edited 5 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 »

The MMSE numbers for the EMERGE showing the absolute scores were, in fact, disclosed.
(See the investors pdf from the CTAD meeting.)

I extracted the data points with the online tool I mentioned above, though I only
included 2 decimals points (not 15). If anyone can do better with the extraction
tool, please keep your achievement quiet.

1.35, -0.00
26.58, -1.70
52.20, -1.90
77.49, -2.70
52.76, -2.31
78.58, -3.32


What we see here is that the placebo decline was ~3.3 at 78 weeks and the 10 mg/kg was ~2.7.
This gives the +0.6 high treatment advantage noted in the pdf (~19% reduction in decline).
All three points (low, high and no dose) overlapped at (0,0) and (26,-1.7).

It is surprising that this is so much smaller than the 2.2 gain noted from PRIME.
The placebo decline seems somewhat light, there might be a point or more
hidden decline lurking there that might account for the difference. If so, this
decline might have been accounted for in the March cognitive assessment.

Why didn't they show us the waterfall?
~Half of the treated patients in PRIME had strong responses (strong response as defined as >- 1 SD above average of amyloid clearance) with a result of stabilization of MMSE score over the course of a year. Non-responders (those with <-1 SD amyloid clearance from average) declined ~3 MMSE points (i.e., benefit). How many placebo patients had no decline after a year?
Yet, placebo patients that do not decline might not even have AD.
Perhaps some of the treated patients who did not decline also were non-AD.
However, mixing the responding and non-responding treated patients greatly obscures the true treatment effect.
It was a missed opportunity not to have the patients all set in a patient registry with established rates of decline etc.
Possibly the entire placebo pool could then have been used for the aducan trials.


Hmm, the EMERGE trial enrolled AD patients that were less advanced than those in the PRIME trial.
With EMERGE, MMSE scores ranged from 24-30 inclusive, in PRIME enrollment was for mild AD 20-26 inclusive and
prodromal 26-30 inclusive. Wonder if that made a difference.
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 hope readers find my comments interesting.

Admittedly, aside from dental surgery there are few things that I have less enthusiasm for than writing for.. er.. fun.
An occasional, "Well said J11", or "I couldn't have expressed that idea better myself" would be greatly appreciated.
I find that the best policy is to write and then run; going through proper revisioning is simply too much.

Nevertheless, I will continue to try and understand the aducanumab results.

Using the online app (again) I calculated the correlation between the SUVR (and the change) at week 54 (also versus baseline for change) and CDR-sb and MMSE for the PRIME study. 98.5% for CDR-sb ( p=0.0022) and ~71% for MMSE. This might seem to be
somewhat obscure; indeed, what relevance does this have? Just as with finding the link between cholesterol and heart disease, it is important to firmly establish the connection. Once established then the biomarker can be a substitute for the other measures. In this context all that would then have been necessary is to report change in amyloid levels which clearly declined.

Also of interest is that notwithstanding the preoccupation in my previous posts with MMSE, the primary psychometric instrument chosen to determine efficacy in the phase III trials was CDR-sb. This is understood based upon the research literature to be the appropriate test for those with early AD (i.e., the patients enrolled in the trials). As can be seen above, CDR-sb followed SUVR almost
exactly in the PRIME study. CDR-sb was largely a proxy for SUVR in PRIME. It is then no great surprise that EMERGE reported a strongly positive top line result on its primary efficacy measure, CDR-sb (0.01 ITT).

Page 15 from the original Biogen pdf shows the CDR-sb for EMERGE and ENGAGE for those who received "sufficient" high dose exposure to aducan (on the right). Of note here is that the patients on high dose in the two trials had almost the same cognitive response on CDR-sb. Almost 100% of the difference from placebo in the two trials resulted from a difference in the change in placebo. If one were to simply switch placebos between the two trials, then ENGAGE would then be significant and EMERGE would not. Perhaps it should be required as standard practice in reporting series of clinical trials to combine placebos.

Also for interest on the left of this same page is the size of the delta for the SUVR in the trials especially at 24, and ~54 weeks.
From Figure 2b of PRIME we can see that at the lowest doses there is minimal SUVR change while there is a large change at 10 mg/kg (~0.20 mean composite SUVR). Yet at week 26, this results in almost 0 delta over placebo for high dose (Figure 3A). Back
to page 15. Emerge showed ~0.10 drop in composite SUV at 26 weeks. [Edit: Week 26 from the CTAD pdf, actually shows a magnified view of the SUVR: it was ~0.07, Given this it was somewhat aggressive to expect a positive result at week 78)]. This highlights how out of temporal alignment the two trials are. As implied by PRIME, there should be virtually no improvement over placebo in EMERGE or ENGAGE even at 52 weeks. This is largely what the figures on the right on page 15 show. There is some improvement, though this is likely a consequence of the treatment of the patients for a year, just not for the full time at a high effective dose.

There is lots of great information in the two pdfs. For example, figures are included that separate out prodromal/mild AD, APOE epsilon 4 +/-, and different brain region response at different dosing levels etc. for PRIME.

I am especially impressed with Figure 2c of PRIME. This figure shows the cognitive responses conditioned upon high versus low amyloid removal. It clearly suggests that the average cognitive change for EMERGE of ~2.7 on MMSE is not an homogeneous result.
With PRIME about half had stable MMSE and half had typical placebo like decline.

I am not sure whether this is allowed, though reporting response by actual amyloid response could be extremely powerful. Merely giving someone a high dose of aducan does not necessarily imply that they will have more amyloid removal than someone on a low dose or even placebo. If amyloid removal is the presumed mechanism of action, then why exactly is the treatment assignment of central importance in the analysis? Treatment class is actually an epiphenomenon: what truly matters, as the independent variable, is amyloid removal. I really want to see this analysis for the phase IIIs. A 2D figure showing patient data for amyloid removal on x axis and cognitive measure on the y axis, color coated by treatment arm, prodromal/AD status, APOE +/- etc. likely would be extremely insightful.

On additional observation that I made is the phase III trials involved patients from many nations. This was not true with PRIME which drew patients only from the US. In fact, the PRIME article saw this narrow geographic focus as a limitation. I am not sure whether that is true. While Biogen is doing its all to embrace wokeness, it is not entirely clear to me that such a perspective aligns well with the nature of Alzheimer's. While science and medicine does what it can to rise above culture and place and strives for an objective truth, I have a suspicion that possibly Alzheimer's is flavored by local context. There might be an American Alzheimer's, a European Alzheimer's .... If so, this would have added noise to the phase III results.
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 »

[img][img]
You do not have the required permissions to view the files attached to this post.
Post Reply