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 wanted to start to get a ball park feel of where the Graduate trials might wind up. The above shows roughly what one of the Graduate CDR-sb readouts could look like by the numbers: 1 point difference on CDR-sb at 116 weeks, 350 per arm to allow for 30% drop out, 2 SD ( as was seen with Lecanemab). The t score of 6.6 puts this well outside of what is reported in the tables. Basically, if you achieve 2-3 you are good; Approaching 7, might just call it a day with <0.0001 as above.

I am just not comfortable calling a supersized result such as this. We have been more in the region of ~0.4 at 18 months. Yet, by going out to 2 years 3 months there is certainly the potential (and probably a near inevitability) that treatment effect will enlarge. 0.7 to 0.8 on CDR-sb, though large, does seem at least conceivable with extended dosing, yet I am still more fixated on the 18 month readout point as this is the time point that we have the most accumulated evidence from other trials. The 2 years plus treatment effect is something of an unknown for me 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 »

I replied on the Lecanemab thread about a patient incident recently reported and I want to add more to that answer here.

One of the more outstanding pieces of news that was recently reported was that:

“3.1% of all patients in the #lecanemab study on anticoagulants developed significant bleeds called MACROHEMORRHAGES compared to 0.46% of patients not on anticoagulants”.

- 6.73 times!

Those on anticoagulants had 6.73 times more macrohemorrhages than those not on anticoagulants? There is a great deal of risk heterogeneity that is not obvious when all patients are grouped together. The anticoagulant issue has been noted for some time, though it is helpful that there is now a quantification of the risk involved.

What is of interest is how much the risk is concentrated in a pre-known subgroup. I commented on this
feature of clinical trials before. Clinical trials can have very non-homogeneous risk across patients. I am
somewhat surprised that this is not prospectively pre-specified. One could have some sort of a risk score
that would quantify the risk that patients faced. Why isn't this done now? Where would the recently reported patient be on such a risk distribution? Some patients are just known or expected to be at higher risk. What patients and health care workers are most interested in is determining what level of a risk that is likely faced by a patient with matching risk to those in the clinical trial. The outlier patients with extensive comorbidities might offer minimal insight to the risks faced by the typical patient. Pretending that such heterogeneities of patients do not exist seems very deceptive.

For the patient report that has recently been publicized, I found the suggestion that the patient might have been dosed through a number of serious medical events problematic. This point is not clear in the report; it is possible that the medical events referred to were much earlier in time than the ARIA events. Providing the timeline could be helpful. The important point is that anti-amyloid mabs will provide a benefit upwards of years in the future; if a patient is struggling medically, it would seem for the best to make sure that medical stability has been clearly reached before reconsidering mab therapy.

Of further note is that the patient was treated in the long term extension round. What is of interest here is that the patient possibly was on placebo during the phase 3 portion and then went to mab treatment in the extension. This is relevant because the inclusion for the trial likely required patients to be reasonably healthy at the start. What might have happened was the patient developed a series of serious of medical problems during the placebo phase and then was allowed to roll forward into treatment. I wonder how stringent the inclusions were in the re-enrollment phase for the extension trial or if there might be some mechanism to proactively remove patients from the trial who develop recurring medical problems. Also of interest is that I believe the extension offered patients the chance to receive subQ dosing. Did this patient receive subQ?

One of the benefits that I had not been aware of with Lecanemab was its short half life. Lecaneamb has a half life of ~7 days. This compares to ~25 days for Aducanumab. The short half life would seem to offer important safety advantages. Consider if one developed brain bleeding/hemorrhaging. With Lecanemab the body could over the course of a few days clear the mab from circulation and possibly bring the concentration down below the symptom threshold. With Aducanumab this might require weeks. Also when dosing, Aducanumab would be dosed after 1 half life; with Lecanemab even when dosed twice a month, would be 2 half lives away from the last dose. I am not sure of the PK involved, though do you really want to be 2 half lives away? This feature, though, would add built-in safety. Perhaps the compromise that was reached is that they did not want to move to weekly dosing. It would then seem more feasible for people to control ARIA symptoms (if present) by simply waiting an extra week or two before another dose. With Aducanumab people might feel more pressured to continue dosing even if ARIA symptoms were present.
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 incidence of amyloid-related imaging abnormalities-edema/effusion (ARIA-E), an adverse event associated with anti-amyloid antibodies, was 12.5% in the lecanemab group and 1.7% in the placebo group. The incidence of symptomatic ARIA-E was 2.8% in the lecanemab group and 0.0% in the placebo group. The ARIA-H (ARIA cerebral microhemorrhages, cerebral macrohemorrhages, and superficial siderosis) rate was 17.0% in the lecanemab group and 8.7% in the placebo group. The incidence of symptomatic ARIA-H was 0.7% in the lecanemab group and 0.2% in the placebo group.


7 = 0.48 x% + 3.23 (1000-x)%

7 = 32.3 - 2.75x%

25.3 = 2.75x%

x% = 9.2

x = 920 0.48% --> 4.416

1000-x = 80 * 3.23 --> 2.584


This is the number crunch for the anti-coagulant vs non-anti-coagulant patients.
With roughly 1,000 patients there would be 7 symptomatic ARIA-H in the treatment arm (.7% *1000).
The patients not on anti-coagulants had a 0.48% risk, while those on anti-coagulants had a 3.23% risk of symptomatic ARIA-H. Weighting out the risk gives us x=920 and 1000-x = 80.
920 patients were not on anti-coagulants and of these patients ~4.5 had symptomatic ARIA-h.
80 patients were on anti-coagulants and of these patients ~2.5 had symptomatic ARIA-H.

Interestingly, since we are dealing with whole number of patients we might be able to determine the exact
numbers involved by searching for the solution that did make the numbers whole.

The above helps to clarify how many of each patients were affected by ARIA-H in the different risk groups.
I had thought that perhaps there might be a mix as many as 300 on anti-coagulant to 700 not on anti-coagulant and then the on antis would form a majority of the those with ARIA-H. As seen above though that the on anti-s are probably a fairly low number on the trial and made up a minority of the symptomatic ARIA patients.

I am not sure whether these numbers are correct, though this is the best that I can figure given the available numbers.
Last edited by J11 on Wed Nov 02, 2022 11:50 am, 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 »

My overall impression about the Lecanemab safety issue is that there is a rapid and ongoing evolution towards reduced risk. Lecanemab has to date shown significant enhancement of safety relative to Aducanumab. In addition the research into the subcutaneous route of administration for Lecanemab suggests that this would be multiple times safer than the IV dosing that had been the standard in the phase 2 and 3 clinical trials. Surprisingly, it is thought that epsilon 34s might have the same (or roughly similar) ARIA risk as the e33s while under treatment with Lecanemab. I am interested to see research that could verify this assertion.

Yet, another safety enhancement that is evolving in the background is the potential for patients to be treated with LMTM in place of mabs. Up till now we have largely been locked into the dilemma on thread of what the millions of people on the cusp of the black hole of Alzheimer's could do. The assumption that we have made is that mabs were the only near term treatment that would be available as an approved medicine from regulators. LMTM now appears to be on the nearish term approval horizon. High risk patients could then be treated with LMTM and could avoid ARIA problems. What we can anticipate happening is that as those AD patients with the most commorbidities and thus highest risk of ARIA will progress through the mab treatment window, and risk for ARIA will decline.

The standard of care for AD will move towards initiating treatment with mabs at least by the time a patient acquires an MCI diagnosis. The risk of ARIA that we have seen relates to the fact that the patients were never treated for their CAA pathology. I will be watching to see whether at some point CAA itself might become an on label indication for anti-ab mabs. I continue to wonder: How much stroke is caused by CAA? How many patients did not survive to be included in the mab trials due to CAA illness and perhaps early coexisting AD? It is also of interest to consider the potentially reduced risk of stroke/ARIA that the mab patients might experience once they have been treated (in comparison to the placebo)? Cleaning out all of that cerebral amyloid in the patients treated with mabs could result in a lasting and significant improvement in their health above and beyond the benefits related to AD.

However, I still see mabs as a central treatment option in AD. Amyloid is the most upstream pathology in AD. Shutting off amyloid early would shut off AD. Pharma has been highly aware of the central importance of amyloid for a long time and they are strategically positioned to turn off the tap on Alzheimer's. For those towards the early end of AD or those interested in prevention amyloid is clearly the best place to focus efforts.

tau is then more of a downstream target that can help (possibly help a lot) though it is still downstream. Those who are primarily focused on symptom control and not disease control might be satisfied with a tau blocker. However, from what I can see an anti-amyloid mab would be my goto treatment.
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 »

Dosing 1.PNG
Dosing 3.PNG
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 »

Stages.PNG
AD Progression.PNG
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 »

ARIA 1.png
Capture 2 Edit 5.png
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 »

ARIA 3.png
ARIA 2.png
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 think I have probably posted all of the above figures to thread before, though I now realize that there are those who have not read the entirety of the thread and need some reminding of past posts. So I will recycle some of the good ones from time to time.

The first pair of figures might be somewhat out of sequence, though it still a good one. It shows how much of a risk reducer subcutaneous dosing could have with Lecanemab. It is fairly startling. The Clarity press report from last month reported the expected 12% ARIA-E; that is what we see for the APOE e4+ on the left hand side for IV dosing ( bottom figure of first post in this sequence); the e4-s on IV in the simulation were ~6%. Then they modeled subQ and the ARIA-E rates can be seen to drop by multiple times. The e34s might even approach the rate of the e33s at ~2%.
What is highlights for me is how difficult it is to have a reasonable discussion about Lecanemab when the yardsticks are moving upfield so fast. This is not avoiding the question so much as it is directly addressing the concerns and getting us to a better treatment.

The top figure in the first post shows how much spiking behavior there is with IV dosing and how this type of dosing adds risk without contributing any marginal treatment effect. I had been unclear why there were the little blue hills with the SC dosing, though I can see now that the hilltops are the Cmaxes for the weekly SC doses.

I like the next two figures that illustrate the disease progression of AD. I am not sure whether these figures are meant to be stylized, though they seem to offer quite a bit of comfort to those in the pre-MCI period. The figures show that AD creeps forward over a multi-decade time frame. Not surprisingly (from the amyloid hypothesis point of view) amyloid pathology begins upwards of 3 decades before disease onset. It is somewhat surprising that memory happens so late (only 10 years before onset?). There is clearly a great opportunity to prevent AD well before the involvement of tau and neurodegeneration.

The next set starts off with the phase 2 safety data. What caught my eye in particular here was how low ARIA-E was for the lower doses. Going from 5 mg/kg biweekly to 10 mg/kg monthly (which is equidose) almost triples ARIA-e risk in e4s, for the e4-s ARIA risk with 5 mg/kg biweekly was reported as 0%. For some patients (especially those with time to spare) the risk reward ratio for slower treatment might favor the slower dosing approach.

The figure below shows the CDR-sb vs SUVR graph for a range of mabs. What I find of interest is how the higher of the equidoses the 10 mg/kg once a month and the 5 mg/kg once a month removed more amyloid then the corresponding equidoses, though showed less cognitive response. I noted that with a 7 day half life on once monthly dosing patients would be 4 half lives away between doses. This would mean that for much of the month, patients would have minimal drug exposure. This might help to explain why the more frequent of the equidose pairs had better cognitive results.

The figure directly above shows the large reduction in ARIA-E risk when moving from Aducanumab to Lecanemab.
It is surprisingly large; and then there is different risk grouping based upon (e.g., anti-coagulant use).
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! They have done an ARIA GWAS for Aducanumab!

Some patients are simply at high genetic risk for side effects.
Even in my own family experience these very very unusual idiosyncratic responses have occurred.
That is what makes medicine so interesting: you give someone a placebo and this causes a medical emergency.
If life were as we expected it to be, there would not be much challenge for medicine.

The ARIA GWAS could be another great way of derisking mabs of ARIA side-effects prospectively and thereby enhancing mab safety. Perhaps all the pharmas could create a common ARIA GWAS database so that the shared knowledge could cooperatively build up a team effort.


Another safety enhancer that I want to note is the pretitration idea. I have already mentioned this one, though I want to restate it as I have previously mentioned it is such low hanging fruit; my perception is that it is not so much low hanging fruit as it is ground lying fruit-- it's an easier grounder to scoop up to score a win.

The idea would be to take patients who were estimated to be one year (or more) from official on label mab and dose them at the maximal dose determined to produce zero ARIA and then roll them into the on label dosing titration and treatment dose schedule. The goal would be to have zero ARIA in the pretitration round and also zero ARIA in the on label dosing. What I find exciting about this idea is that it has near term potential to be doable. Results could be possible over a 1-3 year time horizon. All that is involved is to establish safety; efficacy has already been shown so would not be necessary. A year or two of slow uptitration might be just about right to have no ARIA risk.

The AHEAD trial is 5 years from completion, though it's purpose is to establish efficacy in the pre-MCI stage. That is a more difficult task; you would need much better cognitive tests such as PACC to find a treatment effect over placebo. With the pretitration idea, the purpose is not treatment efficacy but safety. It would be brilliant if they could offer patients a near ARIA free treatment guarantee if they were pre-titrated as suggested above. This proactively addresses safety concerns and potentially could eliminate such concerns for the many people who are pre-MCI.

I think moving ahead on this idea would be a very good strategy. Offering a glimpse of a future without ARIA risk would be a profoundly powerful way of countering concerns about mab safety. There are so many tens of millions of e34s and e44s who are well before their age of onset, easing their concerns about side-effects (and the considerable costs involved in monitoring the risks) would clearly be a large win. It clearly highlights that ARIA risk will continue to evolve in the future and in time could be expected to vanish. The nature of the evolution of treatment risk was one of the 3 core questions that the CMS posed for anti-amyloid mabs. This pretitration idea really should be made a priority. Perhaps a clinical trial could be started even before the PDUFA date.
Post Reply