Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
How long does it take to analyze data?
I'm at a large conference and Sanofi just presented three phase 3 trials on tolebrutinib - 2 in RRMS and 1 in SPMS ---- close to 3000 patients overall. The 2 phase 3 trials had the last study visit 6 weeks ago, and then had a shared database lock 4 weeks ago. They analyzed the primary and key secondary endpoints, checked and rechecked, wrote a PR for the initial PR only 2 weeks later and then presented all three trials at ECTRIMS today. They plan to submit to the FDA for progressive MS by the end of the year. Missling is not a serious manager.
I’m in your town for ECTRIMS. Haven’t run across the chocolate covered herring but I’ll keep looking. The pastries are amazing.
146 and 20 are about the same
Crappy phase two’s lead to poorly designed phase three’s. But this is Dr. Missling’s modus operandus. A great statistician can’t make up for a cheap CEO
I agree for the most part - Dr. Jin will have invaluable input into the study design - especially endpoints, power calculations to decide N, and the statistical hierarchy and how the endpoints will be evaluated but will not actually design the trial. Power calculations are done without much budget consideration. N and cost of the trial are related and the final decision will be up to Missling. However, though I would not consider Dr. Jin to be an AD or PD or Rett expert, I would consider him to be an expert in neurologic clinical trial design and their analysis. He has personally seen many dozen neurology trials and their associated strengths and weaknesses. He will be in charge of the SAP . If Anavex had not hired Dr. Jin they would be spending just as much on consultants (as they likely did with Dr. Ette)
Yes, the only thing leveraged was time to complete an adequate sized phase 3 trial
This paper goes over the thinking used to design the Exellence Rett study based on the good data from the small US Rett trial.
The new big is small: Leveraging knowledge from small trials for rare disease drug development: Blarcamesine for Rett syndrome
https://bpspubs.onlinelibrary.wiley.com/doi/full/10.1111/bcp.15843
They assumed that results from week 7 to week 12 would simply be extending the line from week 0 to week 7 out to week 12. Also that 30 mg would naturally be better than 5 mg (both of these in figure 6) The problem was only partially the initial design but the stupidity not to significantly adjust the N up after the failed Avatar trial (yes a fail using the accepted endpoints) is 100% on Missling. 12 weeks was not better than 7, 30 mg was not better than 5mg. The paper showed a great knowledge of Bayesian statistics to design the initial trial --- only problem is they did not form a new posterior hypothesis by adjusting the prior hypothesis after the Avatar data was available and increasing N to 180 or so. The study was also hampered by the 2:1 randomization. Maybe the new Big is still the old Big.
Ene Ette's resume differs from Dr. Jin but both are expert clinical trial statistician
http://www.anoixiscorp.com/about/ette-ene.htm
and if 5.50 breaks, looks like 4.66 is next support
It's listed.
Here's the link to the actual article: https://ojrd.biomedcentral.com/articles/10.1186/s13023-024-03313-8
on the right, where it says Sections, scroll down to Author Information
Author information
Authors and Affiliations
Anavex Life Sciences Corp, New York, NY, USA
Walter E. Kaufmann, Heather M. O’Leary, Vanessa Vogel-Farley, Katherine V. Barnes & Christopher U. Missling
Emory University School of Medicine, Atlanta, GA, USA
Walter E. Kaufmann
University of Alabama at Birmingham Heersink School of Medicine, Birmingham, AL, USA
Alan K. Percy
Vanderbilt University Medical Center, Nashville, TN, USA
Jeffrey L. Neul
Telethon Kids Institute, The University of Western Australia, Nedlands, WA, Australia
Jenny Downs & Helen Leonard
International Rett Syndrome Foundation (IRSF), Cincinnati, OH, USA
Paige Nues
Rush University Medical Center, Chicago, IL, USA
Girish D. Sharma
Monmouth University, West Long Branch, NJ, USA
Theresa E. Bartolotta
Maastricht University, Maastricht, The Netherlands
Gillian S. Townend & Leopold M. G. Curfs
Pro Rett Ricerca, Sermide E Felonica, Mantua, Italy
Orietta Mariotti
Rett Syndrome Association of Australia (RSAA), Grovedale, VIC, Australia
Claude Buda
National Institute of Mental Health Intramural Research Program, National Institutes of Health, Bethesda, MD, USA
Lindsay M. Oberman
What could possibly go wrong with polynomial extrapolations far beyond the endpoint?
I pulled up the 2a OLE not the 2b/3. My mistake. The 2b/3 does have ADAS-Cog and ADL. There is no mention of MRI or biomarkers. I'll stand by the rest of my post
He and his aliases are the main reason I keep posting
My mistake, I linked to the phase 2a OLE not the 2b/3. Do you have any link as to whether MRI and biomarkers are included?
He will also be invaluable to help design the P3 trial
Generally, OLE's remain blinded to previous treatment when patients rollover. This is done primarily to keep the integrity of the phase 3 (or in this case, 2b/3) which is all important to regulatory agencies, but it also helps to keep the integrity of data gleamed from the OLE. However, patients in the OLE will be aware that they are all on drug - though not always to the dose (once again for integrity). Data is not available for analytic review by the sponsor (i.e. Anavex) in an immediate searchable database manner. However, if per-specified in the SAP, data will be available for analysis at certain time-points - i.e. at one year. If not in the OLE SAP, data should not be available until the end.
Name Ticker Weight Shares Held Date
ANAVEX LIFE SCIENCES AVXL 0.132393 2140578.000 June 21. 2024
ANAVEX LIFE SCIENCES AVXL 0.320167 5210012.000 June 20. 2024
State street is down 3.5 million shares. They were the #1 holder last quarter. Still have 3.7 MM Shares. This is mostly due to the XBI fund cutting AVXL by half. If MC gets stays above 400 MM they will re-add some of these back next quarter
Dew, thank you for the message.
I only see NAION about once every year (though I see inflammatory ON near weekly). I seem to get the most neuro-ophthalmologic referrals in my group, probably because so many neuroimmunologic cases affect the eyes and the academic centers are backed up 10 months for new patient slots. Most of my optic nerve referrals are inflammatory optic neuritis (seen in MS, MOGAD, NMOSD) but every now and then I see an AAION or NAION several autoimmune diseases affect. Clinically, ON, NAION and AAION all overlap though treatments, prognosis and association with other diseases differ. NAION is fairly rare and I do not know how tight the association with semaglutide is based on the recent study (mostly due to profound referral bias in that study). The proposed mechanism for NAION is hypoperfusion of the disc causing edema of the disc head. Then in crowded disc heads (anatomic, drusen) the swelling itself causes a positive feedback loop causing more hypoperfusion and subsequent edema/damage. PDE5-I may increase hypoperfusion.
The association between NAION and PDE5 drugs has been observed in several studies but challenged in others
It looks like they have built up a database of EEG tracings to train an AI model of the EEG pattern recognition. Well trained AI is generally very good at pattern recognition and most EEG software packages already have rudimentary spike detection and some have a power spectrum abilities. The issue would be is there enough value add. AI might be able to read an EEG better than me (I just do a couple a month). My epilepsy fellowship trained partner can read an EEG study in about 5 minutes (she does about 80 standard and a few ambulatory EEG/month). I do not think she would find it that valuable to have a computer do the interpretation (especially since some overread might still be necessary). The technique may have value for non-neurology practices (like psych) with physicians not trained in EEG interpretation. I also do not know how unique there database is. There are many publications reporting the use of AI and some aspect of EEG interpretation (https://pubmed.ncbi.nlm.nih.gov/?term=ai+eeg) . Do they have a special sauce?
Is there a reason you keep referring to the small Phase 2 for donanemab instead of the 6 times larger phase 3? In case you don't have the link:
https://jamanetwork.com/journals/jama/fullarticle/2807533
sab, here is a link to the slides presented at AAIC a couple weeks ago. Look at the safety cohorts slide #22 near the end
https://www.anavex.com/_files/ugd/79bcf7_79c4b8754b9f4eaca6da3e912e11689b.pdf
The 6th row is entitled "TEAE leading to treatment and study discontinuation". TEAE is treatment emergent adverse event. Combined had 103 out of 335 or 30.7% dropouts. Placebo had only 12 out of 168 or 7.1% dropouts
The mAB trials were more heavily screened. If you look at Figure 1 from the donanemab trial:
https://jamanetwork.com/journals/jama/fullarticle/2807533
you will see the reasons patients were excluded. The main reason was MMSE out of range and PET not showing amyloid. Better pre-screening might have avoided some of the MMSE range exclusions.
You are conveniently ignoring the huge number of patients who dropped out during the up-titration. The FDA and EMA won't.
The safety population (all patients receiving one or more dose) was n=167, 168 and 168 for 30 mg. 50 mg and placebo.
After up-titration n was 154, 144 and 164 for 30 mg. 50 mg and placebo.
At 48 weeks patients who had an ADAS-Cog results were n= 108, 83 and 122 for 30 mg. 50 mg and placebo
So patients completing from baseline to end of study were
30 mg : 65%
50 mg: 49% (not a typo)
Placebo: 73%
Your numbers are based on the patients who completed the trial after "surviving" the up titration were
30 mg: 70%
50 mg: 58%
Plac: 74%
And these numbers were over 48 weeks not 72 weeks
Hopefully, patients will do better in the upcoming Phase 3
Data from Dr. Sabbagh's slides https://www.anavex.com/_files/ugd/79bcf7_79c4b8754b9f4eaca6da3e912e11689b.pdf
Going from initial baseline to 52 weeks, dropouts were quite a bit more for the patient's on drug compared to the placebo patients, even more so if just looking at the dropouts during the up-titration. Therefore, COVID-19 issues are unlikely to have played much of a role. I should add that every single drug study running between 2020 and 2023 also had to deal with COVID related issues. The lecanemab and donanemab trials did not have nearly as much disparity between placebo and treated dropouts (the actual IV infusions were fairly well tolerated though of course there were some dropouts due to the ARIA MRI changes)
Regardless of the "KOL's" that Missling quoted, I do not think that there is any chance that the OLE will qualify as a confirmatory study. I also think no sober KOL would have actually told him that. The main purpose of an OLE is to gain additional long-term data for safety . Efficacy data from OLEs can sometimes be used to help with marketing (always nice to show a medication is still working) but I have never seen it used for regulatory approval. In this case, the fact that 40% of the patients dropped out the first year, I doubt there would be more than 200 patients in the 96-week dataset at the end, not necessarily representative of the 509 who started. And these patients would be divided between 30 mg, 50 mg and placebo followed by 30 to 50 mg cohorts. Additionally, I do not think imaging and biomarkers are part of the OLE. .
As I mentioned over the past year and a half, I do not think there is any chance (well maybe 1% to be consistent) of approval unless an additional study is done for 18 months with a lower dropout rate. AD is extremely common, being a small company is no excuse not to do appropriate studies
Most Phase 3's report 6-10 weeks after last patient last visit. If anything the OLE, being very simple should be on the faster side. However, since OLE's never play much a role in regulatory submissions (and also rarely move SP), there is less onus to get the data out quickly. We are at 8 weeks now (maybe even 12+ weeks at 96 weeks should have been end of April 2024). If Anavex was a typical drug company data should be out, but the average time for Missling seems more like 20 weeks.
I have the Donanemab P3 trial (JAMA) on my desktop so quickly pulled it up to look at dates. The 76 week study at 200 sites completed enrollment November 5 2021 and completed the study in April 2023 (database lock was April 28). They reported topline data just one week later on May 3, 2023 and peer review publication was accepted in June 2023 and published July 17, 2023 --- just 12 weeks after the trial ended. That is an extreme example but shows that all you need for topline data is a couple weeks.
Capital was low back in 2013 when the phase 2a was initiated. Capital is not low now for the A371 schizo trial. If he let loose another 10 million, a 150-180 patient P2 (part B) could be performed. Part A will just be 9-21 patients regardless of Part B trial size
I got out today at 36.50 -- ran too far too fast, I suspect it will be back in the 20's soon. I am happy with my triple. If I miss the boat, so be it.
Blood diagnostics continue to improve but are not quite as sensitive as PET or CSF biomarkers. No single biomarker passes muster at this time but panels are rather good My current go to test is the Labcorp ATN panel (https://www.labcorp.com/providers/neurology/atn/primary-care). It measures amyloid 42/40 ratio, pTau 181 and neurofilament light. Many patients follow a pattern (1) reduced AB42/40 ratio (corresponding to FDA Stage 1 and 2 and beyond) followed by (2) elevated pTau (stage 2 to 3 and beyond) and finally (3) elevated NfL (stage 3/4 and beyond). AB 42/40 decrease and pTau increase correlates with PET amyloid pathology >90% of the time. If AB42/40 only, a PET may be necessary to more definitely diagnose (and always required by insurance if a MAb is being considered). Studies show primary care diagnostic opinion only correlates with PET 60-70% of the time and neurology 70-80% so this easy relatively inexpensive test is easy. Other companies have panels using pTau217 instead of pTau181 and they are likely similar
So preclinically or minimally symptomatic (performing normally on screening tests like MMSE (Stage 1 and 2)), the blood test will often show the AB42/40 reduction and sometimes both AB42/40 ratio reduction and pTau elevation (inspiring more confidence in the diagnosis)
If blarcamesine ever gets approval (which I believe would take another trial) or other oral safe med like simufilam is approved, then a test like this every 2-3 years after age 65 would be a great screen. The tests are getting better and there could be >95% agreement with PET in another 2-3 years.
We will get the 13D/F/G updates from end of Q2 in another week and a half. I suspect State Street will be down +/- 3.3 million (down 4% of shares outstanding) knowing XBI shed that number. Other major holders will just be a little down or up.
I didn't see you earlier messages regarding this as I have you on ignore. Please see my response to ignatius.
Gotta love predictions. Here's a great one from you from last November 17
I believe we will get the Rett data, the Alzheimer’s full dataset, and the peer review article all before the end of the year. I also believe that the data will be so positive that we will also see a wild run during this time frame. How far over and above the high of 31 it goes that is to be seen by the momentum crowd. Maybe we get somewhere between 35 and 40 and then a pull back by the shorts may be back to 20-25 where they will load up And then watch this thing go to triple digits over the next few years. I can dream.
Short interest went down 4 MM shares from mid May to June 28, I had expected more. The volume I had predicted for June 28 (In my post I said June 30 but 28 was the last trading day of second quarter) ended up being split into June 21 (much of this XBI related) and June 28 (Russell related). As predicted by both of us, price did go up once all the re-balancing shenanigans were over.
Good thing we invest in share price and not short interest.
I asked a different way and got the following answer.
Me: The FDA describes stages of dementia in this guidance https://www.fda.gov/media/110903/download. The CDR global is often used in clinical trials. Would CDR 0.5 correspond to Stage 2 or Stage 3
Gemini: Couldn't directly answer your question from the FDA document, but according to a web search the stages align with NIA-AA stages.
In the NIA-AA system, a CDR of 0.5 corresponds to stage 3: Mild Cognitive Impairment (MCI) due to Alzheimer's disease.
Let me know if you have other questions!