Skip to Content

Vanderbilt Heads Into the Clinic for Alzheimer’s

There’s an interesting experiment underway at Vanderbilt, where a team from the university is taking a possible Alzheimer’s candidate into clinical trials on their own. Now, you could come up with several headlines for this, along the lines of “Academic researchers in way over their heads”, or “Plucky professors make end run around drug industry”, but neither of those are true.

That’s partly because many of the professors involved (such as Jeff Conn, ex-Merck neuroscience) have drug industry experience themselves, and they know exactly what they’re getting into. This is Vanderbilt’s Center for Neuroscience Drug Discovery, and they previously were working with AstraZeneca on M4 muscarinic compounds. That effort got derailed, though, as AZ got out of the neuroscience field entirely, and another collaboration between Vandy and Bristol-Myers Squibb ran into similar troubles. So this time, they’re going on by themselves for as long as possible.

But the other reason that this has a chance of working out is that they realize that “as long as possible” means “all the way through Phase I, we hope”. As the head of medicinal chemistry at the center put it to Endpoints, “Doing an IND on your own is quite an undertaking, including tasks largely unknown in academic circles“. That’s for sure – there’s a lot of toxicology, formulations, and manufacturing work that has to go into an IND filing, and hardly any academic discovery effort is equipped to get through them. (My bet is that a fair amount of this was contracted out by the Vanderbilt center, just as it is by smaller companies in the industry. Keeping a bunch of GLP/GMP facilities going to where they can withstand FDA scrutiny is not a cheap undertaking).

This may well be the first time an academic center has gone into Phase I like this completely without drug industry collaboration (C&E News thinks so). Does anyone else have any other examples? It’s rare enough even for the clinical compound itself to come from academia. (For those interesting in trying, I recommend this book). Even so, Vanderbilt has no intention of going all the way through the clinic. This is Alzheimer’s, after all, and the later clinical trials would take far more money than any university would (or in many cases, even could) commit. Just to get this compound into Phase I took financing from several foundations. So no, you’re not going to see Vandy-branded drugs down at the pharmacy.

The idea here, just as with a small company, is to take the drug candidate into humans and “de-risk” it to the point that a larger outfit might want to step in and do a deal. The compound itself is a selective allosteric modulator of the M1 muscarinic receptor, and that certainly takes me back. During my first years in the industry in the early 1990s, I spent a lot of time on muscarinic approaches to Alzheimer’s, and the Vanderbilt team will be the first to tell you that it’s not a new field. Previous attempts to target M1 for the disease have run into tox problems – a full agonist at the receptor is probably not the answer, and semi-selective agonists such as xanomeline have had trouble as well (although that approach is still alive). The hope is that the allosteric mechanism will give a cleaner response, but as I recall, even that has been tried before (although this does seem to be a better compound, pharmacologically, than previous attempts).

What could a drug like this do for Alzheimer’s? That’s a good question. All the muscarinic approaches are a rear-guard action, from what I know. You would not expect the underlying deterioration of Alzheimer’s to be affected, but you could improve the function of the cholinergic neurons that remain, buy some time, and improve symptoms and behavior. If I’ve mischaracterized these mechanisms, I’m sure that someone will let me know, but that’s how I’ve always seen them. So this is not any kind of Alzheimer’s cure, but if it works, it could be useful.

First, though, the compound has to get through Phase I. That stage of clinical trials is mainly to check blood levels and pharmacokinetics, to make sure that the appropriate doses have been working out for the actual disease patients in Phase II, but it’s also there to give you a read on gross toxicity. And for an M1-targeted compound, that’s the key. The sorts of side effects (gastrointestinal, cardiovascular, et al) that have derailed previous muscarinic efforts can be monitored, and the Vanderbilt team should know if they really have made a cleaner compound. At that point, they can make a case for an actual drug company to step in. Will anyone? That’s a good question, too, because it takes a lot of money and a lot of nerve to get into this field. Many of the big players in it have been going all the way to what they’re hoping will be disease-modifying therapies, rather than treating symptoms – after all, it’s going to be a long, expensive trial no matter what, and the chances of getting the symptoms right aren’t much better than the chances of hitting the underlying disease mechanism – whatever that is. So it’s no sure thing that even a successful Phase I trial would lead to a deal, although it certainly still could.

I wish the team good luck, of course. It’s a real accomplishment to get this far, and doing it from academia is harder still. I look forward to seeing how the Phase I data shake out, and how the program fares from there.

54 comments on “Vanderbilt Heads Into the Clinic for Alzheimer’s”

  1. Calvin says:

    Eh, not the first academic centre to go through a P1 without industry help. Of the top of my head, Sir Steve Bloom at Imperial took a peptide for obesity through a Phase 1 and a Phase 1b-ish. Developed the compound himself. And the ICR in London have also taken compounds developed in house into the clinic. But I agree that it’s rare and academics typically aren’t tooled up for that type of exercise.

    1. In Vivo Veritas says:

      The Bloom work with PYY3-36 would barely qualify as a clinical trial here in the US, & I’m not even sure it would pass muster in the UK today. No snarkiness intended, but Bloom shopped that molecule around to many a big pharma. I’m not in the field anymore, but my last recollection was that he had no takers. Is that still true?

      1. Calvin says:

        Yeah, you are pretty much correct I believe. But he did discover the peptide, optimize it and take it through 2 Phase 1 studies. So while it might not work (I’m not sure), he is an academic who took a compound from discovery through Phase 1 (and complete both Phase 1s. No terminations). I didn’t say it was any good…..

  2. anon says:


  3. RTW says:

    Carboplatin from Michigan State University with the help of ICR in London. Later sold the rights to BMS.

  4. Anon says:

    “Now, you could come up with several headlines for this…”

    BBC: Vanderbilt scientists set to develop breakthrough cure for Alzheimer’s?

    1. SedatedFMS says:

      Is there a British link?

  5. Anchor says:

    Con act by Jeff Conn! Anything goes for AD as the media is very hungry for any treatment including dirt. Good luck to all.

  6. Isodore says:

    “[T]he later clinical trials would take far more money than any university would (or in many cases, even could) commit.”

    To “in many cases, even could” I would add “or even should”. This arguably goes outside the scope of university research and the mission of an institution of higher learning. By expending funds in such directions would surely shortchange students’ and post-docs’ education and training by taking away money from classroom, lab and infrastructure maintenance and updating. Just consider how many more grad students and post-docs would money spent on Phase I, never mind later phase, trials might fund.

  7. qetzal says:

    Interesting to note that academic trials with novel small molecule drugs are so uncommon, while academic trials with things like gene and cell therapy are quite common.

  8. Cameron Beaudreault says:

    Bruce Silverman coming up with Pregabalin at Northwestern is certainly the only analogous case I’ve ever heard of. Then again, I know very little about drug discovery in general – glad to see Vanderbilt taking the bold step.

    1. anon the II says:

      I think he goes by Richard and not Bruce, which is his middle name. At least that was my impression.

    2. Xiao Bu says:

      Silverman collaborated with Parke-Davis (a subsidiary of Pfizer) during preclinical work (ref linked in my name). Not to say it wasn’t a great invention, but definitely not an example of academics going into clinical trials. I am curious to know how far Arun Ghosh took darunavir before partnering with pharma, but ran out of time to do the research.

  9. Druid says:

    They are doing exactly the right thing. There are many ideas for tackling diseases in the hands of academic champions, so many that they are not gold mines (as those champions often believe) but almost worthless until some steps forward have been taken (“pig in a poke” springs to mind), and the first steps are preclinical & human safety. Looking at their idea, a lot can be learned in healthy volunteers, and not horribly expensive if it is potent. If enough people do this, and if they all publish their findings, good and bad (it is in their professional interests, unlike industry), someone might have a success.

  10. Hap says:

    Alzheimer’s is so hard, and so much of the risk is late, that it seems like it would be really hard to interest a (big) pharma in this, unless you had really impressive PII results, and PII is still going to cost a lot of money for a school to invest in a clinical trial.

    1. Anon says:

      Agreed, especially for this target. Merck had an M1 PAM in the clinic and yes, one can argue differences in the compound profiles. But, we all know Pharma isn’t always interested in something different if there is some dogma behind the target (i.e., side-effect profile). Also, target engagement isn’t easy with PAMs.

      1. Sulphonamide says:

        “Also target engagement isn’t easy with PAMs”

        Interesting statement – could you perhaps explain that one in a bit more detail please, as I am not clear as to why an allosteric site should be any different to an orthosteric site in terms of achieving the requisite binding potency and kinetics?

        1. Anon says:

          Target engagement with PET ligands isn’t as straightforward with PAMs as with other orthosteric ligands. There are a few (mGlu3 (or 2 from JNJ) and the mGlu5 based on the NAMs), but not as many. Some of the issues are the potency/efficacy are tied to the cooperativity and fold-shift, etc. Most of what I know are looking at other biomarkers for clinical applications (EEG, CSF, etc.) rather than PET ligand approach – which has made some Pharma partners a bit wary.

        2. Anon^2 says:


          I suggest you read some reviews on the “Operational Model”. This model uses a different, clearly more complex mathematical formalism to explain the binding/function disconnect seen with GPCR PAMs.

          Medicinal chemists used to optimizing activity at the target based on a single parameter (eg EC50 or ki) will need to “upgrade” their skills to work with up to 4 parameters that impact allosteric effects: alpha, beta, Kb and TauB.

  11. David Antonini says:

    Seems to me that this sort of thing is not out of scope of academia at all. Maybe in methodology, but the acquisition of new knowledge, testing of hypotheses is exactly what research academia should be doing, no?

    1. road says:

      Sure, but GLP toxicology and CMC are not things that could or should be performed by grad students.

  12. drongo says:

    A group at Stanford took an antiCD47 mAb into Ph I. The clinical study is now being run by a company Forty Seven Inc, but if you look at the archive you’ll see that the study was originally sponsored by Stanford (if you click on my name, you should go to the right page).
    Here’s a paper describing their preclinical work: Liu et al, Pre-Clinical Development of a Humanized Anti-CD47 Antibody with Anti-Cancer Therapeutic Potential, PLoS One. 2015 Sep 21;10(9):e0137345. doi: 10.1371/journal.pone.0137345.

  13. DCRogers says:

    While I agree that later trial stages would be of little value to grad students and postdocs, having them exposed to all of the issues going into a PhI seems quite useful!

    Even if much of the work is outsourced, that’s not uncommon in industry now, and skills at working with collaborators and digesting their results is also useful.

    1. Anon says:

      Unfortunately, very few graduate students or post-docs are part of the academic drug discovery group at Vanderbilt. It’s mostly staff scientists that are part of the group, so the “learning” aspect that could be gained by students/post-docs is not present.

      1. Phil says:

        This is accurate. Vanderbilt’s CNDD is academic with an asterisk. It’s more like a privately funded drug discovery group housed within a university.

        That being said, both Conn and Lindsley have research groups that work with the staff scientists, and it seems to me those students are exposed to plenty of valuable drug discovery knowledge and experience.

        (I am in no way affiliated with Vandy, I have just visited and met with some of those involved in the CNDD)

      2. Phil says:

        Sorry, I should clarify, I meant to quote “mostly staff scientists” as the accurate part. It is inaccurate to say there is no learning on the part of graduate students/post-docs.

        1. Anon says:

          Except it’s not privately funded. It’s very much NIH and donor funded with some private funding – but as the article mentioned, some of the Pharma funding dried up due to decisions by the companies.

          But, the post-doc/student learning is not directly tied to the projects and that seems to be done intentionally – from what has been disclosed at meetings from the VCNDD group. Need to keep the students separate for IP reasons and some post-docs work as non-traditional post-docs and others are traditional academic post-docs. Nearly all discovery work is staff scientists – not to say students don’t learn, but it’s not an integrated student/staff group. I believe they are in completly separate sites, if I remember the last talk correctly.

          1. Phil says:

            Donor/foundation funding is not private? It’s not VC funded – I didn’t mean to imply it was. They have NIH funding, sure.

            But we are splitting hairs. We can agree VCNDD is not a purely academic center. It’s a different category.

          2. Anon says:

            You are correct Phil, sorry. I see private funding and immediately think VC/Pharma because there is control/dilution exerted. And I agree, VCNDD is not a true academic drug discovery group, an autonomous hybrid housed in an academic building.

          3. Phil says:

            No worries, I will keep your (entirely understandable) connotation of private funding as VC in mind in the future.

            I talked to a couple students developing tool compounds specifically for studying muscarinic receptors, this must have been 2014 or so. These students were already well-versed in medicinal chemistry concepts like scaffold hopping and the P-chem of binding (I know, these are just the tip of the iceberg but I’m talking about 2nd or 3rd year grad students).

            I did not realize the extent of separation of the research groups from the staff scientists doing the confidential work. It makes a lot of sense once you point it out, otherwise what would the students be able to publish? Judging from the papers up to this point, students are involved in lead ID but must phase out during lead optimization.

            Still, I don’t worry about the quality of their education in drug discovery.

  14. Lane Simonian says:

    I wish more academics could be involved in phase one clinical trials for Alzheimer’s disease, but this particular approach seems to fall into the category of things we have tried before and have not worked.

    Muscarinic acetylcholine receptors are one of several g protein-coupled receptors damaged by oxidation in Alzheimer’s disease. Other such receptors affect smell, sleep, mood, social recognition and alertness. Trying to reverse the oxidation of muscarinic acetylcholine receptors would be a more feasible strategy than to modulate them.

    Inactivation of the human brain muscarinic acetylcholine receptor by oxidative damage catalyzed by a low molecular weight endogenous inhibitor from Alzheimer’s brain is prevented by pyrophosphate analogs, bioflavonoids and other antioxidants.

    “Natural antioxidants and pyrophosphate analogs may improve the effectiveness of acetylcholinesterase inhibitors and prove useful in the treatment and prevention of Alzheimer’s disease since the muscarinic acetylcholine receptor is required for memory, and decreased cholinergic function is a critical deficit in Alzheimer’s disease.”

    Here is another approach coming out of Vanderbilt that may work better: a curcumin aeresol.

  15. Academic researcher says:

    As the project manager of an academic drug discovery group, we also have a program with an NCE that is in preclinical development. With the generous financial support of a UK philanthropy, we have developed industry-standard criteria and milestones (unusual for academia!) based on our target product profile. We have used and are using CROs and consultants to help in areas where we don’t have the expertise. Our current funding allows us to do formulation work, GLP-toxicology and GMP production up to an IND filing. Our frustration right now is that the NIH doesn’t really have a grant mechanism that supports a Phase I clinical trial to the level typically required to do it well.

    As industry looks to academia to derisk new ideas, I think you will see more academics carry things through to the clinic. Stay tuned…

    1. LiqC says:

      I suppose this is also what the Schultz Research Institute (formerly Scripps) is seeking to do post merger with Calibr.

  16. Diver dude says:

    If you are any kind of clinical pharmacologist at all, you will be designing PD endpoints into this Phase I programme, particularly in this case as the pharmacology is so well known. I’d also be finding ways to assess the brain penetration of the drug after peripheral administration (pharmacoEEG would be favourite). You can then start to model whether you are likely to be able to get meaningful drug levels at the effect site before the AEs kick in too hard and what plasma concentrations you need to be able to achieve to deliver this.

  17. Pformer CNS Guy says:

    When I was head of Drug Discovery at a now defunct biotech, our company delivered a very reasonable novel AD candidate to a big pharma during a 5-year cognition collaboration. The pharma company took it through phase 1, where it performed admirably (strong evidence for target engagement, attractive safety profile, desirable dose range, etc), wherein they balked at conducting an AD phase 2 trial (given the high placebo responses they’d gotten in previous efforts). After exploration for re-purposing for PD, they returned the compound as they exited CNS entirely. The point being, it may not matter what the approach…the time/costs/difficulty of conducting a conclusive phase 2 (and then phase 3) trial in AD may prove too much to justify for even big pharma, even with an compound that looks attractive through phase 1.

    I know a couple of people within the Vanderbilt enterprise, and I admire the whole experiment in academic drug discovery there, so I wish them well. But what’s the commercialization endgame going to look like? I’m with Hap: anymore, without clear phase 2 positive results, how does one to attract big pharma to partner with you on an AD therapeutic? (The question might apply to most indications with cognition endpoints).

    1. hn says:

      Great comments! Do you think AD should be avoided altogether by academic drug discovery?

      1. Pformer CNS Guy says:

        Innovation and basic science at molecular target level is still driven overwhelmingly by academia. To the extent that academic drug discovery demonstrates those targets may be attractive to industry by profiling novel interventions, that works and is being done at a high level by many institutions, funded by both the NIH, foundations, and industry partners. But advancing such projects into the clinic without a commercialization strategy in place seems a poor investment of resources.

  18. Mike says:

    Given the recent disclosures of M1-mediated adverse events by Pfizer, BMS, and Merck, it’s not clear to me how Vanderbilt expects to avoid the same problems. The selectivity issue (versus the other muscarinic receptors) has been solved- all of the classic cholinergic symptoms can be induced by highly selective M1 PAMs. My guess is that this is an attempt to create “window dressing” for the VCNDD; that is, they will be able to point to the fact that they advanced a compound into the clinic when courting potential donors. In this view, the improbability of success is almost irrelevant.

    1. Dr CNS says:

      Allosteric ligands are a different kind of beast. We are learning in real time how to qualify these compounds, and their pharmacological effects. We have a steep learning curve ahead of us.
      Let’s give the Vandy colleagues a bit of credit…

      Congrats to Jeff Conn’s team – and best of luck moving forward!

      1. Mike says:

        Allosteric ligands are a nice way to get selectivity, but the intrinsic safety benefits have been overestimated. See for yourself:

        Unless there is some kind of signalling bias or kinetic profile that Vanderbilt hasn’t disclosed, this looks to be beating a dead horse.

        1. Anon says:

          These may not be apples-to-apples comparisons. The Pfizer compound is a PAM-Agonist (from the title) and the JPET paper notes the Heptares compound hasn’t shown the same issues, and that is an agonist as well. It will be interesting to see if VU saw any issues in the IND safety studies and whether the CNS exposure will be high enough for efficacy, something that Merck had issues with – low B:P in their reported M1 PAMs.

          1. Dr CNS says:

            Really? Selectivity in what assays? In vitro? In what species? Did the functional activity translate to more physiological readouts like electrophysiology? Or mechanistic in vivo tests?
            What was the level if receptor expression in the in vitro assays, which impacts the TauB of the PAM?

            Take a look at the mGluR5 PAM program that VU and JNJ ran together and you will see some of the best CNS medicinal chemistry, which is not a bad way to attack these challenging diseases.

  19. halbax says:

    What about the work that is ongoing to develop psilocybin for cancer anxiety and treatment-resistent depression depression? Much of the work is funded by the Heffer Institute and the Usona Institute and has been taking place at UCLA, Johns Hopkins, and NYU. The work is currently in phase I/II, and they are hoping to start multi-site phase III trials sometime during the next few years.

  20. Walter Moos says:

    Three nonprofits collaborated to take pralatrexate (Folotyn) into Phase 1 without biotech or pharma support: SRI International, Southern Research, and Sloan Kettering.

  21. MoBio says:

    The probability of success beyond Phase I in CNS remains dismal [6.2-9.6% depending on neurology vs psychiatry primary indication; Nature Reviews Drug Discovery 15, 447 (2016) ] so it is a good bet that the compound will not advance. FWIW oncology remains the worst (5.1% beyond Phase I).

    Additionally, given that the cholinergic system degenerates during AD, M1-PAMs will perforce lose effectiveness as the disease progresses. I’d be delighted to be wrong, but cannot see this is a viable option going forward. It was an interesting idea a decade ago, though.

    1. Dr CNS says:


      As the cholinergic system degenerates you will need to get more efficacy to restore function. The right kind of PAM may do just that. Additionally, PAMs can make otherwise inactive catabolites active (like choline at M1r) and therefore enhance pharmacodynamic effects.

      1. MoBio says:

        I guess that’s the hypothesis to test…choline is not much of an agonist at muscarinics even with PAMs

        1. Dr CNS says:

          Agreed –
          So, if an hypothesis that, if supported, may change the way we believe drugs work is not worth testing, what is?

          1. MoBio says:

            If I was a betting person (which I am not) I would not bet on it.

            Also, all hypotheses are not worth testing unless one has unlimited resources. This is one I would not test

          2. Dr CNS says:

            If you are interested in learning about this topic, the VU group has now published some of their work in this area in ACS Chem Neurosci. “Diverse Effects on M1 Signaling and Adverse Effect Liability within a Series of M1 Ago-PAMs”

  22. loupgarous says:

    Interesting experiment – can Big Academia (and Vanderbilt’s well-respected in medical research) out-perform Big Pharma in drug development?

    You’d think so, from all the horror stories here about how decision-making happens in Big Pharma. And outfits like Burroughs-Wellcome existed more to develop new drugs for tropical medicine and other indications than to make stellar profits for decades (until they decided to join the victorious opposition).

    I actually had a part in efforts by CROs (independently of big drug firm sponsors) to get NIH funding for clinical research. I could have had a nice job for me with that CRO doing more of the same, had I not wished to keep my wife and kids on the Gulf Coast, where they were pretty happy. Still kicking myself for that one, because a year later I followed the job market to be a contract analyst away from the Coast- with nowhere near the scope for achievement and success. You live and learn.

    1. Anon says:

      Outperform in drug development? Absolutely not. The Development is where academic groups will need Big Pharma. Outperform in the Discovery? I guess it all depends on what you want to measure – and if you measure correctly, you can get any answer you want. An academic group, IMO, will never outperform any Big Pharma because all that matters is what gets to the patient, and that will nearly always come with Big Pharma help.

    2. hn says:

      It shouldn’t be a competition because academia is not set up to work efficiently, at scale, and in large teams. Where academic drug discovery can succeed is through diversity of ideas, finding niches that have been overlooked by pharma, and partnering with industry and former industry scientists.

  23. Scott says:

    Honestly, I could see a lot of educational purpose in having students run some clinical trials. Probably post-docs, though I’m sure you could also get some teaching utility for nursing students and Med Students.

    The rest of the conversation here is waaaaay over my head, but I’m an HR major, not a chemist or health care professional.

Comments are closed.