About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Department of Placebo Effects | Main | Physics, for Dogs and Others »

January 15, 2010

Sirtuin Scenarios

Email This Entry

Posted by Derek

So, after reading what Pfizer has to say about Sirtris (and by extension, about GlaxoSmithKline's heavy investment in them), let's go over the possibilities. What happened, and what's going on?

We'll start out with the first branch point: either Pfizer (and Amgen) are right that there's trouble with the Sirtris assays and compounds (Reality A, I'll call it), or they're wrong (Reality B). For the rest of this piece, I'm going to assume that they're right, because I think that this is almost certainly the case. At least two separate groups of competent investigators have reported trouble, and that's good enough for me. (We'll discuss the implications of that in a bit).

Now we come to the second branch point: either Glaxo did enough due diligence to be aware of the problems (scenario A1) or they didn't realize them at the time of the deal (scenario A2). If A1 is the case, then we'd have to assume that the most likely consequence (A1a) is that Sirtris had other non-public assets that did check out, and that GSK's management felt that these justified the purchase. (A1b would be the scenario where GSK was well aware of the Sirtris problems, knew also that they didn't have anything else to offer, and bought them anyway, which doesn't make sense). These assets could have been other compounds, and/or a leg up on the complicated biology of this field. The difficulty with that line of thinking is that having found the fundamental assay problems with the Sirtris work, the GSK people would surely have been much more cautious about drawing sweeping conclusions about the rest of the company's intellectual property.

If A2 is the case, then we're looking at sheer fecklessness on the part of GSK's upper management. I'd like to be able to rule this out, but there have been other deals in the history of this industry that make that hard to do. I have witnessed at least one such personally. One problem is that these deals tend to be initiated near the highest levels of a company, and these people are not always the most technically savvy (or up-to-date) members of an organization. Even with a science background, the CEO of a large company does not have the time to be a scientist. (I'm reminded of Peter O'Toole's character in My Favorite Year: "I'm not an actor - I'm a movie star!"

Overall, though, I find it hard to believe that no one would have noticed the reported problems at all, which leads me to favor what I'll call scenario A3: the problems with the Sirtris assays may well have been known/realized at the lower scientific levels of GSK's organization, but these concerns may not have made it to the top in a sufficiently timely or vigorous manner. The deal would have gone through under its own momentum, then, in a flurry of last-minute misgivings which would have been hard to distinguish from the usual butterflies that accompany any large transaction or the preliminary stirrings of buyer's remorse. The sorts of reasons advanced in the A1 paragraph above would have been used to justify pushing ahead. With that in mind, this scenario could be broken down further into A3a, where Sirtris also had some other assets that the rest of us haven't seen, and A3b, where they didn't. I think that A3a is more likely, since that would have provided some of the momentum to get the deal done regardless. A3b is basically A2 with different timing and slightly less cluelessness.

So where do things go from here? That obviously depends on which of those three realities obtains. If A1 (specifically A1a) is the case, then GSK plows ahead with their secret Sirtris assets and compounds, and good luck to all concerned. It's worth keeping in mind that sirtuins are quite interesting and important, and that it's an area worth investigating on its own merits. (Pfizer and Amgen, among others, must think so too; that's the only reason that they would have been trying to replicate the Sirtris work).

If A2 is the real story, well, I'm very sorry to hear it. A lot of people seem ready to believe this one, partly because of anger over the layoffs the company has been going through. The most likely consequence of A2 is that $720 million dollars disappears, never to yield anything that's of use to anyone, so I hope that this isn't what happened.

And if, as I think, A3 is what actually happened, then that sort of depends on whether we're looking at A3a or A3b. If the former, then Glaxo overpaid, but has a fighting chance to redeem itself. If the latter, then Glaxo not only overpaid, but (as with A2) is in danger of losing its whole investment as well. We'll all find out.

But we may not find out very quickly. GSK has (like many other companies) a tendency to be rather close-mouthed about the progress of some of its research. When I worked in the nuclear receptor field, we all were very interested in the fate of a particular Glaxo compound, the first selective PPAR-delta ligand to go into the clinic. The company had talked about some animal and preclinical data, but we knew that they were taking it into humans (after all, it was listed that way in their pipeline updates). But it stayed listed like that. . .and stayed. . .and stayed. . .until, as the months and years passed, it became obvious to even the most optimistic observer that the compound's development was (at the very least) extremely complicated, and (more likely) had actually quietly ceased a good while before, albeit with no change in its public status.

In this case, now that these doubts have come up, GSK has a real interest in pointing out any success it may have. If its sirtuin compounds go into the clinic and just sort of hang there, that will probably be an even worse sign than usual. And if no sirtuin compounds even go into the clinic at all, well, the question has answered itself. I hope that's not what happens.

Comments (61) + TrackBacks (0) | Category: Aging and Lifespan | Clinical Trials | Diabetes and Obesity | Drug Development


1. UK Chemist on January 15, 2010 9:21 AM writes...

I'm sure many of us who have been on due diligence activities have seen examples where the team recommend not to do the deal and were over-ruled by senior managment.I wonder if anyone has examples where the team were wrong and senior managment were right.I certainly dont' have any.

Permalink to Comment

2. Ed on January 15, 2010 9:27 AM writes...

If compounds are listed as being in the pipeline, surely NYSE/SEC or LSE/FSA listing rules must stipulate that information provided to investors must be accurate, and thus compounds that aren't being actively developed must be shown to be dropped?

Permalink to Comment

3. Doug on January 15, 2010 9:29 AM writes...

Hi Derek, my apologies in advance. My comment is not related to Sirtuin, but about recent layoffs in the industry. At least one of the scientist in Wyeth (Pearl River) committed suicide, and another one in Groton tried to kill his wife. It has been brutal for the scientists and researchers who were let go. And it seems (correct me if wrong) the scientists and researchers have no saying at all about their own fate in a company. My question is why there are no unions in our industry. Are there any historical reasons? I am not personally agreeing with everything a union represents. But the scientists/researchers need to have a voice in the management of the company. The massive layoffs are partly the results of years' miss management and over expansion in the big pharms.

Permalink to Comment

4. Anonymous on January 15, 2010 9:32 AM writes...

Well, SRT2104 is in clinical trials

Permalink to Comment

5. Tok on January 15, 2010 9:37 AM writes...


Sorry but unions don't work for research. The union's most powerful tool is a strike. It would take about 10 years for a industry wide strike to have enough effect to warrant attention. Every one of those 10 years would bring record breaking profits for these companies since they're not paying for R&D. By that time, everything will be in China and India.

Permalink to Comment

6. Jose on January 15, 2010 9:47 AM writes...

"and thus compounds that aren't being actively developed must be shown to be dropped?"

I think the devil lies in the detail of how exactly one defines "actively" developed in this case. Sure, a janitor could be sifting through boxes for "storage" or, one associate biologist is still running some assays in a quiet corner, or some poor bloke in formulations is playing with Tween 80... or.... yup, that's active!

Permalink to Comment

7. Chemjobber on January 15, 2010 9:59 AM writes...

"At least one of the scientist in Wyeth (Pearl River) committed suicide, and another one in Groton tried to kill his wife."

Link, please?

Permalink to Comment

8. emjeff on January 15, 2010 10:03 AM writes...

Unions worked real well for the auto industry. It is astounding to me that anyone in this day and age thinks that the reply to global competition is to create a class of entitled workers with no accountability.

Permalink to Comment

9. darwin on January 15, 2010 10:12 AM writes...


Permalink to Comment

10. The Pharmacoepidemiologist on January 15, 2010 10:19 AM writes...

The sirtuins have lots of intracellular actions. One such action concerns the regulation of mitochondrial activity; potentially, a signal might be given for mitochondrial inactivation or, perhaps, disintegration. If this happens in the liver, heart or brain, then death of the subject comes into play. Bottom line: the biological effects of sirtuins may be quite different in lab or clinical setting, as compared to a post-marketing approval one. The latter might be fraught with noxious effects missed during clinical development. At that point, no matter what GSK paid, it will have been too much.

Permalink to Comment

11. Hap on January 15, 2010 10:20 AM writes...

5) Other than the fact that (for a variety of reasons) lots of chemists don't necessarily want unions to work, I am sort of surprised that unions managed to work at all. My first thought is that a lot of the cases where unions were effective depended on having conditions that localized the workforce (mining, for example - you can ship people in but it's probably hard and you have to run a gauntlet), which doesn't exist for anything relying on information rather than physical materials.

If companies want to ship drug discovery overseas, well, they'll do it anyway, though it seems like an extended suicide note for everyone other than the people in charge who'll walk away with lots of cash before Chinese and Indian drug companies eat their companies' breakfasts, lunches, and dinners. If there aren't any jobs in manufacturing, and no jobs in info or technical skill, where will the money coming from to pay service employees, and who will actually have the money to buy anyone's products? It'll be harder as well to convince people to gain technical skills if it seems like a long road to penury, and then I wonder where the economy will manage to go (other than "into a deep hole").

It just seems like too many people are indicating that GSK had enough info to not buy Sirtris and ignored it, anyway. Any drug candidates coming from Sirtris would be suspect - if their core technology was so flawed, then it would be hard to trust that the rest of their drug discovery was effective enough to deliver a high-value drug candidate. Torcetrapib should give one an idea what can happen to even the best drug candidates, anyway (the ones that everyone thinks are lead-pipe cinches) - buying candidates you don't know if you can trust at all seems like a really expensive lottery ticket. If you're depending on luck without any knowledge at all, well, you're probably hosed.

Permalink to Comment

13. Mike G on January 15, 2010 10:26 AM writes...

What about a scenario (a variation on A2) in which GSK, having little natural products expertise, relied heavily on Sirtris' data package and/or research methods (assay formats, etc). After all, Sirtris is the expert in reservatrol pharmacology, so it makes sense that GSK might defer to them at least for mechanistic due diligence. I'm wondering if this sort of over-reliance on licensee data is becoming a common practice in areas where pathways and targets are unprecedented?

Permalink to Comment

14. Hap on January 15, 2010 10:27 AM writes...

Unions worked real well for the auto industry. It is astounding to me that anyone in this day and age thinks that the reply to global competition is to create a class of entitled workers with no accountability.

Um, what would CEO's be, at this point? It's not exactly an endorsement of the idea of unions, but it sure looks like the current situation. Apparently, someone didn't mind the idea of creating an entitled class of workers at stockholders' expense with no accountability as a way to make companies "globally competitive".

Permalink to Comment

15. Annonymous on January 15, 2010 10:31 AM writes...

Chemjobber - here is one of the incidents

Permalink to Comment

16. FMC on January 15, 2010 10:33 AM writes...

@ 8, no I am not a liberal. However, what we are appearing to create is a class of companies with no accountability. This is no whining, but I think some checks and balances would be good.

Permalink to Comment

17. alig on January 15, 2010 10:39 AM writes...

It is A2. While I was not in the room when the due diligence was discussed with upper management, I know some of the people who did the due diligence. Within days of the deal being made, they were flabbergasted that they had been completely ignored by upper management. GSK had an internal SIRT1 program before the deal, but all the hits turned out to be artifacts of the assay. So the scientists at the bench knew to look for this.

Permalink to Comment

18. Chemjobber on January 15, 2010 11:10 AM writes...

Um, wow. That's terrible.

Permalink to Comment

19. ex GSK on January 15, 2010 11:11 AM writes...

alig is completely correct.

Permalink to Comment

20. Hap on January 15, 2010 11:29 AM writes...

Do you mean A1b (GSK knew of the problems, at least in general with sirtuin assays, and ignored that possibility with Sirtris), or have I misunderstood this?

It's not really fair to say that all CEOs form a privileged class, but there seems to be an awful lot of incentives for some that make at least a subset of them appear to be a privileged class.

Permalink to Comment

21. darwin on January 15, 2010 11:38 AM writes...

Ward-stop being so hard on the Beaver.

Permalink to Comment

22. retread on January 15, 2010 11:56 AM writes...

"I'll call scenario A3: the problems with the Sirtris assays may well have been known/realized at the lower scientific levels of GSK's organization, but these concerns may not have made it to the top in a sufficiently timely or vigorous manner."

This may just be typical of large organizations. Here's a blast from the past based on personal experience. In 1968 the USA had half a million men in Vietnam. The Army needed lots of docs to take care of them and their motto was "If you can practice medicine outside the army you can practice it inside the army". There was no 4F for docs nor were there medical excuses. This meant that all newly minted MDs would spend two years during or after residency training in the service. Fortunately (for me) the Army was short of neurologists in 1968, so with just one year of residency (instead of the usual 3) under my belt I was sent to one of their best hospitals (Fitzsimons) to work under an excellent and seasoned neurologist (Col. Halbert Herman Schwamb -- whose name alone scared the hell out of me).

The tour of duty in Vietnam was 1 year for everyone, so docs who had been there for their first of two years got their pick of where to go for their final year. Naturally, Fitzsimons was one of their top picks, so the place was full of them.

The army had something called the 'body count' which meant the number of Viet Cong (and possibly civilian) bodies they could find. It gave a number, which was increasing with each passing month. It showed we were winning. However not one of the returning 2 year docs I talked to (and I talked to a lot of them) thought we were winning. Most thought we were losing, and badly. They were of course right.

Since most of the 2 year docs were not career military (like myself) they didn't send anything up the chain of command. I can tell you that it would have been career destroying for a West Point grad coming back as a very junior officer to say this to the higher ups (who were either blind or stupid e.g. Westmoreland). Hopefully this is not the way it was at GSK.

Permalink to Comment

23. okemist on January 15, 2010 12:08 PM writes...

First off my heart goes out to all who have experienced layoffs in recent months, especially those who are suffering emotionally.
Next there are several ways to profit from aquiring new technology, Dereks past employer did a $500M 5 yr deal that got 0.0 compounds developed, but reaped a large profit in equity. There are other ways to profit beside producing a blockbuster api.
And C, all you anti-union, think you are managers because you have PhD's wouldn't be eliligble to be in a union. I have felt for ages that associate scientists should have union protections with far too many reasons than should be listed here.

Permalink to Comment

24. Phoenix70 on January 15, 2010 12:47 PM writes...

@ retread

I have a little different take or "tangent" on scenario A3. I *was* one of those West Point grads ('67) in 'nam in 1970. I worked for an organization called the Phoenix Program. A required report called the HES report, which basically stated what regions were controlled by the good guys vs the bad guys, had to be periodically submitted up the chain of command starting from the grunts on the ground. Regardless of what you envisioned, these reports were factual when generated (and not career destroying - at least to me