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

« Hope Darn Well Springs Eternal | Main | How Not to Do It: Liquid Oxygen Cylinders »

March 9, 2010

A GSK/Sirtris Wrap-Up

Email This Entry

Posted by Derek

Nature Biotechnology weighs in on the GSK/Sirtris controversy. They have a lot of good information, and I'm not just saying that because someone there has clearly read over the comments that have showed up to my posts on the subject. The short form:

The controversy over Sirtris drugs reached a tipping point in January with a publication by Pfizer researchers led by Kay Ahn showing that resveratrol activates SIRT1 only when linked to a fluorophore. Although Ahn declined to be interviewed by Nature Biotechnology, a statement issued by Pfizer says the group's findings “call into question the mechanism of action of resveratrol and other reported activators of the SIRT1 enzyme.”

Most experts, however, say it's too soon to write off Sirtris' compounds altogether, assuming they're clinically useful by mechanisms that don't involve sirtuin binding. And for its part, GSK won't concede that Sirtris' small molecules don't bind the targets. In an e-mailed statement, Ad Rawcliffe, head of GSK's WorldWide Business Development group, says, “There is nothing that has happened to date, including the publication [by Pfizer,] that suggests otherwise.”

We'll see if GSK and Sirtris have some more publications ready to silence their detractors. But what will really do that, and what we'll all have to wait for, are clinical results.

Comments (23) + TrackBacks (0) | Category: Aging and Lifespan | Biological News


1. Hap on March 9, 2010 10:46 AM writes...

It sounds like some people at GSK have severe recto-cranial inversions. (Just because you don't like data doesn't mean it doesn't exist.) The lack of binding of compounds to sirtuin (or their need for fluorescent tags to do so) might not be relevant, if binding of their compounds to sirtuin isn't their mode of action, but to simply say that nothing is happening without some sort of substantial data seems...nuts, unless, of course, you think your stockholders are stupid and explaining why the studies are wrong or irrelevant won't help, in which case, I guess we're all screwed.

I guess being fools must be better than telling stockholders that you got sucker-punched for three-quarters of a billion dollars, even though you're probably going to have to tell them anyway.

Permalink to Comment

2. E on March 9, 2010 11:05 AM writes...

You know this is more about business for GSK than about science when the "WorldWide Business Development" guy is chosen to back their scientific findings.

Permalink to Comment

3. SP on March 9, 2010 11:54 AM writes...

Wait, is the question activation or binding? Surely someone did a biophysical assay at some point?

Permalink to Comment

4. DLIB on March 9, 2010 12:08 PM writes...

Yes. Pfizer used calorimetry. It doesn't bind without the fluorophore

Permalink to Comment

5. Vince on March 9, 2010 4:03 PM writes...

To play devil's advocate, I'm not quite sure if that is necessary in-vivo. My understanding is that Salermide acts in a SIRT1-dependent manner, yet is p53-independent -- which is what Pfizer's metric was, but I could be mistaken as I haven't read it since it came out.

I was happy to see I wasn't the only one who thought the in-vivo work was somewhat weak (Link). The experimental design and results just don't sit right with me, nor do they align with what I've seen in an ischemic-stroke model.

That being said, who knows anymore.

Permalink to Comment

6. OE on March 9, 2010 4:03 PM writes...

Moncef (head of R&D) and Patrick Vallance (head of Medicine dev) DID know. In the fall before the Sirtris purchase 4 senior scientists went down to Florida where GSK was having a worldwide VP's meeting and the presented all the the due diligence that 3 different CEDD units had done. No one believed that the assay was real and had doubts about the experimental design and the targets themselves. The scientists were thanked their efforts and assumed that they had help prevent a bad purchase. Wrong... Just wait... there are more of these coming....

Permalink to Comment

7. sirt1fan on March 9, 2010 5:44 PM writes...

The problem with the Pfizer data is that there's been recent reports about HIF-2 activation driving Sirt-1 levels low with consequent compromise of mitochondria. That's not Pfizer, GSK or Sitris Pharm data, it's from some academicians. No statement Sirt-1 binding anything, either. And the logic behind compromise of mitochondrial integrity in the face of decreased oxygen levels does make sense. It's not something fanciful that requires some major insight to understand.

Maybe, just maybe, GSK got this one right and Pfizer wrong? Actually, isn't Pfizer the company who demonstrated that raising HDLs raised the risk of death? All the other data suggests otherwise, but since it's Pfizer, it must be more credible, right?


Permalink to Comment

8. Hap on March 9, 2010 6:10 PM writes...

1) Pfizer wasn't the only one to come up with the data suggesting the compound of note don't bind to sirtuins - Amgen also had similar results. They could both be wrong, but not based on flaws with Pfizer, unless they're coordinating attacks, which is unlikely.

2) If there are so many reasons why Pfizer's data would be wrong, why would GSK not assail their research on those bases? Having a PR guy claim Pfizer's research is of no consequence or having Sirtrin's leader claim that Pfizer must not have known if their compounds were pure doesn't support the idea that they have substantial data contradicting Pfizer and Amgen's contentions. Telling people not to look at the data behind the curtain (the "tempest in a teapot" comment) doesn't help to gain anyone's trust, either.

It will all come down to having drugs in clinic based on the theory, which is hard enough anyway. The question whether GSK needed $720M to start looking for them will probably be open even then.

Permalink to Comment

9. sirt1fan on March 10, 2010 2:25 AM writes...

Yes, Amgen has looked at sirts, and they also looked at HIF. In both programs, they killed every lab animal exposed. I don't think that means much except that Amgen has some very toxic compounds. Nothing more than that. And as I recall it, it's not a question of what a drug company say, it's what it publishes. That's replication, and so far, there's been no replication.

Permalink to Comment

10. sgcox on March 10, 2010 5:04 AM writes...

sirt1fan - no question that sirtuins are interesting and important proteins which have potentialto become nice drug targets. The problem is Sirtris and GSK topbrass. Sirtris "activators" are screening artifacts. Period. Pfizer, Amgen and couple of academics before them showed it conclusively and very convincingly for people like me (outsider to Sirt1 biology but know a bit about biochemistry). Insisting on opposite simply makes me think Sirtris "fans" are snake oil merchants.

Permalink to Comment

11. Petros on March 10, 2010 9:38 AM writes...

OE's comment (#6) suggests that the internal GSK's politics were worse than had previously been suggested.

Might such points lead to sticky questions of GSK's managment at its AGM?

Permalink to Comment

12. Virgil on March 10, 2010 10:16 AM writes...

@Vince (#5) - care to elaborate on the ischemic stroke results?

On a different topic... one thing I think which is driving a lot of these problems is the free AVAILABILITY of these compounds for academic research.

We (an academic group in a US University) approached Sirtris a couple years ago for some SRT1720 and related molecules, and were told they couldn't give it out due to the GSK buy-out. However, we were told "go ahead and synthesize it yourself for non-commercial use, the method is published". We did so, and got a bunch of negative data (it doesn't do what it should - don't ask me to elaborate, publications are in the pipeline).

The problem is, how do we know if we got negative data because it really doesn't act on SIRTs, or because our synthesis was incorrect? These are very complex molecules with multiple-step syntheses (IIRC SRT1720 is 7 independent steps - outsource it to Eastern Europe and you're talking >$5k for a gram). The potential for contamination and incorrect assignment of NMR shifts is huge. We may just have shoddy material. I suspect the case may be similar for Pfizer - maybe they just didn't make it correctly? Or maybe even Sirtris screwed up the synthesis, and the molecules they report on and give structures for are not actually what's in the bottle? Who knows? Maybe the published method is missing a "voodoo" step or critical detail?

Bottom line - the only way to be sure is to get the stuff from Sirtris itself, but we've since approached them to ask for the "genuine article", and they don't even respond to emails. Thus, Sirtris currently holds all the cards - they retain the ability to claim that others can't repeat their data because maybe they didn't make the drug properly, all the while not giving out their "magic prep" because apparently they don't need to because the synthetic method is published.

Now if only my paltry academic research budget could afford a really really good synthetic organic chemist!

Permalink to Comment

13. SALESnotSCIENCE on March 10, 2010 1:17 PM writes...

To echo #6's comments above:

BD decisions at GSK are driven by ex-sales guys who are more interested in building their resume "deal-sheet" for their next job, rather than buying assets that actually have a shot at succeeding. Many of them have barely a science education, but exert final "buy/no buy" authority over the scientists. Couple of years ago, they sunk a huge chunk of cash over an "oral insulin"...a whole bunch of scientists pointed out a potentially serious toxicity issue, but were overruled by a bunch of ex-Sales guys who were more interested in "making the deal" rather than buying a viable asset. Sure enough, within a month of the deal, severe toxicity issues were unearthed (all pointed out by the scientists beforehand) and the program was terminated. The idiots who ok'd the deal are still in positions of power...Moncef (!) was the head of BD's amazing how badly this company has lost its way...and as for Westphal (from Sirtis) his archived NYTimes profile..the guy was ex-investment banking and surely knows how to dress up a pig to sell to gullible customers..."60 Minutes" segment et al!

Permalink to Comment

14. analyzeit on March 10, 2010 6:36 PM writes...

Hey Virgil, just hire yourself a decent chemist, or at least talk to one.

One can easily send a sample to an NMR lab here in the US and ask them to verify the structure.

Permalink to Comment

15. rtprip on March 10, 2010 9:43 PM writes...

Ad Rawcliffe - can someone tell me what his scientific credentials are that his statement should be believed? What has he ever done other than pick up Moncef's underwear from the copulate jet?

Permalink to Comment

16. NotABOB on March 11, 2010 5:59 AM writes...

Well said "rtprip", you've obviously worked at GSK and understand the culture there.

Permalink to Comment

17. ABC on March 11, 2010 6:33 AM writes...

Their R&D China site has been trying to reproduce Sirtris results for the last two years. You know they would show it if you had it. but ...

Permalink to Comment

18. Anonymous on March 11, 2010 8:04 AM writes...

I saw David Sinclair give a talk recently, where he spoke for the last 5 minutes about the Sirtris compounds. He framed the story as "everyone agrees that the compounds activate Sirt1" but there is just a "debate in the literature" about which Sirt1 assay is the most relevant. (i.e., "teach the debate")

It was so dishonest. He didn't show any data refuting the specific evidence that the compounds are junk, or even acknowledge in a serious way the substance of the criticisms. He also, bizarrely, portrayed himself as the victim ("why does everyone associate me with these compounds? I do so many other things!" and also "one word of advice: never let a reporter in your lab") Of course, Sinclair is OK with being the sirtuin guy when there is a glowing portrait in the NY Times at stake, or when millions of dollars from GSK are on the line. But now that some people are questioning the science, don't blame him! It was really pathetic.

Permalink to Comment

19. processchemist on March 11, 2010 10:37 AM writes...

Bad chemistry, surely awful small molecule candidates? Maybe chemistry MATTERS?

Permalink to Comment

20. Hap on March 11, 2010 12:12 PM writes...

#9: True, but no one's replicated Sirtris's results either - the Pfizer, Amgen, and other papers are all about the inability to replicate Sirtris's original research. The lack of replication of studies is a two-edged sword - if replication is a problem with Pfizer and Amgen's work (you don't know that it can't be replicated because the replications studies haven't been reproduced themselves), it's also a problem with Sirtris's, because Sirtris's work is what couldn't be replicated (and hasn't been by others).

Permalink to Comment

21. Schwantzkopf on March 11, 2010 1:08 PM writes...

If this Sirtis deal turns out to be the crock of crap those in the know think it is, then who's going to take the rap? Which of that pair of wise guys, Slaoui or Vallance or both, is going to get his ass kicked? A lot of poor schmucks have been buckreamed by these assholes but, smile, smile, it might be payback time. Let's hope some of the Brits give them hell at the AGM because the stock holders will be mad as hell when the full story gets out!

Permalink to Comment

22. RenegadeSci on March 12, 2010 7:02 PM writes...

I love how the academic angle is glossed over in your articles.

A big paper came out which was a mouse study.

AMPK-deficient mice are resistant to the metabolic effects of resveratrol;db09-0482v1

Their is also the breakdown in the brain.

Resveratrol is neuroprotective because it is not a direct activator of Sirt1—A hypothesis

Bor Luen Tang

The hypothesis paper lays it out.

Sinclair "solved" this problem in two steps.
1. Screen (questionable, and never random suspects picked)
2. Microarray data, from what I've found. It does not give gene order. If he had neural tissue data, I'd like to see it. I be it'll show that SIRT1 isn't activated.

Microarray data does not give gene order. The screen may not give good data. The NIH paper gives evidence that Sinclair has his gene order wrong. That's science, but he does do other things. The resveratrol studies in mice were great, but the knockout mice take years to make. SRT501 is working well. They'll be fine unless someone sharks them.

Permalink to Comment

23. GSK in a pickle on March 18, 2010 1:49 PM writes...

To #21: it is crap; just a question of how long they will allow the costly game to go on and how GSK try to position themselves toward admitting the mistake.

Interestingly, looks as if he head of R&D is getting set-up to be gracefully pushed aside from the small-molecule world back to proteins, vaccines from whence he emerged. Could this be the first step in retribution for his support of the Syrtris deal?

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry