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

« The Supreme Court Slams Big Pharma? Not Exactly. | Main | The NIH Goes For the Gusto »

March 28, 2011

Value in Structure?

Email This Entry

Posted by Derek

A friend on the computational/structural side of the business sent along this article from Nature Reviews Drug Discovery. The authors are looking through the Thomson database at drug targets that are the subject of active research in the industry, and comparing the ones that have structural information available to the ones that don't: enzyme targets (with high-resolution structures) and and GPCRs without it. They're trying to to see if structural data is worth enough to show up in the success rates (and thus the valuations) of the resulting projects.

Overall, the Thomson database has over a thousand projects in it from these two groups, a bit over 600 from the structure-enabled enzymes and just under 500 GPCR projects. What they found was that 70% of the projects in the GPCR category were listed as "suspended" or "discontinued", but only 44% of the enzyme projects were so listed. In order to correct for probability of success across different targets, the authors picked ten targets from each group that have led, overall, to similar numbers of launched drugs. Looking at the progress of the two groups, the structure-enabled projects are again lower in the "stopped" categories, with corresponding increases in discovery and the various clinical phases.

You have to go to the supplementary info for the targets themselves, but here they are: for the enzymes, it's DPP-IV, BCR-ABL, HER2 kinase, renin, Factor Xa, HDAC, HIV integrase, JAK2, Hep C protease, and cathepsin K. For the receptor projects, the list is endothelin A receptor, P2Y12, CXCR4, angiogensin II receptor, sphingosine-1-phosphate receptor, NK1, muscarinic M1, vasopressin V2, melatonin receptor, and adenosine A2A.

Looking over these, though, I think that the situation is more complicated than the authors have presented. For example, DPP-IV has good structural information now, but that's not how people got into the area. The cyanopyrrolidine class of inhibitors, which really jump-started the field, were made by analogy to a reported class of prolyl endopeptidase inhibitors (BOMCL 1996, p. 1163). Three years later, the most well-characterized Novartis compound in the series was being studied by classic enzymology techniques, because it still wasn't possible to say just how it was binding. But even more to the point, this is a well-trodden area now. Any DPP-IV project that's going on now is piggybacking not only on structural information, but on an awful lot of known SAR and toxicology.

And look at renin. That's been a target forever, structure or not. And it's safe to say that it wasn't lack of structural information that was holding the area back, nor was it the presence of it that got a compound finally through the clinic. You can say the same things about Factor Xa. The target was validated by naturally occurring peptides, and developed in various series by classical SAR. The X-ray structure of one of the first solid drug candidates in the area (rivaroxaban) bound to its target, came after the compound had been identified and the SAR had been optimized. Factor Xa efforts going on now also are standing on the shoulders of an awful lot of work.

In the case of histone deacetylase, the first launched drug in that category (SAHA, vorinostat) has already been identified before any sort of X-ray structure was available. Overall, that target is an interesting addition to the list, since there are actually a whole series of them, some of which have structural information and some of which don't. The big difficulty in that area is that we don't really know what the various roles of the different isoforms are, and thus how the profiles of different compounds might translate to the clinic, so I wouldn't say that structural data is helping with the rate-determining steps in the field.

On the receptor side, I also wouldn't say that it's lack of structural information that's necessarily holding things back in all of those cases, either. Take muscarinic M1 - muscarinic ligands have been known for a zillion years. That encompasses fairly selective antagonists, and hardly-selective-at-all agonists, so I'm not sure which class the authors intended. If they're talking about antagonists, then there are plenty already known. And if they're talking about agonists, I doubt that even detailed structural information would help, given the size of the native ligand (acetylcholine).

And the vasopressin V2 case is similar to some of the enzyme ones, in that there's already an approved drug in this category (tolvaptan), with several others in the same structural class chasing it. Then you have the adenosine A2A field, where long lists of agonists and antagonists have been found over the years, structure or not. The problem there has been finding a clinical use for them; all sorts of indications have been chased over the years, a problem that structural information would have not helped with in the least.

Now, it's true that there are projects in these categories where structure has helped out quite a bit, and it's also true that detailed GPCR structures would be welcome (and are slowly coming along, for that matter). I'm not denying either of those. But what does strike me is that there are so many confounding variables in this particular comparison, especially among the specific targets that are the subject of the article's featured graphic, that I just don't think that its conclusions follow.

Comments (32) + TrackBacks (0) | Category: Drug Development | Drug Industry History | In Silico


1. CYTIRPS on March 28, 2011 9:33 AM writes...

Structural chemists just found another way to get some self-satisfaction. Maybe they think it would bring some job security. Structural chemistry is simply one of the many tools for drug discovery. It is helpful in some cases but it could also put artificial constraints in some cases. Biomolecules are very flexible. Many drugs were discovered without structural chemistry. CambridgeSoft should publish in NRDD showing ChemDraw increases the probabilities of success in drug discovery.

Permalink to Comment

2. anon the II on March 28, 2011 10:33 AM writes...

Why is it that these kinds of articles are always written by guys from the UK? Is it some kind of right of passage over there?

The truth is that these articles are fair reads for a history of recent drug discovery, but really have no statistical validity. It's like every dot on the graph is an explainable outlier. Which is kind of what Derek just did.

Permalink to Comment

3. 7TM on March 28, 2011 10:44 AM writes...

Because one of the authors is head of chemistry at a company using stabilised GPCRs for structure based drug discovery possibly?

Permalink to Comment

4. Rick on March 28, 2011 10:50 AM writes...

Perhaps there's some semantic validity in the comparison that appeals to stock analysts, but it is so scientifically non-sensical on so many levels (biochemical, physiological, pharmacological, just plain logical to name a few), it's like asking whether vanilla ice cream is stronger than the number 327 or the color blue.

Permalink to Comment

5. Mike on March 28, 2011 10:57 AM writes...

Any sphingosine 1 phosphate 1 receptor projects that have stopped may have less to do with structure than the fact that FTY720 beat competitors to the market, and the other S1P receptors are being investigated now.

Permalink to Comment

6. Ed on March 28, 2011 12:03 PM writes...

These guys are in the business of selling, not science - specifically the Heptares STARS platform and their adenosine A2A program.

So, now they can cite themselves and "prove" the superiority of their platform.

"It has been shown that...."


Permalink to Comment

7. leftscienceawhileago on March 28, 2011 12:27 PM writes...

Might be a good time to take a (critical) look back on the investment numbers in "structural genomics".

Some really great technology (I love being able to drive a beamline from my home internet connection), and a few thousand structures (27% of which have new folds)...but did we really get anything out of it?

Permalink to Comment

8. MedChem on March 28, 2011 1:09 PM writes...

It's all very simple, really. Are all structures helpful in a substantial way? No. Some structures are more (and truly) helpful than others. Therefore the overall success rate is higher when structures are available.

Permalink to Comment

9. Commissar on March 28, 2011 2:16 PM writes...

Comrades, the authors are indeed great patriots! The party leaders are impressed with the use of the paradise called Statistics to mask the truth.
Surely, you realize free markets disappeared long ago when certain financial institutions could not fail or tanks would be in the streets. Now we must cut the compensation for The Workers so that the "cheap" labor can spend 95% of their day sifting through the constant stream of green cheese and baloney like this article.

Permalink to Comment

10. chris on March 28, 2011 2:22 PM writes...

I was involved in a programme for many years that struggled with a particular target, after several years we eventually got a crystal structure of the enzyme and it was apparent that there were major problems. The programme was stopped and the effort moved elsewhere, we felt this was a success for the structural science.
I've also seen instances where structural information has driven exploration of new chemical classes.
However in the grand scheme of things binding to the target is often the easy step, sorting out the off-target activity, ADMET, safety etc. is the hard part.

Permalink to Comment

11. MadDog on March 28, 2011 2:22 PM writes...

I agree with the assertions from Derek and others that some of the purported benefits of a structural information are often late in the program. And the authors don’t' have the data to back up their conclusions. However, this is an intriguing analysis to possibly dig a little deeper. Are there simply more hands working on structure based targets? Are they easier to sell to management and get more funding? To make a case for valuation, you need the numbers on how the programs were resourced. Not too easy to get I assume.

Permalink to Comment

12. drug_hunter on March 28, 2011 3:03 PM writes...

I'd like to see if anyone on this board would be comfortable arguing that they would rather NOT have structural information available for their project.

I have heard such arguments before, but they are relatively uncommon.

To be crystal clear [ho ho] note that this is not the same as contending that having structure has been proven to be useful, or guarantees success, or whatever it is that the authors of the NRDD paper were suggesting.

Any takers?

Permalink to Comment

13. Medicamenta Vera on March 28, 2011 4:01 PM writes...

The Parke-Davis (which became Pfizer Ann Arbor... until the neutron bomb was unleashed) FXa discovery team struggled with bioavailability and half-life for our first lead compound.

A crystal structure was available, and guided work for the two backup compounds that followed. The project team benefited from having the crystal stucture which enabled the development of structurally diverse backups within the short time-frame the development team believed was necessary. The modelers did some nice work.

Permalink to Comment

14. hn on March 28, 2011 4:38 PM writes...

I'm involved with some GPCR structural biology. I really hope this is useful to the chemists, not just academic exercises.

Permalink to Comment

15. CMCguy on March 28, 2011 4:47 PM writes...

More info is generally always better however like use of Metaphors everything must be applied with appropriate precautions or limits. As other have mentioned impacting the target is only one consideration leading to a drug.

In response to #12 drug_hunter I have heard of situations where projects struggled against errors or misinterpretation in structural info so those involved did wish not have had such in retrospect (granted exception and all learned to be a bit more skeptical and rigorous).

Permalink to Comment

16. Hap on March 28, 2011 4:55 PM writes...

I think Milkshake referred to some kinase inhibitors where different (but structurally similar) compounds bound to active sites in reversed orientations. That'll put a crimp in trying to design analogs.

You'd rather have info than not, but crystal structures have limitations, and pitching them as a cure-all is not honest.

Permalink to Comment

17. Doglotion on March 28, 2011 5:23 PM writes...

Interesting question. I'd say another confounding factor is that GPCR biology can be significantly more complex than enzyme biology. For the most part, a small molecule will inhibit an enzyme (or not), full stop. With GPCRs you have agonists, antagonists, partial inverse agonists, allosteric modulators, receptor internalization, dimerization, differential activity against endogenous ligands, monkey business in the assays due to overexpression/cofactors/indirect readouts, etc. All of which can complicate figuring out the therapeutic potential of a given approach. Structural information aside, this uncertainty probably contributes to the lower success rate for GPCRs.

Permalink to Comment

18. Rick on March 28, 2011 6:01 PM writes...

drug_hunter (#12),
I certainly hope no one on this board would say they would prefer NOT to have structural data during the discovery process. It would be self-defeating luddism to decline any source of potentially useful info. It's more a question of when, where and how you use it.

The real problem, in my opinion, is the not-uncommon insistence by management (or investors, or potential corporate partners) that a project should not be undertaken or supported without a crystal structure and a docking model or two, or whatever the latest, hottest technology is at the time. Among the various silly reasons I've heard for dropping a project are: no genomic information, no genetic validations for the target, no enzymatic assay for the target, no combichem scheme, no in silico ADME model that predicts the behavior of the hits/leads, "peculiar" behavior in Caco-2 cells, "too high" in vitro serum binding, "too low" in vitro serum binding. These tail wagging the dog kinds of edicts are what really frosts my stones.

Permalink to Comment

19. Anonymous on March 28, 2011 7:01 PM writes...

Derek what drives your rationale? You've made it very clear that you dislike QSAR (a cornerstone of medchem) and now SBDD is of little help?

Does your drug design decision process involve a dart board?

Permalink to Comment

20. BCP on March 28, 2011 7:16 PM writes...

I think I'm pretty much with Derek here. Structural data can be extremely useful for ligand design/finding, but how exactly does it help you overcome a safety problem or poor bioavailability?

Permalink to Comment

21. Chinabonding on March 28, 2011 8:23 PM writes...

It's sometimes instructive to remember a crystal structure is like a snapshot of a movie.

I've found structural information most helpful in rank ordering hypothesis to try out in the lab.

Long long ago, for one project we were able to get a new crystal structure almost every week, lots of hours spent looking at pretty pictures, but not alot of property improvement. Maybe it would have been better NOT to have a crystal for that one!

Permalink to Comment

22. Anonymous on March 29, 2011 1:51 AM writes...

I think the key question is What is the the ROI (Return Of Investment) of SBDD? What will bring a drug discovery project forward most cost efficiently? Is it to invest in structural biology or adding more medchems to the project?
I don't think anyone in the industry has a clear answer.

Permalink to Comment

23. fragment_boy on March 29, 2011 4:09 AM writes...

For a project to be succesful, you need to retain almost a cynics view of the data in hand. Speaking as a crystallographer, over-interpretation of strcutural data is a huge problem, and one that is not helped by articles such as the one in DDT.

Pretty pictures can lead projects down the wrong path, but compleetly ignoring structural data is just as wrong...... what you need is a balanced view of the data and rationalisation of its worth.

Sadly drug discovery projects are not often belessed with balanced rationalisation of data, be it strcutural, biochemical or other.

Permalink to Comment

24. anon the II on March 29, 2011 6:41 AM writes...

to #19 Anonymous

Ah, Youth! It's refreshing to see such naivete in the young ones. I almost hate to see the coming slap of reality.

If QSAR is the cornerstone of Medchem, it's not hard to understand why the structure is tottering. But we all know it's not. QSAR is usually an academic exercise engaged in after a compound has gone to clinic or the project has been dropped.

Please forgive the dangling participle.

Permalink to Comment

25. Corein Hansch on March 29, 2011 9:57 AM writes...

#24. Ah the naive organic chemists who probably applies qsar w/o knowing it (at least you better be). Maybe Derek will let you use his dart board. He's an organic chemist too.

Permalink to Comment

26. drug_hunter on March 29, 2011 10:30 AM writes...

All snide comments aside, how *do* we combine all the available data -- from all sources -- into an intelligible form, and use that understanding to decide what to make next?

Of course QSAR and SBDD and every other tool at our disposal needs to be used -- intelligently of course.

There is a point beyond which scientific caution turns to negativity and cynicism. I think this conversation has stepped over the line. We should all be grown up enough to know that one source of information is not sufficient. Teams have to make some choices - how are you going to decide what compounds to make?

My answer is, You need enough "support services" (modelers, crystallographers, etc) to enable the best decisions to be made by the project chemists. How many is that? Depends on the project. I'd say about one modeler per project team is about right, if the chemistry team is very engaged and proactive and all that. And those modelers should of course be using all available sources of information -- not just xray structures.

Permalink to Comment

27. MoMo on March 29, 2011 11:59 AM writes...

Ask the scientists at Vertex how SBDD has worked for them since its inception. Great idea and sure way to lure in the big $$$ from investors who dont know QSAR from SBDD, but it sure looks good on paper! Makes pretty pictures and J Med Chem Publications too!

And for QSAR ridiculousness check out your own chemical inventories of the creeping lipophiles you all have been creating for the past 2 decades. LOL!

Drug design is situational-specific, although the more situations you create, the better.

I dont know what Pharma has been doing, other than trying to find the EASIEST way to create drugs- and we all know how that worked out all you -OMICS and combichem fans.

Sorry, there is no easy way out.

Permalink to Comment

28. TB drudge on March 29, 2011 2:18 PM writes...

Anyone have a comment on the TB Structural Genomics Consortium ? They have funded over 225 crystal structures of theoretically essential proteins from Mycobacterium tuberculosis.

Why? To aid in the discovery of new drugs, it is said. The TB drug discovery world hardly has any funding at all, and here is a huge chunk of money being dumped into Xtal structures of dubious value. Reason? Makes a nice tight package, easy milestones: Get structure, ring bell, get next grant, etc. Can anyone explain it better?

Permalink to Comment

29. anon on March 29, 2011 2:40 PM writes...

The isoniazid paper came out of the TB structural genomics consortium.

Permalink to Comment

30. MoMo on March 29, 2011 2:49 PM writes...

TB Drudge,

Simple economics here. Research money is spent keeping the 10-fold numbers of biologists (to chemists) employed and busy because the science/Pharma world is run by biologists, and they control the research environment to self-perpetuate themselves, their careers and their self-worth.

Mankind and reality would have preferred to see 225+ different NOVEL compound series directed at the bug then worry about their targets. But this is why TB is the scourge it is, especially if you are poor, incarcerated or institutionalized.

And Pharma knows they have no money for their disease.

Permalink to Comment

31. Jon on March 29, 2011 9:51 PM writes...

It really is disappointing when good research is skewed in favour of commercial interests.
I just wish the investors actually took the time to learn the value and concept of what they are investing in!

Permalink to Comment

32. ano99 on April 2, 2011 7:51 AM writes...

The value of a structure for advancing med chem is entirely situational as has been appropriately pointed out.

Maximum value is obtained when target structures are obtained with early-stage leads that bind in "non-standard" binding sites. Since structural data is the only way to rapidly and accurately determine the molecular topography of the new binding site, the structure provides a rapid means for a refocused screening campaign using compounds expected to adopt conformations that could be compatible with the new site. Unfortunately, these type of projects tended to be rare in my experiences.

In my experiences, the majority of the structural biology resources were placed on obtaining structures of known targets with inhibitor series that binds in an expected manner.

Take the ATP site in the kinase family for example. How much value is there in experimentally determining the structures of say 30 different analogues from a single or related series that all bind in the same way? In my opinion the first structure confirming the expected binding mode has some value, but additional structures from the same series tend to provide very little additional value. So why does the industry place most of their resources here?. I think their are multiple reasons for this but one key reason is improper goal setting my senior management. For example, most chemists want their analogue to be the next clinical candidate and will push a SB team to determine the structure of their candidate, naively thinking that it will show them an obvious simple fix for their series. This rarely if ever happens as the main drivers of a project (PK,ADME, etc) are never solved with xray data, as many here have pointed out. Another reason is the way a typical structural biologists goal set is determined. In many cases, compensation is directly tied to the number of structures solved as opposed to their impact. It is thus in the Structural Biologists best interests to spend a majority of his/her time working on projects with no value.

To me, the whole concept of an iterative structure based design paradigm for lead optimization has little to no value. The industry would be better off using its structural folks almost exclusively for lead ID by discovering and characterizing novel small-molecule binding sites in new protein classes.

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