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

« Godzilla vs. Mothra? Relman vs. Epstein! | Main | You Discover It, We Sell It. Deal? »

July 25, 2007

From the Sequencer to the Drugstore?

Email This Entry

Posted by Derek

A science writer who's read this blog for some time asked me a question which I thought I'd throw out to the readership. I was, in yesterday's post, making reference through gritted teeth to the amount of money the drug industry spent on genomic approaches. So here's the question, verbatim: "What drugs, if any, have been developed thanks in large part to insights gleaned from the human genome project?"

I don't think we're going to have to use many fingers, personally, given what I've seen. The "in large part" clause will take care of a lot of tangential cases that have been claimed mostly for PR purposes. There may be some dispute about the word "developed", since it could still be early for something to be hitting the market from the time of the Human Genome Project. Let's take that to mean "shown substantial and continuing clinical progress".

And I realize that there's room to argue about the "human genome project" part of the question, too, since many small companies (deCODE, Millennium, Incyte, etc.) did a lot of work of their own outside of the official HGP. But for argument's sake, let's throw the question open to all the genomic approaches. Examples?

Comments (16) + TrackBacks (0) | Category: Drug Development | Drug Industry History


1. dlib on July 26, 2007 12:36 AM writes...

this comment number - 1

Permalink to Comment

2. Peter Ellis on July 26, 2007 2:17 AM writes...

I suspect it's a category error. How [i]can[/i] knowledge of a genome help you make a drug? Knowing a protein sequence doesn't tell you whether it's druggable at all, let alone what compund to use.

What it may do (and I'd suspect is likely to do in future) is suggest targets to try. So, while it can suggest which drugs you should be making, it doesn't have much to offer in the way of actually doing the making.

So the "in large part" is an unfulfillable condition. The "large part" of drug discovery is surely rooted in molecular modelling, and good old-fashioned trial and error, and doesn't have much to do with sequence information.

It's a bit like asking "Which books, if any, have been written thanks largely to insights gleaned from the Oxford English Dictionary?" That's not how it works.

Permalink to Comment

3. Kay on July 26, 2007 4:04 AM writes...

#2: There certainly are possible paths from genome knowledge to drug development - e.g. via genome-wide microarrays (expression or CGH), identification of targets (which nobody would have on their bill if it weren't for the genome project), validation of the target - followed by conventional steps of the drug development process.

I don't know of any such examples, but I am optimistic that they will come sooner or later.

Permalink to Comment

4. Derek Lowe on July 26, 2007 5:54 AM writes...

Peter, I take your point - but remember a few years back, when companies were throwing out hundreds of millions of dollars on a category error? There was a real ship's-going-to-leave-without-you frenzy. . .

Permalink to Comment

5. Keith Robison on July 26, 2007 9:08 AM writes...

I agree that it is difficult to really peg drugs as having really been kick-started by the genome project. However, there are some candidates out there.

First, as you suggest, look at the pipelines of the surviving genomics companies, in particular at biologics. Now, the catch is that many of the targetrs / therapeutics being worked on had little if any input from genomics, and in most cases only those close to the story can really tell. But there are definitely proteins & antibodies in HGS's pipeline (such as the lupus drug) which can definitely trace roots back to genomics. A lot of Millennium's current pipeline came from the Leukosite merger, but some of the stuff was genomics-derived and for some targets the genomics data really boosted the interest. Many of the Millennium in-licensing deals fall in this category. I think DeCode in-licensed a Bayer compound based on their genome results.

You can also see development paths changing because of the genome project. Discovery of kinase mutations specific to particular tumor types (e.g. BRAF in melanoma) has shifted research focus, though not always successfully (Nexavar, developed as a BRAF inhibitor, I believe still has yet to prove value in melanoma).

In general, there just hasn't been enough time. The real genomics binge was just over a decade ago, so on real pharmaceutical timescales not enough time has passed. On the other hand, we genome freaks promised better.

Looking on the public side, there are some interesting bits of retargeting drugs based on microarray data -- there is at least one clinical trial running at Dana Farber based on this approach -- take a dull, off-patent drug & put it in a new context.

Genomics can tell you a lot. I do dispute the idea that you can't identify druggable targets from sequence data -- you can with great success identify proteins which are in the families which have been previously successfully drugged. Now, it is a bunch of steps from there to even a high throughput screen, but the success rate of getting to HTS and out to a hit is far better than a random draw from the genome. This is not to claim that there aren't targets that would be missed or bogus pseudo-targets picked up; just that the genome data pushes you forward a bunch.

The big immediate impact of genomics is going to be on diagnosis & prediction. One semi-poster child would be Diffuse Large B-Cell Lymphoma, where the disease has been trifurcated by molecular diagnostics. Those divisions (ABC, GC & PMDLCL) show different responses to therapy, though Rituxan may level the ABC vs GC division a bit.

Permalink to Comment

6. TW Andrews on July 26, 2007 9:18 AM writes...

I think a much better question would be "Which diseases do we understand significantly more now owing to the genome project?"

I think a lot of the PR troubles of the pharmaceutical industry have to do with people misunderstanding the difference between discovering what causes a condition and what cures (or treats) a condition.

For a small molecule company, genomics is obviously not going to drive drug discovery. However, it's not a stretch to imagine it driving the selection of targets which get screened.

Permalink to Comment

7. milkshake on July 26, 2007 10:39 AM writes...

There is sequence information on all kinases in human genome, one can search for kinases similar to his target kinase, and look for selectivity of his inhibitors.

The downside is that with many newly identified kinases one does not always understand what they do and where they are expressed.

Permalink to Comment

8. GATC on July 26, 2007 11:22 AM writes...

Affinium's Fab inhibitors come to mind as do several other new antibacterials that are either in preclinical or just entering the clinic at other small companies. In the case of Affinium, it was a cast-off from GSK's Manhattan-Micro initiative. The spectrum was much too narrow for Big-Pharma (staph only coverage). I know there are others out there too hitting diverse targets (PDF, DNA replication, etc.).

Although not at the scale of the HGP, bacterial genomics did a lot to identify essential targets, gauge spectrum, and evaluate selectivity, etc. There is a large patent estate available for mining if your pockets are deep enough and your management knows what it is doing. Another big problem is having enough chemical diversity to cover such a large target space in addition to the negative financial reasons for not developing new antibacterials.

I would make the argument that bacterial genomics was almost too successful.

Permalink to Comment

9. Ex-ProScript on July 26, 2007 12:45 PM writes...

Re: comment #5

Actually Millennium's pipeline is almost solely derived from a small company ProScript which I worked for. We were acquired by LeukoSite and then when LeukoSite was acquired by Millennium they acquired the ProScript porfolio.
Velcade which is Millennium's only marketed project was discovered and optimized at ProScript and nowhere was genomics to be seen. The IKK inhibitor that Milleniium is advancing to clinical trials also began its life at ProScript and, again, was not in any way based on genomics.
Millennium is actually the example that shows the whole genomics game was a farce as far as producing drugs. Had they stuck to that approach there would be a number of empty buildings on Sidney St. in Cambridge.

Permalink to Comment

10. Chrispy on July 26, 2007 3:40 PM writes...

I'm in early research at a small pharma/big biotech. We use the Human Genome data absolutely all the time in order to simply figure out how to make the relevent proteins. It is a piece of cake to determine counterscreen enzymes, for example (although the tox which gets you is likely to come from somewhere completely unexpected).

I agree with the previous post that we haven't yet seen the real implications come out the other end of the pipeline. But I would add that having this information allows us to do better research, and faster, even if it doesn't, in and of itself, solve the productivity issues of modern drug design. Those problems come largely from a risk-averse industry facing an increasingly risk-hostile environment.

Permalink to Comment

11. Clark on July 26, 2007 11:38 PM writes...

Off topic - Here is an interesting question: if you look at the top 5 drugs at each of LLY, PFE, MRK, and AZN what percentage were discovered in house? Vs bought after phase 1 trials.

I'd bet the number is well under 20%. If that high. For instance I believe for PFE the answer is none.

Permalink to Comment

12. Sue on July 27, 2007 8:28 AM writes...

I don't even work in this sector and I believe the answer is zero, zilch, nada, nothing!! The Genome project has yielded regard to the question.

Permalink to Comment

13. HelicalZz on July 27, 2007 9:45 AM writes...

" "What drugs, if any, have been developed thanks in large part to insights gleaned from the human genome project?"

It's the wrong question.

There are many examples - but they aren't necessarily new pills in new bottles. The human genome project isn't going to show the way to the next penicillin. But it will help (and is helping) with who gets what and benefits from it the most. Herceptin is probably the best example of this. If not for the genome project, this may be a failed drug thrown in the great big dustbin.

The Roche - Affymetrix Cyp card - thank you genome project - this has provided key drug-drug interaction information. While on microarrays - how about the recent FDA approval of Mammaprint ( an array that helps determine the aggressiveness of breast cancers. This will help dictate treatment. No, it's not a new drug - but it's a healthcare improvement.

Infectious disease benefits as well. I like what Cepheid is doing in helping ID particularly aggressive strains of bug with RT-PCR - right in the doctors office (or nearby). Again, not a drug but a boon to patient treatment - and again derived from the genome project.


Permalink to Comment

14. Keith Robison on July 27, 2007 11:05 AM writes...

Ex-Proscript adds an important clarification, though I'd clarify further. Three important compounds (Velcade, IKK inhibitor, and the NAE inhibitor) in Millennium's pipeline originated with Proscript, which had been bought by Leukosite just before Millennium swallowed Leukosite. A lot of the inflammation pipeline targeted chemokine receptors originated with Leukosite itself, and the Aurora program was actually home-grown.

Permalink to Comment

15. Gary McKnight on July 27, 2007 1:35 PM writes...

ZymoGenetics, at which I employed, has 3 drugs in active Phase 2 or Phase 1 trials that were discovered via genomics. These are biologics, specifically IL-21 , Peg-Interferon-lambda 1 (IL-29) and Atacicept (Soluble TACI receptor)

Additional drug candidates discovered via genomics are not yet at clinical phase.

Permalink to Comment

16. Sandy Eggo on August 5, 2007 9:40 AM writes...

Sure. Here's a great example from a conference held in San Diego:
You'll note Dr. Stephen Kaldor's claim to have progressed a compound from the discovery of it's gene into the Phase 3 in 45 months. Hmmm. Scratch the surface and you'll actually discover that this "gene" codes for a protein that has been known for decades (DPPIV) and has been a hot topic of research long before Kaldor or Syrrx were doing chemistry.

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