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

« Foaming Up Over the Edge | Main | Yahoots »

April 29, 2004

An Era Begins

Email This Entry

Posted by Derek

The big news today was the Thursday release of two papers, one from Science and one from the NEJM, from two teams who have done something that we in the drug industry have long been working on and expecting, with equal parts anticipation and fear.

AstraZeneca's drug Iressa, like many other chemotherapies, varies widely in its effectiveness. Some lung cancer patients have experienced dramatic, life-saving remissions, while others have shown no improvement at all. Everyone's suspected that there is some genetic difference responsible for this, and now it's been found. The responders tend to have a set of mutations in their version of Iressa's target protein, EGFR, as compared to the nonresponders. So here it is - the pharmacogenomic era has started for real.

I've written about this issue before. As I've said, this is going to mean, eventually, that we're going to need a lot more cancer drugs as the various forms of the disease are categorized. Of course, the best way to categorize them is for us in the drug industry to keep on delivering therapies that don't work as well as we'd hoped, which is a little nerve-wracking. There's no way we could have predicted these EGFR mutations from first principles, for example. But we're going to end up with a lot more discrete targets to work on.

And as we find drugs for them, the clinical effects are going to be dramatic. If we can keep it up, people are going to gradually, over the next decades, stop dying from cancer. It'll be one tumor line at a time, but I think it'll happen. In an aging population, that's going to be the next thing to a miracle.

But finding all these drug subtypes is going to be tough, and expensive - and the kicker is that we're no longer going to have as large a patient population to sell them to. The not-so-well-kept secret of cancer therapy is that everyone has, for a long time now, ended up prescribing every drug for almost everything. Because you never know, it might work, and what do you have to lose after a certain point? But that whole market regime is going to collapse eventually, and the first tremor just shook today.

This is going to be "creative destruction" in its purest form, because something's going to have to change drastically. We drug companies are going to have to deliver drugs that (at least as far as we can see now) are going to cost just as much to find and develop as the ones we have today - and maybe more. But we're only going to be able to sell them to a fraction of the market that we can now, and this in a world where people are already having rug-biting fits about the prices we charge.

There's celebrating going on in the drug industry tonight - we finally know why a compound really works in some people! We know exactly who to sell the drug to! But there a lot of nervousness, too, because now we also know exactly who we can't sell it to, either.

Comments (7) + TrackBacks (0) | Category:


1. Chris C. on April 30, 2004 4:41 PM writes...

Oddly, does this imply an argument for the nationalization of cancer research at some point in the future? That is, if there is a significantly reduced market, but a continued need, should the gov't step in at some point to underwrite some of the research costs, to insure that the research continues without requiring exorbitant pricing?

Permalink to Comment

2. Oliver Morton on May 1, 2004 5:12 AM writes...

Terrific post. But is there not a somewhat offsetting upside in the downside, too, in that some of the sub-group drugs that will be needed may well already have been developed, but not approved due to toxicity and/or ineffectiveness concerns when tested without pharmacogenomic analysis?

Permalink to Comment

3. Larry M on May 1, 2004 12:20 PM writes...

Incorporating genomics into drug evaluation has a potential to make drug discovery more INexpensive. Most of the costs occur in Phase III trying to show a small benefit that is statistically significant. This requires large trials with long duration. If with genomic data incorporated into the trial design, only active responders could be selected for the trial, the difference between placebo and active drug could be demonstrated with much smaller groups and in much shorter time frames. It is the noise from non-responders that obscures the effects in current research.
That is however a big IF and won't happen soon, IMO.

Permalink to Comment

4. qetzal on May 1, 2004 4:17 PM writes...

Larry, I can also see it making things worse, not better. Seems like it depends heavily on when you figure out how to ID the target population. If you can do it early, then I think you're right that it should allow for smaller Phase III's.

But, what if you have to do a big trial, with several different pharmacogenomic analyses, just to convince yourself that you can ID the target group. Then, FDA would require another trial in the (now prospectively defined) target group.

In addition, I suspect there will be plenty of cases where pharmacogenomics will be unable to ID a target group. Any additional effort that was spent trying to find the (nonexistant) target group will be sunk cost.

It will be extremely interesting to see how this all balances out. (And I mean that very much in the sense of the old Chinese curse!)

Permalink to Comment

5. Derek Lowe on May 1, 2004 9:47 PM writes...

I think Qetzal's on the right track. Larry's correct that a good pharmacogenomic ID would make for instantly compelling clinical data, but I think that it'll be hard to get those early enough to make the whole thing possible.

Not that everyone won't be looking. The thing is, we've already been searching for this sort of thing, both for positive effects and negative ones (identifying people who show impaired tolerance or worse side effects.) And for the most part, it's been hard.

The Iressa/EGFR story is a dramatic one, but we're going to have to pick up the pace on the dramatic ones. If we don't start seeing several more such discoveries over the next 3 to 5 years, it may be that there aren't a lot of such clear markers to be found. I sure hope that's not the case. . .

Permalink to Comment

6. Derek Lowe on May 3, 2004 6:54 AM writes...

As for Chris C.'s question about nationalizing research, the sort of subsidy he describes is actually already in use: the Orphan Drug Act. This gives a company broader patent protection for a small-market drug.

Gleevec, for one, is designated as an orphan drug in its primary indications. The everyone-tries-everything principle mentioned in the post has made it a big seller, but it still took some nerve for Novartis to go ahead with its development.

Permalink to Comment

7. Nick Henriquez on May 4, 2004 3:21 AM writes...

I will write about this in my own blog. People are AGAIN forgetting a very important characteristic of cancer: it's not a stable target!
Although less so than RNA-viruses cancers are genetically variable. Increasing the selection on any single characteristic by adding drug X will mean different clone-sets survive. There may be gains in survival, but with single-target drugs the medium-term outcome ALWAYS is drug-resistance followed by death.
The only way to attack cancer successfully is using multiple drugs attacking different targets.

Permalink to Comment


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