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

« Down In Phase III. Again. | Main | The Pfizer Sandwich Closure »

March 1, 2011

Use Avastin! Don't Use Avastin!

Email This Entry

Posted by Derek

The Genentech/Roche drug Avastin has been in the news a lot lately, mostly about cost/benefit analysis for its uses in oncology. It's nobody's idea of a cheap drug even for those indications where it shows results. But there's one therapeutic area where it's actually the bargain alternative.

That's AMD, wet age-related macular degeneration. Stopping the growth of those leaking blood vessels in the eye is the standard therapy for the condition, so a VEGF-targeted therapy is just the thing. Lucentis is the anti-VEGF antibody that's approved for that use; it showed very impressive results in the clinic, and seems to perform just as well in the real world.

But Lucentis is expensive. And while it's different from Avastin, it's really not that different. It is, in fact, an opthalmic-delivery-optimized version of the same general antibody, and was developed by the same folks at Genentech. Avastin itself isn't packaged in units small enough for AMD therapy, but if you have a practice with a number of patients, well. . .by the time you split it out, an Avastin injection is about $50, versus nearly $2000 for Lucentis. In fact, a great many physicians in the US (possibly a majority) use Avastin off-label in just that fashion. A UK study last fall shored up that practice with some data, and a number of other studies are underway.

One of these, conducted by the NIH, should be reporting soon. And that's putting Roche/Genentech in an odd position. They have not supplied drugs for the trial, for one thing. Last fall the New York Times reported that rebates are now being offered to opthamologists if they'll use Lucentis, which many have interpreted as a preemptive maneuver to deal with the likely NIH results.

This is a mess, no doubt about it. While Genentech did indeed spend the time, money, and effort to develop Lucentis as a separate therapy, there seems to have been an active effort to avoid finding out if Avastin wouldn't have been just as good. The market does provide perverse incentives like this sometimes - this is an instance where I think that the NIH is doing exactly what it should be doing by running the head-to-head trial.

But I don't think that Roche is going to like the results. And they could find themselves arguing, simultaneously, that Avastin should not be used for AMD, even though it's cheaper than the alternatives and may well be just as effective, while Avastin should be used for metastatic breast cancer, even though it's more expensive than the alternatives and may well not be effective at all. And while the company will surely argue that the numbers are not what they appear, and that there are other numbers that say differently, and that it's all quite complex, they're going to be unable to escape the downward slice of Occam's razor: that in every case, they're arguing for the exact position that maximizes their revenue.

This is what companies do, of course. We shouldn't expect any less. But that doesn't mean that the revenue-maximizing path is always the right one, either.

Comments (27) + TrackBacks (0) | Category: Clinical Trials | Drug Prices | Why Everyone Loves Us


1. You're Pfizered on March 1, 2011 10:10 AM writes...

It's no wonder our industry has such a bad reputation.

Permalink to Comment

2. Anonymous on March 1, 2011 10:14 AM writes...

In related news, Regeneron recently submitted a biologic license application to FDA for VEGF Trap-Eye, a potential competitor to Lucentis. VEGF Trap-Eye is dosed once every two months, while Lucentis is dosed once every month. Not sure how much VEGF Trap-Eye will cost relative to Lucentis or Avastin.

Permalink to Comment

3. Anonalso on March 1, 2011 10:18 AM writes...

I wonder how Novartis is liking all this news with Lucentis since they are part of the deal marketing it.

Permalink to Comment

4. qetzal on March 1, 2011 10:28 AM writes...

IIRC, Genentech/Roche have argued that Lucentis is somehow better than Avastin for ocular use.

It's not just the formulation. From the pkg inserts, Avastin is a humanized antibody, molecular weight 149 kDa. Lucentis is an antibody fragment (derived from the same Ab, right?), MW 48 kDa.

Anyone else remember the details of why Genentech claims Lucentis is better for this use? Any data to support those claims?

Permalink to Comment

5. Hap on March 1, 2011 10:39 AM writes...

I think the 40-fold cost factor is the killer, since it's hard to argue to patients that two very similar drugs need to differ so much in price other than "Because we can." If they were closer in price, rebates and other sales tactics might help, but Genentech/Roche is basically trying and failing to enforce a bad pricing scheme (well, bad for everyone else, and easily evadable) by force, and failing, and making themselves look bad doing it.

Companies try to maximize their revenue, but that only works when it's also good for the people on the other end of the deal. When it's not, anything you do to make them take is going to be viewed as the imposition of will it is, and not the workings of a willing agreement.

Permalink to Comment

6. lazybratsche on March 1, 2011 10:41 AM writes...

Why is there such a difference in price? I imagine it has something to do with market size, development costs, and the usual supply/demand calculations, but I have a hard time understanding why optimizing one particular biologic for the eye makes it 40x more expensive.

Permalink to Comment

7. Diocletius on March 1, 2011 10:45 AM writes...

Brilliant argument. I would not be surprised to see it used again as the debate heats up. Congrats!

Permalink to Comment

8. petros on March 1, 2011 11:02 AM writes...

The same thing is underway in the UK where NICE is assessing the relative benefits of the two

Permalink to Comment

9. NoDrugsNoJobs on March 1, 2011 11:04 AM writes...

Its ironic that if Genetech/Roche promoted avastin for the eye indication, they would be paying very hefty fines and be the subject of much tsk-tsking on this board. However, since others are promoting its off-label use and because we see that this off-label use goes against Amgen's putative interests, we think its a really good thing. I guess the fact that the company that made the investment, took the risks and costs of the clinical trials is the same company thats getting screwed on the back end by the off labeling and unapproved use by a different drug makes it all ok......

Permalink to Comment

10. Cellbio on March 1, 2011 11:20 AM writes...

Lazy, the price difference reflects what the market can support, at least in theory, given the benefit of treatment over costs associated with no effective therapy. The problem for Genentech is that the same rationale provided the cost basis for Avastin (a topic of debate for sure), and local delivery of Avastin to the eye demands much lower amounts of drug, so using the current formulation provides very little profit, dictating the "demand" for a distinct formulation that can be priced differently. So what to do? Cleave of the Fc portion of Avastin, put that molecule only through clinical trials, and charge 40x more. It is not 40x more expensive, it is not meeting a medical need that Avastin can't (at least to my knowledge), just meeting a price point that is profitable.

Permalink to Comment

11. Hap on March 1, 2011 11:21 AM writes...

Except they tried really hard to avoid testing Avastin for that indication, almost as if they knew that that it worked no better than their own cheaper drug, kind of like selling two versions of thalidomide for two indications, charging ten times as much for one, and hoping that nobody would notice.

It wasn't like they couldn't have gotten their money back with a modest increase in Avastin's price (considering the enlargement of its target audience), but they got greedy, and caught out. Sorry about that.

Permalink to Comment

12. Anonymous on March 1, 2011 11:46 AM writes...

Lucentis is injected directly into the eye on a less frequent dosing regimen..................for some this is worth the cost even if the risk/benefit use is the same. Not everything has to do with just "effectiveness" - sometimes quality of life and different risk/benefit profile are the reasons.

Permalink to Comment

13. CMCguy on March 1, 2011 12:13 PM writes...

The key is that ``Genentech did indeed spend the time, money, and effort to develop Lucentis as a separate therapy``. Although scientifically connected by a common VEGF target the discovery & development probably where largely on parallel paths that have different goals and different reasons (partnering or collaborations). The formulations and dosage reflect the unique requirements for the targeted patients and if only focus on relative manufacturing cost delta is likely small contribution with estimate 3-5X (with limited annual batch sizes could be 10x). In essence, as with most drugs, cost are reflection for an independent development and the pricing is tied to market-size and patent life to achieve a positive ROI. Although most drugs do go after market expansion post approval during the earlier stages most drugs get narrowly slotted for a specific approval so some of this appears to be somewhat Monday morning quarterbacking/Meta analysis on decisions made many years ago under different circumstances. Its another layer of complexity to drug development and sure 40x sounds unreasonable until it is put in a context that captures the whole picture.

#9 is correct that Genetech/Roche would be raked over the coals if they engaged or hinted any off-label promotion and could even be held responsible for allowing supplies of Avastin to go for unapproved uses. The whole issue of dose splitting is very controversial (contamination risk) with routinely accepted in US would be illegal in many EU countries (thought included UK so studies there curious). Such multi-dosing would have to be expressly studied as part of approval so any MDs who practice this should automatically assume 100% liability for adverse events.

Permalink to Comment

14. Anonymous on March 1, 2011 12:20 PM writes...

I'm a little surprised that on this blog, so little attention is paid to the differences between the molecules.

Avastin is a much larger molecule, and has worse tissue penetration than the Fab (important, since you are trying to penetrate to the back of the eye). Also, Avastin has an intact set of antibody effector domains, that could trigger inflammation in the eye, which is why Lucentis was thought to be safer than Avastin. (And indeed, Lucentis does have a better side effect profile than Avastin, but not terribly better.)

So yes, Lucentis was actually designed for this indication, although I'm sure people at Roche regret going through the process now.

To #10, Lucentis is not produced through enzymatic cleavage of Avastin, it's a separate product produced in bacteria as a Fab and refolded.

Permalink to Comment

15. Hap on March 1, 2011 1:16 PM writes...

I understand that they have to get the money for finding out the drug works and factor that into what they price it at, and that dinging for not testing Lucentis against Avastin is probably not fair, but when the cost differential is that large, they probably aren't going to sell a whole lot of Lucentis in the first place. A sale requires a (willing) buyer and a (willing) seller, and while a factor of 40 over the nonopthalmic variant may ensure that the seller makes enough money, it doesn't make any sense for the buyer without really strong data. If it's strong enough to risk loss of sterilization, side effects from the different formulations, and other issues to use Avastin over Lucentis, you priced Lucentis too high, and when you try to force the issue, you look bad doing it.

I wonder if they couldn't have simply raised Avastin's price somewhat (1.5-2 times?) and priced the opthalmic formulation similarly (again, 1.5-2 times) so that they could make the money back for Lucentis, but their own people obviously have better numbers and a better understand of what works financially.

In a sense, getting approval for Lucentis is like getting a new indication for a generic or a supplement - you need to reward the company for making the investment into the generic and for finding out useful info about it, but you don't have a method to get money specifically for the new uses, and anything you do to try to recoup that money makes you look evil.

Permalink to Comment

16. milkshake on March 1, 2011 2:05 PM writes...

even with expensive-to-make protein stuff like monoclonal antibodies, the cost of the active substance is only a (tiny) fraction of the cost of the drug. By far the most expensive part of the drug development are the clinical trials.

It would be easy enough to reformulate Avastin to make it unusable in eye treatment - by adding a strong dye (for example methylene blue)

Permalink to Comment

17. barry on March 1, 2011 3:40 PM writes...

so what exactly is the site of action of Lucentis? It's injected right into the orb. Is it neutralizing VEGF in the vitreous humor? In the interstices between cells of the vessel walls? Are the differences in MW (and therefore transport rate) relevant?

Permalink to Comment

18. HelicalZz on March 1, 2011 3:42 PM writes...

"an opthalmic-delivery-optimized version"

Well cross my heart and hope to die.


Permalink to Comment

19. cynical1 on March 1, 2011 3:44 PM writes...


No one knows what it's like
To be the bad man
To be the sad man
Behind blue eyes

No one knows what it's like
To be hated
To be fated
To telling only lies

But my dreams
They aren't as empty
As my conscience seems to be

Permalink to Comment

20. CMCguy on March 1, 2011 3:50 PM writes...

milkshake please tell me you are joking about adding a dye as a blocking strategy? Formulations are generally driven by KISS principles and how one could justify use of a dye as necessary or of benefit is beyond me.

On the other hand I do suspect there could be clever formulators who could come up with excipients that would be compatible with IV and not ocular usage however again rationalizing and then doing clinical work to support a reformulation are not insignificant hurdles which seem as non-starters.

Permalink to Comment

21. CMCguy on March 1, 2011 4:11 PM writes...

Derek- can you indicate where you got the $50/$2000 cost figures as the link UK article indicates £50 vs £700 (14.5x) although well could reflect negotiated vs US unregulated prices it would be closer to factor in more important COGs discussion.

I found comments interesting by the Univ of Liverpool head of ophthalmic research did suggest reasons on why Lucentis was developed.

Permalink to Comment

22. RickW on March 1, 2011 5:02 PM writes...

In addition to being a Fab, Lucentis was affinity matured by phage display so it has a much higher affinity than Avastin. It is really a pretty different drug, I guess we'll find out from the NIH trial if it was worth the extra effort.

Permalink to Comment

23. Cellbio on March 1, 2011 10:24 PM writes...

#14, Didn't mean to imply enzymatic production of Fab, cleave it off with a DNA construct, fine.

Can you offer any support for your tissue penetration claims? I have worked with abs for a decade or so, and every ab fragment or alternative scaffold company makes these claims, with no support. Certainly never seen clinical evidence, and don't think you'll see it with Avastin failing to work due to lack of tissue penetration. BTW, what is with this theory anyway, that abs normally fail to penetrate into tissue at significant levels? Might be a problem for immune protection no? How about the VEGF Trap? Larger than a Fab, close to ab size, works?

Yes, it is a different molecule, engineered to be so, but all done for pricing not indication specific needs of the drug. Maybe I'll be surprised by further clinical study.

Permalink to Comment

24. milkshake on March 2, 2011 3:17 AM writes...

The methylene blue was of course a joke. It would be a lot better to use a colorless additive - for example a preservative - one that coincidentally causes eye photosensitivity.

Its too late anyway to reformulate the product past the approval. The only good way now I can think of would be to raise the price tenfold. We would then create a nonprofit foundation (into which the company would contribute on tax-deferred basis). The foundation would subsidize the cost for cancer patients so that they get 90% discount on the compassionate basis.

Permalink to Comment

25. Anonymous on March 2, 2011 5:15 AM writes...

#14 and #22 are right about the differences between the molecules. However, it looks like all this good science didn't translate into wildly greater efficacy/safety/dosing frequency profile for Lucentis (as far as we can tell from separate trials and the few head-to-head comparisons).
So I agree with #15 - it's making them look bad to continue to price Lucentis so high and push it so hard. Surely this happens all the time - a drug is developed, is shown to be effective, but not really any different from standard care - so why should the consumer pay a big premium for it? The only difference here is that 'standard care' is an off-label use of Avastin.

Permalink to Comment

26. Rick on March 2, 2011 7:24 AM writes...

One thing that really bugs me about the furor over avastin vs. lucentis marketing is what it says about how our standards have fallen. The discovery of the VEGF pathway and early research on the effects of blocking it suggested huge potential for drugs working by this mechanism. Compared to other "breakthrough" drugs of yesteryear, like penicillin or lovastatin, these drugs are disappointing. I'm not saying that avastin and lucentis are not effective to some extent, just that they fall far short of what most researchers and clinicians believed was possible a decade ago.

Fights over marketing exclusivity and pricing strategies suggest that financial performance has displaced clinical performance as the key measure of a drug's efficacy. As a result, the industry wastes money fighting over profits for (relatively) crappy drugs when the resources would be much better directed toward using what we've learned about their shortcomings and applying the supposedly vast powers of new drug discovery technologies to discover much better replacements.

Permalink to Comment

27. Murali Nair on June 23, 2013 8:32 AM writes...

I had an Avastin injection yesterday. The results are immediate, to the extent I can send this message by browsing on the net. I personally believe that Roche, its US biotech arm Genentech and its marketing arm Novartis are all so much greedy. First they do not cooperate with NIH study. Second, they willingly fail to apply for FDA approval for Avastin. And third, most important they are offering bribes to Ophthalmologists to prescribe Lucentis! This is bad commercial practice for sure!

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

Adoptive T-Cell Therapy for Cancer: The Short Version
How Much Is Wrong?
The 2013 Drug Approvals: Not So Great?
Positive Rules and Negative Ones
Prices Rising - Every Year, Every Drug?
Easy Aziridines
Back Blogging (Bonus Biographical Begging)
It Just So Happens That I Have A Conference Right Here