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

« Is Genzyme a Buyable Company At All? | Main | Lilly's Gamma Secretase Inhibitor for Alzheimer's: Worse Than Nothing »

August 17, 2010

Avastin: Taking It Back

Email This Entry

Posted by Derek

As if one were needed, here's an example of how rough the state of current oncology therapy is today. Avastin, the antibody-based therapy from Genentech/Roche, had been approved (conditionally) for advanced breast cancer, based on a study showing about a five-month benefit in tumor growth. (Everyone should already know that such numbers, for many types of cancer, are indeed enough to get an indication approved, and everyone has, I'm sure, already decided what they think about that.)

But the approval came with a requirement to follow up on those results. For one thing, the study that led to conditional approval didn't show much of a survival benefit, making the approval itself controversial at the time.. The follow-up work has shown that those initial results were right on target. For metastatic breast cancer, Avastin has something like a month-and-a-half survival benefit. That probably doesn't outweigh the risks, and the FDA is seriously thinking about revoking that earlier approval.

Based on these numbers, I think that they should go ahead and do that. The whole point of conditional or accelerated approval is that it can go either way when the harder numbers come in, and in this case, it seems pretty clear that the benefit isn't there. No one cares about tumor growth if it doesn't affect survival or (at the very least) quality of life. And in this case, the later studies have suggested that even the earlier tumor growth numbers were too optimistic. You have to be willing to abide by the evidence.

Because of Avastin's high cost, this is probably going to turn into a rationing-health-care argument - in fact, it probably has already. But I'm not even talking cost here. Avastin, by the evidence we have, does not seem to help advanced breast cancer patients. It wouldn't help them even if it were free.

Comments (30) + TrackBacks (0) | Category: Cancer | Clinical Trials


1. Mark on August 17, 2010 8:11 AM writes...

What I find so interesting about this case is the dynamics between patients, companies, payers and the FDA.

Does Avastin help at least a few breast cancer patients? Probably. However, if the FDA revokes the indication, payers are likely to yank reimbursement for the drug in breast cancer patients. Hence the uproar by patients.


Permalink to Comment

2. Anon on August 17, 2010 8:16 AM writes...

Derek, seriously? How many months' survival benefit is worth it to you, and who are you -- or the FDA -- to judge such a thing? If the survival benefit is there, and it's statistically significant, the drug should stay approved -- period. The rest is, as you more accurately say, an argument for insurance companies about cost vs. benefit.

Avastin isn't the greatest thing since sliced bread, but I'll wager there are plenty of dying cancer patients out there who would happily give everything they had for the chance to live another month and a half.

Personally I think only looking at survival endpoints is part of the reason that new cancer therapies are progressing forward through FDA approval so slowly. There are still more dying cancer patients who would be happy to live the same amount of time if it meant that the time they had left was spent more comfortably instead of in the hospital debilitated by conventional chemotherapy. I'm not saying that Avastin is that drug -- that's more of a comment about the state of the field in general. If FDA continues to focus exclusively on overall survival as the only meaningful endpoint, we will miss opportunities to improve the lives of cancer patients.

Permalink to Comment

3. InfMP on August 17, 2010 8:31 AM writes...

I agree with comment #2. When it comes to living or dying right now, I think patients should be allowed to choose a drug that might not necessarily help.

It's not like avastin is making these patients die faster right?

Permalink to Comment

4. Brian on August 17, 2010 9:37 AM writes...

This seems to really get to the heart of several issues. A relative was diagnosed with stage IIIb ovarian cancer 10 years ago. She's been on chemo much of that time (never off more than 2 months at a time). Gradually each drug has lost effectiveness or she has become sensitive to them. She started taking Avastin maybe 6 months ago. I don't believe it has been formally approved for this use, though some studies indicate a benefit. It brought her CA125 levels way down and has kept them in check, though they are slowly creeping up again. Getting the HMO to approve the treatment was not easy.

So this is an anecdotal case. And perhaps, on average, Avastin does not help breast cancer patients... or ovarian cancer patients? But what about the outliers? Until "personalized medicine" advances to a point where it can fulfill some of its promises is it not "worth" trying on any given patient to see if there is a response? Admittedly the cost is huge, either way.

Permalink to Comment

5. J-bone on August 17, 2010 9:49 AM writes...

This very topic has already been covered here in recent months.

Permalink to Comment

6. gr on August 17, 2010 10:09 AM writes...

I agree with comment #2. When it comes to living or dying right now, I think patients should be allowed to choose a drug that might not necessarily help.

In which case there are quite a lot for them to already choose from, no?

Permalink to Comment

7. Anonymous on August 17, 2010 11:12 AM writes...

I remember someone asked Franz Humer at an employee meeting how he can justify charging upwards of $150,000 for a drug that adds only a month to a cancer patient's demise. He angrily replied how arrogant the question was and that life is precious. He also said he can pass the "red face test" where he can look a dying cancer patient in the eye and justify the drug's cost, without turning red. I say it's fine with me, but insurance companies shouldn't have to foot the bill for such dubious benefits. Our healthcare industry is notorious for eating up our savings at the end of life. We'd be better off to learn to accept death and bequeath our assets to a more worthwhile cause.

Permalink to Comment

8. alig on August 17, 2010 11:49 AM writes...

Re: "It's not like avastin is making these patients die faster right?"
Actually Avastin does cause some patients to die faster. Avastin has some pretty nasty CV, GI & renal side effects that cause some patients to die before their cancer would have killed them. That may be the reason the overall surival benefit isn't there.

Permalink to Comment

9. bbooooooya on August 17, 2010 12:34 PM writes...

" I think patients should be allowed to choose a drug that might not necessarily help."

No problem, as long as they pay for it.

Permalink to Comment

10. Hap on August 17, 2010 12:56 PM writes...

It seems hard to expect people to make cost/benefit analyses on drugs when neither doctors nor regulators with years of the experience in the field, better technical knowledge, and better emotional separation from the issues involved are able to do so. (This has come up before here with informed consent in trials, as well.) The number of people flocking to use supplements with no data other than the word of liars, rogues, and thieves would seem to indicate that that the assumption that people are easily capable of or have the knowledge to make reasonable cost-benefit analyses on drugs is not valid or correct.

People facing near certain death are likely to take a flyer on something because they have only one life and want to keep it, and opening the permit doors to drugs with no real effect would seem to be a permission slip to turn the cancer field into something more akin to supplements. Opening the opportunity to give all of one's money to the least scrupulous for drugs of nebulous benefit would seem to be a recipe for lowering people's trust in the drug industry even further.

Permalink to Comment

11. dearieme on August 17, 2010 1:05 PM writes...

You are not going to live forever. Die with it.

Permalink to Comment

12. dearieme on August 17, 2010 1:06 PM writes...

You are not going to live forever. Die with it.

Permalink to Comment

13. Anon on August 17, 2010 2:07 PM writes...

Hap (#10),

Going along with my previous comments in #2, how do you judge (or even presume to judge) what is a "nebulous benefit"? What is enough survival for you? One month? Three months? Six months? A year? That is a slippery slope of logic once you start going down it.

Even one week of increased overall survival is a benefit, provided the clinical trial meets the tests of statistical significance. Those are the criteria that the FDA is supposed to make its decisions based on -- scientific ones. Frankly there have been too many FDA advisory committee meetings lately -- especially ODAC -- where the panel has completely politicized the issues and turned it into a cost-benefit analysis rather than a weighing of the facts. That is *not* their remit, and they seem to have forgotten that fact.

If you take your statements any further than what the data analysis says, you are no longer doing science and have entered the realm of either insurance actuary (at best) or politician (at worst). In my mind there is a huge difference between a scientifically proven 6-week survival benefit in a controlled clinical trial and someone taking a flyer on unproven, untested supplements.

Permalink to Comment

14. Norman Yarvin on August 17, 2010 3:13 PM writes...

That one-month survival benefit is an average, not a guarantee. It might mean that half the people who take it live no longer, and half live another two months. Or it might mean that 95% of the people live no longer, and 5% are entirely cured. In the former case, the drug is indeed worthless. In the latter case, the question is more interesting. (Could we, for instance, figure out how to predict who is in that 5%?)

Permalink to Comment

15. Hap on August 17, 2010 3:24 PM writes...

But that (to make a judgement based on weighing of the facts, which as I said, qualified people could not do unambiguously) is what you're expecting random people (sans training in the field) to do. Supplements have no data at all, and yet aren't really distinguished effectively by people from drugs with actual tested claims, something abetted by the FDA (thx Orrin!). If cost-benefit analyses aren't able to be used effectively in their least taxing implementation, I don't see how expecting someone (in their last month and a half of life) to do a full statistical analysis on drug costs and benefits is going to work out well.

The drug industry has problems because it oversells and overgenerally sells small benefits as large ones. Selling drugs with limited benefits for the farm leaves the question of what you'll charge when you can actually talk about remission or cure hanging like the sword of Damocles. It also is likely to make the difference between drugs that cost a lot but do something worth the money and supplements and drugs that don't harder to see. If you think the drug industry needs more expensive drugs that don't do much (and that that will somehow return it to the glory days), well, good luck with that.

Permalink to Comment

16. ronathan richardson on August 17, 2010 3:37 PM writes...

Yup, keep pinning on those pink ribbons, hosting fundraisers, and sending the last bit of cash keeping you out of medical bankruptcy off to the holy overlords at Genentech for being such noble beings as to accept your $100,000 payment for a toxic injectable that gives you a few more weeks of bedridden, vomiting existence. The FDA exists to protect consumers not only from dangerous products, but from business practices that take advantage of vulnerable people. People dying of cancer and their family members are simply not rational actors--virtually all of them. My grandmother was a conservative, level-headed college professor who, nevertheless, spent her life savings of $150,000 on treatments that allowed her daughter to survive maybe a month longer than expected, all of which was spent with tubes in her arm, pallid and listless from the amphotericin keeping fungal infections at bay. $150k, or even a million, is worth it, if it cures you or gives you 2 years of extra healthy life--if something really works, I have no problem with them charging through the roof for it. Avastin puts more people into medical bankruptcy than the number of people it gives another healthy 2 months. The FDA has the power to stop this.

Permalink to Comment

17. David Young MD on August 17, 2010 4:13 PM writes...

Avastin is a horribly expensive drug. If it doesn't do anything for breast cancer patients, it shouldn't be used. Instead, the money should be allocated for other market drugs, that can be occasionally helpful to other cancer patients in an off label use (including Avastin in Ovarian cancer.)

Keep in mind that when you combine a drug such as avastin with another useful anticancer drug, such as taxol, and a patient has regression of tumor, you never know in that individual patient whether or not it is just the taxol producing the anti-cancer effect, or is it the combination.

I have many patients who are denied off label treatment when they have a much better chance of helping them than avastin in breast cancer. Avastin has been a very helpful drug in colon cancer. It is probably much less helpful in breast cancer and in lung cancer.

Permalink to Comment

18. expharma on August 17, 2010 5:51 PM writes...

it doesn't work for this cancer but it does for others, it's not perfect and should be withdrawn for this indication. if the science says no then withdraw otherwise market it as a suppliment

Permalink to Comment

19. C-drug on August 17, 2010 6:25 PM writes...


What do you consider to be enough survival benefit? 1 month, 1 day, 1 hour? I think the slopes are slippery in both directions, hence why someone has to make a call.

Does anyone know if this drug has been, or could be used in combination therapy?

Permalink to Comment

20. Anonymous on August 17, 2010 9:48 PM writes...

Using diagnostics to target the patient population (and identify potential high responders) is the way things are moving. Watch Roche...they are a moving very fast and furious in that direction. They are big on personalized healthcare. It makes a lot of sense ....don't mass market your drug (remember vioxx etc...) run the risk of liability. Instead, focus on and identify patients that are likely to respond to your therapy and where you may ultimately provide benefit.

Permalink to Comment

21. cliffintokyo on August 17, 2010 9:54 PM writes...

Good comments, and especially Norman.
FYI: FDA are only considering the utility of Avastin for treating advanced breast cancer. Their decision will not affect any other approved indications.
Shall we come back to this when the FDA verdict is out?

Permalink to Comment

22. Scott on August 18, 2010 2:37 AM writes...

@20, Yes, I don;t have the facts at hand, but screening breast cancer patients for the HER2 gene allowed Genentech to identify the patients most likely to respond to herceptin.

Permalink to Comment

23. FMC on August 18, 2010 8:56 AM writes...

Here an interesting article in the WSJ on that matter:

Permalink to Comment

24. David Young MD on August 18, 2010 3:04 PM writes...


It has only been used in combination (save kidney cancer) and it is used in combinatio with Taxol in breast cancer. In this setting, the increased response is modest and no survival benefit. Avastin does have some adverse effects (bleeding, possible perforation of intestines, hypertension, encephalopahty), most manageable and costs a great deal. There are probably a lot better ways to use that money for those with metastatic disease. In other words, there are probably other effective medications that would be denied these patients when this highly expensive medication is (was) paid for. I would bet that carbazetaxel, the new drug for prostate cancer would likely work much better (in a late setting)than avastin and do more good. But carbazetaxel would never be paid for.


Permalink to Comment

25. RTW on August 18, 2010 10:57 PM writes...

# 17, 24 Mr Young MD - That is actually part of the reason for many of these trials. Avastin may not be all that great for breast cancer, but since its out there, Oncologists are going to try it in combination with other drugs for other cancers. In fact that is a strategy in getting these drigs to the market place. Oncologist start throwing these drigs at anything...... The e problem is paying for these off lable uses.

I 15 years primarily in anticancer drug discovery until big blue shut it down in A^2. We often did multi drug comparisons of standard agents against the new ones we where working on and sometimes in combination with them. In the animal models I saw some pretty exciting things. One compound in particular CI-1033 in phase II which I saw pictures of one of the patents with skin cancer. It nearly cleared it up! The follow up to that compound is currently making its slow progress through the gauntlet. I am sure once it is out there oncologists will also try it with just about everything they can think of....

So - I don't see the Avastin trial as a failure. Just a case of too much wishful thinking perhaps. So long as its still available perhaps another combination with an even newer drug might be a fantastic combination. We just can't know. The targets look promissinig, but as has been pointed out, Cancer is a very tough nut to crack because it doesn't play by any sort of rules.... Too many mutations... Most peoples cancers are not made up of just one mutation...

Genetic characterization of tumor types would help us to achive targeted therapy, but its very expensive. And coupled with the expense of many of these new MAB type drugs its prohibative I think. We need a lot of targeted inexpensive to make small molecules. In fact many that don't make it to the clinic I think should still be tested over and over again in combinations with other agents. We need more molecular probes.

Just my two cents.....

Permalink to Comment

26. Anonymous on August 26, 2010 10:04 PM writes...

Characterization of tumors is expensive now but like anything else (ie: flat screen TV's) the price will level out and come dowm. it's not about cost's about doing the right thing for mankind!

Permalink to Comment

27. Don M on December 16, 2010 4:22 PM writes...

The federal government has no constitutional authority to regulate medicine.

Avastin may, or may not help some patients. It is the contractual duty of doctors to identify which patients can be helped by a drug, and to then proscribe it. It is the contractual duty of doctors to identify which patients are not helped, or would be helped more by other drugs, and not prescribe ineffective or less effective drugs.

By the federal government taking over unwarranted powers, they prevent the development of private entities which would help doctors make that decision. Further, they slow the private development of new medicines which can be targeted to cover patients not well covered by existing drugs. The entire FDA should be defunded and fired. Today!

Permalink to Comment

28. Indiana on December 16, 2010 8:52 PM writes...

27: The FDA is quite clear on its limits and does not regulate the practice of medicine.

The federal government has all the constitutional authority needed in the Constitution in the Interstate Commerce clause (actually written by the Founding Fathers, not imagined or made up by a bureaucrat somewhere) to regulate interstate commerce of medicines such as Avastin.

Avastin remains available for any doctor to prescribe, or any patient to use, including those with breast cancer, because it has been shown to work in at least one cancer (aside from breast cancer). The FDA is not stopping anyone from prescribing it for any indication. In fact, if your doctor wanted to prescribe Avastin for a sore throat, and you were willing to take it, they would have nothing to say about the matter.

But Genentech can't make a false claim that it works for breast cancer if the data shows it doesn't and introduce it into interstate commerce. If you want to set up a little company to make Avastin entirely within the boundaries of one state, the FDA would have no jurisdiction.

The issue some people are bent out of shape about is that the insurance companies link reimbursement to FDA approval. So, if you want to pay for the drug, take it and pay for it. And complain to the insurance company for their policy of not paying for drugs that are not approved. That is not the FDA's fault or problem.

Even if it were, why would you want insurance companies to pay the big bucks for a drug that has very little value in terms of curing people of cancer or making real prolongation of life? You are being ridiculous if you think that this should be standard practice.

If you think your life is "priceless" then you should be willing to hand over all your money to eek out a few extra weeks/months. But I don't think your life is "priceless" if I am insured by the same carrier, and I am fine with drug companies saying no. Sorry. Pay for it yourself.

Permalink to Comment

29. Render on December 17, 2010 1:09 AM writes...

"Avastin, by the evidence we have, does not seem to help advanced breast cancer patients. It wouldn't help them even if it were free."

So your point is that the FDA did not really make themselves irrelevant by virtue of the fact that they're revoking approval of a drug because of its expense, but because their onerous and unbelievably costly drug approval process can actually result in approving drugs that aren't good enough (by their standards) for all people?

Permalink to Comment

30. Indiana on December 17, 2010 10:33 PM writes...

29: The FDA lowered their tradtional standards for drug approval for certain categories of diseases with very active interest groups who wanted quick access to drugs that were promising in at least some way. These would be incurable diseases such as cancer and HIV/AIDS. This is called being responsive to the public, by many.

So, Avastin got a tentative approval on the basis of some data to suggest that it was going to lead to survival benefit. The tradeoff for such approval is that the manufacturer has to do further studies to prove the survival is improved. When that didn't pan out, the approval was taken away. But, as before, you can use it; the manufacturer can't promote it. And an indirect consequence (not the FDA's fault) is that insurance companies won't pay for it. This works for me.

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