Long-time readers will know that I am skeptical of the FDA. Let’s ignore that for the purpose of this post. Now consider the following two quotes.
The FDA recommended unanimously that Avastin no longer be used to treat breast cancer, saying that the risks of the drug far outweighed any benefits.
…”Even though we have anecdotal information, we don’t have evidence that it prolongs survival or improves quality of life,” said Natalie Compagni-Portis, a patient representative and voting member of the FDA panel. In a series of four questions, the six-member panel voted across the board that the clinical trials conducted by Genentech did not provide evidence that Avastin prolonged life for breast cancer patients, nor did it improve their quality of life. The panel also recommended that FDA commissioner Peggy Hamburg should not continue to allow the drug to be used for breast cancer patients.
A strong statement from the FDA. Now compare:
Medicare will continue to cover Avastin for breast cancer treatment even if U.S. Food and Drug Administration Commissioner Peggy Hamburg decides to withdraw Avastin for such use, according to Don McLeod, a spokesman for the Centers for Medicare and Medicaid.
“The FDA decision, when it comes, does not affect CMS,” McLeod told Reuters.
How does this make sense? Does CMS have information that runs counter to the FDA’s information? If so, let’s hear it. Or is this just a turf war? What does this say about the prospects for cost control of Medicare?
IIRC:
WRT Avastin, it seems that a small percent of patients respond very very very strongly to it, but most don’t really benefit. The FDA doesn’t accept any evidence from studies that don’t fit its criteria so it ignores this. And, any study using standard regression techniques will show that at the mean, avastin isn’t that great, because we don’t know how to separate out the people who respond to it ahead of time.
Personalized medicine is one of the paths forward. What policies are blocking this path? Burn them down.
It’s not any policy. For the most part, we have no clue how to personalize in a way that works. Certainly not with Avastin.
Do any respond very strongly negatively? How long does Avastin need to be administered before doctors can say if a patient is responding or not.
The problem, of course, is that those small few who did “respond strongly” may have done just as well if they didn’t take Avastin. If you had looked at the placebo control group, you no doubt could have found a small subset that responded strongly to getting no Avastin. To prove the personalized effect (something I do believe exists in a lot of cases), you will have to construct different kinds of studies to tease it out.
Never mind. Derek Lowe, below, has already addressed this.
“What does this say about the prospects for cost control of Medicare?”
A great question, especially considering that Avastin costs about $90,000 per year for treatment.
The real question for me is, how much did health insurance companies influence the FDA’s decision?
So if the FDA removes approval shouldn’t the fact that Medicare is willing to still pay for Avastin just be irrelevant? Can a FDA-non-approved pharmaceutical even be sold or used legally in the US?
No, but in the case of Avastin, it is still approved for other cancers, so doctors can prescribe it off-label. Genentech simply can’t promote it in any way as a breast cancer treatment.
If a doctor prescribes off-label, is his subsequent legal liability higher? Just curious.
“If a doctor prescribes off-label, is his subsequent legal liability higher? Just curious.”
If he prescribes it ON-label, does the government assume the liability? (haha)
It depends.
The standard for medical malpractice (way oversimplified) is whether the doctor did something unreasonable. Its really tough to show that prescribing an FDA approved medication for its intended purpose in line with its indications is unreasonable (though it is not impossible, particularly if there is a clearly superior alternative).
Its somewhat easier to show that prescribing a medication for something it is not intended to counter is unreasonable. However even then, if its part of the standard care most doctors would give, the plaintiff is dead in the water.
“If he prescribes it ON-label, does the government assume the liability? (haha)”
The FDA effectively immunizes the doctor from a claim that the drug wasn’t effective.
But why would the FDA assume the doctor’s liability as a whole? Just because its approved doesn’t mean that the doctor followed the indications and directions, apprised the patient of the risks, diagnosed the condition correctly, etc.
a) it ain’t their money.
2) it’s always a turf war.
Doc Merlin, your comment actually sums up why the FDA didn’t approve Avastin for this indication. It does appear that a small number of people benefit from it – but that number is so small that we don’t know if it was Avastin that actually was the crucial ingredient for them. Meanwhile, everyone else (the whopping huge majority) are wasting their time, their hopes, and their money (or their insurance company’s money) by taking it, and exposing themselves to some pretty rough side effects at the same time. For nothing.
If we knew that there really were good responders out there, and if we knew how to find them a priori, that would be a much different situation. Genentech/Roche has a strong motivation to find such people (as do almost all companies with oncology therapies), but it’s been a lot harder than it looked a few years ago. (I’ve seen some of the biomarker efforts myself during my time in the industry).
The FDA is charged with risk/benefit analysis, and in this case, it’s actually pretty clear that if you’re going to have an agency that does that, then it should say “No”. We can argue about whether the FDA should exist, or what criteria it should use – but by the criteria it uses now, this was the right decision.
This is the FIRST generation of anti-angiogenic drugs. The FDA basically says you can’t have large-scale human testing. Fine, but what does that leave us with? Not much.
Q: What can we possibly do? A: The people who don’t respond to first-generation are going to die. Fully inform them and pay them to participate.
Actually, the decision was to give it fast-track approval in 2008. So, they were for it before they were against it. Maybe they did adhere to their internal principles, but even that is not a simple question.
Andrew, the 2008 approval was based on the preliminary data, which did look positive. Larger studies didn’t bear that out: the Avastin-treated groups died at the same rate as the groups that didn’t get it. This happens all the time, and when it does, physicians and regulatory agencies should (and should be able to) change their minds.
The FDA is most definitely not saying that you can’t have large-scale human testing of these drugs. It is saying, though, that when you do such testing and they don’t work, that you should act on that.
>>What does this say about the prospects for cost control of Medicare?>>
It says we really need some independent board to advice on these payment issues. Perhaps we could call it the “Independent Payment Advisory Board” or IPAB. If only someone could get a law passed that included such a feature.
+1
“This time it’s different” says the reality denier. “IPAB is totally unlike all those other independent boards and formulas.” “It will be immune to politics, the thing that prevented the Sustainable Growth Rate from taking effect.’ “It will be immune to the complaints that caused HMO cost control to fall apart.” “Patients will easily accept the idea that a faraway board is not letting their doctors prescribe what they think would be best.” “People will suddenly accept the tradeoff of saving money instead of doing everything we can to save lives, when they’ve noisily rejected it up until now.”
“independent board” – independent from whom? The Earthlings? It seems to me this is the problem with some people – they see something magical in collective action, or in government. Like the old joke about the English: “One Englishman, an idiot; two Englishmen, two idiots; three Englishmen – a great nation!”
Right, because health and medicine works by majority rule.
And we also know that, based on other existing free markets, companies would never, ever try to market something that in essence has no positive effects for 9x% of the general population as being beneficial to them.
You anti-government types have a few good arguments for reducing the size and scope of government but many of them fall flat when met with real-world situations that are obvious to the rest of humanity.
blah,
I think you don’t quite get it. Of course the companies would try to market it. Although why they would choose to go to the trouble and billions in development costs to defraud customers tells me it is you with the oversimplified narrative.
That is after all why we have doctors who read medical journals. Doctors can still prescribe it off-label.
Of course the companies would try to market it. Although why they would choose to go to the trouble and billions in development costs to defraud customers tells me it is you with the oversimplified narrative.
?? I don’t get this at all. After billions in sunk costs that determine the drug does nothing except unpleasant side effects, but which give them federal permission to market it, why would they not market it? The remaining costs are much smaller, and if it brings in more than those costs, they come out ahead relative to not marketing it.
Assuming that it doesn’t turn out to have unexpected flaws that open them to class action lawsuits and horrendous future expenses. But that could happen with any drug. And the natural behavior is to ignore the possibility since it’s hard to quantify. And isn’t it harder to do class-action lawsuits now? So whatever the risk is, it’s lower than it used to be when they were doing OK.
Without the FDA, there *are* no billions of dollars in sunk development costs. We’re back to snake oil sales, late night TV and internet marketing style. You want your senile parents buying their meds from a guy like Kevin Trudeau?
Look: if you’re facing a prognosis of death by cancer in a year, do you really have the time or wherewithal to fuck around with untested shit or would you rather try something that has, say, passed clinical trials as verified by an unbiased third party peer group?
I welcome you to explore the former in your personal life but most sane adults wouldn’t go that route.
(Obviously if there is no tested drug on the market and all that’s available is something in development or in clinical trials, then yes, experimentation might be all that’s available to you and my opinion is that you should be free to pursue that.)
If “most sane adults wouldn’t go that route,” then you’re saying that requiring FDA approval is hardly necessary, because most sane adults would wait for FDA approval or something similar even if it were voluntary. Insurance companies would also wait for such evidence before approving things. (But Medicare might depend on majority rule.)
Doesn’t your argument depend on far too many people able to go the snake oil route, plus insurance and Medicare/Medicaid being willing to charge the rest of us for their decisions?
The latter is actually important, because it doesn’t matter too much if people waste their own money on excess Vitamin C, or D, or what have you. However, when you have a situation like in the UK, where the NHS will pay for homeopathy with tax dollars, that does create a problem for all of us.
One of the silly things with the current regulatory system is that it’s illegal to actually mention real scientific evidence or math if you’re not FDA approved, but you’re free to offer vague claims, so long as there is absolutely no science, math, or hard statements behind them. With only optional FDA approval, things outside the system would be free to offer real science in support of their claims and then– as you claim– most sane adults would prefer the ones with real science.
But IPAB members aren’t going to be elected.
Oh wait. That makes them “unelected, unaccountable, bureaucrats.” So there is no way to have a board that makes intelligent data-based recommendations about medical treatments? According to some, I guess not.
I’m having a lot trouble reconciling the simultaneous arguments of “there’s no such thing as death panels, they’re just a crazy right-wing invention” and “we really need death panels, they are going to be totally awesome and stuff.”
Saying “We’re not going to pay $90,000 to give Grandma something that doesn’t appear to do anything except make her feel worse” and “We’re going to let Grandma die” are two very different things. That you call the first “death panels” is ridiculous.
But the IPAB isn’t just the first. It’s unrealistic to claim that the decision is between two equal things, one cheaper, or to use or not use something that doesn’t do anything at all.
The real case, and the hard decisions, are between treatment A and treatment B, where A is more effective but also far more expensive. There are also cases where “X is really expensive, and it probably won’t work, but it might work, and it’s the last hope we’ve got.” Those are the decisions that are made in other countries, and that the IPAB will make. Sometimes that will mean letting someone die sooner because keeping them alive longer would be really, really expensive.
It’s ridiculous to pretend that those aren’t the kinds of decisions that the IPAB will also face.
But those decisions are being made already, aren’t they? In what way will IPAB lead to worse decisions?
Not really, not in Medicare as this example shows. What the American people have demanded is that insurance covers everything that possibly could help, regardless of cost. When private insurance makes similar decisions, private insurance gets similarly attacked.
I’m not arguing that. What I’m arguing is that IPAB won’t work. The evidence is extremely strong that it won’t work, since things, like CMS, designed to do the same thing already fail.
IPAB won’t control costs, and pretending that it will is silly and unrealistic. Therefore, claiming that Obamacare will actually get costs under control is silly and unrealistic. As soon as it starts to bite, it will be relaxed, just like the Sustainable Growth Formula and its “doc fix.”
The PPACA supporters have traded a guarantee of more spending now for a vague promise of less spending later. The vague promise will never happen. Therefore, contrary to the hopes of its wonky supporters, Obamacare is hopelessly financially irresponsible.
But if IPAB is supposed to do both (as it is) then this is more like saying “death panels are going to be totally awesome and stuff.”
+1000, rather
“This time it’s different” says the reality denier. “IPAB is totally unlike all those other independent boards and formulas.” “It will be immune to politics.”
Your sarcasm meter is totally broken. It is quite obvious that foosion made his comment tongue-in-cheek.
No, it says that the IPAB won’t work, and anyone who thinks that it will is being foolish. The CMS already has independent boards; they don’t work.
This is evidence that all those who believe that PPACA actually will save any money through IPAB are deluded.
We are approaching unmanageable complexity in health care in the US. With so many insurers, so many administrators (every company has one!) so many decision makers at every level, there are too many decision points that can be influenced without scrutiny. Simple systems, like a national government, tend to be transparent. A health care market with multiple government actors at multiple levels and hundreds of self-interested companies, cannot be transparent. And you can rarely lower costs in a system by increasing complexity. I’m not talking regulation – well regulated systems work well, I’m talking complexity that serves no purpose.
Our system has gotten to the point that we invite obfuscation, like health care savings plans and individual state oversight. It’s not a winning strategy. Health care should be treated like infrastructure, because 1) it’s necessary 2) it’s a public good 3) it affects every other market 4) has unique geographic concerns (hospitals, clinics, etc.).
RE: “Health care should be treated like infrastructure because … It’s a public good”
Hmmm… seems pretty excludable to me (if nothing else see doctors refusing to take Medicaid patients). Also seems pretty rivalrous as there’s only so many hospital beds and doctor hours in a day.
You mean to say that there are positive externalities to healthcare, which if it were underprovisioned would be a rationale for subsidizing it, but it’s clearly not underprovisioned. In fact I say that healthcare is overprovisioned (see Rand studies and others). But most everything has *some* positive externalities to it though negatives can outweigh positives (even murder reduces congestion), so I’m not certain what wouldn’t fit you criteria.
But even then, just because something is a public good, that’s not generally a good rationale for public provisioning in and of itself as private provisioning of public goods will occur when there is still sufficient returns to the provisioner to make it worthwhile to do so privately. For example, there’s a guy about 15 miles south that drops about $200k every year on a fireworks display for the 4th. In comparison to healthcare, watching a fireworks display is not rivalrous as my watching it does not stop you from watching it, nor can I easily stop you from watching the display either (non-excludable).
“Simple systems, like a national government, tend to be transparent.”
I, I don’t even know where to start with that. I could start by questioning whether the national government is “simple.” I could probably more easily question whether government is “transparent.”
Didn’t Obama run (partially) on a platform to make government transparent? How’s that been working out?
http://www.washingtonpost.com/politics/is-the-door-closing-on-open-government/2011/06/30/AGiY3YsH_story.html
http://www.nytimes.com/2011/06/27/opinion/27stone.html
http://www.bloomberg.com/news/2011-06-29/lawmakers-chafe-at-secrecy-surrounding-deficit-reduction-talks.html
et al.
Wait, you’re arguing that since there are “unique geographic concerns” that it should be treated exactly the same in all geographic areas, with a one size fits all system?
@Derek Lowe:
Yes, basically we have a massive endogeneity problem in the data that is used. Now, because the state pays for it, this turns what would have been a private good into a rather huge public good problem.
This should have gone the other way. If you want to try it you should be able to, on your own dime. Medicare should be the institution that is complaining about the cost vs. effectiveness.
AFAIK, Medicare does not distinguish cause from treatment. If a drug or treatment is allowed it is covered; appropriateness is left to the doctor/patient. Medicare also has a history of supporting trials even with only experimental approval would would likely support trials aimed at identifying appropriate subjects for it. The FDA decision will have much more effect on doctors likelihood to prescribe such a treatment.
“And oncologists will still have the option of prescribing Avastin “off label” to their patients with metastatic breast cancer.”
http://health.usnews.com/health-news/managing-your-healthcare/cancer/articles/2011/07/01/avastin-rejection-supported-by-cancer-experts
The way it “makes sense” is that the FDA regulates marketing/advertising while doctors still (nominally) have the authority to prescribe the treatments they deem appropriate and the insurance programs reimburse the doctor’s decision.
What about Medicaid? Do they have an independent stand or just toe the Medicare line? Not sure how exactly the system works……
The non-expert commentary on Avastin has been overwhelmingly misleading. There is no evidence whatsoever that Avastin works better for some people than others. What the data show is that persons taking Avastin have variable outcomes that are not significantly different from the control group. Some people do quite while on Avastin, but, because the same can be said of those in the control group, there is no good reason to believe that Avastin actually works.
Admittedly, it is possible that some group of patients, if isolated, would do better on Avastin. For this to have escaped notice, however, it would also have to be the case that an essentially identical number of patients do worse on Avastin than on the control treatment. Everyone readily concedes that this is a possibility, but it is a most unlikely possibility indeed.
The other thing that non-expert commentators are getting wrong is the significance of “progression-free” survival. In the most recent study, Avastin did slightly slow tumor growth, leading a longer “progression free survival” rate. Based upon this, people have made the claim that even if Avastin does not extend life, it produces a higher quality of life for cancer patients taking it. This would only be true if a slower growing tumor meant a higher quality of life. Unfortuantely, ithe most recent study directly contradicted this, finding that Avastin patients — slow growth of tumors and all — had a lower quality of life due to Avastin’s side effects.
However you feel about the FDA, it is quite clear that the FDA’s opponents have intentionally distorted the evidence on Avastin to advance their political agendas.
The other way of saying it is that there is no evidence that there aren’t a subset of people who respond well.
Trying to disprove a negative, Andrew’?
I think you have me confused with other people.
Because, we don’t get to to do experiments on people where we say “you are the control who are going to get just the unproven drug so we can get good statistical power.”
Rather than being highly unlikely, I suspect it is quite likely that some respond favorably and some respond poorly because that’s how our bodies work.
“Specifically, Paclitaxel plus Avastin significantly prolonged progression-free survival compared to paclitaxel alone (median, 11.8 vs. 5.9 months; hazard ratio for progression, 0.60; P<0.001). Adding Avastin to paclitaxel also increased the objective response rate (36.9% vs. 21.2%, P<0.001). Unfortunately, however, the overall survival rate (median, 26.7 vs. 25.2 months; hazard ratio, 0.88; P=0.16). Complications were also more frequent in the group receiving Avastin."
"Although the increase in PFS was impressive, the lack of effect on OS was very disturbing, strongly suggesting that drug toxicity was "taking back" all the benefit the addition of the drug was providing against the cancer. "
So, since the overall survival is due to the complications and not lack of effectiveness that means the actual questions to answer (A) which are which and (B) what is the cause of the poor response and can the drug be improved?
"However you feel about the FDA, it is quite clear that the FDA’s opponents have intentionally distorted the evidence on Avastin to advance their political agendas."
It's not about how I feel about the FDA. I separate those feelings quite easily. I doubt the statistics are as tidy as you present in your narrative and I suspect things like the cost of the drug and politics has muscled it's way into the "science" due to cost, complaints against Roche/Genentech, etc. I wasn't in the meetings, so no I can't present proof.
http://www.sciencebasedmedicine.org/index.php/avastin-and-metastatic-breast-cancer-when-science-based-medicine-collides-with-fda-regulation/
"Over the last decade, Avastin has tended to be a magnet for controversy. The major reason that Avastin has attracted a lot of attention is that it’s so expensive (up to $100,000 a year per patient), and part of the reason it’s so expensive is that it is a protein rather than a small molecule."
But what does the fast track approval followed by revocation mean? Does it mean that the FDA was operating purely as they should? That's one possibility. My guess is that they are simply re-asserting their monopoly authority over the power to do human experimentation because of politics and not because they are doing everything scientific like.
Good reply. I think cost is a major factor in the debate, but I don’t think that it influenced the FDA. Avastin has not been shown to work for breast cancer patients. This is the bottom line. What has happened is that a number of patient groups, relying on anecdotal evidence, have joined forces with people ideologically opposed to the FDA to advance their disparate causes. Again, people who claim Avastin works for some patients have literally no evidence whatsoever to support this contention.
To say that Avastin probably works for some people, because “that is how our body works” is a weak defense. The variability of outcomes was quite similar to the control. If Avastin really did work for some people, this would likely have been observable in the data. There would have been a larger group of patients doing well on Avastin than those doing well in the control group. This was not observed. The most likely — but not only — explanation for this is that Avastin simply does not work for breast cancer patients. Sure, it is possible that toxicity is completely masking the positive outcomes, but, really, how likely is this?
Sure, it is possible that toxicity is completely masking the positive outcomes, but, really, how likely is this?
A priori, not unlikely at all. The usual problem is that cancer cells usually metabolize fast, so any drug that kills cells which metabolize fast will have some effect on the cancer. But it’s likely to also kill important cells in your immune system, and heart, etc. Does it affect the cancer more than it affects the cells you want to keep? Maybe yes, maybe no. This is the usual course of events.
There’s no reason to approve this drug for general use. There is no evidence that it does more good than harm. The chance that it does enough more good than harm is too low to justify its expense.
But if the originators have some theoretical reason to think it ought to work, maybe they could find an improved version, or find palliatives that they hope will reduce the toxicity, and try again. We really ought to allow as much experimentation as anybody wants to fund, as long as there are cancer patients available who aren’t already in experimental groups or control groups.
It isn’t unlikely that the toxicity masks some of the benefits, but rather it is unlikely that the toxicity masks all of the benefits. Moreover, Roche made no effort to demonstrate that Avastin worked better on certain groups of patients than others. If there really was some prospect that this was true, why would Roche not have bothered to investigate it? I know that some people here think this is obviously true, but it was apparently not obvious to the very people who discovered the drug in the first place.
What everyone who supports Avastin is doing is looking at certain specific patient outcomes and claiming that, for some people, Avastin works. You would think that a bunch of smart people on an economics blog would realize that this is poor reasoning.
Rather than being highly unlikely, I suspect it is quite likely that some respond favorably and some respond poorly because that’s how our bodies work.
Using this methodology, I can show that in some people, ordinary water cures cancer.
There is also a problem with how we are going to define whether or not a drug is effective. Avastin did show a small amount of improved progression free survival, but overall survival was not changed. While I prefer the more verifiable, and useful IMO, endpoint of overall survival (or improvement in quality of life), others may differ. We then need to decide if we can afford this different endpoint. (Nice editorial in NEJM 6/27 on this)
This is a problem we have faced and will continue to face. It is one of the reasons why private insurance companies do not control costs. All it takes is an emotional appeal by the right people, and data do not matter. Insurance companies fold. It is a problem with Medicare now, but at least it usually pays less.
Steve
One key problem is that approval or denial helps determine if a drug company can continue development of a promising new strategy. Avastin is both a new strategy (anti-angiogenisis) as well as a protein rather than small molecule. If progression free survival is improved and overall survival is largely unaffected, that would be good enough for me to continue looking into it. It is apparently more clearly beneficial for other types of cancer and I wonder if the original application to breast cancer was a shot in the dark at scale (and the over-emphasis that breast cancer gets).
Another question is how much would the drug cost if it were used at scale? It is a protein rather than a small-molecule which is part of the cost. But since the low-hanging fruit of small molecules has already been largely picked, the protein path is one way forward.
In addition to the lack of “personalized medicine” granularity in the process, there is also no such thing as “breast cancer.” There is Her2+/- among others. So, requiring a drug to cover “breast cancer” is the wrong direction. I noticed that in the followup tests they actually tested the drug against different cancer in combination with a different chemotherapy. This may be Roche’s fault swinging for the fences, the FDA’s fault in having over-generalized categories, or someone else’s fault entirely.
It’s not clear what mechanism the blogger prefers for weeding unsafe drugs out of the pharmaceutical pool but I’d don’t trust self-regulation nor do I want to play guinea pig.
In cancer, especially, a high percentage of drug therapy is off-label. Normally that’s because no one is willing to pay for the numerous clinical studies needed to get all of the possible uses on the label. There is a robust scientific literature that bypasses the FDA, but is used by oncologists to decide on treatment options.
In other words, it would be, in a practical sense, impossible for CMS to deny reimbursement for cancer treatments simply because they are off-label uses. To do so would deny reimbursement for many standard-of-care treatments.
This makes it difficult at best for CMS to deny reimbursement for Avastin simply because it is off-label.
We need, but do not have, a mechanism for Medicare to make reimbursement decisions based on the best available science. The UK has NICE, but that is held up as the epitome of “death panels.”
There are real reasons for the off-label use some of which you align, then you conclude we need to not reimburse for off-label use.
I don’t get the logical leap.
There is a robust scientific literature that bypasses the FDA, but is used by oncologists to decide on treatment options.
If the literature is good enough for the oncologists why isn’t it good enough for the FDA? Does the FDA need to change its approval protocols?
Not all the literature is as “robust” as you say. Rather publications vary in quality all over the map. A single small trial of 10 patients where 7 respond to the drug is not the same as a multi-center trial of 1000 patients where maybe only 10-20% respond. It’s the job of the FDA to insure that the overall data supporting approval is robust enough to assure that the drug will be safe and efficacious as per the approved indications.
A single small trial of 10 patients where 7 respond to the drug is not the same as a multi-center trial of 1000 patients where maybe only 10-20% respond.
At a minimum, when we use off-label drugs that there is inadequate data for, we should report the results. When there has not been enough of a study done, but the drug is getting used anyway….
Ten thousand uncontrolled cases doesn’t give you anything like the same result as a real controlled study. But it’s available. All it needs is the reporting, and we might learn about some ineffective drugs, and about some that have dangerous consequences.
Since we’re going to be using drugs that are for all practical purposes untested, we might as well try to track what happens in practice.
And it will never happen, because Medicare is subject to politics and majority rule. Just see what happened with the recent advisory board decision on mammograms.
But of course it’s illegal for many people to talk about this literature, or even to direct people to it, even if they scrupulously say nothing untrue and stick to the scientific facts. Truth and science is no defense from the FDA.
There is no issue here. A regulation was put in place about two decades ago (I was working at a trade association at the time and filed comments on it) that says drugs that are generally accepted into practice as outlined in a compendium (at the time there were three major ones but I think this has changed over time) it must be reimbursed at the Federal level. This is why a lot of off-label cancer drugs (this is really not limited to Avastin!) are reimbursed by Medicare. More critically is whether the MDs who keep these references up to date change their decision and state unequivocally that Avastin is not a valid treatment. AstraZeneca went through the same thing a couple of years ago with Iressa (gefitinib) for small-cell lung cancer. While it may help a small subset of patients, the drug was withdrawn from the US market in 2009. I suspect FDA will force a label change for the breast cancer indication and the ‘correct’ thing for the company to do is make it available under a treatment IND for all physicians who wish to use it for treating this condition. Of course they would not be permitted to sell this at market price (but could recoup manufacturing costs under the regulations). This would allow access and the company could continue to collect clinical data to show if the drug does or does not work. Any other approach is quite frankly unethical business practice.
Yeah, what Orange 14 said. The law’s the law and it says that if a drug is deemed safe and effective by one of the major compendia, it must be reimbursed by the federal programs, e.g. Medicare, and that the sole reason that Medicare will reimburse for the drug.
As an aside for those who ma seem taken aback that doctors use drugs for conditions not approved by the FDA, this is a very common situation. It costs MILLIONS of dollars (maybe more) for a drug company to get FDA approval for a new drug for a particular indication and that’s the usual initial process – a drug gets approved for a specific condition. Once approved for that condition, researchers and doctors may find that it also works (as in the case of Avastin) for other conditions. Rather than paying millions of dollars again for an FDA approval for the new condition, doctors use the drug “off label”, i.e. for a condition not specifically approved by the FDA, but also not necessarily disapproved by the FDA (since they didn’t look at it for this other condition). Indeed, there are many commonly used drugs that have never been FDA approved such as aspirin. If aspirin was invented today, it’s unlikely that it could get an FDA approval.
“The panel also recommended that FDA commissioner Peggy Hamburg should not continue to allow the drug to be used for breast cancer patients.”
It would actually help if we started using words that made sense. They cannot not allow the drug to be used. That’s not what they regulate.
The claims in these comments that Avastin doesn’t benefit a significant number of women are simply not true. This decision is about a mid-level FDA bureaucrat trying to advance his personal regulatory agenda. readers of this blog need to know that FDA does not dispute that Avastin increases progression free survival for a substantial number of women – an efect all of the practicing brast cancer experts who spoke at the hearing say is in fact valuable clinical benefit. For the record, there were no oncologists on the FDA “expert” panel that treat breast cancer patients. CMS does not have different information than FDA, but they, the European Medicines Agency (FDA’s counterpart in Europe) and the National Comprehensive Cancer Network (a network of leading, practicing oncologist/clinical researchers that make practice recommendations for the field of oncology, and on which CMS relies for coverage decisions) both strongly disagree with the FDA’s Office of Oncology Drug Products. The EMA, looking at the same data, have endorsed continued approval and use of Avastin in two combinations for first line breast cancer. The NCCN also has endorsed its continued use in the US. The FDA is becoming increasingly isolated in its very poorly supported position on this. The FDA is wrong on every count, including its application of the Accelerated Approval provisions put in place by Congress in the Food and Drug Administration Act of 2007. in fact, this attempt to remove the indication for first-line metastatic breast cancer is more about a campaign by Dr. Richard Pazdur, Director of OODP to eliminate Accelerated Approval as a viable approval pathway, than anything else. As for the “panel” vote, stay tuned. Every member of the panel was placed on ODAC by Richard Pazdur – to rubber stamp his campaign to eliminate Accelerated Approval. Stay tuned. This action by FDA is probably going to result in some major changes at FDA. The severe problems with the leadership and policies advanced and rigidly enforced by the FDA’s cancer drug office over the last 8 years is finally getting some light on it. It well could, and we think should, cost some people their jobs at FDA; among them Drs. Richard Pazdur and Patricia Keegan, the two managers in OODP who have crafted and advanced the agenda to eliminate Accelerated Approval. It remains to be seen if Dr. Hamburg should keep her job, but the fact that she let her subordinates conduct a rigged hearing (and it was rigged – stay tuned) and arrive at a recommendation that virtually no other oncologic decision-making body, including CMS, its sister agency within HHS, makes it clear that FDA, Pazdur and Keegan are on an island pretty much by themselves. If Hamburg doesn’t quickly reverse her agency’s decision and take visible corrective action, which really must be of the personnel kind, then she deserves to lose her job. As a political aside, the President really can’t have this kind of nonsensical regulatory decision-making going on at the FDA if he expects to have a second term. If this decision, and the people who delivered it at FDA, remains in place – the President will face opponents who can point to a real “death panel” problem within his administration. This is far from the only good cancer drug the OODP is mishandling right now. Because of their regressive policies, they are mishandling virtually all of the cancer drugs now in the pipeline. Absent an expedited and decisive response to this mistake, more will surface soon. The President should know that Drs. Pazdur and Keegan are not flexible people, and will not stop heading in the wrong directions they have chosen simply because one of their decisions gets neutralized. Despite recent simplistic studies to the contrary, cancer drug development and approval in the US has been dramatically slowed and disincentivized by Drs. Pazdur and Keegan, and that has been allowed to occur despite sharp criticism and pressure from many, very well-informed stakeholders, including patients and their advocates, over the last 8 years. More than a million American cancer patients have died prematurely while waiting for FDA to get out of the way of obvious progress. Now Drs. Pazdur and Keegan are extending their campaign into the removal of useful cancer drugs from the market. We have already lost our momentum for advancing the fight against cancer, and the EMA have Asia have taken it. Their regulatory systems are speeding up, modernizing and recognizing the promise of our advancing knowledge of the molecular biology of cancer. We have slowed to a near complete stop, and are now going backwards. All because of two tenured government bureaucrats in mid-level positions at the FDA – bureaucrats the last 6 or 7 (yes there really were that many during the Bush Administration) FDA Commissioners and Acting Commissioners didn’t have the courage to do something about. It’s time for Congress to weigh in – specifically in the House Energy and Commerce and Government Oversight and Reform Committees. No director of OODP should have the enormous, unsupervised and as a result, unchecked power that Dr. Pazdur has accumulated. At this point, when the entire oncologic regulatory community has come to a conclusion opposite to his, Dr. Hamburg should take this opportunity to make the personnel changes we have needed in the leadership of OODP for so long. It will take years to undo the damage. The sooner we get started, the better.
Final word on the hearing. It was a sham. It was fixed. It really was. As this issue moves forward, that will become obvious. Again, stay tuned.
Steven Walker
Co-Founder, Abigail Alliance for Better Access to Developmental Drugs
Lots of ad hominems without much information on the merits of the drug itself. Libel?
The FDA and EMA have often disagreed on regulatory decisions and will continue to do so; it’s the nature of evaluating clinical data (and safety data as well since there are often different safety labels on products as well). Dr. Pazdur is not trying to eliminate the expedited approval process and in fact has done quite the opposite, made significant reforms in the type of data that can be used to support an approval. There are as many advocates and patients on the opposing side on this issue (count the very large number of women who were treated unsucessfully with Avastin in that number). Dr. Pazdur is not unsupervised; his Office reports directly to Dr. Woodcock who is head of the Center for Drug Evaluation and Research nor is his power unchecked. It might be best if you sit down and take a serious look at the clinical data before casting aspersions on good people.
In essence, the payer is the control on whether or not the treatment is denied or allowed. As such, that payer is the one that bears the public outcry when desperate and dying women tell their stories to Dateline about how the cruel insurance company or faceless bureaucrat at Medicare refused to allow them a chance at survival. This is why CMS is going to continue to support payment for Avastin’s off-label usage. Anyone that believes IPAB would make a different choice is naive.
So, when IPAB converts breast cancer treatment into a single payment based on coding for various fairly objective criteria, say size, number, intrusion, genetic profile of patient and of biopsy, duration, or whatever doctors claim guide their treatment and that indicate the severity, with a bonus for remission for each year of remission?
If the provider pays for Avastin as part of treatment, will there be no cost control?
Let’s say the provider gets $100K bonus for 1 year remission, another 50K for the second, $10K for each of the next 3 years, based on the claim that $90K for 1 year of Avastin will work effectively?
Will the provider figure out who, if anyone, will benefit from Avastin?
So, why hasn’t CMS or any private insurance provider done this already? They’ve already had the power to do so. Perhaps because it’s difficult to do so and too easily gamed, perhaps because politics can eventually alter the decision of any independent boards.
More hand-waving magic. IPAB won’t work. There already are “independent boards” charged with the same sorts of responsibilities. They don’t work.
But of course, “this time it’s different” say the people who deny reality. IPAB supporters are simply not reality-based.
More hand-waving magic. IPAB won’t work. There already are “independent boards” charged with the same sorts of responsibilities. They don’t work.
The political process does not allow any solution that would work. When somebody truly thinks they will die unless whoever gets to choose pays, and it does not come out of the chooser’s pocket, and the chooser looks like a big public grinch who kills people, they will choose to fund that medical care.
There are still ways to keep costs from rising so fast. One is to cut the rate of medical innovation. Cut the patent protection and increase the R&D costs. At some point private companies will stop creating new procedures designed to maximise profits (and costs).
The government can still fund academic or other research that looks likely to cut costs.
If the USA gives up the leading role for medical advance, and nobody else takes up that role, then the economy is better off.
Something that could reduce medical costs faster is another world war. Once we have a big long war going, we draft a whole lot of the MDs and the rest get supply shortages. Somebody comes along wanting complex expensive care, and they say “Hey buddy, there’s a war on.”.
After the war, things will be so different that we won’t pick up where we left off. New problems, new solutions, I can’t predict ahead of time what would happen at all.
The Wikipedia article on Avastin says treatment costs $100,000 / year in the US and CA$40,000 in Canada. Are the prices really so different just across the border? What gives?
Other than the US every other country has negotiated price controls for pharmaceuticals.
In theory, what prevents American insurers from buying Avastin on the Canadian markets? Maybe via shadow buyers? Isn’t this a good arbitrage opportunity? Would FDA again prohibit this?
It is against the law for anyone other than the manufacturer to import from another country. Additionally, there is the distinct probability of getting counterfeit product. Companies might also limit access to the drug if they see it being diverted to the US market.
We get to pay for the R&D because other countries exercise monopsony price controls that only allow unit-variable profit on a post-research basis. (On the plus side, Nature has reported that as a result of this, we get new drugs about 18 months sooner on average.)
Probably we’ll start doing it too eventually, and then everyone will be baffled as to why so few new drugs are being invented anymore. The solution will be to have the government do that, too.
The future is now. People are already baffled by why no new drugs are being invented anymore.
This is a convenient talking point, but there’s no evidence of it. Even your Nature name-drop makes little sense, since drugs developed in the US are still able to enter other markets. Our FDA regulatory structure likely accounts for that 18-month difference, rather than Medicare reimbursement subsidized R&D. Furthermore, pharma profits are quite robust compared to every other business sector, and their underinvestment in R&D compared to marketing is their own choice. Finally, we already give them a HUUUUUUUGGGGGGEEEEEE subsidy through NIH-driven academic research, which creates most of the basic knowledge upon which their products are based. If you are concerned about the future of drug R&D, expand translational NIH partnerships with business and academia rather than subsidizing corporate profits through clear over-payment.
Both R&D and marketing are designed to be profitable. They are complementary. If you can market drugs to make them sell more, then you’ll invest more in R&D to create them.
If marketing isn’t independently profitable, you’re doing it wrong.
Sorry, no:
Nor was this ‘drug gap’ due to faster FDA processing: both agencies have an identical mean approval time of 15.7 months. Instead, said Kaitin, drugs hit the US market first because the sponsors choose to submit them there first. The advantage of the US is almost wholly down to its lack of price controls, says Kaitin. “Investors tend to invest in places where there is less control over prices, and it is always better to do your clinical trials in the countries where you plan to market,” he says.
Price controls lead to shortages.
You also are missing the difference between basic research and productization. NIH does not do the latter, and all the basic research in the world has no practical value without it.
Although, on the plus side, you did fall neatly into “The solution will be to have the government do that, too.”
The short answer is that CA gov., as well as EU ones basically tell the pharma company: “this is how much we are willing to pay for this drug – take it or leave it”. In the US, the price is set by the company, taking into account possible public blow-back if the price is deemed obscene, but also the manufacturing and development costs etc.
I’m kind of bemused by the argument that we should go ahead with this because it’s “first generation”. Haven’t we seen that before? Has it ever worked?
Like, we require all US drivers to use “gasohol” which has no value, because we want to encourage development of a gasohol that would actually have value. Wouldn’t it be better to instead fund research to create something better?
And we had a missile-defense system which did not work, and we were ready to deploy it anyway. “Maybe someday we’ll have missile defense that does work, and we’ll get it faster if we pay for lots of missile defenses that don’t work first.” Again, wouldn’t it be better to do the research?
I just don’t get it. It’s like, we need automobiles that run on water as their fuel. So if we buy a million automobiles a year that are advertised to run on water, that don’t work, will that make it more likely that we get an automobile that really does run on water?
I just don’t get it.
Yes, all the time in the private sector. When the government mandates that we use things, no, it doesn’t always work out. But plenty of things that were fairly bad in their first generation form have been bought in the private sector by early adopters and gotten better.
Automobiles (before the Model T), Cell phones, computers, etc.
John Thacker, when things actually work well enough for people to use them, and sell at a price people can afford, then they have the chance to get better — and make money in the meantime.
When they don’t work, but the government pays for them anyway, is there any reason to suspect they might get better?
This is an example of a product which does not work but which government pays for anyway.
Oh, I completely agree with you. That’s why I said there’s a difference between things in the private sector, but not when government mandates something.
Well, what about things that sell to the private sector but which do not work?
For example, Laetrile. Here was a cancer treatment which apparently did not work, but it sold very well to the public. Even illegally.
Did the market for Laetrile result in Laetrile getting better at curing cancer?
Genome sequencing is another thing that has gotten much, much cheaper since it’s first generation. It was way too expensive to think about covering initially, but the price drop in just ten years has been astounding.
The decision to pay for Avastin reflects a middle ground position that tries to incorporate Republican objections to cutting Medicare costs. From wikipedia on IPAB: “During the health care reform debate of 2009-2010 that ended with the passage of the legislation by the Democratic-controlled Congress, Republicans proposed striking provisions in the bill that would require Medicare cost control.”
IPAB doesn’t issue its first recommendations until 2014 but the expectation is they will be for more integrated treatment and payment, which for something like breast cancer will eliminate payment for the cancer drugs themselves, but instead pay for a drug based cancer treatment program, and the provider will pay for the drugs as needed.
So why doesn’t CMS do this already? Why haven’t private insurance companies, if people like you already know the answer? Insurance companies and CMS have no shortage of independent boards.
IPAB will suffer the same fate as the Sustainable Growth Formula, no matter how much you attempt to deny reality.
Alex, see these:
http://www.nejm.org/doi/full/10.1056/NEJMhpr0807774
http://www.cms.gov/CoverageGenInfo/02_compendia.asp
Comments on this entry are closed.