FDA Tiered Approval

have an interesting op-ed on a tiered approval regime for the FDA:

Instituting a codified approval paradigm based on four tiered levels of clinical effectiveness (biomarkers, clinical signs and symptoms, disease modification and clinical outcomes) — with evidence regarding clinical utility progressively increasing — would greatly reduce the regulatory uncertainty and subjectivity, as well as the time to approval of innovative medicines.

Moreover, the four tiers, coupled with a commitment to apply state-of-the-art technologies (Apple Watch, telemetry and other health monitoring systems) to obtain clinical evidence would allow for additional learnings from use of drugs by practicing doctors treating real world patients. This knowledge would unearth additional uses, information that can be added to the product label to allow safer and more effective use of drugs and the identification of drug combinations that lead to even greater health benefits.

See also Bartley Madden’s work on Free to Choose Medicine which would similarly create dual tracks, one the standard FDA process and a second observational track that would bring drugs to market more quickly with the tradeoff being fewer clinical trials. As clinical trials rise in expense and more treatments are targeted towards smaller patients groups (i.e. personalized medicine) and as statistical techniques improve, we will need and can benefit from reforms to the FDA process along these lines.


The long process for drug approval causes unnecessary deaths. There should be a better way to get new drugs to people who can benefit from them.

Nice to see the Bartley J. Madden Chair in Economics at the Mercatus Center not being shy about promoting Bartley J. Madden's work.

Which is certainly better than when only a select few could read about Bartley J. Madden in the first place. One assumes that this link is unobjectionable these days, as is its opening sentence - 'Bartley J. Madden is currently an independent researcher and a policy advisor for The Heartland Institute.' http://freetochoosemedicine.com/?page_id=13473

And the last sentence, of course - 'His work in public policy has resulted in the Free To Choose Medicine plan, which was developed in journal articles published in Regulation, Cancer Biotherapy & Radiopharmaceuticals, Medical Hypotheses, and Engage.'

By Jove, c_p, you've finally done it! You've opened my eyes with this comment!
Obviously, Alex and Tyler are totally in the pocket of Bartley J. Maddon, and cannot be trusted! From now on, I will view every one of their posts through this lens, including and especially the ones on the best restaurants in the Eden Center, whether Magnus Carlsen is a truly great grandmaster, and what the best classical music albums of the year were!
Thank you, thank you for exposing the lies behind the façade. You are providing a true public service, and best of all, it's a service not remotely connected to the Mercatus Center.

Well, to be clear, I am actually working off my guilt at having done monstrous things in my work in marketing for GMU before I was terminated. Every day I came to work and got paid for selling something, which is probably the worst thing a human can do.

Now I live in what many here call the socialist hellhole of Germany, where the locals share my disdain for capitalism and money. I enjoy posting as a disloyal reader when my posts are not moderated away, and I am in no way obsessed with Tyler Cowen, nor do I have deepfakes of him in various pornographic Nazi-themed photos.

"...as well as the time to approval of innovative medicines."

I was always under the impression that the long lead-time for approval was somewhat part of the whole point in that it allows for evaluation of drug effectiveness and side-effects over a longer period of time. Is that not the case?

That being said I agree wholeheartedly with one of Scott Alexander's points on the issue, that being that things - barring anything else - would be greatly simplified if the USA would just acknowledge the concordant testing regimes in other developed markets (like EU) for approval of the same drugs here. Apparently pharmaceutical science changes once you cross the Azores heading West. Eliminating that redundancy alone would save a lot of time and duplication of effort.

Baby steps first. Europe is so weird, the FDA may start considering first the drugs tested on Canada.

I agree. I commented on Alexander's post (maybe 3 weeks ago?) in this regard about some of the downsides of 'concordant approval' as well

1) It could lead to big-pharma playing different testing regimes against each other.
2) They would probe different testing agencies for weakness or influence to get drugs approved that wouldn't pass elsewhere.
3) There are still tremendous amounts of cultural differences - even close by - regarding the line between a food and drug and what. That alone could create quite a swamp.

But yes, maybe Canada first. Or Japan...I hear quite good things about the drug-testing protocols in Japan.

Reciprocity creates a race to the bottom where drug firms select the country with the lowest standards or, assuming standards could be harmonized, the laxest enforcement.

When did people begin prattling about "learnings" rather than "lessons"?

From the reform proposal of four tiers of approval:

"First, we propose that manufacturers have the option—as opposed to being required—to conduct additional studies to obtain a higher-order effectiveness claim. This decision would be driven by market forces,including consumer (payer, doctor, and patient) demands and competition. The FDA could not demand that a company apply for a particular (higher) order of effectiveness. Second, these approvals would not be conditional, meaning, for example, unlike the accelerated approval pathway, the FDA could not revoke an approval that is based on biomarkers (Category 1 in our example) if a subsequent study did not show a survival advantage (Category 4). "

I'm amazed at the last sentence. Even NEW EVIDENCE is available, they are asking for the FDA to ignore it and stick to the past decision. It's a full rejection of science..........WTF?

Nah, we do this all the time. Suppose we change our shift structure. We follow it up and find that we are getting patients out of the hospital 20% faster. Hooray.

Then we follow this policy for five years. We find that have not changed 90 day mortality of our patients at all. Oof, bad.

Then somebody sits down and says, you know one subgroup of the patients is showing increased mortality, say those with poor social situations, while everyone else is living longer. Hmm.

Now we stratify patients and those who need to stay longer get some more social work and everyone gets the benefit of better handoffs.

This is not an uncommon scenario. Pretty much ALL drugs improve some biomarker, but if you give them to the wrong people, they decrease mortality. There are a few drugs where dosing the general population would be beneficial (e.g. statins); most would fail that "clinical trial".

Sometimes we aim to high. We say this drug should be good for all breast cancers. Instead it is only actually effective for HER2 positive ones or maybe only if given concurrently with antiestrogen therapy.

With all drugs, the name of the game is finding the patient population that will be helped and giving it to them. Guessing wrong about the exact target, should not force you to start over at square zero rather than following the scientific method by utilizing the new data to refine your hypothesis and test again.

The drugs on this new track would, presumably, be drugs that have yet to undergo Phase III clinical trials. 90% of prospective drugs fail these trials due to lack if efficacy or because they do more harm than good. So this proposal would be extraordinarily wasteful and harmful.
Read the blog "In the Pipeline" by Derek Lowe" today about the phase III trial that was halted for increasing mortality by 20 percent.

Yes, and?

Lots of drugs increase mortality in the wrong population. If I put lisinopril in the water supply I will kill people with kidney issues, with hypotension, and of course create a major teratogen risk. If I gave the average person lithium with blind dosing, I would expect a lot of mortality.

Right now one failed clinical trial means the death knell for drugs that might be very, very wrong in one population, but very right in another. If you don't let us try dangerous drugs and find the one, tiny patient population where they are beneficial you are basically dooming most cancer meds from the start.

As far as "lack of efficacy", oh please. I have patients who developed anaphylaxis to the only drug available to treat their condition. They are literally dying because a drug, to which they would not have immune responses, failed to prove itself to be "non-inferior". One of the trial drugs, for instance, worked about 60% of the time. The comparator worked around 70% of the time. Given that the comparator will eventually kill these patients, I would not be choosing to give them "failed" medicine that is 10% less efficacious than standard of care … I would be giving them medication that is 60% more efficacious than anything else I can safely use.

After all, ARBs are most commonly prescribed for the sole reason that the patient cannot take ACEs safely. Nobody gives a rat's ass about their relative efficacy, because, thankfully, the patient population is large enough to run a solid clinical trial. When you are the poor schmuck who has some screwy mast cell disease … well hey I guess you deserved to die because there are literally not enough people with similar pathology to run a clinical trial.

Insurers would not pay for stuff with weak evidence. That's why FDA's "right to try" program has been an utter failure. Nobody is getting the still experimental drugs, because companies don't want to give them away and insurers won't pay.

Either that or the ensurers will not be idiots and pay at some fraction of the normal price for weaker evidence.

Insurers don't pay for medicine to be nice. They do so because medicine is cheaper than the alternatives (e.g. surgery, hospitalization, transplant).

With a single ER visit for some of the patients I have who might benefit running well north of $10K … I could totally see an insurance company offering to take a $1K flutter on avoiding a dozen ER visits even at 70% risk of failure.

Comments for this post are closed