Map of life expectancy at birth from Global Education Project.

Monday, December 20, 2010

A brief primer on drug regulation wonkery

I realize there is a lot of potential for confusion when we talk about the drug approval process and exactly how consumer protection and drug company economic interests interact. I'm not actually a major expert on these issues myself but I think I can at least give you a handle on some basic concepts that will help clarify matters.

First, drug manufacturers, like every enterprise engaged in research and development, can file for patents on new compounds. These generally last for 20 years. As long as a company holds a patent on a compound, no other company can sell it, unless they come to an agreement with the patent holder and (ordinarily) pay royalties.

However, you can't legally sell a drug and make health claims for it without FDA approval -- called a drug license. (Not to be confused with a license sold by a patent holder to another company to make the drug.) Before getting a license to market a drug, a company first gets an investigational license that allows them to test it in clinical trials. By the time this process is completed and a company is licensed to market a drug, much (if not all) of the 20 years of patent protection has expired. So Congress has sweetened the incentive to develop new drugs by also allowing the FDA to grant additional marketing exclusivity periods, typically of 5 years.

If a drug is licensed, then physicians can prescribe it (or it can be sold over the counter if it has that kind of license). However, the drug companies are additionally restricted by the FDA in the specific claims they can make when they market the drug. That won't necessarily stop doctors for prescribing it "off label," for unapproved indications, but the companies are not legally allowed to encourage them to do so. This is a major problem area. Enforcement has been lax, and drug companies have found all sorts of ways of getting around these restrictions.

Once both the patent and marketing exclusivity periods have expired, other companies can apply to manufacture the drug. But they have to satisfy the FDA that their formulations are biologically equivalent to the original. The original license holders can intervene and contest these claims.

Once generic versions of a drug are available, the price typically goes way down, because now there is competition. The unit cost of manufacturing a pharmaceutical is usually trivial. Drugs are expensive because of marketing exclusivity which allows manufacturers to charge whatever they think they can get.

The final layer in this casserole is insurance reimbursement policy. Medicare is pretty much required to pay for drugs prescribed for any approved indication, but private insurers can set their own policies and they often have tiered co-payment systems that encourage use of medications which are cheaper for them. They can also refuse to pay for certain treatments if they choose. This is an endless source of confusion to physicians who have to deal with patients whose insurers use different "formularies," or lists of approved drugs at various levels of copays.

Drug manufacturers try to extend their marketing exclusivity by coming up with just slightly different versions of old drugs that they hope they can persuade consumers and physicians are better and hence worth paying brand name prices for. This is called "evergreening." The whole COX-2 inhibitor fiasco can be understood as essentially an effort to evergreen aspirin. They can also keep advertising the brand name drug and hope to persuade people it is better than the generic, viz. Lipitor. (There is absolutely no reason that I can see for anyone to pay for Lipitor with very cheap generics available that are biologically equivalent. But remember, I'm not a real doctor.)

Here's some information about all this on the FDA's web site.

So, to clarify a recent flapdoodle, the FDA had given conditional approval to market Avastin for metastatic breast cancer on condition the manufacturer conduct additional trials to demonstrate that it really works. It turns out it doesn't, so approval for those specific claims has been withdrawn. However, Avastin is still licensed and approved for other indications, so doctors can still prescribe it to women with metastatic breast cancer if they want to. The question is whether insurers will pay for it. Presumably, once the appeal period is over, and if it is unsuccessful, Medicare will not.

That's why Genentech is unhappy. But it's a) still licensed and b) still under marketing exclusivity.

Then there's the whole question of so-called "supplements," which we'll get to another time. . .


kathy a. said...

what exactly do you mean by "biological equivalence"? i assumed generics were the exact chemical compounds as the originals, but maybe i'm missing a layer or few of understanding here.

we've had some struggles with insurance formularies. sometimes a cheaper but drug does *not* work the same way for individuals. this was a while back, but our insurance denied coverage for singulair, for my son's asthma, saying claritin ought to work. which, it didn't. they changed their mind after he spent a few days in the hospital. [dated example, but nothing i have heard about the operations of insurance companies reassures me that they have suddenly become more responsive to the individual needs of patients.]

Cervantes said...

In addition to the drug itself, medication preparations may contain inert substances to fill space, be packaged in time release forms, the capsules affect how quickly the drug is absorbed, etc. So there can be disputes about biological equivalence. Also you have to show the compounds are pure and identical -- there are left- and right-handed versions of some chemicals, etc. So there can be differences that might matter, in principle, hence the possibility of disputes.

Singulair and Claritin are not the same thing at all, so that's a different kettle of fish.

Vesna said...

@Kathy: In case of biological molecules – proteins (as for example Avastin – monoclonal antibody), generics are never identical molecules, they are actually very similar, but not identical. They have the same biological activity and work in the same way, but there are some differences between them and it is arguable whether one or the other is better, but their effect should be comparable (similar) in order to get approval by state agencies (FDA, EMA, etc.)

Cervantes said...

Now Vesna has opened up yet another can of worms, which I was holding off on. Most of the drugs we have come to know and love are relatively simple compounds, but Avastin is a protein -- a monoclonal antibody, to be exact, which you can always recognize because their generic names end in "mab." These are generally manufactured by genetic engineering of bacteria. As Vesna says, their bioequivalence can be a bit harder to establish, although that's controversial. The drug companies are also bucking for extended market exclusivity for these classes of drugs.

kathy a. said...

you know, i hadn't considered treatments based on proteins or other biological materials. or the "inert ingredients" part of the equation.

yeah, i never understood the insurance formulary's idea that claratin would do it. my son's primary problem was the all kinds of things triggered his asthma, not just nasal allergies -- like every cold he caught. i think the event that led to his hospitalization was the park service burning eucalyptus about 10 miles away. prevailing winds blew the smoke right at us.