The Problems With Generic Medications Go Deeper Than One Company

I.

Like many people, I recently read about Turing Pharmaceuticals’ purchase of anti-toxoplasma drug Daraprim and subsequent price increase of 5000%. Vox and Marginal Revolution have already done some good work addressing this particular case, but have only touched upon the broader issue: that everything about generic medications is approximately this terrible.

As far as I’m concerned, the interesting aspect of this case isn’t just that the CEO of Turing is an asshole who is lining his own pockets with zillions of dollars by gouging AIDS patients. I assume most pharmaceutical company CEOs are assholes who would line their own pockets with zillions of dollars by gouging AIDS patients if the opportunity presented itself. The interesting aspect of this case is that the CEO of Turing got the opportunity. How?

In the United States, pharmaceutical companies that discover a new drug are granted a 20-year term of exclusivity to reward them for the public service of drug research. During this time, they can and do price-gouge as much as they want. After twenty years, the drug becomes public domain and anybody who wants can compete to produce it, usually leading to a precipitous fall in costs. But Daraprim is fifty years old; its patent is long-since expired. So Sarah Kliff from Vox asks the obvious question: why doesn’t someone just produce a competitor?:

Daraprim isn’t a frequently used drug. The New York Times estimates that between 8,000 and 12,000 prescriptions get filled annually. You could only fill about a quarter of a baseball stadium with the number of people who take the drug in a given year.

So think about a generic drug manufacturer looking at the Daraprim situation. There are fixed costs associated with building a new plant (or possible lost revenue on other drugs, if they switch production at an existing plant), getting samples of the drug, and figuring out how to make the generic product…with Daraprim, there simply isn’t a big enough patient population for a competitor to sell a “good amount” to. And this is, more generally, a problem with the markets for drugs that only a small number of patients use. They often aren’t big enough to support two competitors.

Moreover, there’s risk associated with starting a drug price war. Let’s say I decide to launch Sarah’s Generic Drug Company, and I’m pretty sure I can break even by slightly undercutting Turing and charging $700. What happens if Turing responds by dropping its price down to $500, or even back to $13.50? It will keep all its patients — and my nascent drug company is likely going bankrupt.

This is definitely part of the story. On the other hand, what about Longecity group buys? Someone on a drugs forum hears about a cool experimental chemical that sounds fun to try. They get a couple dozen friends in on it and pay a lab in China a few hundred dollars to synthesize a big batch. Then the Chinese ship it over, they distribute it to their friends, and they all get a decent supply of a totally novel drug for a few dollars a pill – compared to the $750 per pill that Turing is charging for Daraprim. I am not a chemist, but the Daraprim molecule does not look very intimidating. I bet if a group from Longecity got a couple of toxoplasma patients together for a group buy, they could all get treatments for maybe a few hundred dollars each instead of the $63,000 Turing is now charging. In fact, I encourage somebody to do exactly that as an act of civil disobedience/political activism and win themselves some free publicity.

So how come Longecity can do this, but real generic pharmaceutical manufacturers can’t? I’m not totally sure, but my best guess is that it involves bioequivalence studies (different from purity studies). Generic drugs don’t need the excruciatingly drawn-out safety and efficacy studies required of new brand-name medications, but they do need to pass a bioequivalency study proving that their drug is absorbed the same way as the original. According to Wikipedia, the most common type of bioequivalence study is to “measure the time it takes the generic drug to reach the bloodstream in 24 to 36 healthy volunteers; this gives them the rate of absorption, or bioavailability, of the generic drug, which they can then compare to that of the innovator drug”.

This might not seem so bad, but it must be harder than it sounds. This site, whose style is overly bombastic but whose information seems mostly correct, says that:

The cost and time involved in the ANDA [generic application] process varies depending on the drug, its safety, how long it has been on the market, etc. To have an ANDA approved, it typically requires an investment of about $2 million, and it takes a total of two to three years to get the drug to market…in addition to these costs, a company should budget 15% for legal fees, because wherever there is a big manufacturer with a sizable market share involved, they will sue, just to try to eliminate more competition from the market.

This adds an important extra dimension to Vox’s theory that it’s just too hard to start making a generic medication. If all you want to do is synthesize an active ingredient in powder form, and you’re not too concerned about staying on the right side of the law, it costs pennies and takes however long you need to FedEx something from China. If you also want FDA approval, it costs $2 million and takes two years.

Remember, Daraprim is used by about 10,000 people per year, and before the recent Turing price markup, it cost $13.50 per pill x eighty pills per treatment. 10,000 * 80 * $13.50 = about $10 million per year, of which maybe $5 million was profit. That means you have to capture a big chunk of the Daraprim market before it’s worth trying to get yourself approved to make Daraprim; the FDA is essentially telling pharma companies to “go big or go home”. Nobody wanted to go big, so they all went home.

In the absence of this barrier, it would be easy for small boutique companies with a couple of chemical engineers on hand to spend a few weeks manufacturing a few thousand doses of the drug whenever it was necessary to meet demand. This is how the supplement and nootropic industries work right now, and nootropics are dirt cheap, even though a lot of “nootropics” are the same chemicals as regular expensive medications except with a “not intended for human consumption” label slapped on the bottle that everyone knows to ignore.

I think this might be what’s going on with generic modafinil. Last week I prescribed some modafinil to one of my patients and got a call back from their insurance company saying it was denied because it cost too much.

I told the insurance company that was silly because modafinil only cost about $60 a month.

The insurance company said no, it cost way more than that.

This surprised me, because half the rationalist community uses modafinil, and even some of the doctors I work with use modafinil on long night shifts, and they all get it for $60 a month from places like ModafinilCat.

But according to Nootriment, a month’s supply of modafinil at real bricks-and-mortar pharmacies costs anywhere from $469.23 (Costco) to $850.84 (RiteAid). I’m not totally sure what’s going on, but my guess is that ModafinilCat (illegally) buys it from people who haven’t gone through the FDA’s bioequivalence testing, and RiteAid buys it from people who have. As far as I can tell, both are made by Indian pharmaceutical companies unrelated to the original American company who discovered the drug, but RiteAid’s Indian pharmaceutical company has put more work into staying on the right side of the US government.

If any of my patients are reading this and are upset because I prescribed them a drug which they couldn’t afford, I unreservedly apologize. I was laboring under the misapprehension that the pharmaceutical market made sense.

II.

No tour of terrible generic medications policies would be complete without a stop for Kesselheim and Solomon’s analyis of the Unapproved Drugs Initiative of 2006.

The FDA wanted to encourage people to study drugs that were already in the public domain and get them up to FDA standards. This is potentially a very noble plan. I’ve written before on how it’s basically impossible to get melatonin to interface with the health care system because it got into the public domain without the relevant FDA standards being met. Likewise, there’s no interest in using minocycline to treat schizophrenia because it’s a public-domain drug and nobody profits off of doing the FDA compliance work. So the FDA was definitely responding to a real problem.

Their solution, though, was to say that if anybody did a good enough study on a public domain drug, they could grab it out of the public domain and have it be their exclusive drug for the next while. This was a terrible terrible terrible idea.

Colchicine is a very popular and very effective gout treatment extracted from the Colchicum plant. It’s been used for so long that its first recorded mention in medical literature is on an ancient Egyptian papyrus. The medievals called it “hermodactyl”; Arabic physician Avicenna recommended it; notable gout sufferer Ben Franklin brought the first Colchicum specimens to North America.

But the ancient Egyptians, being a primitive and barbaric people, had no FDA. And although many different groups had done studies proving colchicine effective, none of them had done so on the official FDA forms. In 2007 a company called URL Pharmaceuticals did an official FDA safety study, showed that yup, it was safe all right, and for this service were granted exclusive right to produce colchicine. After suing all other colchicine producers out of business and establishing a monopoly, they raised the price of colchicine by 5000%, costing gout patients thousands of dollars a year.

According to FiercePharma, something similar happened with hydroxyprogesterone caproate, although the FDA later changed its mind. I can’t find any other examples, but the legal framework is still there if someone else wants to try.

III.

Other times generic manufacturing proceeds smoothly. A drug is popular and many different pharmaceutical companies pass the bioequivalency tests, get in on the action, and compete with one another. Nobody snatches it out of the public domain at the last second and receives a new monopoly on it. The companies are able to sell it to the pharmacies for a reasonable cost.

Now you get to have a completely different set of things go wrong.

Michigan Drug Prices is my state’s official drug price register. You can type in any Michigan ZIP code and any drug and find out how much it costs at all your local pharmacies. It’s pretty neat.

Celexa has been generic for more than a decade, it’s got a reputation for being inexpensive, and I prescribe it a lot. Let’s see how much my patients have to pay.

The closest RiteAid to my office charges $4 for a 30-day supply of Celexa 20 mg. The local CVS sells the same amount for $19.79. The local Walgreens sells it for $24.99. And the local KMart will sell it to my patients for the low, low cost of $88.15. That’s an…interesting…range of prices.

If I try to buy it off GoodRx.com, a site that offers pharmacy price comparisons, I can get it for $3.60 from a mail-order pharmacy. But I can also get it for $6.64 from K-Mart, special offer for GoodRx customers only. $10.00 from Walgreens. $11.99 from CVS. All the same stores that were trying to gouge me before. As soon as you take the basic step of saying “by the way, I’m also comparing costs with other pharmacies” their prices drop 90%.

I am far from the only person to notice this. PBS did a segment on one of the reporter’s mothers looking for a breast cancer drug. She originally paid $400 a month for it, which is steep but perhaps worth the cost as a high-tech treatment for a potentially fatal illness. Then she went to Costco and found the same medication cost $10.

Why does this sort of thing happen? I’m not sure. I expect it has something to do with insurance co-pays; if an insurance looks at some kind of average cost of Celexa and decides that the Celexa co-pay will be $5, then it doesn’t much matter to the customer whether they buy it from a pharmacy charging $10 or $10,000. But why doesn’t the insurance company do one the thing everyone in health care agrees insurance companies do best: send whiny faxes complaining that they’re not going to pay you? I don’t know.

But for now you might want to try using something like GoodRx.com if you’re buying expensive medications. And stay away from cats, because there’s never been a worse time to get toxoplasma.