Pharma $20-sidewalk-bills

This one's been puzzling me since I saw it reported first last week. The underlying study got a lot of press when it was first published in 2013. Here's the original press release; it's easy to find lots of follow-up stories based on it. It's resurfaced with Jezebel figuring the patriarchy was to blame. Siouxie Wills takes it up again in today's Dom.

Long story short, one study suggested Viagra is effective in mitigating menstrual pain - but nobody seems to have known about it or prescribed if for that purpose. And nobody seems to have replicated the initial trial.

So what's up? Nobody typically complains that Big Pharma isn't quick to jump on any profit-making opportunity. The usual complaint goes the other way.

First potential explanation - maybe it isn't nearly as effective as the first published study suggested. There were 25 subjects in the 2013 trial. The trial showed efficacy against a placebo, but didn't test against existing treatments. This 2019 metastudy does not include any follow-up tests of Viagra. Why wouldn't there be follow-up trials or use?

My next instinct's to look at regulation.

The FDA rules say that once something's been proven safe and effective for one use, it can be prescribed for other uses. So there's no particular legal barrier to doctors prescribing Viagra for menstrual pain. But a lot of them might not know about it. It has been legally difficult to market pharmaceuticals for off-label uses, but I'm not sure on the current state of play on that. And doctors might reasonably be reluctant to prescribe based on a single study of 25 subjects.

Efficacy trials aren't the cheapest things in the world, and might be necessary to market Viagra for a novel use - or to convince doctors to prescribe it. But we aren't talking about some rare disease thing where there'd be a small number of customers. So why mightn't they have gone for it?

One potential reason: you can get a patent extension by registering an older drug for a novel use.

In addition to patent protection for the original compound and method of use, patents directed to new uses and treatment indications can be obtained. Developing new methods of use for identified compounds can be a successful strategy for maximizing research dollars and for increasing the commercial life.

Several pharmaceutical companies have successfully obtained patent protection for new methods of use. For example, Merck originally developed, patented, and marketed finasteride as a treatment for benign prostate enlargement under the brand name Proscar. Additional patent protection and FDA approval were sought when a new use for finasteride — treating male pattern baldness — was identified. Finasteride for the treatment of hair loss is marketed under the brand name Propecia. Similarly, the compound atomoxetine was patented in the early 1980s by Lilly and initially investigated as a treatment for depression. Further research and development of atomoxetine led to the identification of a new use for this compound in the treatment of attention deficit hyperactivity disorder. Lilly has obtained patent protection and FDA approval for this new use, marketing it as Strattera. More than two million prescriptions for Strattera were written in its first nine months on the market.7

Ideally, more than one of these approaches should be employed to extend patent protection. For example, in addition to developing a once-weekly formulation, Lilly sought to minimize its losses from the expiration of the Prozac patent by obtaining a patent and FDA approval for a new medical use of fluoxetine in the treatment of premenstrual dysphoric disorder (PMDD). Lilly markets fluoxetine for PMDD as Sarafem and has secured patent protection until 2007 for this new indication.

But that doesn't seem to work here - the patents on Viagra are up, and the research work on the alternative use is from 2013. If it were patent extension, it should have been filed a couple years ago and available by now.

The most plausible candidate explanations then, in descending order of plausibility, although my priors are fairly flat:

• Unpublished follow-up trials showed the first study to have been a fluke;
• The costs of proving efficacy for the novel use are high relative to the likelihood of being granted a patent extension based on the novel use - I don't know enough about practice here;
• Maybe the Phase I trials only included men, so the costs of a new trial are much higher than I'd thought;
• There's some other regulatory barrier in here I don't know about;
• Pharma executives are idiots, and all potential entrants are idiots too. They're leaving a pile of money on the table.

If I had to give odds - at least 2 chance in 3 that the first one's the right explanation. But I've wide confidence intervals here. Anyone have better information?