Here are three more new and noteworthy articles from October and November, including Professor Robin Feldman’s new empirical study of the pharmaceutical industry.

Robin Feldman & Connie Wang, May Your Price Be Ever Green. 

Professor Feldman’s paper reports on an empirical study of the new patents and new regulatory exclusivities that were added to marketed new drugs between 2005 and 2015.  FDA awards a new regulatory exclusivity if the company (1) secures approval of a new condition of use supported by clinical data (three years), (2) secures approval of a new use in a rare disease (seven years), or (3) conducts pediatric research in accordance with a written request from the agency (six months).  The agency lists new patents if the company reports that the PTO has issued a new patent claiming the drug or a method of using the drug; after initial approval, these might relate to new uses, new formulations, and new dosage forms, among other things.

The authors and their five research assistants did a ton of work putting together the database.  Most of the information came from an annual FDA publication known in the industry as the Orange Book, but they had to review every annual edition from 2005 to 2015, and they also reviewed every monthly supplement.  (I’ve gone through these before for similar exercises, and transferring the entries into a database is — as they say — painstaking work.)  The findings are fascinating and important.  Just to give one example, more and more companies are receiving orphan exclusivity for new uses; the number was 9 in 2005, and 27 in 2015.  The number of drugs that added a patent increased steadily over the same time period, from 112 in 2005 to 260 in 2015.  The data will be publicly available when the work is published, which is terrific news for other researchers.

I will be candid, though, and admit that I take issue with Professor Feldman’s interpretation of the data.  The paper starts by declaring itself a study of “pharmaceutical bad behavior.”  Where she sees a “common and pervasive problem endemic to the pharmaceutical industry,” I see innovation.  And I almost never see the “evergreening” that others lament.  Even if an innovator develops and patents a new version of its product, generic companies are free to copy the old version when its patents and exclusivity expire.  The only thing conceivable extended is the innovator’s advantageous position in the marketplace, but this comes through continuing innovation, which is precisely as our system intends.  The generic companies could always innovate — and some do.  If they choose a business model that relies on someone else’s research, they run the risk that the state of the art has moved on.  And when the innovator’s new product isn’t all that much better, payers can (and do) balk at covering it — in which case the sales will go to the generics of the innovator’s original product.  For more on this, see Professor Risch’s terrific blog entry on the article.

But, putting aside my strong disagreement with Professor Feldman’s conclusions from the data, this paper reflects an extraordinary amount of work for which we should all be grateful.

Mason Marks, Psychedelic Medicine for Mental Illness and Drug Abuse: Overcoming Social and Legal Obstacles, 21 NYU J. Legis. & Pub. Pol’y (forthcoming 2018).  

This article relates to a fun — and I think important — topic that has received quite a bit of mainstream media attention in 2017, including this piece in the New York Times and this report on PBS News Hour.  The focus is psychedelic substances — all controlled substances heavily regulated by DEA — including LSD, psilocybin (magic mushrooms), ayahuasca (yage), ketamine (which is actually approved, as an injection, for use in anesthesia), MDMA (ecstasy), and ibogaine.  All showed promise in treating psychiatric illness in the middle of the 20th century, but the research pretty much ground to a halt.  Marks explains this carefully, tying it not only to the War on Drugs but also stigma and the lack of incentives. I think he probable means for the heart of the article to be the next section, but the science and history discussions fascinated me.

The heart of the article identifies and explores several “paths” forward, working from the presumption that this interest ought to be picked up.  These compounds could be put through conventional premarket clinical trials under FDA oversight, he argues; they could be rescheduled at the DEA; and they could be decriminalized at the state level.  He notes major obstacles to each idea and offers ways to mitigate those obstacles.  For instance, premarket clinical trials are expensive and take a long time; he suggests sponsors explore accelerated approval or breakthrough status at the agency.  Rescheduling would require the compound to have a generally accepted medical use, which he argues should be defined broadly and flexibly —  not limited to data from randomized controlled clinical trials.  Although he mostly presents these as alternatives, I tend to view them as all necessary.  And he does note in the conclusion that simply decriminalizing at the state level “may not directly increase access for patients” due to the possibility of federal preemption under the Controlled Substances Act.  As a result, he concludes, “a hybrid” approach might be needed.

Charlotte Tschider, Deus ex Machina: Regulating Cybersecurity and Artificial Intelligence for Patients of the Future, forthcoming in the Savannah Law Review. 

Professor Tschider considers the incorporation of artificial intelligence into medical devices, looking at the patient safety implications and the regulatory response in both Europe and the United States.  She offers detailed information about actual and potential use of AI, explaining how it has the potential to improve health outcomes as well as the potential to introduce new threats without appropriate cybersecurity measures.  Readers who want a tutorial on the scientific landscape may find this helpful.

Her recommendations appear in the final section and mainly relate to the need to consider cybersecurity and artificial intelligence safety concerns during premarket review processes.  As an example of the kinds of issues she identifies: she points out that evolving AI functionality could mean that a cleared device no longer performs the way it originally did.  Devices leveraging AI capabilities, she writes, “will likely require a new model for clinical testing and ongoing  management.”