Antiseizure medications (ASMs) are still the primary way by which we suppress seizures in epilepsy and other “seizure disorders.”

The neat thing about ASMs is that there are so many of them. A few dozen. It really depends on how you count them, though—are benzodiazepines ASMs? Or only 1,5-substituted ones?

(I mean, fosphenytoin basically exists only to serve as a prodrug for phenytoin which suggests it’s not an extra; but primidone makes phenobarbital inter alia and both oxcarbazepine and eslicarbazepine acetate are mere delivery vehicles for licarbazepine, and yet.)

So why do more? And why do the brand names suck?

You will probably be familiar with Kwon & Brodie and the general (and well-reproduced) finding that seizure abolition rates are monotonically decreasing functions of the number of trialled ASMs. What’s the rate of drug resistance in epilepsy in 2022? Around 30%—not so different from when people were still oohing and ahhing over Luminal. (In the aftermath of other wars and pandemics.)

This would seem to suggest that the addition of new ASMs over the years has not brought us any closer to abolishing seizures for all patients. Whether some ASMs are more efficacious than others is kind of an open question because no one really does any head-to-head trials anymore.

And so here’s where AES2022 and, really, AES annual meetings in general show how incremental they can be. During the Epilepsy Therapies Symposium today there was a talk on “New Antiseizure Medications.” The three miraculous new agents? Cenobamate, fenfluramine, and ganaxolone.

Cenobamate was approved in 2019 which does make it somewhat new and shiny still but it’s hardly an emerging therapy. It has two major things going for it: it hits persistent sodium currents as well as GABA receptors; and it killed a person during clinical trials due to DRESS.

Fenfluramine actually is a pretty interesting addition to the ASM set with fairly novel MOA but it’s not new or on-the-horizon.

And in fact maybe my hangup is that I expected something like a mini-Eilat conference and all I got was some sanitized slides from pharma talks.

(Don’t get me started on ganaxalone—its brand name is Ztalmy which sounds like the title of an early Soviet sci-fi novel. In fact I propose a simple metric on the horribleness of brand names: 1 point for every X, Y, Z, K; a point for every consonant in excess of vowels; and a point for using Y as a vowel. Ztalmy scores 5; Xcopri scores 3; Fycompa is only a 2.)

Also of interest were several posters about retigabine (which is the INN; the US generic name is ezogabine). RTG was unceremoniously yeeted from the market a number of years ago after reports of retinal pigment changes and what I perceived was lackluster take-up.

This is a sort of the fenfluramine playbook. Your product had problems, got yanked, but no let’s find a way to rehabilitate it. I imagine everyone wants to start with a developed drug rather than develop from scratch even if clinical trials remain expensive and difficult.

Just in passing, the company that seems to be developing retigabine is Xyzagen, which if it were a drug name would score a 5 on my ad-hoc scale.

But do we need new ASMs? The ASMs that have been specifically developed for orphan conditions in the past few years, like clobazam and cannabidiol, seem to be broadly effective in epilepsy (or at least to the same extent as other ASMs). Drug-resistance rates seem not to have changed. Few people are interested in head-to-head trials—although a few network meta-analyses do exist, so shout out.

The annual course here at AES2022 on Sunday is all about personalized medicine. And I welcome an age where ASM choice is guided by good genomic and phenotypic data rather than familiarity, insurance coverage, and how fancy was the pharma-sponsored dinner’s restaurant.

For the record I don’t think we need more traditional-style ASMs. We got enough. Unless you can replicate the excellent PK and dose form flexibility of levetiracetam without the neuropsychiatric disjaskitness. Or maybe a drug or two with a truly novel MOA we can use in rational polytherapy. (And doesn’t there lie a rub or two?)

But I’m just a simple country epileptologist so who knows.