Cenobamate's Real-World Efficacy Against Its Own Mandated Slow Titration
A single patient with drug-resistant focal epilepsy starting cenobamate. The efficacy isn't in question — the disagreement is about how to get him through the twelve weeks the label won't let anyone skip.
M.O., a 34-year-old man, has had focal epilepsy since age nine and has cycled through carbamazepine, lamotrigine, levetiracetam, and most recently lacosamide, none of which brought him below three to four disabling seizures a month. He works part-time doing inventory for a warehouse on a schedule his manager has quietly rearranged twice already to keep him employed after seizures at work, and he lives with his sister, who has become the person who calls 911 when a seizure runs long. He is otherwise healthy — no cardiac or renal disease, normal liver function — and has never had a rash or hypersensitivity reaction to any AED he's tried.
His neurologist started cenobamate three weeks ago, and the real-world data behind that choice are genuinely strong: published real-world cohorts of patients this refractory report seizure-freedom rates well above what most add-on AEDs achieve in this population, a meaningful outlier in a field where a fourth or fifth agent rarely does much better than the third. What the data don't change is the label's own titration schedule — a twelve-week ramp from 12.5mg to the 200mg target dose. That schedule is not a generic caution: three confirmed DRESS cases, a severe and potentially fatal hypersensitivity reaction, occurred among the first 953 adults exposed to cenobamate during early development, when starting doses were higher and titration faster. Sperling et al. (2020) then ran a phase 3 open-label safety study of a deliberately start-low, go-slow schedule and found no cases of DRESS in 1,339 patients. That study is where the current schedule comes from; it is the specific mechanism that makes the efficacy safe to prescribe at all.
He is currently three weeks in, at 25mg, having had two more seizures since starting — not more than his baseline rate, but not fewer either, and his sister asked directly in clinic today whether there's any way to get him to a dose that's actually working faster, having read the same real-world outcome data his neurologist quoted when starting the drug.
Three weeks in, at a quarter of the target dose
The titration schedule isn't the cautious option layered on top of an otherwise-fast drug — it's the reason cenobamate is safe to use at all. The DRESS cases came directly from the faster titration and higher starting doses used in early development; the twelve-week ramp to 200mg is the specific fix Sperling et al. (2020) tested and validated across 1,339 patients, not a generic safety buffer we have room to trim for a patient who's frustrated by it.
I understand his sister is reading the same real-world seizure-freedom numbers I quoted when we started this drug, and I don't want to undersell them — they're genuinely better than almost anything else available for a patient this refractory. But those outcome numbers were measured in patients who got there by finishing the titration, not by skipping ahead.
I'm not proposing we shorten the label's schedule — I agree that's not on the table.
But treating the twelve weeks as a fixed, costless wash because the schedule itself can't move undercounts what actually happens during them: three or four disabling seizures a month for a patient whose employment has already survived two seizures at work by his manager's goodwill, not by any real margin he has left.
The question I'd actually put on the table is whether we bridge the gap rather than just wait it out — optimize what he's already on, or add a low-dose second agent temporarily, specifically to blunt the seizure burden during the twelve weeks cenobamate can't legally move any faster through.
That's the right question, and I want to answer it honestly rather than reach for the nearest lever. Cenobamate is a meaningful CYP2C19 inhibitor — it raises levels of clobazam's active metabolite, phenytoin and phenobarbital as titration continues, which argues against adding a new interacting drug into the mix. Lacosamide would look like the obvious candidate, since he's already on it and cenobamate doesn't change its plasma levels at all. But the pharmacokinetics aren't the problem here; the pharmacodynamics are. In the phase 3 open-label experience, lacosamide was the concomitant drug most often reduced during cenobamate titration — down roughly 22% on average by a year — and the published expert consensus recommends reducing it early, at low cenobamate doses, precisely because the two together drive dizziness and sedation.
So the honest answer to your question is that there isn't a clean pharmacologic bridge. Uptitrating his lacosamide is the one move the combination evidence specifically argues against, and every obvious alternative either interacts with cenobamate or has already failed him. What I'd do instead is hold cenobamate exactly to the labeled schedule, leave lacosamide where it is and watch it for early reduction rather than increase, and put the effort into a written rescue-medication plan and a documented conversation with his employer's occupational health about the twelve weeks — treating the interval as something to be survived deliberately rather than a gap we can medicate away.
Agreed: cenobamate's titration proceeds exactly per label, no acceleration considered. His lacosamide dose was increased toward the top of his prior tolerated range as the actual bridge through the remaining nine weeks, with his sister told plainly why the titration itself can't move faster — the same trial finding that justified starting the drug is what fixed its schedule.
Not fully settled: how much further headroom exists in his lacosamide dose if seizures continue at the current rate through week six or seven, and at what point accepting a temporarily busier medication regimen becomes worse than accepting the seizure burden it's meant to offset. Reassessment set for week six rather than waiting for the full twelve.