Lamotrigine for BPD After a Large Negative Trial: What Do You Tell a Patient Who's Stable?
A patient stable for 24 years on lamotrigine for BPD is told, for the first time, that a large randomized trial found the drug no better than placebo. The disagreement is whether her own long course should outweigh a definitive population-level null result -- and who gets to decide.
M.R., a 52-year-old retired schoolteacher, became a grandmother for the first time six weeks ago, a role she has been looking forward to since her daughter's pregnancy was announced last year and one she describes, without exaggeration, as the thing she has most wanted in her adult life. She arrives at today's appointment settled and articulate, the same way she has for most of the fifteen years her psychiatrist has known her, and is visibly unsettled by the actual subject of today's visit.
She was diagnosed with borderline personality disorder at twenty-seven, after several years cycling through multiple antidepressants and a brief lithium trial that produced tremor without benefit. Lamotrigine, started at twenty-eight, was different: within several months her mood swings became less frequent and dramatically less severe, and she has remained on it, unchanged, for essentially her entire adult working life, crediting it directly and specifically — not therapy, not any other factor — for the stability that let her build a full teaching career and raise two children largely without the crises that marked her twenties. She has had no hospitalizations and no self-harm since starting it.
Her psychiatrist raised, for the first time today, a 2018 randomized trial that found lamotrigine performed no better than placebo in borderline personality disorder at any measured timepoint over a full year — a trial far larger and more rigorous than anything available when M.R. started the drug twenty-four years ago. That finding doesn't erase what M.R. has actually experienced, but it does raise a real question her psychiatrist can no longer avoid asking: whether continuing a medication a strong trial found ineffective is good practice, however well the patient in front of her seems to be doing on it.
Routine follow-up, fifteen years in care
This is genuinely one of the more definitive negative trials in this whole diagnostic category, and I think it has to change what we recommend. Crawford and colleagues' 2018 LABILE trial randomized 276 patients with BPD to lamotrigine or placebo and found, at 52 weeks, essentially identical scores on the primary outcome measure — 11.3 versus 11.5, not remotely close to significant — and no difference on any secondary outcome either, including self-harm and quality of life. That's not an underpowered or ambiguous result; it's the largest, most rigorous trial ever run on this question, and it came back flatly negative. Lamotrigine also carries a real, if uncommon, serious rash risk. Continuing a drug a trial this size found ineffective, with a real safety signal attached, isn't something I'd recommend defaulting to just because nothing has gone wrong yet.
The trial is real and I'm not going to argue with its rigor — it's a well-designed, adequately powered study, and the result is what it is at the population level.
But a population-level null doesn't mean zero individual responders, and M.R. has been stable for twenty-four years on this specific drug in a diagnosis where stability is often hard-won and precarious. Worth noting directly: only about a third of the LABILE trial's own participants were actually taking the medication as prescribed by the end of the year, which may have diluted any true signal that exists in patients who, unlike a meaningful share of that trial population, actually take it consistently — which she clearly has, for over two decades. Disrupting a regimen that has coincided with the best-functioning period of her adult life, on the strength of a population average that may not describe her, carries its own real risk that isn't captured in the trial's own outcome measures.
I don't think either of you is wrong, and I don't think this needs to be decided today. What I owe her is the actual information — the trial, what it found, and honestly, that I don't know whether she's a true responder or whether twenty-four years of stability happened for other reasons entirely. What I don't owe her is a unilateral decision made in this room without her. If she wants to attempt a slow, carefully watched taper, we can do that on a timeline she chooses — not this visit, six weeks into being a grandmother, when destabilization risk is a real, additional cost layered on top of everything else being asked of her right now.
Agreed: lamotrigine continues unchanged today. M.R. was given the LABILE trial's actual findings directly — what it measured, what it found, and its real limitations — and asked to think about whether she wants to pursue a taper, on her own timeline, rather than being told what to do in the room. The plan is to revisit explicitly at her next routine visit in three months, or sooner if she initiates the conversation herself.
It will be slow and closely monitored, scheduled at a time she chooses rather than during a period of major life transition.
That choice is documented as informed and hers, not a default the team failed to revisit — the rash risk and lack of trial-level support remain on record for future review.
Not agreed: the Clinical Pharmacologist maintained, for the record, that deferring a recommendation to taper risks treating patient comfort as equivalent to clinical evidence; the Psychiatric Pharmacist and Attending Psychiatrist held that an individual patient's twenty-four-year course, honestly disclosed alongside the trial, is itself a legitimate part of an informed decision — not something the trial result should simply override.