Trofinetide for Rett Syndrome: A Modest Signal Against a Fragile Gut
A girl with Rett syndrome is offered trofinetide, the first drug ever approved for the condition's core features. The trial's effect size was modest and its most common side effect is diarrhea — a real problem in a girl whose nutrition and growth are already precarious.
E.F. is eight years old and has lived with a confirmed MECP2 mutation and a clinical diagnosis of Rett syndrome since she was two and a half, when a period of regression — the loss of the few purposeful hand movements and single words she had briefly had, replaced by the repetitive hand-wringing that eventually prompted her workup — followed the pattern her mother had read about online before any doctor said the word Rett out loud. She communicates through eye gaze and an AAC device, uses a wheelchair for community outings though she can take supported steps at home, and has a seizure disorder reasonably well controlled on her current regimen.
Her nutrition is the thread running through all of it, and the detail that matters is not that she has a G-tube but what it took to keep her growth curve where it is. Placed at five for poor oral intake and slow weight gain, it is used now for supplemental feeds on days her oral intake falls short — which is to say her curve is stable because it is actively held there, not because it has stabilized on its own. She has no chronic diarrhea at baseline. That last point is worth stating plainly, because it means any new stool frequency on treatment will be attributable rather than confounded.
Trofinetide, approved for patients two years and older, is the first therapy ever approved for Rett's core features rather than for a single associated symptom, working through IGF-1-related signaling on synaptic maturation and glial function. Its pivotal trial, LAVENDER, enrolled girls and young women aged 5 to 20 — at eight, E.F. sits inside that range, near the younger half where most participants were — and found a statistically significant RSBQ improvement over placebo at twelve weeks, a change of about 4.9 points against 1.7 on a 90-point caregiver-reported scale. That is a real result and a modest one, and clinicians who accept the statistics have argued about the size.
The complication specific to E.F. is that the trial's dominant adverse effect lands precisely on the system her care team has spent three years protecting. Diarrhea occurred in roughly 82 percent of treated patients against 20 percent on placebo, vomiting in 29 percent, and 17 percent discontinued for adverse events against 2 percent. More pointed for her: 12 percent of trofinetide-treated patients lost 7 percent or more of body weight, against 4 percent on placebo. For a girl whose stable curve depends on supplemental feeds she does not always tolerate, that is not a background safety statistic. It is the specific mechanism by which this drug could undo the thing that has taken longest to secure.
At the developmental-behavioral pediatrics follow-up
I think we should offer trofinetide. This is the first therapy ever approved specifically for Rett syndrome's core features, not for one associated symptom at a time the way everything else in her regimen is — that's a genuinely new thing for families who have never had a disease-directed option before. LAVENDER showed a statistically significant improvement on the RSBQ, a validated caregiver-reported measure, against placebo. I won't pretend the effect size is dramatic — it isn't, and I think it's important we're honest with her parents about that rather than oversell it. But 'modest and real' is still a meaningfully different offer than 'nothing,' which is what every family before this approval has had to accept for the core disease itself.
I want to be clear I'm not dismissing the GI concern — I'll let the gastroenterologist speak to that directly. I'm specifically arguing that the historic absence of any disease-directed option is a real reason to take a modest real effect seriously rather than hold it to a higher bar than we'd apply in a disease with more existing options.
You're right that a real, disease-directed option is worth taking seriously even at a modest effect size, and I'm not arguing families shouldn't have access to this drug in general. But I want us to be specific about E.F., not about Rett patients as a category. Her growth curve is stable right now because of years of active vigilance and a G-tube we placed specifically because her oral intake couldn't reliably sustain her — and trofinetide's single most consistently reported finding in LAVENDER is diarrhea in the large majority of treated patients, with vomiting in a substantial minority, a tolerability burden that drove meaningful discontinuation across the trial population. That's not a rare or unpredictable side effect we're weighing against a modest benefit in the abstract. It's the drug's dominant, expected effect landing directly on the one system we've spent three years protecting in this specific child.
'The effect is modest but real, so we should offer it' treats every patient's baseline as roughly equivalent, but hers isn't — a GI side-effect burden that's a manageable inconvenience for a child with an unremarkable feeding history could genuinely destabilize a growth curve that's already precarious in her case specifically.
I don't think this needs to be a binary yes-or-no decided today and locked in indefinitely, and I don't think either of you is actually asking for that. I'd propose a defined trial period — six to eight weeks — with explicit stopping rules agreed in advance. I want to be straight about why that number and not another: LAVENDER read out its efficacy endpoints at twelve weeks, so six to eight weeks is not long enough to tell us whether she is an RSBQ responder, and I'm not going to pretend otherwise. What it is long enough for is the tolerability question, which is the one that actually threatens her — the diarrhea in that trial came on early and disproportionately, and if her stools and her weight are going to destabilize, we will know inside that window rather than at the end of a full course: a specific threshold of stool frequency or a specific degree of weight change that triggers discontinuation without it being treated as a failure or a difficult renegotiation in the moment. That gives her a real chance at the benefit the pediatrician is describing, while giving the gastroenterologist's concern a concrete, pre-agreed off-ramp rather than an open-ended commitment to monitor and hope.
I want to name directly that this doesn't resolve whose read of her risk-benefit balance is right — it just means we don't have to fully agree on that question before giving her family a real, bounded way to find out.
Agreed: start trofinetide for a defined six-to-eight-week trial with explicit, pre-agreed stopping rules — a specific stool-frequency threshold and a specific weight-change threshold — rather than an open-ended commitment, and increase the frequency of nutritional monitoring for the duration of the trial. Her parents were told plainly, in the gastroenterologist's own terms, that the drug's most likely side effect is the one most relevant to E.F. specifically, and in the pediatrician's own terms, that the potential benefit is real but modest, not transformative.