Rett syndrome has become a commercial market, but it has not become a solved disease. The first approved treatment generated $391 million in 2025 sales for Acadia Pharmaceuticals, evidence that a small patient population can support a substantial specialist franchise. A gene therapy from Taysha Gene Therapies has moved towards pivotal testing after early data in a small group of girls and women. Behind those programmes sits a biological constraint that Huda Zoghbi has spent decades defining: the brain is harmed when it has too little MeCP2 protein, yet too much causes a different severe neurological disorder.
In March 2026, Zoghbi’s laboratory at Baylor College of Medicine and the Jan and Dan Duncan Neurological Research Institute reported a preclinical route around that constraint. The team redirected processing of the MECP2 gene away from a less abundant protein form, increasing production of the main form. In normal mice, the approach raised MeCP2 by roughly 50 to 60 per cent. In patient-derived cells carrying certain Rett mutations, increasing the amount of partially functional protein restored some or all measured cellular features, depending on the mutation. Morpholino molecules demonstrated the mechanism in mice, although the researchers explicitly said those molecules were too toxic to become the treatment and pointed instead to related antisense strategies.
The result is early, but strategically important. It suggests a therapy may be able to amplify residual function rather than deliver another gene. Such an approach could be adjustable and, if necessary, stopped—attributes with clear value when the target has a narrow therapeutic window. Zoghbi’s leadership challenge is to turn an unusually precise understanding of dosage into a development path that can compete for capital without overselling preclinical evidence.
A market forms around an unsolved mechanism
Rett syndrome is rare, affecting mainly girls, typically after a period of apparently normal early development. Loss of acquired skills, motor impairment, communication difficulties, seizures and breathing abnormalities create lifelong care needs. The US Food and Drug Administration approved trofinetide in 2023 for patients aged two and older, the first medicine specifically for the condition. Its pivotal 12-week study enrolled 187 patients and showed statistically significant differences on caregiver- and clinician-rated measures. Diarrhoea and vomiting were common.
Acadia’s sales show both demand and limitation. Daybue’s $391 million of 2025 net product sales rose 12 per cent from the prior year; the company forecast further growth for 2026. A new powder formulation broadens practical choice. Yet trofinetide treats symptoms rather than correcting MECP2. Families and investors therefore continue to support programmes that promise a deeper effect.
Taysha’s TSHA-102 is designed to deliver a functional MECP2 construct with a regulatory element intended to control expression. In 2025 the company reported that all ten participants included in an early efficacy dataset gained or regained at least one defined developmental milestone, with no treatment-related serious adverse events or dose-limiting toxicities in the larger safety set at that cutoff. Those are company-reported results from a small, open-label programme. The pivotal design relies partly on natural-history comparisons because patients aged six and above rarely regain such milestones without treatment.
The competitive landscape is therefore not a simple race between modalities. An approved symptomatic medicine can generate recurring revenue and accumulate real-world experience. A one-time gene therapy could command a high upfront value if gains are durable, but it carries vector, dosing and long-term follow-up risks. An antisense-based method inspired by Zoghbi’s 2026 work could offer controllability, though repeated administration may be costly and burdensome. Each model allocates risk differently between developer, patient and payer.
The narrow window is the asset
Zoghbi’s central contribution was not only linking MECP2 mutations to Rett syndrome. Her work also established that neurons are highly sensitive to the amount of MeCP2. Duplicating the gene produces MECP2 duplication syndrome; restoring too much protein is therefore not an acceptable route to treating deficiency. This turns dosage knowledge into a development asset. It tells drug designers what must be controlled and provides a basis for measuring whether a therapy is approaching danger.
The new strategy exploits two versions of the protein, MeCP2-E1 and MeCP2-E2, produced from the same gene. E1 is the principal form required in the brain; E2 is less abundant and has not been associated with Rett syndrome. By blocking the genetic segment used to make E2, the researchers shifted production towards E1. The concept is especially relevant to the estimated majority of patients whose mutations leave some protein function but reduce abundance or DNA binding.
Commercially, that creates a possible genotype-defined segment rather than a universal Rett product. Developers may initially prefer the largest addressable label, but precision can improve trial power and reduce biological risk. Patients whose mutations produce no useful protein may not benefit from amplification; those with partial function may respond differently depending on baseline expression. A smaller, well-selected trial can create more value than a broad study diluted by people the mechanism cannot help.
The difficulty is measurement. Clinicians cannot routinely sample brain tissue to quantify MeCP2. Blood or other accessible biomarkers must reliably reflect the relevant neuronal state. Functional scales in Rett syndrome can be variable and heavily dependent on caregiver observation. Zoghbi’s laboratory has made biomarker discovery one of its priorities, because dose control without a practical gauge is an engineering claim rather than a clinical capability.
Female mosaic biology changes trial design
A second Zoghbi-led study publicised in June 2026 examined female Rett biology at cell-type resolution. Because MECP2 sits on the X chromosome, female brains are mosaics: some cells use the healthy copy and others the mutated copy. Bulk tissue analysis averages those populations and can hide the earliest changes. The new work found strong gene disruptions in specific mutant cell types that were not apparent when mixed cells were analysed together.
This has direct development consequences. A treatment may raise average MeCP2 while leaving the most vulnerable cell population undertreated or pushing healthy cells too high. A biomarker based on a mixed sample could report success even as the relevant neurons remain abnormal. Cell-type signatures can guide endpoints, dose selection and combination strategies. They may also explain why two patients with the same mutation differ clinically because X-chromosome inactivation patterns vary.
The approach raises costs. Trials may require more sophisticated molecular profiling, and diagnostic laboratories must validate tests across sites. Rare-disease developers already face small populations and expensive follow-up. Adding stratification can slow recruitment. But failure to account for mosaicism could be more expensive: an apparently rational programme might fail because its signal was averaged away.
Zoghbi’s research institute model is well suited to that translation. The Duncan NRI sits inside a paediatric hospital and connects genetic discovery, model systems and clinical observation. Its value is not a single patent but the ability to identify where a therapeutic hypothesis breaks between molecular mechanism and a child’s function. For partners, that can reduce target risk before the most capital-intensive stages.
Asia’s missing patients
Rett syndrome is global, but diagnosis and access are not. Genetic testing, specialist neurology and long-term multidisciplinary care are concentrated in wealthier urban centres across Asia. Girls with developmental regression may be diagnosed late or assigned a broader disability category. That weakens prevalence estimates, clinical-trial recruitment and the business case for launching high-cost therapies.
Japan, South Korea, Singapore and Australia can support advanced genetic medicines, while China and India offer large populations and growing genomic capacity. The region’s diversity is scientifically valuable: mutation spectra, background genetics and care pathways may affect observed outcomes. Yet developers often treat Asian expansion as a post-approval commercial exercise. For a mosaic X-linked disorder, regional natural-history data and validated translations of caregiver measures should be built before pivotal conclusions harden.
An adjustable treatment could have particular relevance where health systems hesitate to fund irreversible gene therapy on limited evidence. Repeated antisense dosing spreads expenditure over time and allows payment to stop if benefit is absent. It also requires ongoing specialist delivery and reliable supply. A one-time therapy may be operationally simpler after administration but creates a large upfront budget shock. Neither model solves affordability without earlier diagnosis and outcomes infrastructure.
Partnerships with Asian academic centres could also test whether cell-type biomarkers travel across populations. Companies may view this as an extra development expense. It is better understood as protection against a narrow evidence base. Rare-disease products depend on trust from small, highly informed patient communities; a programme that excludes major regions limits both its market and its ability to detect variation.
Control must become clinical proof
Zoghbi’s 2026 finding does not yet compete with an approved drug or a clinical-stage gene therapy. Morpholinos used in the proof of concept are unsuitable because of toxicity. An antisense candidate would need chemistry, delivery, dose-ranging, long-term safety studies and a commercial sponsor. Increasing a mutant protein is sensible only when that protein retains enough function and does not create new harm.
Its importance lies in the development logic. Rett therapies will not be judged solely by whether they add MeCP2. They must add the right amount in the right cells for long enough to change meaningful function. A reversible modality may offer an advantage, but only biomarkers and clinical outcomes can demonstrate control. Zoghbi has given the field a mechanism for thinking about that precision and a potential route to achieve it.
The next proof point should be a drug-like molecule that produces a predictable, mutation-informed increase in MeCP2 and connects that increase to functional improvement without pushing healthy cells beyond their safe range. If that can be measured in patients, Zoghbi’s dosage insight will become more than foundational biology. It will set the commercial standard for a market in which power without control is not a cure, but another form of risk.
