Metachromatic leukodystrophy, the four million dollar treatment that needs screening

In March 2024, the FDA approved atidarsagene autotemcel, marketed in the United States as Lenmeldy, for pre-symptomatic late-infantile and early-juvenile metachromatic leukodystrophy. The list price at launch was $4.25 million per dose, the highest in the world at the time. Lenmeldy is a one-time autologous gene therapy. The cells it modifies are the patient's own bone marrow stem cells, transduced ex vivo with a lentiviral vector encoding the missing enzyme. The therapy works only if it is given before neurological symptoms have begun. By the time a child with metachromatic leukodystrophy shows the first signs that something is wrong, the window has usually closed.

The United States does not screen newborns for metachromatic leukodystrophy. The condition was on the agenda when the federal advisory committee responsible for the Recommended Uniform Screening Panel was dissolved in 2025. The arithmetic is now on the states.

What MLD is

Metachromatic leukodystrophy is an autosomal recessive lysosomal storage disorder. The enzyme arylsulfatase A, encoded by ARSA on chromosome 22q13, breaks down sulfatides inside the lysosome. Sulfatides are sulfate-containing lipids concentrated in the myelin sheath of central and peripheral nerves. When ARSA activity is lost, sulfatides accumulate in oligodendrocytes, Schwann cells, and macrophages, and the myelin progressively demyelinates. A small fraction of MLD is caused by deficiency of saposin B, encoded by PSAP, which functions as an activator of ARSA, with the same downstream consequence.

The clinical presentation depends on age of onset. Late-infantile MLD, the most common form, presents between 6 and 24 months of age. Affected children, who have developed normally to that point, begin to lose acquired motor skills. Walking becomes unstable. Speech regresses. Cognition declines. Within months to a few years, the picture progresses to severe neurological disability: spasticity, seizures, blindness, loss of communication, and a vegetative state. Death typically occurs in childhood.

Juvenile MLD, with onset between roughly 4 and 16 years, presents with cognitive and behavioral changes that often look psychiatric or developmental before the motor and neurological course makes the underlying disease unmistakable. Adult MLD, with onset after 16, often presents with psychiatric symptoms, cognitive decline, or motor abnormalities that can be misattributed to schizophrenia, dementia, or multiple sclerosis for years. Brain MRI in any form shows progressive demyelination concentrated in the periventricular and central white matter.

Reported live-birth incidence is roughly 1 in 40,000 to 1 in 160,000 in unselected populations. Higher rates have been reported in some Israeli and Arab populations and in Habbanite Jews. Diagnosis uses ARSA enzyme activity in peripheral blood leukocytes, urine sulfatide measurement, brain MRI when neurological symptoms have begun, and ARSA or PSAP gene sequencing. A pseudodeficiency allele in ARSA produces low enzyme activity in vitro without disease, and distinguishing pseudodeficiency from true MLD is part of the workup.

Why screening matters here, specifically

MLD is the canonical example of a disease where the timing of intervention determines whether the intervention works. Hematopoietic stem cell transplantation has been used for decades in MLD. The premise is that donor-derived microglia, the brain's resident macrophages, can carry functional ARSA into the central nervous system and reduce sulfatide accumulation. The clinical evidence is mixed. HSCT before symptom onset can stabilize or slow the late-infantile and juvenile forms in some children. HSCT after symptom onset rarely halts progression. Standard care for symptomatic late-infantile MLD has been supportive rather than curative for that reason.

The ex vivo gene therapy approach changes the arithmetic. Atidarsagene autotemcel, which Orchard Therapeutics developed under the name Libmeldy in Europe and which gained EU approval in 2020, takes the patient's own bone marrow CD34-positive cells, transduces them ex vivo with a lentiviral vector encoding functional ARSA, and reinfuses them after myeloablative conditioning. The transduced cells engraft in the bone marrow, produce ARSA, and a fraction of their progeny migrate into the central nervous system as microglia. The published results in pre-symptomatic late-infantile and early-juvenile children show preservation of motor and cognitive function over years of follow-up, in contrast with the relentless decline of untreated controls. The same therapy in symptomatic children performs much less well. The neurological damage that has already occurred does not reverse.

Without newborn screening, presymptomatic identification of infants with MLD is generally accidental. A previous affected child in the same family is the most common path to early diagnosis. Sporadic cases, which are most cases, present clinically when the regression has begun, and by then the child is typically outside the labeled indication for Lenmeldy and outside the window in which HSCT meaningfully helps. The treatment is approved. The treatment exists. The question is whether the diagnosis arrives in time.

The screening case and the federal vote that did not happen

The Advisory Committee on Heritable Disorders in Newborns and Children, ACHDNC, was the federal advisory committee that recommended additions to the Recommended Uniform Screening Panel. Conditions were nominated, evidence was reviewed by an external evidence review group, the committee deliberated, and the Secretary of Health and Human Services accepted or did not accept the recommendation. The process was slow, transparent, and deeply tied to ASRT-style evidence reviews.

MLD was nominated to the RUSP in 2023 by a coalition that included the MLD Foundation and several academic centers. The evidence review was underway, focused on the analytical validity of dried blood spot ARSA enzyme activity assays plus second-tier sulfatide measurement and on the natural history baseline against which screen-detected outcomes would be compared. A vote was anticipated. In March 2025, ACHDNC was dissolved as part of broader reductions to federal advisory infrastructure. The MLD nomination did not reach a vote. The Recommended Uniform Screening Panel did not gain MLD as a federal recommendation. Duchenne muscular dystrophy, also under review, did not reach a vote either.

State newborn screening panels operate independently. New York and Illinois had begun state-level pilot screening for MLD in advance of expected federal action. Other states cited the absence of a federal recommendation as a reason for not adding the condition. The infrastructure to add a condition state by state, without a federal anchor, is much weaker than the federal process. Each state public health department must conduct its own evidence review, fund the laboratory work, navigate its own legislative or regulatory route, and manage parental notification and follow-up. The cost and complexity discourage adoption.

What this looks like for a family

A child with late-infantile MLD born today in a state that does not screen will be diagnosed after regression has begun. The diagnostic workup may take months. By the time ARSA enzyme activity is measured, the lentiviral gene therapy that could have preserved motor and cognitive function is no longer indicated. The family is offered HSCT in some centers, palliative care in others. The neurological trajectory does not change.

A child with late-infantile MLD born today in a state that does screen, and whose family acts on the result, can be referred to a center that administers Lenmeldy or Libmeldy before symptoms appear. The treatment costs $4.25 million in the United States. Insurance coverage and access programs determine whether the family ever sees a bill in that range. The child preserves motor and cognitive function on the published evidence so far.

The difference between the two children is the screening test. The screening test exists. The federal recommendation that would have placed it on every state's panel is not coming. Whether MLD is detected in time is now a question of geography and the willingness of a state public health department to add a condition without federal cover.