PKU in the developing world

In a hospital in Nairobi or Lagos or Dhaka, a child born this morning with phenylketonuria has roughly the same prognosis as a child born with PKU in Birmingham, England in 1951. The biochemistry was solved seventy years ago. The screening test was solved in 1962. The drugs followed in 2007 and 2018. None of that reaches most of the children born with the disease today.

Hillert and colleagues, in a 2020 systematic review in the American Journal of Human Genetics, estimated that around 450,000 people worldwide have PKU. They found published prevalence data for 64 countries. For most of Africa, much of South and Southeast Asia, and parts of South America and the Caribbean, no national prevalence data existed at all. The map has holes in it because nobody is screening, which means nobody is counting, which means the question of how many children are losing IQ points to a treatable disorder this year cannot be answered.

This is the part of the PKU story that does not get told in the United States. The American version ends in 1963, when Massachusetts mandated screening, or in 2018, when the FDA approved pegvaliase, or in 2025, when sepiapterin received approval for sepiapterin-responsive disease. The global version has not started.

Where screening exists

PKU newborn screening is universal or near-universal across Western Europe, North America, Australia, Japan, South Korea, and most of the former Soviet bloc. It is also universal in a handful of middle-income countries that built programs on consanguinity grounds, where the prevalence was high enough that the cost arithmetic forced action.

Türkiye is the clearest case. Persistent hyperphenylalaninemia is detected in roughly 1 in 4,000 live births in Türkiye, an excess attributed to consanguineous marriage rates near 24 percent. The national PKU screening program began in 1986 and was extended to all 81 provinces in 1994. Congenital hypothyroidism was added in 2006.

Iran's national PKU screening program launched in 2006. Coverage scaled up over the following decade and the program now reaches most of the country. A 2020 systematic review and meta-analysis in BMC Pediatrics by Vela-Amieva and colleagues placed Iranian PKU prevalence at roughly 1 in 5,000, also tied to consanguinity.

Saudi Arabia, the United Arab Emirates, and Qatar run expanded MS/MS screening programs. The UAE began PKU screening in 1995. Qatar reports an expanded panel covering more than 25 conditions. Egypt has a national congenital hypothyroidism program that reaches around 75 percent of newborns, and a tertiary program at Cairo University Children's Hospital screens for an expanded metabolic panel, but national PKU coverage in Egypt is partial.

China's program began in 1981 with PKU and congenital hypothyroidism in two cities. National coverage rose from 3.86 percent in 2003 to 59 percent in 2009, and screening for PKU and congenital hypothyroidism is now reported as nationwide, although rural and migrant populations remain undercounted. Hillert and colleagues estimated Chinese PKU prevalence at 1 in 15,924, close to European rates.

Latin America runs the most patchwork system on the planet. Borrajo, in a 2016 review in the International Journal of Neonatal Screening, mapped the region. Cuba, Costa Rica, Chile, and Uruguay reach more than 99 percent of newborns. Argentina mandated PKU screening in 1986. Costa Rica's national program launched in 1990. Brazil's national program began in 2001 and reports more than 80 percent coverage with PKU on the panel. A 2022 follow-up by Borrajo in Frontiers in Genetics found that 14 Latin American countries include PKU on their panels and that average PKU coverage across those countries is around 92 percent of births. The other countries in the region, Bolivia, Honduras, Peru, run partial or pilot programs.

Where screening does not exist

India has no national newborn screening program. The state of Chandigarh launched a program in 2007 for congenital hypothyroidism, congenital adrenal hyperplasia, and G6PD deficiency, screening around 15,000 births a year across four government hospitals. Kerala screens roughly 140,000 births a year through more than 90 government facilities. Goa runs a smaller program. PKU is included in some pilot panels but not in routine state coverage. The country has roughly 25 million births a year. The pilot programs together cover a small fraction of one percent of them.

Sub-Saharan Africa, with around 40 million births a year, has almost no PKU screening. Pilot programs for PKU and congenital hypothyroidism ran in South Africa in the 1960s and 1980s and were discontinued in the face of competing priorities. The continent's working newborn screening efforts focus on sickle cell disease, where the disease burden is enormous and the intervention is comparatively cheap. A 2023 review by Satekge and colleagues in the African Journal of Laboratory Medicine described organized newborn bloodspot screening as concentrated along the northern coast and effectively absent across most of the continent. Most of Southeast Asia outside of Japan, South Korea, Taiwan, and Singapore is similar.

The structural pattern across both regions is the same. Hospitals deliver babies. Mothers go home with newborns. By the time a child with classical PKU is identified, around the age of one or two, the IQ is already at 50 or below.

What untreated PKU produces

A 2018 systematic review in the Orphanet Journal of Rare Diseases by Van Spronsen and colleagues reviewed the outcomes of late-treated and untreated PKU. The classical presentation is severe to profound intellectual disability, with mean IQ in the range of 40 to 50 by the second year of life, plus seizures, behavioral disturbance, and a characteristic musty odor from accumulating phenylpyruvic acid. About 1 to 2 percent of people with classical PKU appear to escape intellectual disability without treatment, a fraction small enough that it does not change the population picture.

The damage is permanent. A child diagnosed at age 5 in a Lagos hospital with profound developmental delay and a positive ferric chloride urine test cannot be rescued by starting the diet at age 5. The diet prevents accumulation. It does not undo it. This is the same fact that drove Robert Guthrie to spend the late 1950s building the bacterial inhibition assay and the early 1960s fighting the AMA over mandatory screening. The window closes in the first weeks of life.

The medical food problem

Even where screening exists in a low- or middle-income country, treatment access often does not.

Treatment requires a phenylalanine-free amino acid mixture for protein, modified low-protein staple foods for calories, and ongoing biochemical monitoring. The annual wholesale cost of the formula in the United States ranges from around $1,200 for an infant to more than $8,000 for an adult, with end-user pricing typically two to three times wholesale. The market is concentrated. Three companies, Nutricia, Vitaflo, and Cambrooke, supply most of the global formula volume. Their distribution maps prioritize high-income markets where insurance reimbursement, mandates, or national health systems pay for the product.

In a country without a reimbursement system, a family pays cash. A formula bill of several thousand dollars a year, rising as the child grows, is not survivable for most households on most continents. PKU clinicians in Türkiye, Iran, and Latin America have published repeatedly on the same problem. The Vela-Amieva 2022 meta-analysis in the Orphanet Journal of Rare Diseases reported that metabolic control across Latin American PKU populations was substantially worse than in high-income comparator cohorts, with poor adherence driven by formula cost and intermittent supply.

Cross-border supply is unreliable for a separate reason. Medical foods sit in a regulatory category that varies from country to country. Some countries treat them as food, some as drugs, some as nothing in particular. A formula that ships freely within the European Union may face import duties, customs holds, or outright bans crossing into a neighboring country. PKU families in countries without a domestic distributor have, for years, relied on personal couriers, charitable shipments organized by national PKU societies, and irregular goodwill imports from manufacturers. None of that scales.

The newer drugs make the access gap larger. Sapropterin, a synthetic BH4 cofactor that lowers blood phenylalanine in the responsive subset of PKU, was approved by the FDA in December 2007 and by the EMA in 2008. BioMarin reports approvals in around 51 countries. Generic sapropterin became available in the United States in 2018. Pegvaliase, an injectable enzyme substitution therapy approved by the FDA in May 2018, is a high-cost product with a Risk Evaluation and Mitigation Strategy program built around anaphylaxis monitoring. Its global footprint is narrow. Sepiapterin, approved by the FDA in July 2025 for sepiapterin-responsive disease at any age from one month, is at the start of its global rollout.

A child born with PKU in Türkiye in 2026 has a meaningful chance of being identified at birth, started on the diet, and offered sapropterin if responsive. A child born with PKU in Tanzania has none of that.

Why the global frame matters for data

The high-income PKU literature has spent forty years arguing about the optimal target range for blood phenylalanine. United States guidelines call for sustained levels between 120 and 360 µmol/L. European guidelines accept slightly different ranges by age. The arguments hinge on cohort studies of treated populations, mostly in Europe and North America, who entered care in the 1960s and 1970s and are now adults.

Those cohorts are not representative of the global PKU population. They were screened early, treated continuously, and followed by clinics with biochemical monitoring infrastructure. The metabolic phenotype distribution Hillert reported globally, 62 percent classic PKU, 22 percent mild PKU, 16 percent mild hyperphenylalaninemia, was estimated from a sample weighted toward those same high-income cohorts. The genotype distribution outside Europe is different. Türkiye, Iran, and the Arabian peninsula carry a different spectrum of PAH variants, with a higher proportion of severe alleles linked to consanguinity. China carries another distinct spectrum. Sub-Saharan African allele frequencies are largely unknown.

The natural history of PKU in untreated and partially treated populations exists, somewhere, in the medical records of metabolic clinics in Tehran, Istanbul, São Paulo, Mexico City, Cairo, Riyadh, and a long list of other cities. None of it is aggregated. The Iran national program has been running for 20 years. The Türkiye national program has been running for 32 years. The combined longitudinal record on PKU treatment outcomes from those two countries alone would be one of the most important datasets in metabolic medicine, and it does not exist as a dataset. It exists as paper charts and disconnected hospital information systems.

Without that data, the next generation of PKU therapies will be developed and approved on a population that is geographically narrow. Pegvaliase response rates are reported on cohorts enrolled in the United States and Europe. Sepiapterin response rates are similar. Beam Therapeutics' base-editing program, which advanced in 2026 with an Investigational New Drug filing for in vivo PAH editing, will run its early trials in the same geographies. The variants the trial enrolls will be the variants common in those geographies. The variants that drive PKU in Türkiye, Iran, India, and East Asia may or may not be on the trial at all. Approval will follow enrollment. Access will follow approval.

The pattern is older than PKU. Most rare disease drugs are developed using high-income cohorts and approved on the basis of evidence collected from those cohorts. Insurers in low- and middle-income countries then face a label that does not include local genotypes, no local outcome data, and a price set against the income of a different population. Coverage decisions stall.

What would change this is data infrastructure that crosses borders. A longitudinal record of people with PKU, structured the same way in every country, owned by the families who contribute it, with continuous follow-up that does not depend on a single grant or a single ministry of health, would do for the global PKU population what the Türkiye and Iran national programs have done for screening within their borders. It would give the next set of regulators the natural history data they need to extend approvals beyond the cohorts the trials enrolled. It would give clinicians in countries without domestic screening programs a baseline against which to argue for them.

The biology of PKU is the same in every country where a child is born. The infrastructure that determines whether the child develops normally is not. The 17 years between Følling's discovery in 1934 and the Birmingham diet in 1951, and the 12 years between the diet and the first state mandate in 1963, were the cost of building each piece of the high-income system one country at a time. Doing it again, country by country, for the rest of the world is a project measured in decades. Aggregating the data across borders is a project measured in years.