Sizing a Disease You Catch at Birth: What Newborn Screening Does to the Forecast

Newborn screening is the best thing that can happen to a patient with a rare disease, and one of the hardest things that can happen to a forecast built on prevalence. When a condition enters a screening panel, it stops being a stock of prevalent patients a launch penetrates over years and becomes something different in kind: a thin, steady flow of infants identified at birth, sitting on top of a prevalent backlog that a one-time therapy empties as it treats. Spinal muscular atrophy shows what that does to the number.

Screening changes the market's shape

In a chronically diagnosed disease the market is a pool you convert at a rate set by access and adoption. Screening replaces that pool with two objects that behave differently. The first is the prevalent backlog, the patients already alive with the disease, finite and, for a one-time therapy, depleting. The second is the incident flow, the newly screened infants each year, small and fixed by the birth rate. A forecast that blends them into one prevalence figure gets both the size and the timing wrong.

The incident and prevalent split, in SMA

Spinal muscular atrophy has an estimated 8,000 to 10,000 prevalent patients in the United States, a pool that shrinks as one-time gene therapy treats it. Against that, newborn screening identifies roughly 300 pre-symptomatic infants a year, the incident flow that will carry the market once the backlog clears. A one-time therapy priced around $2.1 million cannot be modelled as an annuity penetrating a stock. It is a depletion curve, a catch-up wave on the prevalent backlog settling to the incident birth-cohort rate.

Coverage is the variable that decides the number

How many patients are found at all is a policy question, not an epidemiological one, because screening panels differ by geography. The US recommended panel covers roughly 35 core conditions plus 26 secondary, and SMA reached 48 of 53 US programmes by late 2022, but European panels vary country to country and added SMA only recently and unevenly. The Gulf moves fast where the state builds it: Saudi Arabia added the enzyme assay for Pompe disease to its national panel in 2021 and projects 25 to 35 infantile-onset detections a year. Addressable volume tracks screening penetration by market, not raw incidence.

Size the flow and the depletion, not the pool

A defensible forecast models two curves, not one number. The prevalent backlog is a depleting stock, sized by the catch-up wave a one-time therapy can clear given treatment-centre capacity. The incident flow is screened births multiplied by panel coverage and genetic penetrance, built up geography by geography off each market's screening policy. Get the split right and the forecast shows the market as it behaves, a bolus followed by a trickle. Blend it into prevalence and you overstate the near term and misjudge exactly when it falls.

Screening rewrites the commercial question

The question stops being how large the prevalent market is and becomes how fast the panel finds new patients and how quickly a one-time therapy clears those already waiting. This is one of the shifts behind the rare-disease sizing discipline, and it is why a one-time therapy is modelled as a depleting bolus: when the disease is caught at birth, you size the flow and the depletion, not the pool.