International experiences of genomic newborn screening: Lessons from over 10,800 newborns.

Genomic sequencing integrated into newborn screening from dried blood spots: comparison of four implementation studies conducted in the US, Belgium and Australia

Genomic sequencing could transform newborn screening for rare diseases but raises significant practical, clinical, psychosocial, ethical and policy issues, and evidence is urgently needed to guide health systems. In 2025, four major genomic newborn screening (gNBS) studies, totalling over 10,800 newborns from the US, Belgium and Australia, published initial results, and their different approaches to key implementation issues allow them to be compared here. All reported successful DNA extraction from dried blood spots, which supports integration with existing newborn screening infrastructure, while experience with automation of genomic data analysis and reporting was variable, with insights on balancing sensitivity, specificity, scalability and resource use. The screen-positive rate ranged from 1.6% to 3.7%, with G6PD deficiency by far the most common condition identified, but further comparison of results was limited by wide variation in the number and type of conditions included and by inconsistent definitions of outcomes. The ability to deliver gNBS at the scale and pace of a public screening programme remains untested, data on long-term outcomes and costs are still needed, and concerns persist about inequitable access and outcomes for families at socioeconomic, linguistic and geographical disadvantage.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

The most useful result is not a figure but a finding of feasibility paired with a finding of incomparability: extraction from dried blood spots works, but screen-positive rates of 1.6 to 3.7%, led by G6PD, cannot be compared while condition lists and outcome definitions differ. For a decision-maker, the lesson is therefore the need to harmonise protocols before any extrapolation, rather than an argument for deployment. This is a comparison of four studies chosen as major ones, the abstract reporting neither a pooled estimate nor any performance measure (sensitivity, predictive value), and scaling up remains, by the authors' own admission, untested.

Analysis by Dr Thibaut Benquey