Diagnostic sequencing in sick newborns is not newborn screening
Twenty-five symptomatic neonates and infants under six months, all referred with a suspected genetic disorder and no molecular diagnosis. Seventeen had exome sequencing, eight had genome sequencing. A definitive or likely molecular diagnosis came back in 7 of 25, or 28 per cent, rising to 32 per cent once a RANBP2 susceptibility finding is counted. Three of the seven diagnoses carried established management implications.
The Rare-ID authors are careful about what this does not show. Allocation between exome and genome was chronological rather than randomised, so the two cannot be compared. And they state plainly that a yield drawn from phenotype-selected, symptomatic infants says nothing about what population newborn screening would find in an unselected one. That distinction runs through the rest of this edition.