A dispute over the single number most often quoted to at-risk sons and daughters in families with genetic prion disease has now played out in print. In European Journal of Neurology in December 2025, Izaro Kortazar-Zubizarreta and colleagues in Vitoria-Gasteiz, Spain reported transmission ratio distortion in inherited prion disease: across 24 pedigrees (12 per variant), 65 nuclear families and 151 offspring, the variant was passed on in 67.1 percent of offspring for PRNP p.D178N (95 percent confidence interval 56.3 to 77.9) and 70.5 percent for p.E200K (60.4 to 80.6), against the textbook 50 percent of autosomal dominant inheritance. On 2 March 2026 the journal published a letter from Akin Nihat, Tze How Mok, John Collinge and Simon Mead of the MRC Prion Unit at UCL and the NHS National Prion Clinic, with Eric Vallabh Minikel of the Broad Institute, arguing that ascertainment bias could account for the excess.
The Spanish work was a retrospective analysis of existing family trees, not a prospective study of pregnancies, and transmission ratios were compared with the expected 50 percent using generalised estimating equations. Sex-specific figures were 67.5 percent maternal and 66.7 percent paternal transmission for p.D178N (the paternal comparison at p = 0.056), and 67.3 percent maternal and 78.3 percent paternal for p.E200K. Families were recruited through specialist neurology and genetic counselling services at a single site and through the Spanish Prion Disease Foundation. Inheriting a PRNP variant is not the same as developing illness; age at onset and lifetime risk differ by variant, and the two questions are counselled separately.
The letter's objection is about who ends up in such a sample. Reaching a specialist service, having a familial pedigree recognised and having offspring tested are all, the authors write, more likely in families where transmission has already declared itself, so offspring whose genotype became known only because they fell ill are over-represented. They say the concern would be reduced if every member of each nuclear family had been genotyped, or if only people undergoing predictive testing had been counted, and that neither was done. Until the finding is repeated in large, prospective, population-based cohorts, they argue, "the 50% Mendelian risk should remain the ethically and statistically sound standard for genetic counselling".
The authors' reply, published online on 25 February 2026, states that "we must respectfully but firmly disagree that ascertainment bias substantially undermines our findings". They distinguish how families entered the study from how transmission was counted inside it: large pedigrees were split into nuclear units, each contributing all carrier and non-carrier offspring with known genetic status, so branches with low transmission were counted alongside the branch that brought the family to attention. They report that 35.4 percent of nuclear families had transmission at or below 50 percent, some between 25 and 33 percent, within pedigrees that also contained branches above 67 percent, and offer a worked example in which, assuming 80 percent penetrance by age 80 and three offspring, the chance of at least one affected child differs by about 12 percentage points between a 50 percent and a 67 percent family, which they say is too small to filter out lower-transmission families.