Genetic prion disease: research

How far has PRNP lowering reached in animals?

Three approaches from the same research community have now reported results, all in animals.

An and colleagues, in Nature Medicine in January 2025, delivered a dual-AAV PHP.eB system carrying the base editor BE3.9max and a single guide RNA that installs a PRNP R37X edit. They report 37 percent average installation of the intended edit, a 50 percent reduction in brain prion protein, and a 52 percent extension of lifespan in transgenic human PRNP mice inoculated with pathogenic human prion isolates representing the most common sporadic and genetic subtypes. Engineered variants gave 63 percent average brain prion protein reduction from a 6.7-fold lower viral dose. They report no off-target editing of anticipated clinical significance in human cells or mouse tissues, and they describe in vivo base editing as potentially a one-time treatment.

Chou and colleagues, in a 2025 bioRxiv preprint not yet peer reviewed in that record, used adeno-associated virus to deliver zinc finger repressors that block prion protein production at the transcriptional level. They report prion messenger RNA reduced by more than 95 percent in vitro, neuronal prion protein lowered across the central nervous system for at least 17 months in wild-type mice, substantially extended lifespan in inoculated mice treated early or late, and repression in all 35 brain regions assessed in cynomolgus monkeys after a single intravenous dose of a blood-brain-barrier-penetrant capsid. The abstract describes the survival extension without giving lifespan figures.

The third approach is epigenetic. A BioEssays review of precision epigenetic tools describes CHARM, reported by Neumann and colleagues, as an editor that methylated and silenced the prion gene in mouse brains and then switched itself off. That review does not give a percentage reduction in prion protein, so the quantitative claim for CHARM should be taken from the primary paper rather than from this source.

None of this is evidence of benefit in people. The clinical programmes that follow from it are listed in CJD research.

Which assays detect which mutations?

Assay sensitivity in genetic prion disease depends on the mutation and on the substrate, which bears directly on how a negative test should be interpreted.

In the Myskiw and colleagues panel, the genetic and atypical cohort broke down as fatal familial insomnia with D178N 13, Gerstmann-Straussler-Scheinker with P102L 10, E200K 7, octapeptide repeat insertions 4, V210I 2, genetic CJD with D178N 1, V180I 1, P105S 1 and A117V 1. Detection differed sharply by substrate. Fatal familial insomnia was detected in 12 of 13 by the deer mouse recombinant end point assay and in 0 of 13 by truncated hamster RT-QuIC. P102L was detected in 10 of 10 against 4 of 10. A117V and P105S were each detected by the deer mouse assay and not by truncated hamster RT-QuIC. E200K, V210I and the octapeptide repeat insertions were detected by all three formats tested.

That paper carries internal inconsistencies between abstract and tables, including an inherited case count of 50 in the abstract against 40 inherited plus 10 atypical sporadic cases in Table 2, and the authors list independent validation, inter-laboratory reproducibility and protocol harmonisation as outstanding.

Steinacker and colleagues included 9 genetic CJD patients in their neurofilament cohort, with mean serum neurofilament light of 223 pg per ml against roughly 26 to 49 in controls, and reported that in a single presymptomatic P102L carrier, cerebrospinal fluid neurofilament light and phosphorylated neurofilament heavy were already above the control 75th percentile while CSF tau was normal and RT-QuIC was negative, turning positive at onset. One carrier cannot establish a presymptomatic marker. The wider marker work is in biomarkers.

What did the transmission ratio study find?

Kortazar-Zubizarreta and colleagues examined whether PRNP variants are passed on at the expected Mendelian rate. They analysed 24 pedigrees, 12 for D178N and 12 for E200K, comprising 65 nuclear families and 151 offspring with known genetic status, determined by direct testing for most and by obligate-carrier inference for 18 percent.

They report 49 of 73 offspring carrying D178N, or 67.1 percent with a 95 percent confidence interval of 56.3 to 77.9 percent and p equal to 0.0035; 55 of 78 carrying E200K, or 70.5 percent with an interval of 60.4 to 80.6 percent and p below 0.001; and 104 of 151 combined, or 68.9 percent with an interval of 61.5 to 76.2 percent. They state that embryo selection strategies in preimplantation genetic diagnosis programmes may need reconsideration, because equal allele representation may not hold for PRNP variants, and they call for clinical studies of the effect on reproductive outcomes in those programmes and for single-cell genomics to determine whether any distortion occurs at the gamete or embryo level.

Their stated limitations are substantial and should travel with the result: a retrospective design, recruitment through clinical services, 18 percent of genotypes inferred rather than tested, exclusion of 15 of 39 identified families, post hoc power analysis only, mechanisms untested and described by the authors as speculative, and clustering of D178N in the Basque population, where founder effects or population-specific modifiers could influence the figures. This is a pedigree-level statistical finding about two variants in one study. It is not a figure that applies to an individual, and what any family’s situation means is a question for a clinical genetics service.

What reproductive options are described in the literature?

Uflacker and colleagues reported in JAMA Neurology in 2014 what they describe as the first case of preimplantation genetic diagnosis with in vitro fertilisation used to prevent genetic prion disease, in a 27-year-old asymptomatic woman from a family with Gerstmann-Straussler-Scheinker syndrome due to PRNP F198S. The cycle produced 6 embryos free of the variant, two were transferred, and a twin pregnancy resulted in healthy, slightly premature infants developing normally at 27 months. The authors conclude that in vitro fertilisation with preimplantation genetic diagnosis is a viable option for couples who wish to avoid passing on the disease, and that neurologists should be aware of it so they can advise at-risk families. The author institutions were Duke University, the Reproductive Genetics Institute in Chicago, the Institute for Human Reproduction and the University of Chicago, and the UCSF Memory and Aging Center.

Shir and colleagues set out a practice framework from Israel, developed by 29 professionals across neurology, medical genetics, genetic counselling, nursing and medical psychology, with a family representative revising the draft, agreed at a concluding online meeting in July 2025. Its recommendations include that asymptomatic relatives be offered genetic counselling, that counselling be provided by genetic health professionals or by neurologists familiar with the disease, that testing not be offered under the age of 18, that people in acute distress be encouraged to delay testing by at least six months, that pre-test psychological appraisal be required, that results be delivered in person by the clinician or counsellor who gave the initial counselling, and that structured support programmes and research opportunities be made available to carriers.

Recommendation 5 states that preimplantation and prenatal genetic testing for genetic CJD should be considered case by case with individualised decision-making. The framework describes exclusion testing, in which embryos are selected to exclude the allele from the affected grandparent while the at-risk parent’s own status remains unknown, as available in Israel at tertiary medical centres to first- and second-degree relatives of people with genetic CJD, and states that such testing is publicly funded there for up to two healthy offspring. It notes that prenatal diagnosis is possible by chorionic villus sampling from about 11 to 12 weeks or amniocentesis from about weeks 17 to 23, that the decision after a positive prenatal result presents significant ethical challenges, and that national and cultural approaches vary and are beyond the scope of that paper.

These arrangements are country-specific. Hospital Clinic de Barcelona describes its PICOGEN programme of information and genetic counselling for familial dementias, which explains to relatives of people with genetic prion disease the chance of inheriting the condition, offers them the choice of learning whether they carry the variant, and offers steps to keep it from passing to their children, without naming a specific technique. Its Alzheimer’s and other cognitive disorders unit lists cerebrospinal fluid 14-3-3, PRNP genetic study, brain MRI, glucose PET, electroencephalogram, polysomnography and post-mortem study among its investigations.

Nothing here indicates what any individual should do. Genetic counselling comes before any genetic test, and choosing not to be tested is a valid decision.

Where is this work funded and done?

The 2025 transmission ratio study came from Bioaraba and CIC bioGUNE, with funding listed from the Spanish Agencia Estatal de Investigacion, Interreg POCTEFA grant EFA031/01 NEURO-COOP, Instituto de Salud Carlos III grants forming part of a JPND award, the Basque Foundation for Innovation and Health Research EiTB Maratoia grant BIO22/ALZ/012, and CIC bioGUNE’s Severo Ochoa accreditation. The paper discloses that one author is employed by ATLAS Molecular Pharma and states that the company had no role in the study.

The organisations and programmes involved are described in Bioaraba, Hospital Clinic Barcelona, ATLAS Molecular Pharma, BIOEF, REDPRION, EUROCJD and the AP-HP Pitie genetics department. The disease group as a whole is set out in disease research.