Creutzfeldt-Jakob disease (CJD): more detail

Forms and epidemiology

Classic CJD causes about 1 to 2 deaths per million people per year worldwide, with about 500 to 600 cases reported annually in the United States. Around 85 percent of reported cases are sporadic. Inherited disease, caused by a pathogenic variant in PRNP and transmitted in an autosomal dominant pattern, accounts for about 5 to 15 percent of cases, and MedlinePlus puts the inherited share of prion disease generally at 10 to 15 percent. Iatrogenic cases make up 1 to 2 percent of prion disease. The CDC records six known cases attributable to contaminated surgical or medical equipment, with none reported since 1976.

Mean age at onset is in the late 60s. Presentation under 30 is rare and should prompt consideration of genetic and acquired forms.

Clinical course

The core syndrome is a rapidly progressive dementia accompanied by some combination of myoclonus, cerebellar ataxia, visual disturbance including cortical visual failure, and pyramidal or extrapyramidal signs. Akinetic mutism, a state of wakefulness without speech or voluntary movement, appears late. Median survival from first symptom is a few months, and most people die within a year of onset.

Rate of decline is measured in research and monitoring settings on the MRC Prion Disease Rating Scale, which the UK National Prion Monitoring Cohort uses for serial assessment and which supplied the matched controls in the PRN100 programme described below.

Molecular subtypes

Sporadic CJD is not uniform. Subtypes are defined by the codon 129 genotype of PRNP, which may be methionine homozygous, valine homozygous or heterozygous, together with the type of protease-resistant PrP found in brain. The resulting subtypes differ in presentation, duration and test performance.

Test sensitivity follows subtype, which has a practical consequence at the bedside. The 2022 review by Poleggi and colleagues records lower RT-QuIC sensitivity in the atypical forms, naming variably protease-sensitive prionopathy, MM2-thalamic, MM2-cortical and VV1, and in some genetic forms including fatal familial insomnia and Gerstmann-Straussler-Scheinker disease. A negative assay in a patient whose phenotype suggests one of these does not exclude prion disease.

Diagnostic criteria and their accuracy

The CDC criteria classify cases as definite, probable or possible. Definite requires brain tissue, by standard neuropathology, immunocytochemistry, western blot for protease-resistant PrP, or scrapie-associated fibrils. Probable is met by a neuropsychiatric disorder with a positive RT-QuIC in CSF or other tissues, or by rapidly progressive dementia with at least two of myoclonus, visual or cerebellar signs, pyramidal or extrapyramidal signs and akinetic mutism, plus a positive EEG, CSF 14-3-3 or MRI result, and no alternative explanation on routine investigation.

The 2017 revision of the International CJD Surveillance Network criteria added cortical ribboning on MRI and the RT-QuIC assay. Watson and colleagues validated it in 2022 against 501 autopsy-confirmed sporadic CJD cases and 146 non-cases from surveillance centres in the United Kingdom, France, Germany and Italy. Overall sensitivity rose from 77.5 percent to 92.2 percent, while specificity was essentially unchanged at 81.6 percent before and 80.8 percent after. In the subgroup that had the full investigation panel, 223 cases and 52 non-cases, sensitivity rose from 76.2 to 97.8 percent with specificity 69.2 and 67.3 percent. Among individual tests in that study, RT-QuIC was 91.6 percent sensitive and 100 percent specific, and cortical ribboning on MRI was 67.9 percent sensitive and 86.5 percent specific.

The 2021 Lancet Neurology review that set out the amended criteria describes RT-QuIC as a major advance for confident diagnosis before death, and notes that conflicting studies have made some of the indirect CSF markers controversial. How prion disease is diagnosed: more detail covers the individual tests, and diagnosis in plain words is the version to give a family.

Management

No treatment alters the course of the disease. Management is symptomatic and supportive: myoclonus, agitation, pain, seizures, dysphagia, insomnia and the care needs that follow rapid functional loss. Early involvement of palliative care and of the community team is the practical consequence of the short prognosis, and most families need help arranged faster than services usually move. What to expect sets out the course for families and carers.

Where the form is or may be genetic, relatives’ questions belong with a genetic counsellor, before any testing is arranged and with the decision not to test treated as a full and reasonable answer.

The treatment work so far

PRN100, a fully humanised IgG4κ monoclonal antibody against cellular prion protein, was given to six patients with probable CJD at University College London Hospitals between 9 October 2018 and 31 July 2019, under a Specials Licence, outside the regulated trial route. Patients were titrated from 1 mg/kg and 10 mg/kg, then given 80 to 120 mg/kg intravenously every two weeks, for 7 to 260 days. Progression was compared with severity-matched, subtype-matched and codon 129-matched untreated controls from the National Prion Monitoring Cohort.

Repeated dosing was well tolerated, with no clinically significant adverse reactions. Four of the six reached the target CSF concentration of 50 nM, after 22 to 70 days. All six showed progressive neurological decline on serial MRC Prion Disease Rating Scale assessment, and the paper reports no evidence of slowing relative to controls. Two patients were autopsied, with no cytotoxicity found; in one, treated for 140 days, the pattern of disease-associated PrP was altered, which the authors attribute to drug effect, and brain tissue drug concentrations ranged from about 66 to 182 nM across regions. The authors conclude that the programme appeared safe, reached encouraging CSF and brain concentrations, and justifies formal efficacy trials at the earliest clinical stages and as prophylaxis in people at risk. The CJD Foundation currently lists PRN100 as not enrolling.

What is in the clinic now

The CJD Foundation’s drug development page, carrying statements through October 2026, lists an Ionis antisense oligonucleotide targeting PRNP messenger RNA, with enrolment of one regimen reported complete in a statement of 18 September 2026; the PRiSM study run through NeuroNEXT with the Broad Institute, reported as recruiting in a statement of 22 April 2026, with the presymptomatic arm reported open on 14 October 2026; a Regeneron divalent small interfering RNA programme for adults with symptomatic prion disease, listed on the trials register; preclinical work at Gate Bioscience; and a Sangamo zinc finger programme marked not currently enrolling. These are sponsor statements on those dates and say nothing about whether any of these agents works.

Current trials tracks eligibility and status, and CJD for researchers holds the mechanistic and preclinical work.

Classification, interpretation and eligibility for any study are decisions for the treating neurologist and the trial site.