Fatal familial insomnia (FFI): more detail

What defines the condition genetically?

FFI is a clinicopathological phenotype rather than a separate gene defect. It is produced by the PRNP D178N substitution when codon 129 methionine sits in cis with the mutated allele. The same substitution with 129 valine in cis typically produces a genetic CJD phenotype instead. Reporting a D178N result therefore requires the phase of codon 129, and a report that gives the substitution without it is incomplete for counselling purposes.

Codon 129 on the other allele modifies the course. Reviews note that homozygotes have a shorter mean disease duration than heterozygotes, so a 129MM carrier and a 129MV carrier in the same family are not in the same position prognostically.

GARD describes almost all cases as arising from this one variant, and it is the rarest of the inherited prion diseases, with at least 25 families described worldwide. D178N is concentrated in some populations, notably the Basque country, which is relevant when interpreting cohort data drawn from a single region.

What does the clinical phenotype look like?

Onset is typically in midlife, and disease duration is typically longer than a year, which separates FFI from most sporadic CJD.

The syndrome combines three elements. Sleep disruption comes early and is progressive, with loss of normal sleep architecture rather than ordinary insomnia. Autonomic instability is prominent, including hypertension, hyperhidrosis, pyrexia and tachycardia. Cognitive impairment, ataxia and myoclonus follow, overlapping with the familiar CJD picture. GARD also lists anxiety, depression, hallucinations, weight loss and loss of speech.

The autonomic features are a practical point in ward care, because they are often attributed to agitation or to medication before the diagnosis is considered.

Why do the standard investigations underperform?

Because the phenotype is driven by thalamic pathology with a different aggregate conformation, and the standard tests were calibrated on sporadic disease.

Reviews report that periodic sharp wave complexes on EEG, CSF 14-3-3 and typical MRI changes are usually absent in FFI. Where family history is unknown, this makes clinical diagnosis difficult. The investigations that do support it are polysomnography, which shows disrupted sleep architecture, and fluorodeoxyglucose PET, which shows thalamic hypometabolism. Neither has published sensitivity figures in the review literature, so both are suggestive rather than confirmatory. Genetic testing is definitive.

Seeding assay behaviour is the sharpest illustration. A 2026 retrospective study from the Public Health Agency of Canada’s National Microbiology Laboratory and Case Western Reserve University compared three assays on 116 diagnostic CSF samples, comprising 61 sporadic CJD, 50 inherited prion disease and 5 variably protease-sensitive prionopathy. A modified end-point assay using a deer mouse recombinant substrate reached 96.6 percent sensitivity against 87.1 percent for the standard hamster end-point assay and 72.4 percent for second-generation IQ-CSF RT-QuIC.

For the 13 FFI D178N samples the figures were 12 of 13 with the deer mouse substrate, 4 of 13 with the hamster end-point assay and 0 of 13 with IQ-CSF RT-QuIC. The same paper reported 10 of 10 for GSS P102L with the deer mouse substrate against 4 of 10 by IQ-CSF, and the single V180I case was detected only by IQ-CSF.

Two cautions attach to these numbers. The samples were diagnostic submissions tested retrospectively, not a prospective series, and the paper’s own abstract reports 111 of 116 alongside 96.6 percent, which corresponds to 112 of 116. Anyone quoting the headline sensitivity should check that arithmetic against the table.

The clinical consequence is that a negative CSF seeding result in a suspected FFI case is close to uninformative if the laboratory runs IQ-CSF, and that the assay and substrate should be named when a result is recorded. Reviews also note that reported RT-QuIC sensitivity in FFI appears to differ by population, with negative results common in European cases. How prion disease is diagnosed sets out the test panel generally.

Is FFI a distinct prion type?

The neuropathological evidence suggests it does not fit the sporadic CJD typing scheme. A 2021 study from Homburg examined aggregates from six FFI brains by western blot, a guanidine hydrochloride stability assay and paraffin-embedded tissue blot. The protease-resistant fragment was about 19 kDa in every case, which corresponds to sporadic CJD type 2, while the deposition pattern was fine synaptic in grey matter, resembling type 1, and most aggregates lost stability at low denaturant concentrations, again resembling type 1. One case was more stable than the rest.

The authors concluded that fragment size alone is insufficient for typing and that FFI may represent a conformation of its own, sharing properties with both types, with the caveat that the cases may not all share a uniform template. The series is six brains, so this is a well-argued proposal rather than a settled classification.

What should counselling cover for well relatives?

The published sequence applies here as in the other inherited forms: counselling before any test, time between sessions, and results given in person. Genetic counselling comes first and predictive testing set out the protocols, and declining a test is a recognised choice.

One open question deserves mention because families now find it themselves. A 2025 European Journal of Neurology study of 24 pedigrees, 65 nuclear families and 151 offspring reported transmission of 67.1 percent for D178N, with a 95 percent confidence interval of 56.3 to 77.9, and 70.5 percent for E200K, interval 60.4 to 80.6, against the 50 percent in current guidance. The authors proposed meiotic drive, gametic competition or embryonic selection as hypotheses. The design is retrospective, 18 percent of genotypes were inferred rather than tested, and the D178N families are largely Basque; the authors call for prospective validation across populations before counselling protocols change. A family who has read the paper should hear both the figure and that caveat.

Where next?

Fatal familial insomnia is the plain version for families, and FFI for researchers the scientific one. What to expect covers symptom management and palliative involvement, and current trials lists studies with sponsor-stated eligibility.

This page summarises published literature and is not clinical or genetic advice. No figure here describes the position of any particular person or family.