Fluid Biomarkers in Individuals at Risk for Genetic Prion Disease up to Disease Conversion

Abstract

Objectives: To longitudinally characterize disease-relevant CSF and plasma biomarkers in individuals at risk for genetic prion disease up to disease conversion.

Methods: This single-center longitudinal cohort study has followed known carriers of PRNP pathogenic variants at risk for prion disease, individuals with a close relative who died of genetic prion disease but who have not undergone predictive genetic testing, and controls. All participants were asymptomatic at first visit and returned roughly annually. We determined PRNP genotypes, measured NfL and GFAP in plasma, and RT-QuIC, total PrP, NfL, T-tau, and beta-synuclein in CSF.

Results: Among 41 carriers and 21 controls enrolled, 28 (68%) and 15 (71%) were female, and mean ages were 47.5 and 46.1. At baseline, all individuals were asymptomatic. We observed RT-QuIC seeding activity in the CSF of 3 asymptomatic E200K carriers who subsequently converted to symptomatic and died of prion disease. 1 P102L carrier remained RT-QuIC negative through symptom conversion. No other individuals developed symptoms. The prodromal window from detection of RT-QuIC positivity to disease onset was 1 year long in an E200K individual homozygous (V/V) at PRNP codon 129 and 2.5 and 3.1 years in 2 codon 129 heterozygotes (M/V). Changes in neurodegenerative and neuroinflammatory markers were variably observed prior to onset, with increases observed for plasma NfL in 4/4 converters, and plasma GFAP, CSF NfL, CSF T-tau, and CSF beta-synuclein each in 2/4 converters, although values relative to age and fold changes relative to individual baseline were not remarkable for any of these markers. CSF PrP was longitudinally stable with mean coefficient of variation 9.0% across all individuals over up to 6 years, including data from converting individuals at RT-QuIC-positive timepoints.

Discussion: CSF prion seeding activity may represent the earliest detectable prodromal sign in E200K carriers. Neuronal damage and neuroinflammation markers show limited sensitivity in the prodromal phase. CSF PrP levels remain stable even in the presence of RT-QuIC seeding activity.

Clinical trials registration: ClinicalTrials.gov NCT05124392 posted 2017-12-01, updated 2023-01-27.

Figure 1. Fluid Biomarker Changes in the Cohort

(A) RT-QuIC kinetic curves for N = 149 CSF samples tested (98 from carriers and 51 from controls), showing 6 positive CSF samples (each with 4/4 replicates positive). (B) RT-QuIC endpoint titration of N = 10 CSF samples from 4 unique individuals who developed disease, including the 6 positive CSF samples from 3 E200K converters, with codon 129 genotypes of converters indicated. (C) CSF PrP concentrations represented as changes (∆) relative to individual baseline, shown for the 4 converters plus all individuals with at least 3 years between first and last available CSF sample. N = 76 CSF samples from 19 unique individuals, see eTable 4. (D–H) Biomarkers plasma GFAP (N = 158 samples from 61 unique individuals) (D), plasma NfL (N = 160 samples from 62 unique individuals) (E), CSF NfL (N = 155 samples from 60 unique individuals) (F), CSF T-tau (N = 151 samples from 60 unique individuals) (G), and CSF β-syn (N = 150 samples from 60 unique individuals) (H) are represented by 2 views each. Left: individual age vs absolute concentration in pg/mL, with sequential samples from the same individual connected by thin lines, while thicker lines represent the separate log-linear best fit curves for controls and for nonconverting carriers. Right: years from disease onset vs change (∆) relative to individual baseline in converters, with the same for controls and for nonconverting carriers shown on a separate x-axis. Dashed lines connect timepoints before and after symptom onset. For further breakdown and statistics, see eTable 5.