Predicting Symptom Onset of Genetic Prion Diseases Using Fluid Biomarkers

Prion diseases are rapidly progressive neurodegenerative conditions that are invariably fatal and are caused by an abnormal conformation of the normally produced prion protein. There are no available treatments for prion diseases, and because of diagnostic delays and rapid disease progression, clinically meaningful treatments for these conditions may prove difficult. Previous clinical trials examining the use of flupirtine, quinacrine, pentosan polysulfate, and doxycycline for treating symptomatic prion disease were unsuccessful in prolonging life.¹ However, the putative mechanism of these treatments was aimed at reducing the conversion of the normal prion protein (PrP^c) to the disease-associated prion protein (PrP^Sc). Because PrP^c is a necessary substrate for disease pathogenesis, developing treatments are aiming at reducing its levels, such as a current phase 1/2a study examining the use of antisense oligonucleotides in early symptomatic prion disease.²

Up to 15 percent of human prion diseases are due to a pathogenic genetic sequence variant of the prion protein gene (PRNP). Many of these PRNP sequence variations demonstrate high penetrance, with a high likelihood of disease development for many carriers. Importantly, disease-modifying treatments may have the highest yield for this patient population, given the possibility of delaying or preventing clinical onset of the disease. However, designing effective clinical trials in this study population could prove challenging. Approximately 50 different pathogenic sequence variations of PRNP are known to cause genetic prion disease, all of which are characterized by different ages of disease onset, durations, and clinical disease progressions making clinical trial modeling challenging. In fact, modeling a preventive clinical trial in genetic prion disease based on delaying age at onset would be nearly impossible to accomplish reliably and could last one or more generations.³ These challenges may be ameliorated, however, if there were reliable markers of impending symptomatic prion disease. Such markers would allow the identification of imminent phenoconverters and could significantly improve the efficiency of preventive clinical trials.

In this issue of Neurology®, Vallabh et al.⁴ present the results of a single-site longitudinal natural history study of asymptomatic individuals at risk of genetic prion disease. This study focused on identifying fluid biomarkers that precede clinical onset. The study population comprised PRNP sequence variant carriers (n = 41) and controls (n = 21) who underwent annual cerebrospinal and blood sample analyses and clinical assessments. The biomarkers that were assessed included neurodegenerative and neuroinflammatory markers in plasma (NfL and glial fibrillary acidic protein [GFAP]) and CSF (NfL, total-tau, and beta-synuclein). Markers of abnormal prion protein seeding activity (real-time quaking induced conversion, RT-QuIC) and total PrP levels were also measured in CSF.

Over the course of 4 years, 4 participants in the study of Vallabh et al.⁴ developed symptomatic prion disease. Three participants carried the most common sequence variant that causes genetic Creutzfeldt-Jakob disease (gCJD) (E200K), and one carried the most common sequence variant seen in Gerstmann-Sträussler-Scheinker syndrome (P102L). Unfortunately, none of the neurodegenerative or neuroinflammatory biomarkers were consistently elevated across converters in a way that reliably predicted symptom onset. However, all the E200K carriers that developed disease developed positive RT-QuIC tests at least a year before clinical onset (range: 1.0–3.1 years). The differences in time from RT-QuIC positivity to clinical onset differed by PRNP codon 129 polymorphism, with methionine-valine heterozygotes having the longest time interval, a characteristic that would be expected, given that heterozygotes are known to have longer disease durations and longer incubation periods in acquired prion diseases.⁵ The P102L carrier did not develop a positive RT-QuIC, but this is not unexpected as RT-QuIC is less sensitive in this mutation.⁶ These findings are congruent with previously published studies, suggesting that RT-QuIC positivity can precede clinical onset, especially in E200K carriers.^7,8

There are several limitations of the Vallabh et al.⁴ study. The small sample size precludes any conclusive results from the study. However, the study findings largely demonstrate convergent validity with what is known about prion diseases and with results of similar studies. Additional limitations include the exclusion of neuroimaging biomarkers, which can also be present before symptom onset,⁹ and other tissue sample types, such as olfactory epithelium, tears, and skin biopsies.

One of the most important findings of the Vallabh et al.⁴ study is that it suggests that total PrP can be accurately measured throughout progressive disease stages. Given the challenges of using conventional approaches to preventive clinical trials in prion disease, demonstrating a reduction in a surrogate biomarker may be the most efficient and reliable initial approach to such a study. Underneath the Food and Drug Administration's Accelerated Approval Program, demonstration of total PrP reduction may serve as a surrogate end point for provisional approval that would have to eventually demonstrate clinical benefits. This study provides more data toward such an approach.

Study Funding

The author reports no targeted funding.

Disclosure

B.S. Appleby has received research funding from CDC, NIH, Ionis, Alector, and the CJD Foundation, and has served as a consultant to Ionis, Sangamo, and Gates Biosciences. Go to Neurology.org/N for full disclosures.