The Generative Brain

Field guide for newcomers

Prion disease research spans human CJD, animal prion diseases and protein misfolding biology. Seven completed trials before 2015 were mostly negative, and current work centres on lowering prion protein, with the strongest animal data from gene editing in mice.

This guide gives a researcher from another field the facts that most change how prion literature reads. Deeper treatment is in epidemiology, genetics, mechanisms and the resources page. The entry point for the whole space is Research.

What is being studied?

Prion diseases are fatal neurodegenerative conditions caused by misfolded prion protein. In humans, CJD occurs at about 1 to 2 deaths per million per year. The CDC puts about 85 percent of cases as sporadic, 5 to 15 percent as genetic and a small remainder as acquired. Animal prion diseases (BSE, scrapie, chronic wasting disease) are studied in parallel, both for their own sake and as models of transmission between species.

What did earlier trials show?

A 2015 review by cureffi tabulated seven completed trials. Flupirtine (28 participants, randomised, double-blind) reported slower cognitive decline on ADAS-Cog but did not become standard care. Pentosan polysulfate was tested observationally in the UK (7) and Japan (11), and quinacrine in France (32), the UK PRION-1 trial (107) and a US randomised trial (54). Doxycycline was tested in 121 participants in France and Italy in a randomised, double-blind, placebo-controlled trial and was negative. The review’s design lessons are:

  • Some drugs lacked a strong preclinical basis, and some trials continued after discouraging animal data.
  • Survival is a workable endpoint because median duration in sporadic CJD is about 5 months, but genotype, subtype, age and sex make comparisons noisy.
  • Historical controls were biased by who chose treatment.
  • The MRC prion disease rating scale may give more power than survival.
  • Trials in people with PRNP mutations who have no symptoms would need a predictive biomarker and a registry.

What are the current approaches?

The strongest recent animal result is from the Broad Institute. In a Nature Medicine paper (2025), a dual-AAV system with the PHP.eB capsid delivered the base editor BE3.9max and a guide RNA that installs the PRNP R37X change. In transgenic mice carrying human PRNP and inoculated with human prion isolates, average editing was about 37 percent, brain PrP fell by about 50 percent and lifespan rose by about 52 percent. Improved editors lowered PrP by about 63 percent at a lower viral dose. This is a mouse result. See therapeutics for human trial status.

Who works where?

  • The Alliance BioSecure Foundation, recognised in 2006, funds biological safety research on food and health risks, with prions named in its statutes.
  • The Canadian Food Inspection Agency campus at Fallowfield in Ottawa has about 200 staff, level 2 and 3 containment, and is a World Organisation for Animal Health reference laboratory for chronic wasting disease and scrapie. Other labs are listed under research labs.

What should a newcomer read with care?

Prion terminology varies. Check whether a paper means human CJD, an animal strain or a mouse model, and whether results come from symptomatic or pre-symptomatic animals. The 2015 review also asks for stronger preclinical evidence, such as efficacy against human prions in humanized mice, before trials start.

Sources for this page