These are notes from the CJD2014 conference hosted by CJD Foundation at the Washington Court Hotel in Washington, D.C. on July 11-14, 2014.
As of August 7, 2014, CJD Foundation has posted videos of most of the presentations. As of August 19, 2014, I have embedded these videos below and updated the text to include some details that are mentioned in the videos. If you have any concerns about content of this blog post, please contact me
Dr. Collinge opened with the history of the MRC Prion Unit: it was founded in 1998 by the British government as a response to BSE / vCJD with the mission of developing therapeutics. The Unit has now spun off a company called D-Gen Therapeutics which is responsible for commercializing therapeutics developed by the unit.
He then gave a brief overview of some relevant biological principles including the central dogma, and noted a few reasons why prions are unique, including: because PrP is one of our own proteins, we don’t have a natural immune response to prion disease. He described PrP as being “very highly conserved in evolution”, a statement with which I don’t particularly agree. He reviewed the basics of prion disease: the structure of PrPC, the idea of conformational change, the differences between strains, the effects of codon 129 on disease risk, and so on.
He gave an update on the MRC Prion Unit’s small molecule discovery program in collaboration with GlaxoSmithKline. They screened 2 million compounds and have found a few different leads. They are currently pursuing funding for medicinal chemistry optimization on two compounds called GSK01 and GSK02 which have a unique mechanism of action and do not bind PrP. He presented preliminary in vivo efficacy results for GSK02.
He then discussed the MRC Prion Unit’s PRN100 therapeutic antibody, which Simon Mead introduced at last year’s conference. As with last year, the plan is to first pursue intravenous infusion of the antibody, and if they discover that not enough of the antibody is reaching the brain, they will proceed to direct infusion into the cerebrospinal fluid. ICSM18 was humanized as a G4 antibody (as oppposed to IgG1) because G4 heavy chains are not very active in turning on the human immune response, so this antibody will be less likely to cause an autoimmune response. Dr. Collinge provided several unpublished updates on safety studies for the antibody. These are apparently remaining confidential, as a video of Dr. Collinge’s talk has not been posted.
He proceeded to discuss the question of how clinical trials will measure efficacy in patients. He introduced the MRC Prion Unit Rating Scale [Thompson 2013] and showed that stratifying patients by PRNP codon 129 genotype will increase statistical power for trials. This part of the talk was almost the same as Simon Mead’s talk at Prion2014. Dr. Collinge then went on to discuss the need for a biomarker for therapeutic efficacy in addition to a clinical rating scale. He argued that the ability to measure viral load as a quick biomarker was a crucial factor in the rapid development of HIV drugs in the 1990s. By analogy, MRC Prion Unit has developed a test for infectivity that can detect prions in human tissue at dilutions out to 10-10 and has developed robotics for automating this assay [Edgeworth 2011]. They plan to use this system to monitor prion load in patient tissues such as CSF over the course of a trial to see if infectious titer is reduced. He also mentioned the Scrapie Cell Panel Assay but it was not clear, to me at least, how this related to the methods for measuring titer in human tissues.
Dr. Collinge hopes to learn whether the Phase I trial is funded within the next two months. If it is funded, then the Unit will seek regulatory approval for the trial. He anticipates that funding will be the larger of the two obstacles, and that if funding can be secured, the trial could begin next year.
In the Q&A, Dr. Collinge clarified that the Phase I study will last ~18 months and will only recruit U.K. patients. If results are promising, they will consider international recruitment for a Phase II.
One audience member asked whether the method to measure prion titer could also be used for diagnosing the disease. Dr. Collinge said that yes, it can be used to diagnose vCJD, but it does not work well on sporadic CJD. Titer in the periphery is much lower in sCJD and so is more difficult to detect, but they are beginning to have some ability to detect sCJD in some samples.
In an introduction. Pierluigi Gambetti noted that Dr. Moreno has left Giovanna Mallucci’s lab and is now working for Glenn Telling at Colorado State University. Dr. Moreno was here today to introduce her work from the Mallucci lab on the unfolded protein response (UPR) as a drug target for treating neurodegenerative disease. This work is published in [Moreno 2012, Moreno 2013]. There is a summary of Dr. Mallucci’s presentation on the same subject at Prion2014 Day 2, and I have previously reviewed this work in this blog post.
In the Q&A I suggested that Dr. Moreno might consider using the term “delayed” rather than “prevented” with regards to the effect of PERK inhibitors on prion neuropathology. She agreed this might be appropriate.
Dr. Zanusso reviewed his work, going back more than a decade [Zanusso 2003], showing that PrPSc can be detected in the olfactory epithelium and olfactory mucosa. Zanusso therefore partnered with Byron Caughey to create a diagnostic test for prion disease based on olfactory mucosa brushings plus RT-QuIC. This work has since been published in New England Journal of Medicine [Orru 2014].
In the Q&A, Ilia Baskakov asked how early in the disease the test would work. Dr. Zanusso replied that most of the patients in the study had the nasal brushings taken 1-2 months after the CSF draw. He is currently working on validating the test in patients with nasal brushings and CSF taken on the same day. This is a work in progress.
Dr. Orru introduced the RT-QuIC technology and its usefulness in diagnosing prion disease. Following on the collaboration with Dr. Zanusso [Orru 2014], she is working on further improving the technology. A few points she touched on were:
- Trying to substitute a gentler nasal swabbing procedure in lieu of nasal brushings to make for an even less invasive test
- Determining whether CJD patient nasal brushings and/or CSF used for RT-QuIC are infectious. The experiment is ongoing but so far they see no infectivity in CSF and very little in nasal brushings.
- Developing a new protocol to reduce RT-QuIC reaction times to as little as 5 hours.
- Developing an enhanced QuIC assay or “eQuIC” to detect prions in human blood plasma.
For full details see the above video.
In the Q&A, Ilia Baskakov asked how reproducible the results are when using recombinant PrP from different labs. Dr. Orru replied that the recombinant PrP preps from other labs have not been as reliable in the past, and that the assay is becoming more robust over time but you are still best off following the exact protocol used by the Caughey lab.
Dr. Zou gave an update on his study of induced pluripotent stem cells (iPSC) from asymptomatic patients carrying genetic prion disease mutations. Sonia donated cells for this study, as documented here. There are 23 patients in the study, including genetic mutation carriers, sCJD patients, and negative controls such as Alzheimer’s patients.
Dr. Zou’s results are evidently remaining confidential, as no video of his talk was posted.
In the Q&A, I asked Dr. Zou to make sure that when he writes this up for publication it will be clear to physicians who are not experts on prions that the results do not imply that skin cells from mutation carriers are infectious. At the family session yesterday, we heard from several families who had experienced difficulties in finding a funeral home or even getting a blood draw for genetic testing, due to funeral home directors’ and physicians’ unjustified paranoia about prion infectivity. (CJD Foundation has worked hard to fight these misperceptions – for instance it distributes a pamphlet for funeral directors and Michael Geschwind and Brian Appleby continue to work to educate other physicians.) Zou’s results show biochemical signatures of misfolded PrP in cells from mutation carriers, but do not imply that mutation carriers pose any biosafety risk. Dr. Zou agreed and said that he will be sure to make this clear in the publication.
Dr. Watts began with an introduction to modeling prions in cell culture. He explained the importance of having a faster, cheaper model for studying prions rather than doing everything in mice. He reviewed the Prusiner lab’s results showing that IND24 can double the lifespan of mice infected with RML prions but that this compound does not work on human prions [Berry 2013]. This motivates a need to be able to study human prions in cells and screen drugs against them. He reviewed several attempts to propagate prions in human cells – expressing HuPrP in N2a cells or RK13 cells for instance – all of which have been unsuccessful.
It had been shown that bank voles are uniquely susceptible to prions from a wide range of different species [Nonno 2006], so Watts created mice expressing bank vole PrP. He has now shown that these mice can be infected with a wide range of different prion strains/isolates and that most of the strain properties of prions appear to be maintained in the BvPrP mice [Watts 2014a].
Based on these results, Watts expressed BvPrP in RK13 cells and tried to infect them with human prions passaged in BvPrP mice. Unfortunately, the cells could not reliably be infected.
Watts then reviewed the history of attempts to model genetic and sporadic prion disease mutations in mice. Genetic mutations in mice do not always cause spontaneous disease, or if they do, the disease proves not to be transmissible, or not to have classic biochemical signatures of prion disease. Watts created mice expressing BvPrP with the 109I polymorphism and found that they developed a rapid onset, transmissible prion disease [Watts 2012]. He believes this may be considered a model of sporadic CJD. The mice have a faint but reproducible ~9 kDa band of PK-resistant PrP.
In his most recent (unpublished) work, Watts has created mouse models of three genetic prion diseases on a BvPrP transgene: E200K, D178N, and ΔGPI, a model of the truncating mutations in PRNP in humans. Mice expressing these mutant transgenes all became spontaneously sick. Moreover, most of them only had expression levels of PrP of about the same level as wild-type mice, or even less (the range was 0.3 - 1.3x wild-type levels). These mice had neuropathology that closely resembles the corresponding human prion diseases, suggesting these are true models of the disease. The diseases are transmissible, and mice inoculated with prions from spontaneously sick mice of the same genotype will have a disease course of only about 40 days.
In the Q&A, someone asked why bank voles don’t get spontaneous prion disease if their PrP is so unstable. Watts speculated that the mice only get spontaneous disease due to overexpression, though he said there aren’t yet any BvPrP knock-in mice so we don’t know for sure. Watts speculated that bank voles would indeed get spontaneous prion disease if they lived long enough, but that their lifespans in the wild are too short. He said they had obtained a few brains from very old bank voles from Romolo Nonno, and that some of the brains appeared to show traces of infectivity, suggesting that the bank voles did have the beginnings of subclinical prion disease.
Castilla presented the same talk he gave at Prion2014 Day 2. Castilla was interested in the fact that bank vole susceptibility to prions depends in part on the 109M/I polymorphism. He noted that mice and many other animals have L at this codon, humans have M, and a handful of animals such as Mongolian grazing horses have I. No other amino acids are seen naturally at this codon in mammals. Castilla wondered whether other amino acids at codon 109 would support prion formation. He therefore created recombinant BvPrP with every possible amino acid at codon 109. He performed PMCA on each substrate, and found that I, L and V formed spontaneous PK-resistant PrP. M did not do so, and neither did any of the other amino acids. He next introduced a 109I substitution into PrPs from a wide range of different mammal species, and found that some but not others were able to form spontaneous PK-resistant PrP aggregates in PMCA when they carried the 109I substitution. MoPrP 108I was among those that did form aggregates, so Castilla then created mice expressing MoPrP 108I. Depending on expression level, the mice became spontaneously sick by 140 days (5-6x expression), 260 days (4x expression) or 300 days (3x expression). His current project is to create mice expression HuPrP with this substitution as well.
Pierluigi Gambetti introduced Dr. Notari, who formerly worked for Piero Parchi and has now joined Dr. Gambetti’s group. He is working on characterizing the role of Tau protein in the atypical prion disease known as variably protease-sensitive prionopathy (VPSPr).
VPSPr has been observed in 55 cases worldwide since its discovery in 2008 and is thought to account for ~3% of sporadic prion disease cases, but it may actually be more common, as it is easy to misdiagnose as a different from of dementia. There is hyperphosphorylated Tau in VPSPr brains. He is characterizing the Tau aggregates by their immunoreactivity and electrophoretic mobility and comparing them to Tau in Alzheimer’s, progressive supranuclear palsy, Pick’s disease and other tauopathies.
Robert G. Will
Dr. Will founded and leads the UK CJD Surveillance program and offered an overview of CJD surveillance efforts in Europe. He showed a pie chart of types of prion disease in the UK – 10% are genetic, and the vast majority are sporadic. He showed that more CJD cases are being reported in the U.K. each year, with the rate of increase far exceeding the increase expected based on population growth and population aging. He believes this reflects more complete diagnosis and ascertainment rather than an increase in the actual number of cases. In support of this interpretation, the curve looks the same in Australian CJD surveillance even though Australia does not have BSE or scrapie. The distribution of age of onset is starting to shift more to the right, i.e. more older cases of CJD are being reported, reflecting improved diagnosis in older people. Across Europe, most countries have a CJD incidence of 1.0 to 1.5 cases per million population per year. A few countries have rates far lower, such as 0.3 – these are countries where surveillance has only recently begun, and ascertainment is not very complete yet. Will noted that CJD is not contagious in the classical sense even though it can be transmitted under very specific circumstances. Epidemiological evidence does not suggest that health care workers are at any increased risk for CJD [Alcaldo-Cabero 2012].
He reviewed the recently published negative results of the Italian-French clinical trial of doxycycline in CJD [Haik & Marcon 2014].
He then showed a plot of vCJD deaths per year, and noted that vCJD cases are very rare now. As of 2013, BSE has almost disappeared as well. However, pointing the recent possibility of iatrogenic exposure to CJD in New Hampshire, he emphasized the need for continued surveillance and public health efforts.
In the Q&A, I asked whether the sporadic CJD cases in the UK dataset had all tested negative for PRNP mutations. He said that about 60-70% had undergone testing, and that for the other 30-40% the classification as “sporadic” is a presumption. He argued, however, that it is unlikely very many of these cases are genetic. First, of the cases that do undergo genetic testing, only ~10% are genetic. Second, the cases classified as sporadic are those in which the clinical presentation matches sporadic CJD rather than, say, FFI or GSS. Third, of the cases classified as genetic, about 50% have a family history, whereas the cases classified as sporadic do not. Fourth, in countries with higher rates of genetic testing than those in the UK, it is still found that about 10% of cases are genetic.
Gambetti provided an overview of U.S. CJD surveillance. Despite declining funding, the NPDPSC continues to test a constant or even slightly increasing number of cases each year. They perform tissue tests on >400 samples per year, of which 250-275 are positive each year. This suggests that if the true incidence is 1 case per million population per year, then the NPDPSC is ascertaining about 88% of them. The NPDPSC tests ~3000 CSF samples for 14-3-3 protein and Tau each year, and here the vast majority are negative, with only about 20% (600-700) positive. This is because CSF testing is an early diagnostic step for physicians making differential diagnoses, and so is ordered in many cases where CJD is unlikely. In January this year, the NPDPSC began RT-QuIC testing on CSF for all cases positive for 14-3-3 and/or Tau, and in January through June it has tested 618 samples, 234 of which tested positive.
Gambetti reviewed recent advances in diagnostics using RT-QuIC and PMCA, and then discussed the recently reported vCJD case in Texas. This individual had onset at age 41 and an 18-month disease course. NPDPSC had correctly diagnosed this individual with vCJD based on performing PMCA on a urine sample during the individual’s life (Gambetti cited an abstract by Fabio Moda and Claudio Soto, apparently from a past Prion conference), and autopsy later confirmed that it was indeed vCJD. This individual had grown up in Lebanon and traveled extensively in Europe; there was no travel to UK or Saudi Arabia (the two other countries where Americans had become infected with vCJD) but he may have traveled to France, which has also had BSE.