hsg logo white

HD InsightsHD Research Profiles (Fall 2023)

Lowering Mutant Huntingtin to Treat Huntington’s Disease

Optimizing an assay to track levels of mutant huntingtin in clinical trials

Huntington’s disease (HD) is caused by a mutation, in this case a polyglutamine expansion, in a single copy of the huntingtin (htt) gene. Stopping the expression of the htt gene (htt lowering), and therefore ameliorating the mutant htt (mhtt) protein, is currently the main therapeutic strategy being explored in clinical trials.

One issue with clinical trials is the lack of reliable biomarkers. Biomarkers are biological signals that can be measured accurately and reproducibly outside of the body, for instance in a blood sample, and correlate to a medical state. In the case of HD, a biomarker that can indicate that htt lowering is working after therapeutic delivery of htt-lowering drugs would greatly enhance clinical trial efforts.

There are current research grade assays that have been shown to be able to detect levels of htt in cerebrospinal fluid (CSF) and it is known that levels of htt in CSF correlate with disease stage and symptom severity. The levels of htt in CSF are relatively low, and therefore ultra-sensitive techniques are required to make this assay sensitive enough for use in the clinic. A recent study took on the task of improving this assay to validate it for clinical use1.

The group optimized the bead-based sandwich ligand assay and were able to validate the assay to a point where is fulfills regulatory requirements and can detect relative quantities of htt in CSF of patients. The assay was validated in two independent laboratories, meaning this assay can be transferred to different laboratories. This facilitates clinical development on therapies focused on lowering of htt.

There are a few limitations to the study. The assay requires reference material. Reference material for this assay is a kind of synthetic htt protein that is not identical to htt protein found in the person, and therefore the assay cannot accurately quantitate the actual concentration of htt in CSF. Ideally, future assays that do not require reference material can be designed. Additionally, sampling CSF is a very invasive procedure, so the authors aim to get the assay working in blood samples.

Improving delivery of huntingtin-lowering therapeutics to the brain

Although htt lowering is a promising approach, there has been limited success to date in clinical trials. The main therapeutic used in htt lowering is called an antisense oligonuclueotide (ASO). ASOs can be chemically modified to allow them to be stable, soluble and have low toxicity, however they are still limited in their ability to cross the blood brain barrier (BBB). Because of this, ASOs have to be delivered by injection into the CSF via the spine to get the ASOs to the brain. This method of delivery has so far proven to be ineffective in HD patients, causing worsening of symptoms, and it is likely that the ASOs are not reaching the parts of the brain it needs to.

New research aims to define new delivery routes and improve delivery to the brain2. One method for getting drugs to the brain is intranasal, which is a non-invasive direct route to the brain through the nose. However this method of delivery is very inefficient. To improve efficiency, molecules called nanoparticle vehicles can be used. Apolipoprotein A-I nanodisks (ApoA-I) are biological molecules that have been shown to act as carriers for therapies, enhancing their delivery. The study shows that ApoA-I can bind and transport the htt-lowering ASOs. They tested intranasal delivery in the BACHD mouse model, finding that ApoA-I would rapidly enter the brain after intranasal administration in the mice, and was found minimally in other tissues, which means there will be less chance of toxicity, and the treatment led to htt lowering in both the cortex and striatum in the brains of the treated mice. No immune response was noted, which is also important for therapeutics. The group also noted a modified version of ApoA-I; ApoA-I K133C entered the brain more effectively than unmodified ApoA-I, indicating that additional modifications may be able to further enhance this delivery method. Of note, the ASO was distributed equally across almost all brain regions, which is something that is not noted with other delivery methods.

Overall, being able to deliver ASO to HD patients in a non-invasive, safer and more effective way will greatly enhance the clinical effectiveness of these therapies.  Further work in larger animal models needs to be done
as next steps to validate this delivery modality.

Vauleon, S. et al (2023) Quantifying mutant huntingtin protein in human cerebrospinal fluid to support the development of huntingtin-lowering therapies. Scientific Reports: 13:5332

Aly, A.E et. al (2023) Delivery of mutant huntingtin-lowering antisense oligonucleotides to the brain by intranasally administered apolipoprotein A-I nanodisk. Journal of Controlled Release

Advanced Imaging Techniques to Support Huntington’s Disease Clinical Trials

Hippocampal atrophy in MRI imaging of HD patients

Image techniques and technologies and ways to capitalize on them to assist in HD management and research continue to become more advanced.

A pathophysiological hallmark of Huntington’s disease (HD) is cell loss in different parts of the brain. This degeneration happens over the course of disease and becomes more severe and widespread as disease progresses. Because of this, tracking degeneration or other disease-related physiology in the brain is an important tool for scientists and clinicians so that intervention can happen at the right time. It is also a way to track how potential therapies in clinical trials may be effective or not effective in slowing disease progression.

There are different parts of the brain that are exquisitely sensitive to HD, the striatum being most affected. However other parts of the brain are also affected, which may be a direct or indirect consequence of having HD. Either way, monitoring this degeneration may be beneficial to clinical efforts.

The hippocampus has shown to degenerate or change during HD, and using new imaging analysis techniques1, one group set out to further identify which parts of the hippocampus degenerate in HD.

Utilizing existing MRI data from the longitudinal IMAGE-HD study, which has imaging from 36 symptomatic, 40 pre-symptomatic, and 36 control patients, the group compared the volumes of different subregions of the hippocampus.
In this comparison, the group found accelerated atrophy in symptomatic patients within certain subregions of the hippocampus. Specifically, a region called the perforant pathway showed degeneration associated with repeat length or disease severity within the symptomatic patient population. This pathway is associated with memory, and the degeneration in this part of the brain may contribute to memory deficit symptoms in HD.

This automated MRI segmentation of the hippocampus technique can now be used to translate structural findings from animal models compared with humans, as well as be used to monitor the impact of treatment on brain atrophy during clinical trials.

Non-invasive method for tracking mHtt levels in the brain through advanced imaging

To enhance the usefulness of imaging, compounds and dyes can be used to accentuate image-based techniques. Radioligands are radioactive biochemical substances that are used for diagnosis or for research-oriented study of the receptor systems of the body and are generally delivered intravenously prior to imaging.

A recent study used a radioligand called 11 C-CHDI-180R, which is a ligand that is specific for mutant huntingtin (mHTT) and, through PET imaging, was hypothesized to pick up areas of the brain where mHTT aggregates. Therefore, it represents a non-invasive way to track mHTT in the brain2.

To investigate this in a large animal model, researchers used the radioligand in a macaque monkey model of HD. Control monkeys were injected with viral Htt constructs harbouring 10 CAG (10Q) repeats, and to mimic HD, study group monkeys were injected with Htt constructs harbouring 85 (85Q) repeats. They found that the injection of the virus is in the brain of live monkeys leads to expression of mHtt throughout the brain, mimicking the disease in humans.

After delivering the radioligand to the 85Q animals, PET scans showed a clear binding and signal in the expected areas of the brain, but not in the 10Q animals. Post-mortem high-resolution brain imaging studies confirmed mHTT aggregates in areas of the brain where signal was noted in the PET scan. There was additionally an increase in noted signal in 85Q animals that had progressed further in motor and cognitive disease symptoms, indicating this imaging assay may be able to delineate disease severity.

This work supports the use of the non-invasive 11 C-CHDI-180R radioligand for tracking and quantifying mHTT aggregates in the brain of HD patients and for use in future clinical trials.

Wibawa, P. et al. (2023) Selective perforant-pathway atrophy in Huntington disease: MRI analysis of hippocampal subfields.
Eur J Neurol. 30:2650–2660.

Bertoglio, D et al. (2023) In Vivo Cerebral Imaging of Mutant Huntingtin Aggregates Using 11C-CHDI-180R. J Nuclear Med.
00:1-7

Computational-Assisted Research in Huntington’s Disease

Machine-based learning can be used to predict patient outcomes.

Most biomedical research is done at the bench, however advances in databases, computer programs, and machine learning are assisting research efforts in new and exciting ways.

Although machine learning has already proven to be a very powerful tool in medicine, it requires large patient datasets and therefore, in more rare diseases like Huntington’s disease (HD), machine learning has not been implemented in the recent past. The Enroll-HD study has been collecting longitudinal data on patients for the past 20 years. With over 20,000 patients, there is now a dataset large enough to change this scenario.

In HD, age of onset (AOO) describes when a carrier of the HD mutation starts exhibiting symptoms, and thus when many interventions and changes in lifestyle need to be considered. Understanding AOO is additionally very important for timing of clinical trials. Predicting AOO is currently done based on the length of the CAG repeat, however this does not give an accurate AOO, and a lot of variability exists.

A recent study1 shows how the Enroll-HD database can be exploited using machine learning to make projections about patient outcomes. This study in particular aimed to predict the AOO of patients, as well as testing if machine learning could predict change in a temporal dynamic behavior — in this case,
a patient’s ability to safely operate a motor vehicle as HD symptoms onset.

The machine learning model for predicting AOO outperformed the current best-in-class model. The model was additionally able to assist in predicting a patient’s future driving capacity. This kind of data will have critical importance for timing of clinical trials as well as assist clinicians and patients in making important decisions in terms of patient functional capabilities, which impact quality of life. The machine learning outlined in this study can be implemented to assess many additional outcomes and be used for personalized care in HD.

Datasets, coupled with new computational tools, can identify novel pathways for HD drug discovery.

Another recent study utilized a systems bioinformatics approach and different computational tools, coupled with pre-existing publicly-available genetic and miRNA data, to understand molecular mechanisms of HD2. The information that can be compiled and distilled out of these pre-existing datasets can be used to further experimental studies, infer novel treatments, and give clues to drug and biomarker discovery.

The study aimed to identify common pathways that are affected by HD and, specifically, genes and pathways that differ between different stages of HD. The study identified common pathways in the three different datasets analyzed as well as genes and miRNA that can be targets in therapeutic development work or in biomarker studies.

The pathways the group identified included apoptosis (cell death) pathways, synaptic transmission (brain-related) pathways, and immune system pathways. The group was able to link the genes and pathways they discovered with other studies and HD literature, which helps validate this approach for novel discovery. Novel genes and pathways found using this methodology will need to be validated using cell and animal models in future work.

Ouwerkerk, J et.al. (2023) Machine learning in Huntington’s disease: exploring the Enroll-HD dataset for prognosis and driving capability prediction. Orphanet Journal of Rare Diseases: 18:218

Papanicolaou, Z et al (2023) Integrated Bioinformatics Analysis of Shared Genes, miRNA, Biological Pathways and Their Potential Role
as Therapeutic Targets in Huntington’s Disease Stages. Int. J. Mol. Sci., 24:4873.

About HD Insights

Our mission is to promote, disseminate, and facilitate research on Huntington’s disease. To fulfill this mission, we are guided by an outstanding editorial board that includes representatives from three continents, academia, industry, and the HD community.