CE17 - Recherche translationnelle en santé

Innovative Imaging and cognitive BIOmarkers to predict Huntington’s Disease progression – I2BIO-HD

Innovative Imaging and cognitive biomarkers to predict Huntington’s Disease progression

To demonstrate that combining high-level multimodal imaging techniques, newly validated cognitive assessments and an innovative statistical approach will model and identify distinct subtypes of patients’ profiles related to disease progression. This combination could be useful to track pathological trajectories at different disease stages from very early prodromal stage.

To model disease progression and define an enrichment strategy to improve patient selection for future therapeutic trials

The disease modifying treatments that are currently under assessment in manifest patients (HD patients) defined by a score equal to or greater than 5 at the motor evaluation of the Unified Huntington’s Rating Scale (UHDRS). Most trials use as an endpoint this universally accepted scale (which quantifies motor, cognitive, functional and psychiatric symptoms) or the cUHDRS, which combines a subset of UHDRS items. However, demonstrating an improvement in disease progression relative to placebo requires hundreds of individuals with the best current clinical measures and enrichment strategies. For instance, the use of cUHDRS or caudate atrophy, the best biomarker of disease progression to date, would respectively necessitate about 1000 and 400 patients, for a 2-years two-groups trial. Moreover, even if quantifiable clinical and biological alterations were identified at least 10 to 15 years before the formal diagnosis in large observational studies, few show significant change in preHD compared with controls over reasonable time windows (e.g. 24 months), thus excluding therapeutic trials in this population. However, specific peripheral blood markers of the neurodegenerative process, TEP and MRI imaging and novel tests including calculation and language, cross-sectionally validated in small preHD cohorts, hold great promise for preHD longitudinal follow-up. Finally, criteria for patient inclusion in HD clinical trials are so far sparse and mostly based on the number of CAG repeats or functional capacity, not allowing stratification nor precise prediction of their progression rate. Altogether, novel assessments and novel methods are mandatory for future intervention in preHD.<br />Our main objective is thus to demonstrate that a multimodal approach combining novel imaging markers, biological data and innovative cognitive measures including language developed by us are sensitive enough to detect the progression of the disease in small cohorts of not only manifest HD but also preHD participants. We also aim at modelling disease progression and define an enrichment strategy to improve patient selection for future therapeutic trials

We will conduct a prospective observational cohort in 20 HD patients, 40 preHD, and 20 healthy volunteers. Each participant will be followed for 24 months repeatedly with the UHDRS measurements and new designed tasks including clinical and cognitive evaluations at VM0, V+ 1 month (VM1), VM12, and VM24. PET and MRI imaging, and peripheral blood markers will be carried out at M1 and M24. Duration of research will be 4 years.
The I2BIO-HD project is organized in three work-packages including new strategies in each. WP1 will optimize tasks that have been already designed for manifest patients to the requirement of this new cohort and promote language analyses. WP2 will evaluate the sensitivity of new imaging markers to follow gene carriers acknowledging that all previous HD studies did not include imaging approaches with the most recent PET and MRI techniques. WP3 will exploit multivariate statistical analyses and machine learning approaches to identify distinct subtypes of patients’ profiles related to disease progression, to determine predictors of trajectories and establish models of stratification, and ultimately help to improve results’ sensitivity in small cohorts despite the variability inherent to HD.

No results are yet available. First results will provide from the transversal study, which is not completed.

Our project will be an important step forward in the field by providing powerful tools to follow, and also predict the clinical evolution of patients from the very early phase of the disease (pre-HD). Our ultimate goal is to extract from these data specific combined scores that will allow identifying the best population to evaluate new therapeutic strategies in small cohorts of HD but also in pre-HDs in future clinical studies.

Our validated early and sensitive (isolated or combined) biomarkers will also reduce the number of patients required in a clinical trial and the study duration (XX phase2), yielding to a reduction of clinical costs, which is a major hindrance in the biotech and pharmaceutical industry, especially those developping gene therapies.
Furthermore, our protocol will give the unique opportunity to conduct a study in large HD and pre-HD patient’ cohorts to improve disease progression models and will provide diagnostic and prognostic indicators at an individual level. These tools will guide patient’s medical care and will establish criteria for inclusion stratification and identification of the best population to test a therapy.
Finally, the validation of our innovative and sensitive markers of cognitive disorders might be generalized to other neurodegenerative diseases.

NA

Huntington’s disease (HD) patients suffer from motor, cognitive and behavioral troubles, with heterogeneous phenotypes and variable time course both at individual and group levels. This heterogeneity induces high variance in disease markers and currently none of those is sensitive enough to measure disease progression in small cohorts, predict onset in premanifest patients (preHD) or show significant change in preHD compared to controls over a reasonable time window. The biomarkers used to measure HD progression have been defined years ago and do not allow patient stratification nor precise prediction of their progression rate. Therefore, there is a need for new more specific tools which must be thoroughly tested.
Our aim is to define a model of disease progression and establish an enrichment strategy of pre-HD to improve patient selection for future therapeutic trials. For that purpose, we will combine newly validated cognitive assessments, high-level multimodal imaging techniques, biological parameters and an innovative statistical approach. New cognitive tools rely on: i) eCOG a digitized rapid battery (15 min) that has been shown by Partner2 to be sensitive enough to reduce by a factor of 5 the number of HD patients required to demonstrate a clinical effect of a treatment when compared to the cUHDRS; and ii) spontaneous speech recordings which, in our hands, accurately classify subjects into preHD, HD patients and healthy controls. High-level imaging techniques on a MRI-PET camera: a new PET tracer of phosphodiesterase 10A ([18F]-MNI659) -which is among the earliest biomarkers of the disease- and multimodal MRI techniques, such as morphometry, cortical depth and sulcal anatomy, classic and new DTI sequences to assess white matter microstructure. Biological parameters will focus on Neurofilament light chains, recently found to be the earliest peripheral marker of HD. Innovative statistics will follow several steps: i) a composite score combining multidomain measures will be defined as it may better fulfill sensitivity criteria to detect clinical changes than individual outcomes; ii) we will then use unsupervised clustering algorithms to identify specific patients’ profiles at baseline and typical trajectories over time, applying a two-part methodology depending on the nature of data analysed (both cross-sectional clustering using conventional approaches and novel techniques such as the SUSTAIN methodology and longitudinal clustering); iii) based on the clusters and typical trajectories identified during the previous step, machine learning approaches will identify the best predictors in isolation and combination and establishing their prognostic value on the disease evolution.
These biomarkers will be analyzed both cross-sectionally and longitudinally in a cohort of 40 preHD, 20 HD and 20 healthy controls. They will be examined using imaging, cognitive and biological tools at baseline and 2 years later and will have a complete clinical follow-up in addition to the biomarkers described above.
The chances of success of our project are high since partners have fruitfully collaborated over the past 20 years and have reached excellence in their respective domains: Partner 1 has a long experience in functional imaging in HD and has recently hired an expert in PET quantification of the phosphodiesterase 10A; Partner 2 (Henri Mondor) is the French national center for HD and has developed an international expertise in cognition of HD.
If our project succeeds, it will be an important step forward in the field by providing powerful tools to follow, and also predict, the clinical evolution of patients from the prodomal phase. Our ultimate goal is to extract from these data specific combined scores that will allow identifying the best population to evaluate new therapeutic strategies in small cohorts of HD and preHD in a cost-effective way.

Project coordination

SONIA LAVISSE (DRF/IBFJ/MIRCen/Laboratoire de Maladies Neurodégénératives: mécanismes, thérapies)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partner

MIRCen DRF/IBFJ/MIRCen/Laboratoire de Maladies Neurodégénératives: mécanismes, thérapies
BIO DMU Médecine

Help of the ANR 591,015 euros
Beginning and duration of the scientific project: March 2021 - 48 Months

Useful links

Explorez notre base de projets financés

 

 

ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter