Defining and predicting variability in early Parkinson's disease using quantitative MRI

One of the biggest challenges facing developed countries in the 21st centaury is the increasingly aged population and rising life expectancy. In the UK nearly a quarter of the population will be aged 65 or over within 20 years. As a consequence, more people will be living with chronic, long-standing health problems. This poses significant challenges, particularly in healthcare, and will require a shift towards pre-emptive treatments designed to prevent or slow the progression of chronic diseases in order to cope with this changing demographic.

Parkinson's disease is a common degenerative disorder of the brain that becomes more likely with age. A combination of muscle stiffness, tremor and slow movements alert the clinician to its presence. These symptoms begin to be detectable once half of the nerve cells within a brain region called the substantia nigra have already died. Once diagnosed, disease progression is highly variable. About half the people diagnosed will develop significant problems within four years.

There is evidence that Parkinson's disease actually begins up to 20 years before it is diagnosed; it starts in a different part of the brain and progresses slowly causing more subtle problems. This intervening period is called "pre-clinical Parkinson's". During this period there are certain problems, such as loss of smell or certain sleep disturbances, which are more likely to be experienced by individuals with the condition. However, it is not currently possible to identify who amongst these individuals have pre-clinical Parkinson's or how quickly the disease will progress once diagnosed.

This work seeks to use non-invasive brain-scanning techniques to understand why Parkinson's disease is so variable, develop ways to predict how quickly the disease will progress based on an individual's brain structure and diagnose the condition during the pre-clinical phase. By achieving these aims, strategies aimed at slowing the condition can be researched more accurately, and the disease can be better characterized within an individual, allowing current treatments to be tailored to their present and future needs.

Aims:
This work will recruit individuals with established PD, possible prodromal PD and healthy controls into a longitudinal MRI study to answer these principal questions:

1. Can in vivo quantitative disease staging be achieved in PD using quantitative MRI (qMRI)?
2. What neuroanatomical features explain the clinical variability observed in PD?
3. Can future outcomes be predicted from a single baseline time-point?
4. Can qMRI detect PD during the pre-clinical phase of the illness?

Through these, a quick, non-invasive way to quantify patterns damage in PD will be developed.

Methodology:
This work will recruit three cohorts:
- Healthy controls
- Idiopathic PD
- Rapid Eye Movement Behavioral Disturbance
Both clinical groups will undergo assessment and MRI scanning at baseline and two further annual assessments with repeat scanning at the final time-point. At baseline, the controls will be matched to the PD cohort, and a proportion of this group will undergo the same follow-up as the clinical cohorts. Detailed clinical assessment will be performed in all groups at each time-point.

Scientific opportunities:
This work is specifically powered to detect changes within brainstem structures involved at the earliest stages of PD, using imaging methods developed to allow subcortical nuclei affected by PD to be mapped in vivo. This allows a priori hypotheses based on post-mortem data, linking clinical variability to anatomical differences in PD, to be tested and developed further to predict future outcomes of disease. These approaches will also provide an accurate way to discriminate PD from normal ageing, that can then be applied to the RBD cohort to detect pre-clinical PD. Combined, this work has the potential to provide a quick, non-invasive means to diagnose pre-clinical PD, identify more homogenous PD subtypes, help tailor clinical decisions to individual patient trajectories and provide sensitive quantitative metrics for clinical trials.

Who will benefit from this research?
Groups who would directly benefit over the short-medium term include:

- Patients with established PD
- Patients with possible prodromal PD
- Academics engaged in PD research and clinical trial design
- Clinicians with an interest in movement disorders
- Neuroimaging researchers
- Academics and clinicians with an interest in neurodegenerative diseases.
- Academic and clinicians with an interest in accurately defining in vivo functional anatomical regions e.g. functional neurosurgeons for deep brain stimulation targeting

Over the longer term, this work fits into the published NHS and government strategies for managing an increasing aged population.

How will they benefit from this research?
Short Term (2018-2023)
The immediate benefit of this work will be achieved through teaching and training, both of local interested researchers and also the research assistant for this project. Additionally, the study motivation and rationale would be discussed with local patient groups to gain feedback into the experiment design and data-sharing proposals.

Over the grant, the results would be presented to patient groups, specialist movement disorder meetings and international conferences. The techniques would be more widely disseminated via teaching on the SPM course, and through the release of open-source tools and datasets. The Internet would be used to disseminate the clinical findings, methodological advances, and available open-source resources. In addition to the study website, these would also be promoted via related online forums (e.g. dataset releases via the NITRC or clinical findings via the PREDICT-PD blog).

Finally, over the first two years of the fellowship I will develop a public engagement event centered on the wider issues and practical techniques of preclinical diagnostics in Neurodegeneration. Specifically, this will be developed in conjunction with allied professionals (e.g. genetics and dementia), funding stakeholders, policy-makers and patient groups (e.g. Parkinson's UK Network, Fifth Sense). The aim would be to organize this event for the third year (2020), allowing it to provide a basis for follow-up grant applications and foster wider collaborate efforts.

Medium Term (2023-2028)
In the medium term, this proposal will form the foundation for future work targeted towards understanding the complex interactions contributing to clinical variability in vivo. This will eventually require close collaborations between allied research disciplines such as genetics, epidemiology, molecular and ex vivo imaging and those engaged in clinical trials, and I will seek to foster these over this period. Taking advantage of the availability of 3T MRI and improved inter-site reliability of the qMRI acquisitions, I will work to disseminate the methods used in this proposal and promote larger multi-center collaborations. Finally, I plan to continue to develop and contribute open-source resources through my work to accelerate development in this field.

Long term (>2028)
Developing strategies aimed at slowing disease progression is an essential step, identified by the "NHS - Five Year Forward View", to manage the increasing numbers of individuals surviving to old age. Aligning with this aim, my research seeks to identify individuals during the earliest stages of disease, better understand clinical variability to more accurately phenotype disease and predict its trajectory in individuals and provide quantitative biomarkers to monitor progression in the absence of motor signs. By translating these tools, together with collaborators, I aim to help develop a clinical service designed to assess, investigate, diagnose and stratify pre-clinical and early PD. This will allow appropriate, patient-specific disease modifying therapies to be developed and commenced early in the disease course, prior to significant loss of neural tissue.