Phase 1

The Subjectivity Problem.

PD and PSP are disorders that are notoriously difficult to diagnose and track. Symptoms in the prodromal phase are slight in their magnitude and slow to progress, meaning initial stages of the conditions are not picked up or simply pushed aside as factors of ageing. Overlapping symptoms between PSP and PD (among other movement disorders) results in misdiagnoses. The effect of medication, while alleviating many symptoms, can affect the ability to accurately track the natural progression of PD.

The current gold standard in clinical practice for the assessment of PD and PSP are clinical rating scales. Rating scales, such as the Movement Disorder Society’s Unified Parkinson’s Disease Rating Scales (MDS-UPDRS) and Progressive Supranuclear Palsy Rating Scales (PSPRS), are used to assess the condition by asking the patient questions about their cognitive and emotional state and to perform a range of motor tasks. Based off the patient’s answers and relative performance in the associated motor task, the clinician provides a score based on specific clinical criteria that reflects the disease severity of the condition according to a discrete scale:

0 Normal

1 Slight symptoms evident

2 Mild symptoms evident

3 Moderate symptoms evident

4 Severe symptoms evident

…highlighting the key problem of subjectivity. Rating scales are subjective, based on the opinion of a select number of raters. Subjective opinion will be different between clinicians, can be affected by a range of factors including mood and alertness at the end of the working day, and can be biased by previous experience with patients. This subjectivity can impact understanding of the true disease progression, if and when medication is prescribed, and the success rate of clinical trials. A new approach is required to overcome the problem of subjectivity and advance our understanding of PSP and PD.

Rating scales, while being useful, introduces many flaws into the process of diagnosing and tracking parkinsonian disorders:

  • Most symptoms are scored by simple observation, many any symptoms that are extremely small in amplitude by the rater.

  • The use of a discreet scale limits statistical analyses required to appropriately interrogate the disease data.

  • A simple point scoring scale is prone to ceiling/floor effects when rating symptoms.

  • The opinion of any two raters may not be the same, introduces inter-rater reliability…

The Objective Solution.

Digital technology enables us to quantify symptoms associated with PSP and PD in an objective manner. A range of features can be measured on a granular level, allowing rich data to be obtained for robust analyses. Technology expands the clinical experience beyond the data obtained. By collecting data via specified, active tasks, clinical visits can be streamlined to both the benefit of researcher and patient. OxQUIP Phase 1 began in 2016 with the intention of uilising digital technology on a large cohort of PSP, PD and healthy control (HC) participants to extract reliable, objective biomarker for disease progression.

OxQUIP is comprised of six main groups of participants

Group 1 — de novo PD participants i.e. medication-naive Group 4 — Deep Brain Stimulation (DBS) PD participants

Group 2 — PD participants <8 years since diagnosis Group 5 — PSP participants

Group 3 — PD participants ≥8 years since diagnosis Group 6 — HC participants

The digital technology recorded measurements from three main movement categories:

White man with brown hair resting chin on headrest and staring at a red dot on a computer screen being recorded by a camera underneath the screen.
White hand extending out with the thumb at a right angle to the fingers. Black sensor on index finger.

Saccadic eye movements

Fine motor movements

Person standing with black sensors attached at the chest, waist, wrists, and feet.

Gait movements

Participants attended nine in-person visits to the John Radcliffe hospital at three month intervals, for a total of 24 months.

During these visits, participants performed a range of tasks measured by digital technology to extract features associated with their disease condition.

On their second visit, group 2 participants partook in a levodopa challenge in which they attended clinic off their medication before repeating the tasks on their medication.

V = Visit

  • Schematic of eye looking at central fixation point, represented as a blue star. Gaze indicated by orange dotted line.

    Saccades

    Saccades are rapid, ballistic eye movements from one fixation point to another. They can be measured precisely and non-invasively, meaning they are good markers for neurological function. Previous research has demonstrated saccades to be affected in PD and PSP.

    Saccades represent a promising biomarker as they are stereotyped movements that are easy to elicit, they can be measured accurately and objectively with the appropriate equipment, large numbers of data points an be collected, and longitudinal studies are not confounded by learning effects.

  • Schematic of eye changing gaze from central fixation point (blue star) to a target on the right (green dot). The change in gaze is represented by a thin orange dotted line towards the start to a thick orange dotted line to the green dot.

    Prosaccades

    Prosaccades are eye movements towards a target. The angular displacement of the eye towards the target is the saccadic amplitude and the time to look towards the target from the fixation point is the latency. Prosaccadic latencies have been shown to be greater in PD vs. HC.

  • Schematic of eye changing gaze from central fixation point (blue star) away from the target on the right (red dot). The change in gaze is represented by a thin orange dotted line towards the start to a thick orange dotted line to the red X.

    Antisaccades

    Antisaccades are eye movements from a fixation point away from a target. They are usually performed as part of an instructed task and involve inhibition of a normal reflexive response. Antisaccadic latencies are usually larger in PD and more prone to errors i.e. accidentally looking towards the target.

  • Left: White man with brown hair resting chin on headrest and staring at a red dot on a computer screen being recorded by a camera underneath the screen. Right: White woman with brown hair sitting on chair wearing VR headset and holding VR remotes.

    Saccadometer

    Saccadic eye movements can be measured using saccadometers; infrared head-mounted oculometers. Targets are projected as laser points on a plain surface in a horizontal line and eye movements are recorded from the central fixation towards/away from the target. Developments in camera technology can now allow saccades to measured using computer screen or VR goggles.

Fine Motor

Upper limb bradykinesia (slowness of movement) was captured using a single inertial sensor. Participants wore the sensor on the index finger and performed the MDS-UPDRS finger tapping, hand movement, and pronation-supination tasks (see video). Features extracted from this sensor include speed, amplitude, and rhythm; all of which can be affected by disease progression in PD/PSP.

Fine motor movements recorded by sensor on index finger. UPDRS tasks performed include: 1) Finger Tapping: tapping the index and the thumb together repeatedly, 2) Hand Movements: Clenching the fist and opening the hand repeatedly, 3) Pronation-Supination: Arm extended outwards with palm facing towards. Rotation of the palm towards the ceiling and floor repeatedly.

Gait & Balance

Body-worn sensors, such as inertial measurement units (IMUs), have been used to measure a large number of kinematic features derived from walking and postural sway. Data derived from IMUs have been used to successfully classify between PSP, PD, and HC participants.

Schematic of an individual wearing IMU sensors. Left: front view highlighting trunk, wrist, and foot sensors. Right: back view highlighting lumbar sensor.

Participants wore a six sensor IMU array to explore the progression of walking and swaying kinematic features longitudinally over the course of the study. While wearing the array, participants performed a set of tasks described below:

Postural Sway

Participants were instructed to stand as still as possible for 30 seconds with their eyes closed.

Two-minute walk

Participants performed a 2-minute walk in a quiet, uncarpeted 15-metre corridor, making turns when necessary.

Timed-Up-and-Go

Participants rose to standing from a seated position, walked forward three metres, turned, walked back to the chair and sat back down.

A footplate was utilised to ensure the distance, and angle between the feet was standardised for all participants before trial initiation

Levodopa Challenge

Levodopa is one of the most common antiparkinsonian medications. This symptomatic treatment is dopaminergic, supplementing for the loss of dopamine-producing cells as a result of the progressive neurodegeneration of the substantia nigra in PD. In spite of some common side effects of levodopa, this therapy is usually quite successful at managing a range of symptoms associated with PD. However, from a research perspective, antiparkinsonian medication can mask progression signals of the condition, making it difficult to understand the true progression of PD. Additionally, antiparkinsonian medication can in fact worsen kinematic features, such as altering saccadic latencies and affecting postural sway. This will impact the search for useful biomarkers for PD.

For the second visit of the study, Group 2 participants were asked to omit their antiparkinsonian medication on the morning of testing. Participants were tested twice; first when they arrived (off-meds), and then they were asked to take their medication before being tested for a second time one hour later (on-meds). This allowed us to identify features that were affected by levodopa (important biomarkers for symptomatic therapies) and thus not affected by levodopa (important biomarkers for disease modifying therapies).

Blue and white pills falling to the ground.
Translucent brain pulsing and emitting blue electricity.

Deep Brain Stimulation

Deep brain stimulation (DBS) is the dominant surgical treatment for PD. The procedure involves implanting a pulse generator into specific brain regions to stimulate deep brain structures via small current pulses. DBS of the sub thalamic nucleus is a particularly effective treatment, often improving tremor, bradykinesia, rigidity, and motor abnormalities. It can also lead to the reduction in necessary levodopa medication, in turn reducing unnecessary side effects.

DBS participants (Group 4) undergo the previously described experimental protocol. DBS participants are tested prior to DBS surgery both on- and off-levodopa medication, and a year post-surgery on- and off-stimulation. This allows us to elucidate the effect of both medication and DBS stimulation on PD motor symptoms.