Investigation of low-dose PET-MR imaging for neurodevelopmental disorders

Abstract

Background
The demand for mental health services is rising and, in recent years, the hypothesis of a genetic neurodevelopmental continuum linking psychiatric diseases with neurodevelopmental disorders (NDDs) has emerged. Positron Emission Tomography (PET)- Magnetic Resonance (MR) is an imaging modality with potential for obtaining a better understanding of the aetiology and pathophysiology of these conditions. However, the amount of the radiopharmaceutical injected must be optimised to allow for longitudinal investigations to take place.

Aim
To investigate the feasibility of reducing the radiopharmaceutical dose in patients with neurodevelopmental disorders who undergo brain imaging in the PET-MR scanner, without compromising the quantitative accuracy of outcome measures.

Methods
This project comprised two parts: a retrospective research element and a prospective service evaluation element.

The retrospective research element focused on simulating low dose imaging datasets (1/2, 1/3, 1/4, 1/5, 1/6, 1/10 and 1/15 of the originally administered dose) from [11C]PHNO and [11C]UCB-J PET-MR scans, and compare them to the full dose datasets. For the [11C]PHNO scans, the striatum, caudate, putamen, globus pallidus, thalamus, substantia nigra and accumbens brain regions were assessed. For the [11C]UCB-J scans, the substantia nigra, thalamus, striatum, caudate, putamen, hippocampus, insular cortex, temporal lobe, parietal lobe, frontal cortex, cerebellum and accumbens were assessed.

The service evaluation element focused on disseminating a prospective survey with multiple choice, open-ended and closed questions, to 51 European PET-MR institutions, to understand and define the current practices related to radiopharmaceutical dose optimisation, establish baseline data and generate evidence of service effectiveness.

Analysis
Image data from the retrospective element was analysed with an imaging software (MIAKAT™). Descriptive statistics were calculated for each structure, per simulation. Bias and statistically significant differences were calculated, to determine the impact of low dose imaging on clinical outcome measures, such as non-displaceable binding potential (BPND) and standardised uptake values ratio (SUVR).

Descriptive analysis, and filtering and segmentation techniques were used to analyse the fact-based quantitative and qualitative data obtained from the survey.

Findings
The results obtained for the [11C]PHNO BPND suggest that a decrease of the injectable dose may be possible however, further investigations are required. The results obtained for the [11C]UCB-J suggest that, when the outcome measure is the BPND, a decrease of the injectable dose to 1/2, is feasible; when the outcome measure is SUVR, a decrease of the injectable dose to 1/3, is possible. The methodology deployed appears to be well suited for investigating low dose scans for radiopharmaceuticals with cortical and striatal uptake, when the outcome measure assessed is the SUVR. These findings were further corroborated by the answers to the prospective survey which revealed that optimisation techniques should be radiopharmaceutical dependent.

Moreover, the findings from the prospective survey revealed that, whilst different optimisation methods are used for different clinical indications, non [18F]FDG doses for PET-MR brain scans are frequently not optimised.

Contribution to Knowledge
Low dose PET-MR imaging is possible however, institutional investigations must be carried out. Dose optimisation and low dose imaging are intrinsically related with the field of radiation protection and play a significant role in Sustainability.

Conducting low doses investigations will foster an environment where the health and wellbeing for patients, carers and staff is placed first, by limiting their radiation exposure; promote quality education and advanced practice amongst the Nuclear Medicine community; reduce the discharge of waste products; and optimise clinical operations.

Date of Award	25 Jun 2025
Original language	English
Awarding Institution	University of Bath
Supervisor	Stephen Husbands (Supervisor) & William Hallett (Supervisor)