Benchtop NMR Spectroscopy for Assessment of Clinical Human Pathologies (BRANCH-P STUDY)

Overview

This research programme seeks to combine the resources of NHS primary care, with the leading spectroscopic work in low-magnetic fields of the Wilson Group (Nottingham Trent University) to demonstrate the potential for benchtop Nuclear Magnetic Resonance (NMR) spectroscopy in human clinical pathology. This is an instrument assessment study for point of care viability which will also result in enhanced patient care (pending their consent) in blood screenings and metabolic health data.

The research project aims to gauge the efficacy of benchtop Nuclear Magnetic Resonance devices in a primary health care setting. The study design and methodology is as follows: Typically, before a patient attends their routine standard of care blood test, they will speak with their local General Practitioner (GP) or a Registered Nurse (RN). At these visits, the medical professional will gauge the interest of the patient and test if they are eligible. If the patient is interested in becoming involved with research, and they are eligible for the study, they will be given a 'Study Information and Informed Consent Form' to read before their next visit. This allows for ample time to become fully informed on what the study entails, as well as to ask any questions they may have. The 'Study Information and Informed Consent Form' document contains all information detailing: the purpose of the study, the significance of the research, and the patient's right to withdraw at any time (as in accordance with General Data Protection Regulations \[GDPR\] 2018), potential risks and benefits of taking part in the study and who to contact if the patient had any further questions to be asked. At the patient's routine standard of care visit, they will present their signed and dated consent form to the medical professional taking the biological samples. If the patient loses their consent form before this visit, they will be given another to complete in full before any residual biological samples are taken. This will be counter-signed by the medical professional present at their visit. The patient will then be anonymised by being given a unique identifier in the form of a barcode. This barcode will be applied to the consent form, and the form will act as a linking document. The patient will only be referred by their unique identifier for the remainder of the study. This anonymisation of identifiable patient information (such as name, Date of Birth \[DoB\], etc.) extends to any data produced by analysing the patient's samples. The patient will keep a copy of the consent form for their own records, whilst the original document will be held by the research team to be stored on-site behind several layers of security. A copy of the consent form will also be uploaded onto the patients SystmOne profile. SystmOne is a clinical computer system used by medical professionals to log patient visits, review tests results and upload any new relevant data. After sample collection, no further actions are required of the patient for the remainder of the study. The patient will be given a debriefing document thanking them for their contribution. Within this document there are contact details of the researchers involved and other contact telephone numbers to a variety of organisations and helplines specialising in emotional support, if the patient feels in any way affected by the study. The patient will also be offered the opportunity to be given a copy of their bio fluid spectral data should they be interested in this. This will be accompanied by explanatory notes. The biological samples collected from the patient will go to two different locations: samples taken as part of the patient's routine standard of care will go to a local diagnostics lab where they will be tested on a variety of metrics. Residual biological samples will be processed and tested at the local Willows Health Near Patient Testing laboratory, before being frozen at -80oC. The test results provided by the diagnostics lab will be uploaded to the patient's SystmOne account, where they will be scribed and used for the study, as outlined in the 'Study Information and Informed Consent Form' document. Results gathered from the Willows Health Near Patient Testing laboratory will also be uploaded to the patients SystmOne account and stored for later use within the study. The combination of these two sets of data will form the Near Patient Testing data set. The residual biological samples frozen at Willows Health will then be transferred in batch format to Nottingham Trent University's (NTU's) benchtop Nuclear Magnetic Resonance device. They will then be thawed, prepared further and analysed using the benchtop Nuclear Magnetic Resonance device. The resultant spectra will be stored on the password protected device to form the benchtop Nuclear Magnetic Resonance data set. With both sets of data gathered (Near Patient Testing suite and benchtop Nuclear Magnetic Resonance spectra), a machine learning algorithm will cross-compare the data sets to 'teach' a program to 'read' the benchtop Nuclear Magnetic Resonance spectra and output data in the form of Near Patient Testing machines. On average, \~42 patients attend the Willows Health Medical Centre for routine standard of care visits per week in which biological samples are collected for pathology lab testing. This equates to \~168 samples per month. With the sample collection window planned to take a year, this calculates to \~2016 potential participants being involved within the study. This is an estimated maximum, as not all patients who come for their annual review may not consent to the research study. Additionally, some religious events and holidays may limit the uptake of potential participants, for example: during Ramadan, although it is not strictly prohibited, blood tests may be avoided due to the weakening effect it may have. This, in conjunction with fasting, may dissuade potential participants from joining the study. Given the substantial volumes of data required to 'train' software when utilising machine learning algorithms, the sample size needs to be large. This is imperative to produce accurate and reliable results from training and test sets for machine learning algorithms. This research will result in several advantages for both clinical care and patients. For instance, efficient and rapid analysis of biological fluids will help to overcome the limitations associated with diagnosis of a variety of diseases such diabetes, chronic kidney disease, cancer and any other conditions that might go asymptomatic in the early stages. This work has the potential to pave the way for new clinical pathological techniques to support the early diagnosis of diseases.