Using medical-detection dogs to identify people with SARS-CoV-2. Phase I-III studies. Application for participant results details.

ARCTEC, the commercial arm of The London School of Hygiene and Tropical Medicine (LSHTM) is requesting details of study participants’ COVID test results obtained at Pillar 2 sites and NHS hospitals. The data request is being made to support an HRA-approved study “Using Medical Detection Dogs to identify people with SARS-CoV-2. Phase 1, Proof-of-concept studies”. This study, initiated in 2020, will assess the sensitivity and specificity of medical detection dogs in identifying SARS-CoV-2 infection. The success of this study would aid in preventing future outbreaks of the virus, as the medical detection dogs could provide a high-throughput screening tool in key public risk areas such as airports.

ARCTEC is a wholly owned subsidiary of the LSHTM, all staff are employed on LSHTM contracts, housed in LSHTM building and supported by the same infrastructure [IT and Data Protection Office]). ARCTEC staff have acted as the Trial Coordinating Centre for this study, leading study activities, and recruiting all volunteers for the current data application.

Following the changing path of the epidemic in the UK, the Department of Health and Social Care (DHSC) has made a request for evidence that:

~ The medical detection dogs can detect COVID-19 in people with new viral strains.

~ The dogs can detect COVID-19 in people with low viral loads, and asymptomatic individuals.

The aim of this data request is to identify key subsets within the existing samples (people with variant of concern, low viral load), and test these samples to assess whether dogs’ screening accuracy remains consistent under these variations.

The SARS-CoV-2 ‘Kent’ viral strain (SARS-CoV-2, variant of concern 202012/01) shows S-gene target failure in commonly used Polymerase Chain Reaction (PCR) assays (a method of amplifying tiny amounts of genetic material e.g. DNA, from a sample so it can be detected more easily. It runs in cycles with a higher amount of genetic material present after each cycle is complete). The data requests will give details of assay outcomes, which can be used to infer which participants may have been infected with the ‘Kent’ strain.

Recruitment to the study was conducted between August 2020 and January 2021. As the ‘Kent’ strain became common in the UK in December 2020, it is highly likely that the study contains a mix of samples representing different viral strains.

London School of Hygiene and Tropical Medicine requires further data from the COVID-19 UK Non-hospital Antigen Testing Results (Pillar 2) Dataset to understand which strain of SARS-CoV-2 participants were infected with when the individual tested positive and the onset of symptoms for approximately 900 study participants.

Test outcomes, symptom start dates and PCR cycle thresholds (Ct thresholds) will be used to assess claims around screening success in people with no symptoms/ low viral load.

Release of data will be needed for modelling studies assessing the detection dogs’ utility in different screening scenarios (low/ high infection prevalence, and at ports of entry with novel viral strains present). Release of data will impact the study team’s ability to obtain funding for phase 2. The scope of this data request is to provide data useful to the current study phase (phase 1, proof of principle), next phase (phase 2, in-person training with healthy volunteers) and input into the accessory modelling study.

The LSHTM is the sole Data Controller and Processor. Medical Detection Dogs (MDD), University of Durham and DHSC will not process the NHS Digital pseudonymised data. They will however be given access to sections of aggregated data (in line with the HES analysis guide) in order to assess and report on study outcomes (chiefly the dogs’ sensitivity and specificity in screening key subset groups). The aggregated data provided to MDD will contain group information (viral load, strain type), sent via encrypted file transfer.

The Trial Steering Committee have an advisory role, they do not have any control over how NHS Digital data is used or access to these data (except in the form of aggregated (with small numbers suppressed) trial results produced for or presented at these meetings). They provide advice, guidance, and support to the trial. The Trial Steering Committee meet virtually, and are chaired by an LSHTM staff member.

The research is funded by the Department of Health and Social Care (DHSC). DHSC does not have access to data, except for aggregated reports (with small numbers suppressed) on study outcomes, which do not contain identifiable information.

The data is being processed under the following legal basis:

GDPR 6 (1) (e): Public Task; The study is performing research on a disease of high public health importance, and the data request is required to as part reporting requirements put in place by government stakeholders and GDPR 9(2)(j): Public health research; This research is on a screening tool for COVID-19, a disease of public health importance.

All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract ie: employees, agents and contractors of the Data Recipient who may have access to that data). There will be no requirement nor attempt to re-identify individuals from the data. The data from NHS Digital will not be used for any other purpose other than that outlined in this agreement.

LSHTM will supply a file of approximately 900 trial participants containing participant identifiers (Study ID, Name, Gender, Date of Birth and Test date) to NHS Digital. Data will be transferred to NHS Digital twice, once when the agreement is signed off, and once 2 weeks after the newsletter has been distributed to cohort members to give them sufficient time to withdraw from the study.

Data subjects are trial participants, who have provided informed consent to participate in the trial. Participants are adults aged 16+, who sought a COVID test for reasons of symptoms or probable disease exposure (via household or occupational risk). Participants provided odour samples to the trial, and were subsequently given follow-up for adverse events (up to 30 days after first participation). At 30 days after participation, participants’ involvement in the study ends.

The participant identifiers provided by LSHTM will be linkable to COVID-19 PCR results data in the COVID-19 UK Non-hospital Antigen Testing Results (Pillar 2) Dataset.

The detailed sample results will be returned to LSHTM where the data will be pseudonymised using Participant ID numbers such that it cannot be linked to individual participants without use of a participant ID log.

The dataset will be immediately stored on an LSHTM secure server. Data will only be accessible to named individuals, who are substantively employed by LSHTM. Access must be approved by the Data Manager.

This data will be securely encrypted on the Secure Encrypted Server at LSHTM to form a list of participants IDs, infection status, Date of Birth, Gender, Ethnicity, Enrolment Site, Medical History (if relevant). This fully anonymised encrypted dataset will be shared with Medical Detection Dogs. The fully anonymised decrypted dataset, including participant IDs, will be shared with statisticians.

All data requested is essential to address study aims, and has been minimised to contain only relevant outcomes. The subjects involved have been minimised- no data will be requested on participants who have been withdrawn from the study, or whose samples have been classified unusable (e.g. due to improper packaging, or excessive time lag between day of test, and day of sample collection).

It is expected that there will be a number of different outputs with the results of the study data:

PUBLICATIONS

A scientific report on phase 1 findings is already in draft for submission to peer-reviewed scientific journals (open access publication) and data provided by NHS Digital will feed in to results and discussion of this paper. Publications will only contain data aggregated with small numbers suppressed in line with the HES analysis guide. Target date for publication: June 2021.

GENERAL PUBLIC

Research findings will also be disseminated to the public through various routes including the study websites, newsletters and through the active public engagement programmes. The LSHTM will work with its media department and its lay representatives to ensure the findings are accessible to the broader public. Target date: periodic updates to be provided throughout course of study.

PARTICIPANTS

Participants will be appraised of study findings through a newsletter, to be sent out at study close, July/ August 2021.

ENGAGEMENT WITH POLICY MAKERS

LSHTM will rapidly disseminate evidence on medical detection dogs as a disease-screening tool to key stakeholders (policy-makers, Border Force) through regular meetings and provision of interim reports. Government stakeholders sit on the Trial Steering Committee, and are therefore appraised of trial progress twice monthly in committee meetings.

In addition to engagement with policy-makers, the study will provide reports on key outcomes to stakeholders with a role in translation of the tool to operational deployment (Innovation Leads).

HANDBOOK

At the conclusion of phase 3 of the study, a handbook on dog training for COVID-19 detection will be produced, offering guidance to other teams both in the UK and internationally, on rolling out this strategy, thereby maximising impact of the study. Target data for completion: July 2021.

STUDY DATA SHARING

The research conducted in this study may provide valuable evidence on the use of dogs as a screening tool for other viral infections and health conditions. To enhance future research in the area, anonymised study data may be published in a data repository (only aggregated data with small numbers suppressed in line with the HES analyses guide will be published). This will be beneficial in generating high quality research evidence related to viruses and outbreaks similar to the COVID-19 pandemic in the future. Thus, the potential benefits of this study may be substantial. Target date: To be published alongside publication, June 2021.

No identifiable personal data will ever be published.

The expected benefits of the study are the testing and validation of a medical detection dogs as a screening tool for identification of COVID-19. If found to be an accurate screening tool, medical detection dogs could be deployed at key ports of entry, and high risk public gatherings to quickly and accurately identify people with COVID-19.

Preliminary results from phase I of the study have been promising in indicating that the Medical Detection Dogs are able to detect samples from positive participants.

Medical Detection Dogs are able to screen up to 250 individuals per hour, and are suitable tools for a number of settings, including but not limited to border points, public events, and screening of groups for whom nasal swabs are unfeasible or distressing.

The application of dogs could provide a key tool in preventing further outbreaks of the SARS-CoV-2. If dogs are established as screening tools in such locations, this could have a noticeable impact on future considerations to alleviate lockdown measures.

The data from NHS Digital will be crucial ensuring that the dogs are trained not just to detect the virus, but also picking up variation in virus strain, viral loads, and asymptomatic infected individuals. The knowledge of these details in individuals can enable targeted strategies for treating them and preventing spread.

Yielded Benefits is not a requirement for new applications.