Study protocol for the St James's Hospital, Tallaght University Hospital, Trinity College Dublin Allied Researchers' (STTAR) Bioresource for COVID-19

Study design

This is a prospective dual-site cohort study across two tertiary acute university teaching hospitals. The primary objective of this study is to build a biobank, to understand the clinical characteristics and natural history of COVID-19 infection with the long-term goal of research into improved disease understanding, diagnostic tests and treatments. We utilised a variety of existing infrastructures to support recruitment, clinical data collection, sample collection and storage, including the Department of Immunology, the Department of Infectious Diseases and the Wellcome Trust Clinical Research Facility at SJH and the Department of Respiratory Medicine in Tallaght University Hospital; and the Trinity Translational Medicine Institute (TTMI) to store and process these samples.

In common with other biobanks, study end points evolve over time as new research questions emerge. They currently include patient survival, occurrence of severe complications of the disease or its therapy, occurrence of persistent symptoms following recovery from the acute illness and vaccine responses.

Research timeline

The start date for this study was April 2020 with the first recruit on the 27th April 2020. The duration of the bioresource has been open-ended and will evolve with time and need. Use of existing infrastructure and re-allocation of clinical and research staff allowed the early start of the study.

Study setting

The STTAR Bioresource is situated in St James’s Hospital and Tallaght University Hospital in Dublin, Ireland. Our hospitals have a broad catchment area including parts of Kildare, South and West Dublin. Patients are referred for inpatient COVID-19 care by their general practitioner, emergency ambulance services or by self-presenting to the emergency department. Each hospital also receives inpatient transfers from secondary hospitals linked to them in the Dublin Midlands Hospitals Group⁵.

In both sites, all patients are tested for COVID-19 on admission to hospital, regardless of their presenting illness. Staff members are also tested by the hospital if they develop symptoms suggestive of COVID-19 infection or in the case of hospital outbreaks. Moreover, patients may be referred to the hospital ‘long COVID’ follow-up outpatient clinic, for example from general practitioners (GPs) or following discharge after an admission for COVID-19 infection. These three patient groups – inpatients, hospital staff, and ‘long COVID’ outpatients as well as healthy and disease controls– were eligible for recruitment and inclusion in the Bioresource study population.

People with acute COVID-19

Inpatients with polymerase chain reaction (PCR) confirmed SARS-CoV-2 in the two recruiting acute hospital sites are invited to take part. Hospital staff with confirmed diagnoses of COVID-19 disease are also recruited. Further samples and data are collected at clinical follow-up in post-COVID clinics.

Control group

Staff, patients, and members of the public who do not have acute COVID-19 are invited to participate. In order to facilitate effective matching to cases, controls from a range of ages and varying socio-economic status are sought. Control patients’ COVID-19 infection and vaccination status are recorded. Disease controls will be recruited from patients admitted with other respiratory infections (e.g. influenza).

Other specific disease groups

Patients with polymerase chain reaction (PCR) confirmed SARS-CoV-2 infection with underlying complex immunodeficiency disorders and vasculitis are invited to take part in smaller subgroup studies. Their results are compared to the general COVID-19 positive study group and to the control group. There are no specific exclusion criteria.

Time points

A practical approach to recruitment was put in place during the various waves of the pandemic. It is acknowledged by the study team that there is value in both samples and high-quality patient data. In general, samples are taken the morning following diagnosis or admission during the routine phlebotomy round. The study team monitors the patient record and where there is a clinical deterioration marked by an increase in Fi02 or a requirement for additional ventilatory support or intensive care assessment, a second sample set is sought. Further samples at discharge or convalescence are sought. For practical reasons samples at each time point are not always available or required. Emphasis on the relative importance of each time point is dictated by the steering group in response to the pandemic environment. Convalescent patients are also recruited from clinics during attendance for routine clinical care (Image 2).

Image 2. STTAR Bioresource Procedures Infographic.

Sampling protocol: patients with incident COVID-19

New patients with positive SARS-CoV-2 tests are identified by the STTAR Bioresource Study team from daily internal communications of new positive SARS-CoV-2 PCR tests and COVID-19 ward admissions. Patients are approached for informed consent and provided with the STTAR Bioresource patient information leaflet. The project and use of the samples are explained in detail, with a specific focus on genetic aspects of the study. All patients are given an opportunity to review documentation and ask questions prior to signing the informed consent form. Patients are made aware that there are no consequences to not being involved in the study, informed that they can specifically opt out of genetic aspects of the study and that they can withdraw their data and samples from the study at any time. In cases where informed consent from a patient is not possible, next of kin assent is sought in line with our ethics approvals. If/when they regain capacity, we will approach them for retrospective consent.

A unique bioresource study ID is assigned to each consented patient and recorded in recruitment logs. Completed paper informed consent forms for both hospital sites are held centrally and securely in the Wellcome Trust Clinical Research Facility.

To minimise patient discomfort and reduce the numbers of contacts, study blood samples are taken at the same time as routine clinical samples. A case report form (Appendix A, Case Report Form) detailing demographic and medical background is completed and saliva samples taken by the research nurse obtaining consent (Appendix B, Informed Consent Form). This data is then entered to the secure Dendrite database. Blood samples in SJH are transported to the central processing lab after phlebotomy, where they are retrieved directly by a member of the bioresource team, for immediate processing in the TTMI laboratory. In TUH, blood and saliva samples are taken by the TUH bioresource research nurse and placed in a morning courier to TTMI. If genetics is not consented to, Paxgene samples are not taken from the patient.

Withdrawal procedure

If the patient decides to withdraw from the study at any stage, the research team member documents this decision clearly in the patient’s medical notes and CRF detailing the reason if known. A withdrawal form is then completed.

Participants have the following options:

Sample processing

Samples are received by the STTAR team in a bag labelled ‘COVID BIORESOURCE’. Each bag contains 6 sample types and 8 tubes – the maximum volume collected is 46mls (see Table 1 and Image 3).

Image 3. STTAR Bioresource Study laboratory sample strategy.

Safety

Process

PBMC isolation using SepMate method

Cryopreservation

All steps should be performed immediately with minimal pipetting.

Documentation in sample spreadsheet

Study number, sample type, aliquot number, position and row are documented in the COVID BIORESOURCE 2020 spreadsheet. Information on all samples received for processing and the aliquots generated, including Study ID, aliquot ID, processing date and location of aliquot storage are logged on the STTAR Biobank Patient Recruitment and Sample Log.

Dispensing of samples from biobank

TTMI biobank staff are notified by the Steering committee Chair that a sample request has been approved and the samples are dispensed to the recipient. All samples dispensed from the biobank are logged on STTAR COVID-19 Biobank Patient Recruitment and Sample Log and the following information is recorded:

These are interim processes pending implementation of Module Bio Biobank Information Management System.

Clinical data collection

Patient characteristics (e.g. age, sex, level of education, co-morbidities, pre-admission medication) and features of COVID-19 infection (e.g. date of symptom onset, oxygen requirement, COVID-19 treatment, requirement for admission to intensive care, radiological findings, results of blood tests) are extracted from the electronic patient record in SJH and from the paper medical chart in TUH by clinical research nurses and/or clinical research fellow after completion of the acute inpatient episode. A subset of variables are recorded for the date of blood sampling for storage in the bioresource including symptoms, diseases severity oxygen requirement and medication. and (e.g. symptomatology, disease severity) Peak severity is defined using the WHO clinical progression score. Clinical data quality control is done by the PIs on 10% of cases selected at random.

The case report form (Appendix A) is completed by the research nurse on paper or using tablet computers with the patient at the time of recruitment (or as soon thereafter as possible) and captures additional features not routinely recorded in the patient record including patient level of education, country of origin, smoking/alcohol history, social exclusion status (homelessness, injecting drug use, having been in prison), COVID-19 vaccination status, accommodation, employment status and travel history among other fields. Pre-morbid frailty status is assessed by the clinical research nurse using the clinical frailty score⁸. Patient self-perceived social status is measured using the MacArthur Scale or ‘ladder score’^9,10.

Patient recovery is characterised using a case report form (Appendix C, Convalescent Case Report Form) completed at the time of clinical outpatient review, on paper or ideally directly on the clinical database. This includes back-to-work status, six-minute walk test results (conducted by our physiotherapy team) and Chalder fatigue scores. The six minute walk tests is an exercise test that measures distance walked over six minutes. It is a measure of functional status or fitness.

Clinical data is stored under a unique participant identifier on a bespoke STTAR Bioresource Clinical Data database on Dendrite, a secure electronic data capture (EDC) web platform for building and managing online databases and surveys. The Dendrite database operates on multiple-access levels and only the Principal Investigator (PI), Research Nurse and Clinical Research Fellow have complete access to the database and permission to modify. The STTAR Bioresource Clinical Data database is hosted by St James’s Hospital and security and access permissions are managed by the hospital’s information management system team. The database is protected behind host and institutional firewalls. The data controller for St. James’s Hospital patient data is St. James’s Hospital and for Tallaght University Hospital patient data held on the Dendrite database hosted by SJH, both institutions are data controllers. Inter institutional data sharing agreements between St. James’s Hospital, Tallaght University Hospital and Trinity College Dublin are in place.

Coded biological samples are processed and stored centrally at the STTAR Bioresource and archived by the biobank technician using industry standard Freezerworks software which only the biobank technician and lead study PI have access to. The registry database is shared with TCD IT service providers who manage their security and access permissions. This database is also periodically backed up on an external third-party server located in Dublin.

Experimental data collection

TCD researchers who generate experimental data deriving from analysis of STTAR Bioresource samples or data, store this in dedicated TCD SharePoint folders, which is the preferred storage solution of the university. Each researcher has their own access-controlled folder. These data include spreadsheets (excel, CSV, .xlsx), Graphpad Prism files (.pzfx), flow cytometry files (.fcs), word documents, image files (.jpeg, .png, .ndpi), PowerPoint files (.pptx), transcriptomic data files (.cel). All coded experimental data pertaining to the STTAR Bioresource study is stored on TCD’s Microsoft OneDrive service. OneDrive is Microsoft's cloud-based file storage service, which allows syncing and sharing of files between computers and mobile devices and is the cloud computing service of choice of TCD. Data hosted on OneDrive is securely hosted by Microsoft in Europe in compliance with relevant legislation (see TCD Data Protection Procedural Guidelines).

Strengths of this study

Patients began being recruited to the study in April 2020, only short weeks after the first confirmed case of COVID-19 nationally. This early recruitment has allowed for an invaluable look at the first group of inpatients in Ireland with the disease. Over 1000 patients have been recruited to this study so far. This large study number increases the statistical value of any results. Detailed demographic data are being collected including patient occupation, number of people living in patients’ home, number of people sharing a bedroom with patient, home place (own accommodation, nursing home resident, homeless shelter resident) among other details. This allows us to analyse in detail a patient’s premorbid level of social deprivation, something not analysed in previous similar biobanks. The biobank participants are recruited from inpatient wards, intensive care units and outpatient staff groups. This allows for a wide range of disease severity (asymptomatic to critically unwell), disease acquisition (community, occupation and hospital-acquired) and age groups to be studied. Large numbers of healthcare workers are recruited to this study. This uniquely vulnerable group will be compared to the usual inpatient group. Looking more closely at infected health care workers may help inform future infection control procedures. Over 100 control patients have been recruited to the study. Vaccination status as well as usual case report form demographic details are collected. These healthy age matched control patients will allow comparison of results from studies conducted within STTAR Bioresource.

Limitations of this study

Most patients who are enrolled in this study are hospital inpatients. Patients hospitalised with COVID-19 likely have severe disease requiring high level care. Recruiting most of our patients from the inpatient cohort means we may have a biased sample population and miss those with asymptomatic or mild disease. We follow our recruited patients in post COVID-19 outpatient clinics. Those who cancel their appointments due to resolution of symptoms are more difficult to capture in our study and so there may be a bias towards long COVID symptoms in those we see.

Underlying data

No data are associated with this article.

Extended data

Zenodo: Study Protocol for the St James's Hospital, Tallaght University Hospital, Trinity College Dublin Allied Researchers' (STTAR) Bioresource for COVID-19, https://doi.org/10.5281/zenodo.6278295.

This project contains the following extended data:

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).