Innovating to detect infection
Dr Alex Richter, Professor and Consultant in Clinical Immunology, Institute of Immunology and Immunotherapy, led a team from the University of Birmingham to develop a test that could detect COVID-19 in people with mild disease or who didn’t show symptoms.
The SARS-CoV-2 pandemic has had an unprecedented effect on all our personal and work lives. Many of us were redeployed, we learnt to manage our patients from afar, our laboratory work patterns and flows were thrown upside down and we have rapidly evaluated and introduced new assays within extraordinarily short timelines. But there has never been a time when pathology specialties were needed more. Conversations around testing, sensitivity and specificity of assays, and false positives and negatives have all become common national parlance.1
I direct the Clinical Immunology Service at the University of Birmingham. We are a UKAS accredited NHS-facing immunology and haemato-oncology laboratory that also delivers central trials analysis and has a track record in immunodiagnostic development. Before lockdown we ordered in a number of SARS-CoV-2 antibody assays. Verification of their performance highlighted that they could detect antibodies in patients who have been critically ill in hospital but were woeful at detecting individuals that had suffered milder disease. At the time there was no viral RNA testing for staff or community cases and sick rates were high in our hospitals. There was a real demand for antibody testing among healthcare workers. They needed to know if they had already been exposed and healthcare leaders needed this information for occupational risk analysis.
We repurposed our lab and expanded into the unutilised university teaching laboratories to focus on developing a SARS-CoV-2 ELISA test that was designed to detect individuals with mild disease and even those who didn’t show symptoms. Using a systematic academic approach, we evaluated the best test reagents to produce a high sensitivity assay which detects individuals who have been exposed to the virus and have had mild or asymptomatic disease. Having achieved a high sensitivity and specificity we approached a local Birmingham-based diagnostic company, The Binding Site, to partner us to produce this test commercially. Compressing conception to test launch into four months was a steep learning curve and an amazing testament to teamwork and cooperation.2
Following a call from the Birmingham Women’s and Children’s Hospital one Sunday in May, we met to discuss the first suspected cases of paediatric multisystem inflammatory syndrome temporally associated with SARS-CoV-2 (PIMS-TS). We had a cluster of children who had developed a sepsis/ Kawasaki-like syndrome but had a negative SARS-CoV-2 swab test. Using our assay, we established that all these children had been infected some time ago. This suggested a delayed immune-modulated condition rather than being a direct result of the virus.3
We have found that antibody concentrations are highest in patients that have been hospitalised and have been on the intensive care unit.
Alongside working on the assay, I also worked with the University Hospitals Birmingham Trust to set up a research study to determine the immune response to SARS-CoV-2 infection in healthcare workers who were recovering from infection. In April, the first arm of this study recruited 554 members of staff in just 20 hours. The demand for testing was humbling. This first study, recently published in Thorax,4 found healthcare workers were at increased risk of prior exposure to the virus than the general public, with nearly a quarter already exposed to the disease by this time.
Cleaners and acute medical specialties had higher rates than workers in the intensive care unit, highlighting the importance of differing infection control procedures. We also found higher seropositivity in black and ethnic minority healthcare workers. The study established the presence of asymptomatic carriage in healthcare workers and showed that individuals can be identified in the pre-symptomatic phase.
We have found that antibody concentrations are highest in patients that have been hospitalised and have been on the intensive care unit. Lower levels have been found in those with milder disease and the lowest levels were seen in those that seroconverted following asymptomatic infection.
Understanding the antibody response is essential for predicting long-term protective immunity and designing vaccine schedules.
In total, we have recruited approximately 4,000 hospital and dental healthcare workers and have ongoing studies to determine how good SARS-CoV-2 antibodies are at protecting against the virus and establishing their longevity. Understanding the antibody response is essential for predicting long-term protective immunity and designing vaccine schedules.
Taking blood for so many staff members has been a challenge, so as part of the study, we evaluated alternative sampling solutions. Our assay, dried blood spot (DBS) testing – which uses a pinprick of blood and is easier to obtain and store – compared well with taking a blood sample from a vein5 and we have switched to DBS for recent studies to reduce the burden on research teams. Saliva is still under evaluation, but initial analysis has found that you can readily detect antibodies. There may be some instances where the immune system produces antibodies that can be detected only in saliva or by a blood test.2
While it is difficult to see light in these difficult times, I do hope to contribute towards a positive legacy from this pandemic. Pathology services have been given a profile and invested in. The quality and clinical utility of testing have been rightly scrutinised, and we have a model for monitoring and tackling other viruses that subject the NHS to yearly winter pressures and large-scale loss of life. Academic and NHS organisations have worked together to rapidly translate basic science, and multidisciplinary teams and organisations have collaborated in ways that have not only been effective, but highly rewarding.
2 Faustini SE, Jossi SE, Perez-Toledo M, Shields A, Allen JD, Watanabe Y et al. Detection of antibodies to the SARS-CoV-2 spike glycoprotein in both serum and saliva enhances detection of infection. medRxiv 2020;2020.06.16.20133025.
3 Perez-Toledo M, Faustini SE, Jossi SE, Shields AM, Kanthimathinathan HK, Allen JD et al. Serology confirms SARS-CoV-2 infection in PCR-negative children presenting with Paediatric Inflammatory Multi-System Syndrome. medRxiv 2020;2020.06.05.20123117.
4 Shields A, Faustini SE, Perez-Toledo M, Jossi S, Aldera E, Allen JD et al. SARS-CoV-2 seroprevalence and asymptomatic viral carriage in healthcare workers: a cross-sectional study. Thorax 2020; doi:10.1136/thoraxjnl-2020-215414 (Epub ahead of print).
5 Morley GL, Taylor S, Jossi S, Perez-Toledo M, Faustini SE, Marcial-Juarez E et al. Sensitive detection of SARS-CoV-2-specific-antibodies in dried blood spot samples. medRxiv 2020;2020.07.01.20144295.