Anna Wedderburn speaks to clinicians at St James’s Hospital, Dublin, who are engaged in medical and scientific research to better understand Covid-19 and the differing immunological responses to SARS-CoV-2 infection
It is primarily the immune response of an individual that determines the outcome of SARS-CoV-2 infection. At St James’s Hospital, Dublin, there are a number of ongoing research projects investigating both the clinical implications of this disease, and the immunological reasons behind the clinical manifestations of Covid-19.
Dr Liam Townsend, Specialist Registrar in Infectious Diseases at St James’s, is involved in a number of these projects. One project, which Dr Townsend hopes will soon be published, has used machine learning to discover clinical measurements that predict whether a Covid-19 patient will require ICU attention. Dr Townsend explained the key points of this study to the Medical Independent (MI).
The first step in this project was to determine the clinical differences at peak severity of Covid-19 (defined as the maximum oxygen required during inpatient stay) between patients requiring and not requiring ICU care, examining blood results, x-ray changes, and oxygen levels.
Dr Sultan Imagaliyev, Data Scientist at Trinity College Dublin, developed a machine learning programme that compared the measurements of patients requiring and not requiring ICU care. This programme identified a number of clinical measurements that differed between those in ICU and not in ICU.
The second step was to develop a programme that could predict ahead of time who would be in need of ICU care. The researchers wanted to determine whether the differences discovered in part one of this study could be of predictive value.
For this part of the project, clinical measurements were taken from patients, on average 36 hours prior to the onset of peak illness. The reason that clinical measurements were taken at this time was it tended to be the stage at which many patients presented themselves to the hospital during their Covid-19 illness.
The machine learning model was applied to the clinical measurements, alongside the outcome of each patient, ie, whether they progressed to requiring ICU care or not. The programme or ‘machine’ used this coded data to train itself. It was then able to predict, using the clinical measurements it was given, who would require ICU attention in the next 36 hours.
The machine’s predictions were then tested by giving it ‘unseen’ data; the machine did not know which patients progressed to requiring ICU care, but the research team did. This performance test showed the machine could predict with “excellent accuracy” whether or not a patient would require ICU care within the next 36 hours. From this, the team was able to derive clinical measurements that are predictive of requiring ICU attention.
At the moment, this information is useful, as it offers healthcare workers a glimpse into what the next 36 hours may look like. However, Dr Townsend said that the real hope is that this information will be of assistance in the future, once treatments become available. Being able to predict who will require ICU attention may inform clinicians which patients will benefit from specific treatments.
While this project is very focused on the clinical side of Covid-19, Dr Townsend is also involved in projects that are not as patient-centric, and instead focus on understanding the immunological responses underpinning the spectrum of Covid-19 severity.
Two of these projects are being carried out in collaboration with the Pasteur Institute in Paris, France. These projects are focused on the immune response to SARS-CoV-2 at both systemic and local tissue level — the reason for this being that the local immune environment often differs from the systemic immune environment. SARS-CoV-2 is a respiratory tract infection; the immune response in the nose may differ from the immune response detected in the peripheral blood. Although these are two separate projects, they overlap, as the same patient cohort is being used, and the results will be used in conjunction with one another.
In looking at the local immune response to the virus, Dr Townsend and his colleagues are taking nasal swabs from Covid-19 patients. These nasal swabs only require a quick sample from the inside of the nostril; they do not go deep into the nose, and therefore are not uncomfortable. The swabs will allow researchers to characterise the nasal microbiome, or the microbes living within the nose. It is possible that certain bacteria may be associated with severe or non-severe versions of Covid-19.
The nasal swabs will also provide information on the inflammatory markers within the nose, and whether SARS-CoV-2 antibodies are present. Dr Townsend explained “a blood sample might not show any significant information, whereas a nasal swab might show that actually in the nasal pharynx (the anatomical site of where infection occurs), there is quite a lot of inflammation.” The swabs are sent to the Pasteur Institute for analysis.
Mr Jamie Sugrue, a PhD Candidate at Trinity College Dublin, is involved in the other collaborative project between St James’s Hospital and the Pasteur Institute. Although he usually investigates the immune response to hepatitis C, his research interest took a turn when SARS-CoV-2 began spreading in Ireland. Alongside his hepatitis C work, he is now also uncovering the immunological differences observed in the different severities of Covid-19 infection.
Mr Sugrue works with Dr Townsend and told MI they are using blood from patients with mild, moderate, severe, and critical Covid-19 cases in an attempt to understand the immunological cause of these differences. They hope to determine whether these groups’ immune systems react differently, not only to SARS-CoV-2, but to other immunological stimulants as well.
“The immune response is widely variable from person-to-person,” Mr Sugrue explained. This is something that SARS-CoV-2 has really emphasised; some individuals develop an asymptomatic infection, while others require critical care. This project aims to gain a better understanding of Covid-19 patients’ immune responses as a whole, as opposed to honing-in specifically on their immune response to SARS-CoV-2.
Mr Sugrue will challenge each donor’s blood with four different immunological stimuli, each in its own tube. These tubes are then sent to the Pasteur Institute, where the activity of 560 immune genes and the level of inflammatory markers in the blood are measured. This will shed light on whether individuals with the same level of Covid-19 symptoms share similarities in their immune responses, and if they do, it will demonstrate what these similarities are.
This project will not only enhance our understanding of the different immune responses observed in Covid-19 — it will also help inform therapy design. In order to continue replicating in our bodies, viruses often shut down certain parts of the immune response. If SARS-CoV-2 shuts down a part of the immune response, designing a therapy that attempts to use this part of the immune response would be a waste of time, money, and effort.
For example, it has been demonstrated that the immune system of individuals who suffer only mild Covid-19 symptoms mount a strong type one interferon (IFN) response against the virus. This response is not seen in severe and critical Covid-19 patients. One potential therapy for these individuals could be injecting type one IFN into their bodies. However, we do not know if SARS-CoV-2 has ‘switched off’ these individuals’ capability of responding to type one IFN or not. If it has, using type one IFN as a therapy would be ineffective.
This project will inform researchers what aspect of the immune response they should, or should not, be targeting when developing therapies for this disease.
Each of these projects is using the same cohort of patients. As a result, a rather holistic view of Covid-19 will be formed: From the clinical measurements, to the systemic immune response, to the local tissue immune response.
Dr Townsend noted that none of this research could have been done without the willingness of Covid-19 patients to take part in these studies. Given that this is a new disease and patients are getting blood tests done anyway, many people have said that they are happy to do “anything they can to help”.
Patients monitored at Covid-19 outpatient clinic
Due to the decrease in new Covid-19 cases across Ireland, ICUs are not as busy as they once were, and researchers are finding it more difficult to get new Covid-19 samples. However, there is still plenty of work to be done.
An outpatient clinic has been set up in St James’s Hospital for Covid-19 patients who required hospitalisation, who are no longer infected but are still in recovery. This is to ensure that from a medical perspective, these patients are recovering well. It is not yet fully known what complications may arise from SARS-CoV-2 infection.
In this outpatient clinic, Dr Townsend and his colleagues assess the respiratory function of recovering Covid-19 patients; they want to be aware of whether these individuals have had any breathing issues since leaving hospital. They are also investigating chronic fatigue in these individuals, as this is something they have noticed in many of these recovering patients. This is noteworthy, as chronic fatigue is not typically seen with other infections.
The outpatient clinic is also watching for signs of post-traumatic stress disorder in patients recovering from severe Covid-19, as this would have been a near-death experience for many individuals who otherwise would have been “quite well”.
In terms of scientific research, many patients in the Covid-19 outpatient clinic are donating their blood in order to demonstrate whether this viral infection is leaving any ‘immune scars’ post-resolution of infection. It is possible that, similar to SARS-CoV-1, the immune response may become somewhat altered in some individuals.