“We’re trying to define the disease at the same time as treating it”

Professor Amitava Banerjee tells Tom Ireland about initial insights from the largest clinical study of long COVID to date

Amitava Banerjee is a professor of clinical data science at University College London (UCL) and a consultant cardiologist. At the start of the pandemic, Banerjee was interested in the links between cardiovascular disease and poorer health outcomes for people with COVID-19. But as he began to see more and more patients experiencing chest pains, palpitations and breathlessness for weeks after bouts of COVID-19, he and colleagues at UCL’s specialist long COVID clinics started work on a proposal to study the long-term effects of the disease.

The resulting project, STIMULATE-ICP (Symptoms, Trajectory, Inequalities and Management: Understanding Long-COVID to Address and Transform Existing Integrated Care Pathways), is the largest clinical trial on long COVID to date, involving over 4,500 people and £8 million of funding from the National Institute for Health Research. The two-year study will look at all aspects of long COVID, from its underlying biology and disease course to diagnosis, treatment and care pathways, as well as the impact of the disease on people’s lives and the health system.

In the UK, over one million people have reported COVID-19 symptoms lasting 12 weeks or more, and over half a million have reported symptoms lasting over a year. Ninety NHS clinics have been set up to try and help people suffering with the range of chronic health problems that have together become known as long COVID.

Banerjee believes that Governments have underestimated the long-term impact of COVID-19 as policies and funding continue to be directed at reducing hospitalisations and deaths. He hopes that the STIMULATE-ICP project can improve how chronic health conditions are treated and considered in pandemic preparedness in future.

Ami, firstly can you give us a sense of what range of symptoms you are seeing within the UK’s long COVID clinics? Can you talk us through what someone experiencing long COVID typically might be going through?

We know of probably 200 or so symptoms that people might have. Patients have admirably led a lot of research themselves in terms of symptoms surveys, and the term “long-COVID” was actually coined by patients themselves on social media in around April-May of 2020.

The most common symptoms are fatigue, breathlessness, muscle pain or weakness, fever, sweats, excessive sleepiness and what people have called brain fog – a kind of cognitive impairment where your head feels fuzzy and you're not able to concentrate on anything.

These symptoms might cluster in particular ways that suggest there are different mechanisms at work. The PHOSP-COVID study led by the University of Leicester for example has suggested that there may be a more inflammatory phenotype, and there are people with more neuro-cognitive and neurological-type symptoms, but there's still work to be done there.

Is what you are seeing comparable to any other post infection conditions, or is it completely unique? 
                                                                                                                                                            Viral post infectious syndromes are not uncommon. They've been seen with Epstein Barr Virus, which causes glandular fever, and the common flu. What is really different is the scale and the speed of the onset here. You are never looking at 10% of cases having persistent symptoms, as we are with COVID-19.

From the beginning of the pandemic people have made comparisons with ME, also known as chronic fatigue syndrome – and I think there are some similarities. People with chronic fatigue have been under recognised, under-diagnosed and under-managed in the past, despite having very severe symptoms, and I hope with our study we provide the opportunity to shine a light on these types of illnesses.

Treatment for people experiencing long COVID symptoms currently involves helping them manage their symptoms at home, for example using breathing techniques to help with breathlessness. The STIMULATE-ICP study will look at the efficacy of using anti-histamine, anticoagulant and anti-inflammatory drugs.

We still, today, have this debate – or body of opinion – that feels we can live with high infection rates without any implication, and that's trouble because we've only been looking at hospital admissions, ICU admissions and death. But if we have people with chronic complications, even if it's three months, or six months, that's a substantial effect. We're seeing that already in terms of the workforce in healthcare, people are off work or having to cover colleagues.

I think we've also reached an interesting anthropological stage in the illness in the UK now – something happened from September 2021 where everybody knows somebody who's got an uncle, their mum, a friend or so on who is still experiencing symptoms, and my inbox reflects that. And when people are closer to it in that way it's in their consciousness.

What are some of the unique challenges of studying long COVID compared to other health conditions?

Before the pandemic, as a cardiologist with little business looking at infectious diseases, I was dealing with heart failure, where there's a defined disease. We know what outcomes to measure in clinical trials. We have a lot of experience of doing trials so you know about rates of recruitment, participation and so on. And you've got lots of pre-clinical work to underpin what the mechanism is for the underlying disease.

Compare that with long COVID, where the disease is still being defined – all we've got so far is that it's 12 weeks or more of symptoms post-recovery from COVID-19, with exclusion of any other underlying cause. We have more knowledge and more ideas about the underlying mechanisms than we did six months ago, but we're definitely not there.

We don't have a biomarker for the disease, and so inevitably, the people we're recruiting to a trial like this will be heterogeneous – there's likely to be syndromes rather than a syndrome, and then defining the outcome of that syndrome or syndromes is why we were looking at several outcomes.

The other thing that's unique is the scale: one million people in the UK is as many as any god-fearing chronic disease. So, there's a lot of patients and clinicians who are scratching their heads, really frustrated, and really want answers. They really want to progress on a timescale that we're not used to doing in science.

The scientific community has developed a vaccine at breakneck speed which is incredible, probably Nobel Prize winning material. We have seen this in various ways, whether it's the review bodies, the ethics panels, the regulators – the pandemic has changed ways of working, generally in a good way. We're working as quick as we can, but in a chronic disease setting there's still there's still roadblocks along the way.

We have a risk in that groups of patients who are vulnerable and have been waiting for 22 months to find treatments – they’re being offered all sorts of treatments that are not scientific and not-evidence based. But we should not lower the bar of science or biology or necessarily roll things out based on preliminary data.

You mentioned the science is still not there yet in terms of a mechanism or the underlying biology of these conditions. But what are the best ideas at the moment about what is actually going on in the bodies of people whose symptoms are persisting after the infection seems to have cleared?

All I can tell you is that there are several drugs that we're looking at – and we are both testing whether the mechanisms are important at the same time as testing the drugs. These are all drugs that are already being used and we hope to repurpose for long COVID, because we're going for things which might be quickly scalable at population level. They are drugs that have either been used by patients or have been used in small numbers in the clinic.

There are the antihistamines loratadine and famotidine. The idea there is that there's a mast cell activation, a histamine release that we want to control. Rivaroxaban is an anticoagulant, as there is substantial evidence that micro-clots and clotting have been important, not just in acute COVID but in long COVID.

Then the other underlying mechanism we're looking at is inflammation. Colchicine is an anti-inflammatory which has can help with inflammation of the lining of the heart and lungs, and again, that's found use in the treatment of COVID patients.

Looking at who might be most susceptible to getting long COVID, presumably the unvaccinated are more susceptible than the vaccinated. Is there any other evidence about who might be most likely to experience prolonged symptoms?

Well, I should re-emphasise that the best way to avoid long COVID is to avoid getting infected in the first place. It's terribly trite and obvious, but that's where we in the UK have tripped up. You can't do the studies that we're doing in New Zealand, or in Japan, because they've kept their infection rate down.

But there is now increasing good quality observational data that suggests there is a dose dependent effect. If you've had your booster you are more protected than if you just had the two vaccinations, and two offers better protection than one, and so on.

Anecdotally, some patients have reported that their long COVID symptoms might have been marginally worse than for a little while after they had their vaccine. But generally there's less risk of long COVID after vaccination – you find lower rates and less severe long COVID.

In terms of other risk factors, severity of initial illness seems to be related to more severe long term symptoms. So age might be a factor, as well as certain underlying conditions such as asthma and diabetes that we know is associated with more severe cases. Also, there seems to be a female preponderance too. But there is lots of data to work through and I’m sure we will find out more.

Tell me a bit about the STIMULATE-ICP project and what you hope to achieve.

The first thing we wanted to do was look at current care, to understand the underlying course or trajectory of people with long COVID – what is the burden or cost of this and what proportion of people are recovering? That will be our baseline.

We realised early on that if we do trials of one aspect of the disease, and then another, we'll get our answers slowly and not in a very holistic way. At this kind of scale, where we're trying to define the disease at the same time as treat it, we need to study everything across the entire care pathway.

That's how we came upon the idea to have an investigation – a multi-organ MRI; an embedded drug trial – the "RECOVERY trial" for long COVID; and a digital rehabilitation platform, all in the same project. The trial is aiming to recruit 4,500 people in less than a year with three and six month follow ups. We're looking at a barrage of outcomes, but we're particularly looking at fatigue, returning to work, various psychometric assessments, physical function measures.

We are also developing a biobank in terms of blood analysis, to look at the underlying mechanisms, whether it's immunology, genomics, proteomics, and so we hope to get some answers both about mechanism and the treatment pathway.

The third element is about inequalities. We wanted to look at current care and make sure that within our trials we are not worsening inequalities. Finally, we wanted to see beyond the pandemic and find out what this pathway for long COVID can tell us about what we could or should be doing for other chronic conditions.

The consortium has brought together over 30 partners, universities, hospitals, policy organisations like the Royal College of General Practitioners, and several industry partners. The patient and public involvement is not a token gesture, particularly in this context – they are absolutely involved, if not leading. I think that's really, really important to make sure that we're answering the right questions.

We haven't got time for me to luxuriate in scientific enquiry – we have to do stuff that's practically useful for patients and the health service quickly. So the only way to do that efficiently is to have them front and centre.

Amitava Banerjee is professor of clinical data science at University College London and Honorary Consultant Cardiologist at University College London Hospitals and Barts Health NHS Trust. He is the chief investigator of the STIMULATE-ICP long COVID study.