Covid-19 long-hauler treatment trials will likely rely on established endpoints and trial design features used in other conditions, and will be based on a specific group of symptoms to prove clinical utility, experts said.
Covid-19 long-haulers, or long Covid, are terms used to encompass patients experiencing long-term symptoms in the months after a SARS-CoV-2 infection, including fatigue, as well as respiratory and cognitive complications. A growing number of such cases has prompted research efforts, mostly from academia, including large observational studies and some therapeutic studies.
Rather than starting from the beginning, experts expect the field to exploit endpoints and trial design features from other conditions with similar symptoms, like chronic fatigue syndrome (CFS) and myalgic encephalomyelitis (ME). Endpoints like the six-minute walk test (6MWT) for respiratory symptoms and executive function scores for neurological symptoms are also rational choices, experts added. One recently initiated long-hauler trial, PureTech Health’s Phase II study of LYT-100 (deupirfenidone), uses the 6MWT as its primary endpoint. The trial has a planned primary completion date in August, according to ClinicalTrials.gov.
Still, the lack of any standardised definition to classify patients and the wide variability of symptoms make targeting the ideal patient population for Covid-19 long-hauler treatments particularly challenging, experts said. Additionally, limited data on the potential correlation between acute symptom severity and long-term presentation of symptoms following infection further complicate the trial design process, they added.
Patients may need to be stratified based on their prior hospitalisation history and comorbidities to ensure an efficacy signal is not lost, some experts said. However, neurological symptoms in hospitalised and nonhospitalised patients appear to be anecdotally similar and likely have similar underlying mechanisms, two intensive care unit (ICU) clinicians added.
The FDA recognises patients who have recovered from Covid-19 infection may have residual clinical symptoms, and will continue to work closely with sponsors and researchers to design appropriate development programmes to address Covid-19, an agency spokesperson said. The NIH has also launched a 900-patient longitudinal trial to study Covid-19 sequelae and immunity.
Diverse symptoms prompt various design considerations
There is no consensus on how to best define Covid-19 long-haulers, said Dr Peter Rowe, director, Children’s Centre Chronic Fatigue Clinic, Johns Hopkins University, Baltimore, Maryland. Given the range of symptoms, the endpoints for such trials need to be tailored to the outcome a therapy can change, he said. However, the commonalities between ME/CFS and long-term Covid-19 symptoms help in deciding outcomes for long-haulers trials without having to start over from the beginning, he added. Still, while repurposing existing measures is efficient, they need to be validated in the Covid-19 context, said former FDA employee Laurie Burke, founder of the pharma consultancy firm LORA Group, Royal Oak, Maryland. For example, patient-reported outcome (PRO) questionnaires in CFS may ask about the patient’s ability to perform specific activities and how fatigue impacts them, Burke said. However, whether the specific scale is appropriate for the level of fatigue seen in Covid-19 is unknown, she said.
One important takeaway from ME/CFS trials is the way studies need to have a large sample size, which allows evidence of modest but clinically important improvements to be collected, Rowe said. Rowe referred to fibromyalgia trials as a potential benchmark for appropriately powered studies in terms of appropriate trial size. AbbVie’s (NYSE:ABBV) Savella (milnacipran HCL) received an FDA approval in 2009 based on a Phase III 2,084-patient study following a 125-patient Phase II trial.
The deupirfenidone study is enrolling 168 patients with postacute Covid-19 respiratory complications who were treated with, but no longer require, mechanical ventilation, extracorporeal membrane oxygenation, noninvasive ventilation (eg CPAP or BiPAP), high-flow nasal oxygen therapy, or a combination thereof. The trial was powered to capture change in 6MWT based on emerging data from Covid-19 patients and historical data from severe acute respiratory syndrome, said a PureTech spokesperson, adding the study is powered while keeping in mind the inherent variability in the 6MWT.
Deupirfenidone is intended to address the disease’s long-term respiratory complications and related sequelae. Typically, lung function measured by forced vital capacity is used to assess efficacy in respiratory disorders like idiopathic pulmonary fibrosis, but is known to already be reduced in hospitalised Covid-19 patients and is expected to naturally recover with time, said deupirfenidone investigator Dr Toby Maher, director, Interstitial Lung Disease programme, Keck School of Medicine of USC, Los Angeles. Moreover, spirometry measurements have been difficult during the pandemic due to the inherent risk of spreading viral particles. Muscle strength, cardiovascular disease, respiratory disease and general fitness influence 6MWT results, which may improve naturally in patients after hospital discharge, Maher said. However, an investigational drug will likely have a large enough impact on the lungs to see a greater improvement in Covid-19 long-haulers, he added. A difference of 27 meters in the 6MWT compared to placebo is considered significant enough in fibrotic lung disease to judge a treatment’s impact, Maher said.
Nonetheless, there is an element of uncertainty in designing studies on Covid-19 long-term symptoms since the impact on 6MWT without treatment is unknown in this situation, Maher said. The 6MWT has been used as a primary endpoint in other pulmonary trials and is a direct measure of patient functioning, said the PureTech spokesperson, adding the trial will also assess dyspnea as a secondary endpoint in addition to other respiratory-related PROs.
Similarly, to evaluate long-term Covid-19 neurological symptoms, repurposing existing scales for cognitive impairment is the most efficient research strategy, said James Jackson, PsyD, assistant director, ICU Recovery Centre, Vanderbilt University, Nashville, Tennessee. For patients experiencing long-term neurological symptoms after Covid-19, a sharp decline in executive functioning and an inability to return to work are among the most common and harmful symptoms, Jackson added.
Neurological symptoms in Covid-19 long-haulers include loss of memory, concentration difficulty, anxiety and depression, among others. Because these are difficult to measure biologically, patient-assessed scales of cognitive function are best equipped to determine symptom improvement, he explained. The Minnesota Multiphasic Personality Inventory scale for depression, the Hopkins Symptom Checklist and the Chalder Fatigue Scale are among the symptom scales used for diagnosing and measuring improvements in these syndromes.
Scales used for diagnosing mild cognitive mild impairment, such as the Montreal Cognitive Assessment, could also be used, added Dr Jin Ho Han, associate professor, Department of Emergency Medicine, Vanderbilt University, Nashville, Tennessee. However, concerns remain about these scales lacking the sensitivity to detect subtler changes seen in Covid-19 long-haulers, he explained.
Additionally, reducing acute disease severity and inflammation could have a positive effect on addressing long-term neurological symptoms, Han said. Anecdotally, the higher severity of more severe acute Covid-19 symptoms, particularly a longer time in the ICU, is correlated with worse, more persistent long-term cognitive impairment, he added. As such, any therapy improving disease outcomes could theoretically reduce long-term symptom severity, Han explained.
General challenges on study populations
In choosing study populations, experts said there is an important distinction between patients who have been previously hospitalised and those who have not. Even outside of Covid-19, patients on mechanical ventilation for an extended period tend to have long-term symptoms, experts noted. Intubated patients may develop prolonged tachycardia and fatigue linked to their horizontal positioning, which is different from the same long-term symptoms seen in nonhospitalised Covid-19 patients, Rowe said. However, ICU delirium in both hospitalised and nonhospitalised Covid-19 long-haulers likely involve similar underlying mechanisms, Han said.
If the focus is only on hospitalised patients, the results with any intervention may not hold true for the much larger population with chronic symptoms, Rowe said. In a survey of 3,700 respondents who reported long-term symptoms beyond 90 days, only 8.4% were hospitalised for Covid-19 (Davis; et al; 27 December 2020; medRxiv preprint).
Another challenge with identifying the right trial participants is the way differences in standard treatments for other comorbidities can interfere with the efficacy signal of the investigational treatment, Rowe said, referring to a similar experience with fludocortisone in CFS patients. Among Covid-19 patients, those with comorbid conditions like diabetes and hypertension are at a higher risk for severe disease. Enrichment strategies allowing for the stabilisation of comorbid conditions before randomising patients, or incorporating crossover trial designs where each patient can act as their own control, both have the potential to detect effective therapies, Rowe said.
The field is still trying to understand the prevalence of these long-term symptoms and their relation to the initial infection severity, Maher said, adding the unknowns with long-term Covid-19 symptoms makes trial design challenging. Overall, the lack of observational data, including its incidence and possible risk factors, makes it particularly challenging to study the effect of potential therapies, Han and Jackson agreed.
Manasi Vaidya is a Senior Reporter and William Newton is a Reporter for Clinical Trials Arena parent company GlobalData’s investigative journalism team. A version of this article originally appeared on the Insights module of GlobalData’s Pharmaceutical Intelligence Center. To access more articles like this, visit GlobalData.
