Sarilumab for Patients With Moderate COVID-19 Disease

1. Aims/Objectives

The effectiveness of blockade of IL-6R in treating patients with COVID-19 disease of
moderate severity will be tested in a pragmatic and adaptive randomized trial. The
primary outcome is intubation or death within 14 days of enrollment OR administration of
a rescue dose of IL-6 inhibition if the patient is do-not-resuscitate
(DNR)/do-not-intubate (DNI) and the primary team determines it is appropriate based on
clinical deterioration. Secondary outcomes include time to hospital discharge if alive,
time to clinical recovery as was defined in a recent study, ICU length of stay, time to
return to normal or baseline oxygen saturation, and changes in laboratory biomarkers.

2. Background Information

SARS-COV-2 is a novel human pathogen that emerged at the end of 2019 and rapidly spread
worldwide. Clinical disease caused by this novel infection has a high mortality rate,
particularly in older adults and in patients with underlying cardiopulmonary disease,
populations that are highly prevalent within the VA healthcare system. This creates an
urgent need for the VA to quickly and efficiently identify effective therapies to reduce
mortality and spread of the disease. Based on in vitro and in vivo data, potential
treatment options include anti-viral drugs (such as antimalarials or medications
originally developed to treat HIV infection (e.g., lopinavir/ritonavir, remdesivir)),
anti-inflammatory drugs, including IL-6 inhibition, and innovations in supportive care,
such as placing patients in prone positioning while awake and not intubated.
Importantly, in the context of the global pandemic and exponential spread of the
disease, clinical trials must be designed in a manner that 1) optimizes outcomes for
Veterans, and 2) produces rapid answers, so that effective treatments can be deployed
quickly to reduce unnecessary deaths. Based on a review of 384 ongoing clinical trials
by the Centre for Evidence-Based Medicine, and a daily search of open clinical trials,
many conventional controlled clinical trials of antiviral drugs and cytokine-blocking
drugs are in progress and are likely to be completed within a few months; however, it is
possible that these trials, which may take months to implement and report results, will
not yield any information about how to treat patients until after the peak of the
pandemic is over. A secondary challenge is that it is important to identify what works
and what does not work quickly, so that manufacturers can understand what medications
need to be produced so that supply chains are adequate for treating all patients who
might benefit from an intervention. Thus, innovative clinical trial methodologies are
needed in order to produce answers in a highly compressed time frame to inform medical
decision-making and supply chain management.

This novel and innovative proposal concerns use and comparison an existing drug
(sarilumab) that blocks the receptor for the inflammatory cytokine IL-6. Most studies of
IL-6 pathway blockade are using IV medications and include restrictive eligibility
criteria and are directed towards patients with severe disease. The investigators
hypothesize that patients earlier in the disease course - prior to severe respiratory
decompensation - might benefit the most from IL-6 inhibition, and thus are the
population included in this pragmatic trial. This study will expand our knowledge and
understanding about how IL-6 inhibition can be used to prevent disease progression in
patients with moderate disease. It will use a more readily available subcutaneous,
longer-acting formulation directed towards patients with moderate disease, of whom
25-35% can progress to severe disease, often rapidly. Eligibility criteria allow for a
broad range of underlying choices of other medications, including antiviral medications
such as remdesivir and hydroxychloroquine, among others, baseline laboratory values, and
comorbidities, reflecting likely real-world clinical use. Thus, this pragmatic project
would fill a major gap in research about the clinical care of patients with COVID-19.

Release of preliminary results to the popular press already necessitated revision of
this protocol after only 9 subjects had been enrolled. Interim analysis of a study of
sarilumab by its manufacturer has led to revision of the study population and treatment
arms. Originally, two doses (200 mg and 400 mg) were tested, and patients with either
severe (oxygen requirement and other poor prognostic features) or critical (mechanical
ventilation) disease were included. Interim analysis indicated benefit only with 400 mg
in critical disease, so that will be the design of the remainder of that trial. In
contrast, a study of of a relatively high dose of tocilizumab (8 mg/kg) was reported to
be beneficial in a controlled trial of patients varying in severity from critical
illness to merely requiring supplementary oxygen. In response, the current study has
been modified to allow routine escalation of dosing among a broad range of "moderate"
severity short of critical illness. For patients who are critically ill, it is expected
that off-label use of a wide range of "heroic" therapies will be tried, including but
not limited to cytokine blocking therapies, so exclusion of such patients from this
trial does not limit the therapeutic options for such patients, and from a research
perspective, such patients have become the focus of the largest ongoing study of
sarilumab.

3. Rationale and Purpose

Severe COVID-19 disease is characterized by a severe acute respiratory distress syndrome
(ARDS) followed by a severe cardiomyopathy in many cases. Based on currently available
data, it is unclear how much of the respiratory damage in COVID-19 is due to direct
effects of SARS-COV-2, or if the primary driver of severe disease is the inflammatory
response generated in response to the virus. The key role of inflammation in the
progression of the disease is highlighted by recent studies, which demonstrate that
patients who progress to requiring ICU-level care no longer have detectable virus
present in their respiratory tract. Thus, identifying drugs that could dampen the
inflammatory response prior to severe clinical decompensation without adversely
affecting clearance of virus would be clinically useful and lead to improved outcomes.
Elevated biomarkers of severe inflammation (such as IL-6, CRP, ferritin, and
lymphopenia) have been strongly associated with severe disease and increased mortality
in patients with COVID-19, and early development of a strong IgG response may
paradoxically be a poor prognostic sign during the acute phase of the disease. In
contrast, viral load in upper respiratory secretions at baseline is only modestly
associated with clinical outcomes, and viral load declines in most patients after
presentation, independent of the clinical course of the disease. Severe disease
typically arises more than 2 weeks after symptom onset, a time at which virus is usually
no longer detectable. These findings suggest that progression to severe disease may be
driven more by the inflammatory response to the disease than a direct viral effect, and
point to anti-inflammatory drugs as potentially critical therapeutic options.
Importantly, high dose glucocorticoids have been found to be ineffective in severe
COVID-19 and are associated with somewhat worse outcomes, so other more targeted
anti-inflammatory options are urgently needed.

Blockade of IL-6 signaling is a particularly attractive approach. Sarilumab and
tocilizumab are antibodies to the IL-6 receptor (IL-6R), FDA-approved for long-term
treatment of rheumatoid arthritis. Tocilizumab has been used open-label with enough
anecdotal success in enough patients in China that the manufacturers of sarilumab and
tocilizumab have rapidly opened clinical trials in multiple countries at assess efficacy
in patients with severe COVID-19 disease. Tocilizumab has also demonstrated efficacy in
reducing mortality associated with the cytokine release syndrome in patients receiving
CAR-T treatment for cancer, in which the pathogenic pathway is widespread T-cell
activation, which leads to production of pleiotropic inflammatory cytokines such as IL-6
by monocytes. Although trials attempting to rescue patients with severe COVID-19 with
ARDS and/or cardiomyopathy are essential and are in progress, the ideal population for
clinical use may be patients who are not yet critically ill but are at high risk of
clinical deterioration due to secondary inflammation.

4. Relevance to Veterans Health

COVID-19 has a high mortality rate of 2-3% among symptomatic patients. It is expected to
have a particularly high mortality in Veterans, because age >60, underlying
cardiovascular or lung disease, and obesity are significant risk factors for mortality
and clinical deterioration. Thus, advances in care of patients sick enough to be
hospitalized yet prior to intubation are expected to have particular benefit for
Veterans.

5. Study Design

Design. Prospective, randomized, unblinded interventional clinical trial.

Two arms: standard care, based on established practices within the medical center, or
standard care plus subcutaneous sarilumab.

Assignment Strategy: Randomized play-the-winner design, such that randomization becomes
weighted toward the arm that was more effective in previous enrolled subjects. The
probability of randomization to a specific arm (standard of care or standard of care plus
intervention) will be updated based on outcomes in blocks of 10-20 subjects (see Planned
Statistical Analyses). The protocol does not include administration of sarilumab or
tocilizumab as a "rescue" medication if a patient randomized to standard of care alone
deteriorates to the point that the primary outcome (intubation) is met, or if the patient is
DNR/DNI and the patient's primary clinical team determines intubation would be performed if
the patient's goals of care included intubation. Such patients may receive any treatment per
the judgment of their treating physician, and such treatment might include sarilumab or
tocilizumab. Use of sarilumab or tocilizumab in this setting will simply be regarded as a
treatment failure and not as a protocol deviation.

Data will be extracted remotely from the EHR. No data will be generated specifically for
study purposes.