Advances in Rheumatology

As the number of patients with COVID-19 began to overwhelm New York City hospitals, rheumatologists Elana J. Bernstein, MD, MSc, and Iris Navarro-Millán, MD, MSPH, at NewYork-Presbyterian/Columbia University Irving Medical Center and Hospital for Special Surgery, respectively, started noticing a very familiar pattern of symptoms, namely, those that represent cytokine storm syndrome – a severe immune reaction in which the body releases too many cytokines into the blood too quickly. Along with their rheumatology colleagues, Dr. Bernstein and Dr. Navarro-Millán, who is also a member of the Division of General Internal Medicine at NewYork-Presbyterian/Weill Cornell Medical Center, were taken aback by the similarities of patients with COVID-19 to their patients with autoimmune disease.

Dr. Elana J. Bernstein

Dr. Iris Navarro-Millán

“While I was an attending on the medicine service, several of our patients admitted for COVID-19 had very high inflammatory markers – elevated C-reactive protein levels, erythrocyte sedimentation rates, and ferritin levels,” notes Dr. Bernstein. “And a high percentage of them went on to develop acute respiratory distress syndrome [ARDS] requiring intubation and mechanical ventilation. Many developed renal failure requiring hemodialysis and hypotension requiring vasopressor support. The whole picture was just screaming cytokine storm syndrome.”

“When COVID-19 arrived, I didn’t think that there could be a role for rheumatologists,” says Dr. Navarro-Millán. “But once I redeployed to the medicine service, it turned out we were actually among the specialists at the front lines of this crisis. As I gathered more data about the phenotype of these patients, one particular pattern that I identified was the presentation of symptoms characteristic of macrophage activation syndrome [MAS] that can present in patients with autoimmune diseases. MAS can cause a significant amount of fever and profound pancytopenia and is associated with a high level of mortality. However, while I found some resemblance to the phenotypes that we see in rheumatology such as a significant amount of fever, elevated levels of ferritin and C-reactive protein, and left ventricular hypertrophy, I also saw some differences. Patients with COVID did not develop the profound pancytopenia that is generally part of the spectrum of MAS. While COVID-19 symptoms did not necessarily fit the perfect picture of MAS, it had many of the same specs of cytokine storm.”

“IL-1 is a potent pro-inflammatory cytokine, so IL-1 blockade has broad anti-inflammatory effects. It therefore seemed reasonable to trial anakinra for the treatment of cytokine storm in COVID-19.” — Dr. Elana J. Bernstein

As rheumatologists, Dr. Bernstein and Dr. Navarro-Millán are well-acquainted with the prototypical cytokine storm and MAS, which occurs most commonly in children with systemic juvenile idiopathic arthritis.

“One of the primary therapies that we use for MAS is anakinra, an IL-1 inhibitor,” says Dr. Bernstein. “IL-1 is a potent pro-inflammatory cytokine, so IL-1 blockade has broad anti-inflammatory effects. It therefore seemed reasonable to trial anakinra for the treatment of cytokine storm in COVID-19. And, there was an urgent need to identify effective therapies for the treatment of COVID-19.”

Illustration of a cytokine storm response to infection with the new coronavirus SARS-CoV-2

Dr. Bernstein notes that the majority of COVID patients were getting anakinra when they were very sick…already intubated, septic, and in renal failure. “They may have been getting it too late in the game for it to have much of an effect.” This led Dr. Bernstein and her colleagues to consider administering anakinra to COVID patients at the earliest sign of cytokine storm as they do for children with MAS.

Dr. Navarro-Millán agrees. “We decided on the basis of our observations to initiate treatment with anakinra before patients begin to decompensate and require a ventilator or admission to the ICU. Many of these patients required four to five days of these medications before they were able to decrease the amount of supplemental oxygen. The window of opportunity to administer anakinra was no more than 36 hours after the onset of severe respiratory failure (requiring a high level of supplemental oxygen to keep oxygen saturation above 90 percent). In our experience, patients required several days to improve their respiratory status once they received anakinra.”

“We decided on the basis of our observations to initiate treatment with anakinra before patients begin to decompensate and require a ventilator or admission to the ICU.” — Dr. Iris Navarro-Millán

Both Dr. Bernstein and Dr. Navarro-Millán stress that rigorous precautions were taken in selecting patients who might benefit from anakinra – particularly in the absence of controlled clinical trials. “We were very careful about weighing the risks versus the benefits, and set up very detailed inclusion/exclusion criteria,” says Dr. Navarro-Millán. “For example, patients on another biologic or on chemotherapy were excluded as the risk for other complications was too high. Fortunately, anakinra is a very short-acting medication so the immune system rebounds much more quickly than it does with other immunosuppressive drugs. This enables us to more readily customize the amount of the drug in the body.”

For patients on anakinra who did go on to respiratory failure, the physicians questioned whether keeping them on anakinra could help with extubation. “Patients on ventilators have an underlying increased risk of developing other infections, especially bacterial infections, so suppressing the immune system is not the best approach,” says Dr. Navarro-Millán. “The lesson learned was if we were able to start anakinra early on and the patient still progressed to mechanical ventilation, then we should stop anakinra at that time in order to prevent further complications.”

The Case for Clinical Trials

“While we were able to gather data during this time and have anecdotal reports of drugs that might work for patients with COVID-19, we knew we needed the type of tangible evidence for their safety and efficacy that could only be achieved through controlled trials,” says Dr. Bernstein.

With this in mind, Dr. Bernstein and her colleagues at Columbia proposed a clinical trial to determine whether hospitalized patients with early cytokine storm due to COVID-19 would have better clinical outcomes when treated with an IL-1 antagonist paired with standard of care than when treated with standard of care alone. They hypothesized that a greater proportion of patients treated with anakinra within 72 hours of admission plus standard of care would be discharged or achieve a two-point improvement in the seven-point ordinal scale at day 14. This ordinal scale is a modification of the FLU-IVIG ordinal scale and is being used as an outcome measure in clinical trials of other cytokine inhibitors in COVID-19.

Courtesy of Dr. Elana Bernstein

One particular challenge during the early weeks of the pandemic was that the standard of care was a moving target. For example, notes Dr. Bernstein, “If this trial had been conducted back in March, standard of care would have included hydroxychloroquine and azithromycin but not corticosteroids. But now at Columbia, corticosteroids have become very much standard of care, so if the trial were to start now, standard of care would likely include corticosteroids.”

Ideal candidates for the proposed trial will be SARS-CoV-2 PCR-positive adults who have been hospitalized less than 72 hours and present with hypoxia requiring supplemental oxygen, have chest X-ray abnormalities consistent with COVID-19, and who have elevated inflammatory markers. “For the purposes of this study, we have defined an elevation in inflammatory markers as a high-sensitivity CRP greater than or equal to 150 milligrams per liter,” explains Dr. Bernstein. “Patient exclusion criteria includes enrollment in another interventional trial for COVID-19; use of a cytokine inhibitor or JAK inhibitor within one month; use of an anti-CD20 monoclonal antibody within six months; or requiring mechanical ventilation. To help guide the science and provide preliminary data for our grant application, we identified 2,310 patients with COVID hospitalized at Columbia with sufficient follow-up data as of April 16, 2020.”

A small case study conducted by Dr. Navarro-Millán and her colleagues showed that anakinra appeared to reduce the effects of systemic inflammation in people with COVID-19 who are in severe respiratory distress and may allow certain people with COVID-19 to avoid having to go on a ventilator. In the June 30, 2020, edition of online-first in Arthritis & Rheumatology, the investigators report that of the 11 patients who received anakinra after exhibiting signs of being close to requiring intubation and mechanical ventilation, the seven patients who received anakinra within 36 hours of increasing oxygen requirements were able to avoid mechanical ventilation and were later discharged home. The four patients who did not receive anakinra until four or more days after the onset of respiratory failure were placed on ventilators. Three of these patients were eventually extubated and one patient died.

“This is a small study, but we think it could provide guidance to other doctors to help them identify which patients might benefit from anakinra treatment,” adds Peggy Crow, MD, who worked on the study with Dr. Navarro-Millán. “For those showing evidence of cytokine storm syndrome and increasing oxygen requirements, the timing of the intervention is essential in avoiding the need for intubation. We thought it was important to publish our experiences and get this information out and available.”

According to the authors, the data suggest that anakinra could be beneficial in treating COVID-19 patients with evidence of cytokine storm syndrome when initiated early after onset of acute hypoxemic respiratory failure. Based on their preliminary data, Dr. Navarro-Millán is now developing a controlled study of anakinra in patients with COVID-19 who fit the criteria she developed.

Better Together

“At Columbia, several working groups have since been established for developing and leading COVID-related trials,” says Dr. Bernstein. “These involve physicians and scientists from rheumatology, infectious diseases, oncology, pulmonology and critical care, and general medicine. We attend weekly meetings to review all the clinical trials for COVID that have been proposed or are underway by Columbia investigators as well as industry sponsors. With multiple clinical trials in the same space, this structure helps to ensure that one trial does not compete with another trial for enrollment.”

“I think the most important take-home from our experiences is the need for good communications across disciplines,” adds Dr. Navarro-Millán. “There is a great opportunity for us as physicians to become more open and receptive to the insight and perspectives of every single member of the medical community. We can do better together.”