Kayte Andersen, MSc, and G. Caleb Alexander, MD, MS, with John Jesitus
According to a recent study, clinical outcomes for chronically immunosuppressed patients hospitalized with COVID-19 mirror those of non-immunosuppressed peers. The analysis included immunosuppressive medications commonly used in dermatology.
When the COVID-19 pandemic began, said first author Kayte Andersen, MSc, it was uncertain whether infected people who were chronically immunosuppressed would fare worse because immunosuppressed people are more likely to contract viral infections in general. Ms. Andersen is a doctoral candidate in the Johns Hopkins Bloomberg School of Public Health in Baltimore, Maryland. Conversely, some observers theorized that immunosuppressed people might experience better outcomes because immunosuppressive medications would prevent their immune systems from going into overdrive, which can complicate COVID-19. “There were arguments in either direction,” said Andersen, “and given how many people need a chronic immunosuppressive medicine to manage an important health condition, we thought it was a good question to follow up.”
Study findings
To that end, Andersen and colleagues retrospectively analyzed medical records of 2121 consecutively treated adults hospitalized with confirmed or suspected COVID-19 who were admitted to one of 5 hospitals contributing to the Johns Hopkins COVID Precision Medicine Analytics Platform Registry (JH-CROWN) registry through August 2020.1
Using a World Health Organization definition, researchers categorized patients as immunosuppressed if they had current prescriptions for immunosuppressive drugs—namely, selective immunosuppressants, antineoplastic agents, or prednisone greater than 7.5 mg daily or equivalent—on the date of COVID-19 hospitalization. Altogether, 108 patients (5% of the sample) met immunosuppression criteria.
After controlling for a host of potential confounders reflecting patients’ baseline health and sociodemographic status, investigators found no significant differences between immunocompromised and immunocompetent patients in terms of survival (88% vs 91%, respectively, P = 0.28), length of hospital stay (6.9 vs 5.1 days, P = 0.09), or need for mechanical ventilation (16% vs 15%, P = 0.75).
“There’s no indication that people who need an immunosuppressive medication for another diagnosed condition should stop the medicine because they’re worried about severe COVID,” Andersen said. “Once these people are hospitalized, do they need to go on to a ventilator, stay in the hospital longer, or are they more likely to die? We found none of those.”
Median time to ventilation was slightly longer for immunocompromised patients (3.0 days vs 2.6 days, P = 0.02). But with a small, immunosuppressed cohort and a betweengroup difference of around 10 hours, she said, investigators did not consider this difference clinically meaningful.
Unadjusted regression analyses and inverse probability of treatment weighting (IPTW) showed no significant differences in use of mechanical ventilation, likelihood of in-hospital death, and length of stay by immune status. Restricting the analysis to patients with at least one health-system encounter before COVID-19 hospitalization (to eliminate possible misclassification of patients without pre-hospitalization data) showed similar results.
“These findings are important because of the magnitude of continuing morbidity and mortality attributable to the pandemic,” investigators write, “as well as the frequent use of immunosuppressive medications for the management of a range of chronic conditions.” When investigators limited their analysis to patients on more than 10 mg prednisone daily, the only difference between the 2 groups was that immunosuppressed patients were discharged sooner (HR 0.72, 95% CI 0.60-0.85).
Additionally, immunosuppressed patients had a significantly shorter length of stay (HR 0.75, 95% CI 0.66-0.86) in analyses that defined immunosuppression in terms of diagnoses and/or medication use. It is unclear whether this difference was clinically significant, Andersen said, because the study was not designed to explore the impact of diagnoses. “Adding diagnoses added a lot more people into our ‘immunosuppression’ definition (n = 232 or 11% of the sample),” she added, “so this may be a reflection of increased sample size.”
Further applications
The study included immunosuppressants frequently used in dermatology, including biologics, methotrexate, and cyclosporine, Andersen said. “We also considered chemotherapy and medicines for solid organ transplantation.” However, she said that because of the study’s limited sample size, investigators were unable to separate out medicines specific to dermatology or any other specialties. “But that’s something we’re doing now, using a national data set with hundreds of thousands of hospitalized COVID patients.” This study will include Johns Hopkins among a total of approximately 40 healthcare systems. “It’s a very diverse, rich sample, with people of different ages, geographic regions, races, and ethnicities.” Investigators were completing data analysis in early April, and she said that the team hopes this study will be published this spring or summer.
Strengths of the present study include its examination of real-world experience in a large cohort of patients hospitalized with COVID-19 in the geographic region served by Johns Hopkins hospitals. Additional strengths include the comprehensiveness of the JHCROWN registry and the electronic medical records stored therein.
Investigators moreover used a variety of methods to maximize causal inference by eliminating patients who were not at risk of the primary outcome—mechanical ventilation. The study protocol excluded patients who were ventilated on admission or had advance directives placed within 24 hours of admission.
Furthermore, investigators used stabilized IPTW with doubly robust adjustment and accounted for the competing risk of death where death was not the primary outcome, Andersen said. “Once we applied state-of-the-art statistical and epidemiological principles, we didn’t find any difference. We believe that we did the best job we possibly could to provide appropriate analysis.”
Study limitations include the fact that investigators characterized a limited set of short-term outcomes. Andersen and colleagues therefore recommend further research into the association between chronic immunosuppression and longer-term COVID-19 morbidity and mortality. Additionally, performing the study during a period of rapidly changing clinical protocols may have impacted results, although authors are unaware of whether such protocols were differentially applied to immunosuppressed patients.
“It’s reassuring that we’re all seeming to find that immunosuppression doesn’t necessarily make people do any worse once they are hospitalized.”
Other groups worldwide have investigated the relationship between immunosuppression and COVID-19 outcomes and found similar results.2-4 “It’s reassuring that we’re all seeming to find that immunosuppression doesn’t necessarily make people do any worse once they are hospitalized,” Andersen said. Senior author G. Caleb Alexander, MD, MS, added that the study was not designed to examine whether chronic immunosuppression facilitates SARS-CoV-2 infection in the first place. “Further work is needed to examine this important scientific and clinical question.” Dr. Alexander is a founding co-director of the Johns Hopkins Center for Drug Safety and Effectiveness and professor in the Department of Epidemiology at Johns Hopkins Bloomberg School of Public Health. The study appeared online January 7 in Clinical Infectious Diseases.
REFERENCES
- Andersen KM, Mehta HB, Palamuttam N, et al. Association between chronic use of immunosuppresive drugs and clinical outcomes from coronavirus disease 2019 (COVID-19) hospitalization: a retrospective cohort study in a large US health system [published online ahead of print, 2021 Jan 7]. Clin Infect Dis. 2021;ciaa1488. doi:10.1093/cid/ciaa1488
- Colmenero J, Rodríguez-Perálvarez M, Salcedo M, et al. Epidemiological pattern, incidence, and outcomes of COVID-19 in liver transplant patients. J Hepatol. 2021;74(1):148-155.
- Sanchez-Piedra C, Diaz-Torne C, Manero J, et al. Clinical features and outcomes of COVID-19 in patients with rheumatic diseases treated with biological and synthetic targeted therapies. Ann Rheum Dis. 2020;79(7):988-990.
- Cavagna L, Bruno R, Zanframundo G, et al. Clinical presentation and evolution of COVID-19 in immunosuppressed patients: preliminary evaluation in a North Italian cohort on calcineurin-inhibitors based therapy. medRxiv. Posted online 2020 May 1 (preprint) https://www.medrxiv.org/content/10.1101/2020.04.26.20080663v1.
DISCLOSURES
Ms. Andersen reports no relevant financial interests.
Dr. Alexander has been a paid advisor to IQVIA and is a consultant and equity holder in Monument Analytics, a consultancy whose clients include the life sciences industry and plaintiffs in opioid litigation. He is also a member of the OptumRx National Pharmacy & Therapeutics Committee.
