To the Editor: The COVID-19 pandemic raised concerns about the management of patients with immune-mediated inflammatory diseases treated with immunosuppressive biologics. A third of patients with psoriasis who discontinued their medications had disease progression.1 As population-level analyses of this patient group remain limited, we compared the incidence of COVID-19 and subsequent mortality in a large cohort of patients prescribed biologics and matched controls.
We identified all patients aged 18 years and older with at least 1 prescription for a biologic from July 1, 2019 to February 29, 2020 in the Massachusetts General Brigham Enterprise Data Warehouse. The primary and secondary outcomes for this study were risk of COVID-19 and subsequent mortality, respectively. A multivariable logistic regression was used on matched data to calculate the odds ratio (OR) for COVID-19 diagnosis between the 2 groups, adjusting for age, sex, race, Charlson Comorbidity Index severity grade, median income, and local infection rates. A multivariable Poisson regression was used to compare all-cause mortality among patients diagnosed with COVID-19, adjusting for age, sex, Charlson Comorbidity Index severity grade, median income, and local infection rates. Detailed methods and sensitivity analyses are included in the Supplemental Materials (available via Mendeley at https://data.mendeley.com/datasets/w4478kftkk/1).
We identified 7361 patients who received biologics and 74,910 matched controls. Patient baseline characteristics are presented in Table I . Tumor necrosis factor inhibitors (adalimumab [28.4%], infliximab [15.6%], and etanercept [11.9%]), CD20-directed antibody (rituximab [15.6%]), and interleukin-4A inhibitor (dupilumab [8.6%]) were the most frequently prescribed biologics. Rheumatoid arthritis (27.5%), psoriasis (27.3%), psoriatic arthritis (16.2%), Crohn's disease (24.9%), and ulcerative colitis (18.9%) were the most common indications for biologics in our study.
Table I.
Demographics and clinical characteristics of patients on biologics and matched controls
| Demographic or clinical variable | Biologic group N = 7361 |
Matched controls N = 74,910 |
P value |
|---|---|---|---|
| Age group (years), N (%) | >.99 | ||
| 18-44 | 2783 (37.8%) | 28,321 (37.8%) | |
| 45-64 | 2838 (38.6%) | 28,881 (38.6%) | |
| 65-74 | 1135 (15.4%) | 11,550 (15.4%) | |
| ≥75 | 605 (8.2%) | 6157 (8.2%) | |
| Sex, female, N (%) | 4124 (56.0%) | 41,968 (56.0%) | >.99 |
| Race and ethnicity, N (%) | >.99 | ||
| White non-Hispanic | 6223 (84.5%) | 63,329 (84.5%) | |
| Asian or Pacific Islander non-Hispanic | 263 (3.6%) | 2676 (3.6%) | |
| Black non-Hispanic | 332 (4.5%) | 3379 (4.5%) | |
| Other non-Hispanic | 139 (1.9%) | 1415 (1.9%) | |
| Hispanic | 223 (3.0%) | 2269 (3.0%) | |
| Unknown | 181 (2.5%) | 1842 (2.5%) | |
| Charlson comorbidity index grade, N (%) | >.99 | ||
| Mild (1-2) | 4050 (55.0%) | 41,215 (55.0%) | |
| Moderate (3-4) | 1591 (21.6%) | 16,191 (21.6%) | |
| Severe (≥5) | 1720 (23.4%) | 17,504 (23.4%) | |
| Medical comorbidity, N (%) | |||
| Hypertension | 2147 (29.2%) | 21,561 (28.8%) | .49 |
| Congestive heart failure | 355 (4.8%) | 4867 (6.5%) | <.001 |
| Diabetes | 818 (11.1%) | 11,234 (15.0%) | <.001 |
| Chronic pulmonary disease | 933 (12.7%) | 10,738 (14.3%) | <.001 |
| Other pulmonary disease | 1529 (20.8%) | 17,546 (23.4%) | <.001 |
| Renal disease | 561 (7.6%) | 5797 (7.7%) | .72 |
| Liver disease | 1156 (15.7%) | 11,821 (15.8%) | .86 |
| Hematologic cancer | 593 (8.1%) | 3043 (4.1%) | <.001 |
| Solid organ cancer, not metastatic | 1270 (17.3%) | 15,949 (21.3%) | <.001 |
| Solid organ cancer, metastatic | 145 (2.0%) | 3592 (4.8%) | <.001 |
| Indication | |||
| Asthma | 1428 (19.4%) | 13,162 (17.6%) | <.001 |
| Atopic dermatitis | 2022 (27.5%) | 15,112 (20.2%) | <.001 |
| Chronic lymphocytic leukemia | 65 (0.9%) | 263 (0.4%) | <.001 |
| Non-Hodgkin lymphoma | 483 (6.7%) | 1421 (1.9%) | <.001 |
| Giant cell arteritis | 186 (2.5%) | 159 (0.2%) | <.001 |
| Granulomatosis with polyangiitis | 104 (1.4%) | 105 (0.1%) | <.001 |
| Microscopic polyangiitis | 20 (0.3%) | 6 (0.01%) | <.001 |
| Systemic lupus erythematosus | 233 (3.2%) | 521 (0.7%) | <.001 |
| Pemphigus | 32 (0.4%) | 59 (0.1%) | <.001 |
| Hidradenitis suppurativa | 166 (2.3%) | 387 (0.5%) | <.001 |
| Psoriasis | 2012 (27.3%) | 3054 (4.1%) | <.001 |
| Psoriatic arthritis | 1192 (16.2%) | 369 (0.5%) | <.001 |
| Rheumatoid arthritis | 2027 (27.5%) | 1930 (2.6%) | <.001 |
| Ankylosing spondylitis | 452 (6.1%) | 183 (0.2%) | <.001 |
| Uveitis | 211 (2.9%) | 594 (0.8%) | <.001 |
| Crohn's disease | 1829 (24.9%) | 515 (0.7%) | <.001 |
| Ulcerative colitis | 1388 (18.9%) | 1094 (1.5%) | <.001 |
| COVID-19 positive, N (%) | 87 (1.2%) | 1063 (1.4%) | .10 |
| Died, N (% of COVID-19–positive patients) | 7 (8.0%) | 71 (6.7%) | .79 |
| COVID-19 town or county positivity rate per 100 mean (SD) | 1.4 (0.9) | 1.6 (1.1) | <.001 |
| N = 7317 | N = 74,389 | ||
| Median income in $1000s mean (SD) | 82.0 (29.2) | 79.7 (29.2) | <.001 |
P values <0.05 appear in bold.
Overall, biologics were not associated with COVID-19 (OR, 0.88; 95% confidence interval [CI], 0.71-1.09; P = .25), adjusting for demographics, comorbidity burden, and local infection rates (Table II ). Patients treated with tumor necrosis factor inhibitors were less likely to be diagnosed with SARS-CoV-2 infection compared to matched controls (OR, 0.69; 95% CI, 0.48-0.98; P = .04). Similarly, those treated with dupilumab had lower odds of diagnosis (OR, 0.38; 95% CI, 0.12-1.18), although this difference was not statistically significant (P = .10). Mortality rates were also similar between the 2 groups after adjusting for demographics, comorbidity burden, and local infection rates (OR, 1.13; 95% CI, 0.57-2.76; P = .57).
Table II.
Multivariable logistic regression of the risk of COVID-19 infection and subsequent mortality for patients treated with immunosuppressive biologics
| Variable | OR | 95% CI | P value |
|---|---|---|---|
| Risk of infection for all immunosuppressive biologics | |||
| Biologic use | 0.88 | 0.71-1.09 | .25 |
| Age group (years) | |||
| 18-44 | ref∗ | ref∗ | ref∗ |
| 45-64 | 0.92 | 0.79-1.07 | .28 |
| 65-74 | 0.67 | 0.53-0.85 | .001 |
| ≥75 | 1.22 | 0.96-1.56 | .11 |
| Sex, female | 0.95 | 0.85-1.07 | .43 |
| Race and ethnicity | |||
| White non-Hispanic | ref∗ | ref∗ | ref∗ |
| Asian or Pacific Islander non-Hispanic | 0.36 | 0.21-0.62 | <.001 |
| Black non-Hispanic | 2.10 | 1.73-2.56 | <.001 |
| Other non-Hispanic | 1.36 | 1.04-1.79 | .02 |
| Hispanic | 1.39 | 0.99-1.94 | .06 |
| Unknown | 0.28 | 0.13-0.58 | .001 |
| CCI grade | |||
| Mild (1-2) | ref∗ | ref∗ | ref∗ |
| Moderate (3-4) | 1.32 | 1.11-1.56 | <.01 |
| Severe (≥5) | 1.88 | 1.56-2.26 | <.001 |
| COVID-19 town or county positivity rate | 1.24 | 1.19-1.29 | <.001 |
| Median income in $1,000s | 0.98 | 0.95-1.00 | .06 |
| Risk of infection by immunosuppressive biologic class | |||
| Class | |||
| B-cell activating factor inhibitor | 0 | 0.00-Inf | .98 |
| CD20-directed cytolytic antibody | 1.16 | 0.73-1.83 | .53 |
| Integrin receptor antagonist | 1.27 | 0.61-2.68 | .52 |
| Interleukin-1 receptor antagonist | 2.23 | 0.31-15.82 | .42 |
| Interleukin-4A receptor antagonist | 0.38 | 0.12-1.18 | .10 |
| Interleukin-6 receptor antagonist | 1.35 | 0.60-3.02 | .47 |
| Interleukin-12/23 receptor antagonist | 0.88 | 0.33-2.34 | .79 |
| Interleukin-17A receptor antagonist | 1.75 | 0.83-3.69 | .14 |
| Interleukin-23 antagonist | 1.60 | 0.23-11.40 | .64 |
| Selective T-cell costimulation modulator | 1.63 | 0.61-4.36 | .33 |
| Tumor necrosis factor inhibitor | 0.69 | 0.48-0.98 | .04 |
| Risk of subsequent all-cause mortality for all immunosuppressive biologics | |||
| Biologic use | 1.13 | 0.57-2.76 | .57 |
| Age | 1.06 | 1.04-1.09 | <.001 |
| Sex, female | 0.53 | 0.34-0.83 | <.01 |
| CCI grade | |||
| Mild (1-2) | ref∗ | ref∗ | ref∗ |
| Moderate (3-4) | 2.12 | 0.69-6.51 | .19 |
| Severe (≥5) | 2.96 | 0.99-8.86 | .05 |
| Median income in $1000s | 0.90 | 0.80-1.00 | .06 |
| COVID-19 town or county positivity rate | 0.93 | 0.78-1.11 | .45 |
CCI, Charlson Comorbidity Index; OR, odds ratio; ref, reference.
Reference variable. P values <0.05 appear in bold.
Despite the ongoing vaccination efforts, COVID-19 remains a top health concern. The major finding of our study is that biologics did not increase the risk of a positive COVID-19 diagnosis, which is consistent with published literature.2, 3, 4 Additionally, distinct biologics classes are known to cause varying susceptibilities to other viral infections. In our study, tumor necrosis factor inhibitors were associated with lower odds of COVID-19 diagnosis, consistent with reports of this class of biologics being associated with less-severe disease among large cohorts of patients.2 , 4 Furthermore, we did not identify an association between biologics and mortality.
Our results must be considered in light of the real-world data it is based upon, because these patients may have altered their behavior to decrease their risk of infection, as has been reported in surveys of patients with inflammatory bowel disease and rheumatic diseases.5 Dermatologists and patients should prioritize the well-established risk factors for COVID-19 when making decisions to continue therapy.
Conflicts of interest
None disclosed.
Acknowledgments
The authors thank Stacey Duey and Celina Li of the Research Patient Data Registry for their help with access to patient chart data and Bernard Rosner of Harvard Medical School for his valuable guidance in the study design and analysis.
Footnotes
Drs Pahalyants and Murphy are cofirst authors.
IRB approval status: Approved by the institutional review boards of Mass General Brigham (Protocol 2020P001191) and Massachusetts Department of Public Health (Protocol 1606024-2).
Funding sources: None.
References
- 1.Burlando M., Carmisciano L., Cozzani E., Parodi A. A survey of psoriasis patients on biologics during COVID-19: a single centre experience. J Dermatolog Treat. May 25, 2020 doi: 10.1080/09546634.2020.1770165. [DOI] [PubMed] [Google Scholar]
- 2.Mahil S.K., Dand N., Mason K.J., et al. Factors associated with adverse COVID-19 outcomes in patients with psoriasis–insights from a global registry-based study. J Allergy Clin Immunol. 2021;147(1):60–71. doi: 10.1016/j.jaci.2020.10.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Strangfeld A., Schäfer M., Gianfrancesco M.A., et al. Factors associated with COVID-19-related death in people with rheumatic diseases: results from the COVID-19 Global Rheumatology Alliance physician-reported registry. Ann Rheum Dis. 2021;80(7):930–942. doi: 10.1136/annrheumdis-2020-219498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gianfrancesco M., Yazdany J., Robinson P.C. Epidemiology and outcomes of novel coronavirus 2019 in patients with immune-mediated inflammatory diseases. Curr Opin Rheumatol. 2020;32(5):434–440. doi: 10.1097/BOR.0000000000000725. [DOI] [PubMed] [Google Scholar]
- 5.Hooijberg F., Boekel L., Vogelzang E.H., et al. Patients with rheumatic diseases adhere to COVID-19 isolation measures more strictly than the general population. Lancet Rheumatol. 2020;2(10):e583–e585. doi: 10.1016/S2665-9913(20)30286-1. [DOI] [PMC free article] [PubMed] [Google Scholar]