key: cord-0978985-d84ckia2
authors: Tanimoto, Takuya; Tada, Shinpei; Fujita, Suguru; Hirakawa, Tetsu; Matsumura, Mirai; Isoyama, Shoko; Ueno, Sayaka; Hamai, Kosuke; Tsuji, Naoki; Hirosawa, Hideyasu; Taniguchi, Tomohiro; Okamoto, Takeshi; Omoto, Takuji; Kusunoki, Shinji; Maeda, Hiroyuki; Ishikawa, Nobuhisa
title: Effect of baricitinib in patients with coronavirus disease 2019 and respiratory failure: a propensity score–matched retrospective cohort study
date: 2022-03-07
journal: Respir Investig
DOI: 10.1016/j.resinv.2022.02.004
sha: 3954eac7d1e8da4ccbf65340bcbf61fe18f3fad7
doc_id: 978985
cord_uid: d84ckia2

In this retrospective cohort study, we evaluated the efficacy of baricitinib in the treatment of coronavirus disease 2019 (COVID-19). Among 404 adult patients with COVID-19 who were admitted to our hospital between October 23, 2020, and July 31, 2021, 229 patients with respiratory failure were included. Among these, 41 patients in the baricitinib group and 41 patients in the control group were selected by propensity score matching to adjust for background factors. We compared the survival rates of the two groups at 30 and 60 days after admission. The 30-day survival rate was significantly higher in the baricitinib group than in the control group. However, there was no significant difference in 60-day survival in the two groups. Baricitinib may improve the early prognosis of patients with respiratory failure associated with COVID-19. However, efforts should be made to improve the long-term prognosis.

Hiroshima Prefectural Hospital in Hiroshima, Japan, between October 23, 2020 (the date when 1 admission of patients with COVID-19 in our hospital was initiated again after being temporarily 2 suspended), and July 31, 2021 (when none of the patients had received antibody medicines for 3 COVID-19 or completed two doses of the COVID-19 vaccine). The data were extracted from 4 electronic medical records. Survival at 30 and 60 days after admission was confirmed using 5 electronic medical record information or by a telephone call to the patient. 6

The diagnosis of COVID-19 was confirmed by the presence of symptoms and/or contact with 7 patients with COVID-19 and a positive SARS-CoV-2 genetic or qualitative antigen test . 8 Baricitinib was approved by the Ethics Committee of Hiroshima Prefectural Hospital for use 9 in patients with respiratory failure associated with COVID-19, and its use was initiated on 10 December 15, 2020. Except for initiation of baricitinib, there were no changes in the treatment 11 protocol between October 23, 2020 (start of the study) and December 15, 2020. Baricitinib was 12 subsequently approved for use in patients with respiratory failure associated with COVID-19 13 throughout Japan on April 23, 2021. As per the indications for baricitinib use, only patients with 14 respiratory failure associated with COVID-19 were included in this study . 15 The following exclusion criteria were applied: death or transfer to another hospital within 3 16 days, treatment at other higher medical institutions and transfer to our hospital with improved 17 condition, and history of advanced chronic kidney disease (estimated glomerular filtration rate 18

[eGFR] <15 mL/min/1.73 m 2 ), decompensated cirrhosis, or administration of biologics or other 1 JAK inhibitors. 2

Other than baricitinib, the following drugs were given in various combinations at the 3 discretion of the attending physician: favipiravir (off-label use), corticosteroids (dexamethasone 4 6 mg or methylprednisolone 125-1000 mg, depending on symptoms), remdesivir, heparin, and 5 tocilizumab (off-label use). Patients who received baricitinib (before April 23, 2021), 6 favipiravir, and tocilizumab provided informed consent for off-label treatments. 7

The primary endpoint was the survival rate 30 and 60 days after admission. The propensity 8 scores for treatment with baricitinib were calculated using a logistic regression model with the 9 following covariates: age, sex, severity of disease at admission (Mild, Moderate I, Moderate II, 10 and Severe), body mass index (grouped into four categories: <20, ≥20 to <25, ≥25 to <30, and 11 ≥30), number of comorbidities at risk for severe disease (grouped into three categories: 0, 1−2, 12 and ≥3), and if the patient was treated with or without favipiravir, remdesivir, and corticosteroid 13 (grouped into four categories: none; dexamethasone, 6 or 6.6 mg; methylprednisolone, 125 or 14 or presence of pneumonia; "Moderate II" is defined as SpO2 93% or requirement of oxygen 1 administration; "Severe" is defined as admission to the intensive care unit or requirement of 2 mechanical ventilation. As some patients did not wish to be intubated or ventilated, we added 3 the following to the definition of "Severe": SpO2 93% despite administration of >5 L/min 4 oxygen by a nasal cannula or oxygen mask. The following diseases were considered 5 comorbidities that are risk factors for severe disease: cancer, chronic kidney disease, chronic 6 obstructive pulmonary disease, asthma (moderate to severe), interstitial lung disease, dementia, 7 diabetes, heart diseases, hypertension, immunocompromised state, solid organ transplant, 8 cerebrovascular disease, and dyslipidemia [7, 8] . Based on these propensity scores, the 9 baricitinib group was matched with the control group with a 1:1 nearest-neighbor algorithm 10 using a caliper of 0.2 standard deviations. 11

Comparisons between the baricitinib and control groups were carried out using the Mann-12 J o u r n a l P r e -p r o o f Results 1 Figure 1 depicts the patient selection process. Propensity score matching was performed in 2 41 patients in both groups to reduce the differences in the characteristics between the two 3 groups ( Table 1 ). Despite reduced differences in the patient background, there were significant 4 differences in the "do not intubate" (DNI) order and tocilizumab administration between the two 5 groups. 6

The Kaplan-Meier survival curve is shown in Figure 2A (30-day survival) and B (60-day 7 survival). Log-rank test results showed significant differences in 30-day survival (p = 0.03), but 8 there was no significant difference in survival at 60 days after admission (p = 0.07). 9 Table 2 shows the clinical course of the two groups. There were significant differences 10 between the two groups in the method of oxygen administration at the time of maximum 11 severity of the disease and in the outcome at the time of release from isolation. No significant 12 differences in adverse events were observed between the two groups. 13 14 4

Discussion 15

In this study, we showed that the survival rate at 30 days after admission was higher in the 16 baricitinib group than in the control group. This is the first report of this finding in a real-world 17 situation in Japan. In a phase III trial evaluating the efficacy of baricitinib (COV-BARRIER), 18 J o u r n a l P r e -p r o o f mortality was lower in the baricitinib group at both 28 and 60 days after the date of 1 randomization [5] . In this study, the reason for the lack of significant difference in survival at 60 2 days may partly be due to the small sample size but more likely because deaths after the 30th 3 day were more common in the baricitinib group. Although baricitinib may reduce early 4 mortality due to its anti-inflammatory effects, in some older adult patients who had severe 5 pneumonia despite treatment with baricitinib, respiratory failure did not improve due to residual 6 pneumonia or progressive fibrosis of the lung even after the acute phase was over, and death 7 occurred. To prevent these cases, immunosuppressive therapy including baricitinib should be 8

given from an early stage before pneumonia becomes severe or treatment to improve pulmonary 9 fibrosis caused by prolonged severe pneumonia, such as administration of anti-fibrotic drugs, is 10 necessary. In addition, baricitinib is contraindicated in patients with renal failure and pregnant 11 women; therefore, a drug that can be administered to these patients is desirable. 12

Tocilizumab was used only in the control group, as it cannot be combined with baricitinib. 13

Because of the small number of patients who received tocilizumab, its contribution to survival 14 remains unknown. Further, whether the low number of patients requiring ventilation in the 15 baricitinib group resulted from baricitinib administration is unclear, as it might have been due to 16 the high number of DNI order cases. Although extracorporeal membrane oxygenation (ECMO) 17

is not performed at our hospital, one patient in the baricitinib group underwent ECMO after 18 being transferred to a higher medical institution. 1 No significant differences in adverse events were observed between the two groups. infection may worsen the long-term prognosis in patients receiving baricitinib. With regard to 8 the latter, two patients developed thromboembolism in the baricitinib group (four in the total 9 cohort), and both developed thromboembolism despite receiving heparin, which reaffirms the 10 need to monitor for thromboembolism while administering baricitinib. 11

The generalization of this study for future COVID-19 treatment is limited because it is a 12 single-center study with a small sample size. Although we used propensity score matching to 13 homogenize the differences between the groups in this study, it is impossible to homogenize all 14

Despite advancements in vaccinations and the availability of antibody medicines and oral 16 antiviral medicines, patients may continue to experience severe illness due to the lack of early 17 and appropriate treatment. Therefore, baricitinib is expected to contribute to COVID-19 18 treatment. 1 2 5

Conclusions 3

Baricitinib may improve early survival in patients with respiratory failure associated with 4 COVID-19. However, efforts should be made to improve the long-term prognosis. Liver dysfunction was defined as an increase in aspartate aminotransferase or alanine 1 aminotransferase to at least twice the level prior to baricitinib administration and at least three 2 times the upper limit of the reference values without other apparent causes. 3

Renal dysfunction was defined as an increase in creatinine to at least 1.5 times the level prior to 4 baricitinib administration without other apparent causes. 5

Categorical data are presented as numbers (percentages) and analyzed using Fisher's exact test 6 or chi-square test. 7

A p value <0.05 (in boldface) was considered statistically significant. 

COVID-3 19: consider cytokine storm syndromes and immunosuppression

Cytokine release syndrome in severe COVID-19

Mechanism of baricitinib supports artificial intelligence-predicted testing in COVID

Baricitinib as 12 potential treatment for 2019-nCoV acute respiratory disease

Efficacy and safety of baricitinib for the treatment of hospitalized adults with COVID-19

a randomized, double-blind, parallel-group

Clinical Management of Patients with COVID-19: A guide for front-line healthcare 2 workers

Clinical 6 epidemiology of hospitalized patients

Report of the COVID-19 registry Japan

Investigation of the freely available easy-to-use software "EZR" for medical 13 statistics

Group Manager Medical Translation Division