RESULTS
Overall, 384 patients with moderately severe COVID-19 were included in our study. There were 212 males (55.2%), and the median age was 60 years (IQR 48–70). All patients were Caucasians. Median duration of hospital stay until death or discharge was 16 days (IQR 14-20). 131 patients (34.1%) received tofacitinib in addition to standard of care, and 253 patients (65.9%) were treated with standard of care alone. Baseline clinical and demographic characteristics of patients with low and normal oxygen saturation who received tofacitinib or standard of care alone are presented in Tables 1 and 2.
Patients with reduced oxygen saturation. At enrolment, oxygen saturation at rest on ambient air was 93% or lower in 257 patients (group 1), of whom 72 received tofacitinib on top of standard of care treatment. Patients in the tofacitinib and control groups were well-matched by age, gender, body mass index, comorbidities, oxygen saturation, and various laboratory parameters, including white blood cells, neutrophils, lymphocytes, serum lactate dehydrogenase and eGFR. Median CRP levels were high in both groups. Patients who were treated with tofacitinib had less extensive ground-glass opacification on CT compared to controls, although almost all patients maintained target oxygen saturation using nasal oxygen and did not require high-flow oxygen or non-invasive ventilation. Baseline differences in both groups were observed. In particular, two-times more patients received glucocorticoids in the tofacitinib group than in the control group (43.7% and 21.1%, respectively; p=0.001).
Overall, the primary composite end-point of death or mechanical ventilation was reached in 9 (12.5%) of 72 patients who were treated with tofacitinib and 26 (14.1%) of 185 patients who received standard of care treatment alone (HR 0.92, 95% CI 0.33-2.56). Invasive mechanical ventilation was initiated in 5 (6.9%) patients in the tofacitinib group versus 24 (13.0%) patients in the control group (HR 0.46; 95 CI 0.11-1.99), whereas 8 (11.1%) versus 21 (11.4%) patients in the two groups, respectively, died in the hospital (HR 1.25; 95 CI 0.44-3.54).
Unweighted Kaplan-Meier estimates showed no beneficial effect of treatment with tofacitinib compared with standard of care only (Fig. 1). Also, we found no differences between groups for the mortality end-point or requirement for invasive mechanical ventilation. In Cox regression analysis, addition of tofacitinib to standard of care did not result in a reduced risk of either primary or secondary outcomes (Table 3).
Patients with normal oxygen saturation. At baseline, oxygen saturation at rest was normal (94% or higher) in 127 patients. 59 of them were treated with tofacitinib in addition to standard of care, and 68 patients received only standard of care treatment. Various parameters, including age, gender, body mass index, comorbidities, oxygen saturation, white blood cells, neutrophils, and lymphocytes counts, serum creatinine level, were well balanced across groups. Like in group 1, patients in the tofacitinib group had less extensive ground-glass opacification on CT and a lower median CRP level compared to the control group. The proportions of patients who received treatment with glucocorticoids and low-molecular weight heparins in the tofacitinib group were significantly higher than in the control group.
The primary composite end-point of death or invasive mechanical ventilation was met in 1 (1.7%) of 59 patients who received tofacitinib plus standard of care treatment and 3 (4.4%) of 68 patients who were treated with standard of care alone (HR 0.83; 95 CI 0.07-9.44). No patient in the tofacitinib group required invasive mechanical ventilation, whereas 3 (4.4%) controls were intubated during hospital stay (p=0.25). One patient (1.7%) in the tofacitinib group deceased compared to 3 (4.4%) patients in the control group (HR 1.10; 95 CI 0.10-12.46).
Kaplan-Meier analysis showed no beneficial effect of tofacitinib added to standard of care treatment compared with standard of care alone (Fig. 2). The mortality rates and requirement for invasive mechanical ventilation also did not differ between the two groups. In Cox regression analysis, addition of tofacitinib to standard of care treatment was not associated with a reduced risk of either composite primary or secondary end-points compared with standard of care treatment (Table 3).
Patients treated with glucocorticoids. In total, 118 patients from the study population were treated with intravenous dexamethasone (supplemental table S1). Among 70 patients with reduced oxygen saturation, the primary end point of death or mechanical ventilation occurred in 4 (12.9%) of 31 patients in the tofacitinib group and 11 (28.2%) of 39 patients in the control group. Mechanical ventilation was required in 1 (3.2%) and 10 (25.6%) patients, respectively, whereas 4 (12.9%) patients treated with tofacitinib and glucocorticoids and 9 (23.1%) patients who received glucocorticoids alone died. All the differences in the end points rates between the two groups did not reach statistical significance in the univariate Cox analysis. Among 48 patients with normal oxygen saturation, the composite end point was met in 1 (3.0%) of 31 patients who were treated with tofacitinib in addition to glucocorticoids and in 2 (13.3%) of 15 patients who received glucocorticoids alone. The differences in the primary and secondary end point rates between the two groups were insignificant in the univariate Cox analysis.
Safety. Adverse events were reported in 34 (26.0%) of 131 patients treated with tofacitinib (Table 4). Treatment with tofacitinib was discontinued in 7 (5.3%) patients due to rapid respiratory deterioration (n=2) or serious adverse events (n=5) that included ST-elevation myocardial infarction (n=1), bacterial sepsis (n=2), jugular vein thrombosis (n=1), and bacterial colitis (n=1).