Stone 2020.

Study characteristics
Methods	RCT Blinding: double‐blinding Date of study: from 20 April 2020 to 15 June 2020 Location: multicentre / USA Follow‐up duration (days): 28
Participants	Population: patients with COVID‐19 (mild to severe) Randomised 243 participants (n1 tocilizumab arm = 161 / n2 control arm = 82) Characteristics of participants Mean / median age: 60 years 141 males Admitted to ICU: n = 11 (4%) Severity: mild: n = 38 / moderate: n = 194/ severe: n = 10 / critical: n = 1 Patients on oxygen without intubation: n = 204 (84%) Intubated: n = 1 C‐reactive protein (median) : 94.3 to 116 mg/L Inclusion criteria Patients were eligible for enrolment if they were 19 to 85 years of age and had SARS‐CoV‐2 infection confirmed by either nasopharyngeal swab polymerase chain reaction or serum IgM antibody assay Patients had to have at least 2 of the following signs: fever (body temperature > 38°C) within 72 hours before enrolment; pulmonary infiltrates; or a need for supplemental oxygen in order to maintain an oxygen saturation higher than 92%. At least one of the following laboratory criteria also had to be fulfilled: C‐reactive protein level higher than 50 mg per litre; ferritin level higher than 500 ng per millilitre; D‐dimer level higher than 1000 ng per millilitre; or lactate dehydrogenase level higher than 250 U per litre. Exclusion criteria Unable to provide verbal informed consent or have verbal agreement to participate through attestation and signature of a witness required, as outlined in the Partners IRB’s Table for Consenting in COVID Research that is More than Minimal Risk. Patients between the ages of 79 and 86 will be excluded if they have: NYHA Class III/IV heart 32 of 92; pulmonary infiltrate on chest X ray; need for supplemental O2 to maintain saturation > 92% AND at least 1 of the following: ferritin > 500 ng/mL; CRP > 50 mg/L; LDH > 250 U/L. D‐dimer > 1000 ng/mL failure, insulin‐dependent diabetes mellitus, angina, or treatment of a malignancy (excluding nonmelanoma skin cancer) within 6 months uncontrolled bacterial, fungal, or non‐COVID viral infection active tuberculosis (see appendix B) any prior investigational immunosuppressive therapy within 28‐days or 3 half‐lives of the agent (for instance with biologic or JAK inhibitor) any concurrent immunosuppressive medication that the PI believes would put the patient at higher risk receipt of intravenous tocilizumab for the treatment of a non‐COVID condition within 3 weeks of the first COVID symptom history of hypersensitivity to tocilizumab any concurrent immunosuppressive medication that the PI believes would put the patient at higher risk treatment with other biologic or small‐molecule immunosuppressive therapy such as IL1R‐antagonism, JAK inhibition, or other agents. treatment with convalescent plasma history of diverticulitis or bowel perforation ANC < 500, platelets < 50,000 AST/ALT > 5X ULN Dropouts and withdrawals: 1/243 (1%); 0 withdrawal due to AEs
Interventions	Intervention: tocilizumab (8 mg/kg infusion up to 800 mg max) single dose Control: placebo
Outcomes	Primary outcome of the trial: the primary outcome was intubation (or death, for patients who died before intubation) after administration of tocilizumab or placebo, assessed in a time‐to‐event analysis. Note: improvement was defined as an decrease in score by at least 2 points on the ordinal clinical improvement scale.
Notes	Funding: private (supported by Genentech) Conflict of interest: yes, declared. Quote: “Dr. Stone reports grants from Genentech, during the conduct of the study; grants and personal fees from Principia Biopharma and Roche, grants from Viela, personal fees from Sanofi, Chemocentryx, Celgene, Abbvie, Chugai, Grunenthal, Glaxo Smith Kline, InflaRx, INSmed, Regeneron, Roivant, outside of submitted work.” Protocol: yes, available. Statistical plan: yes, available. Data‐sharing stated: yes, following approval of proposal. Overall comment: in addition to the published article, the trial registry, study protocol and statistical analysis plan were used in data extraction and assessment of risk of bias. The study did not achieve the sample size recorded in the trial registry. There were no other notable differences in study population, procedures, treatments or outcomes between the published article and the trial registry, study protocol and statistical analysis plan.