key: cord-0765173-7apjcf3h
authors: Duijvelaar, Erik; Schippers, Job R; Smeele, Patrick J; de Raaf, Michiel Alexander; Vanhove, Arthur L E M; Blok, Siebe G; Twisk, Jos W R; Noordegraaf, Anton Vonk; de Man, Frances S; Bogaard, Harm Jan; Aman, Jurjan
title: Long-term clinical outcomes of COVID-19 patients treated with imatinib
date: 2022-02-17
journal: Lancet Respir Med
DOI: 10.1016/s2213-2600(22)00052-2
sha: 9beec69a30a680e548f06c222c989ee160bb7e12
doc_id: 765173
cord_uid: 7apjcf3h

nan

Clinical data was obtained from the electronic medical records, and patients were called by a team of physician scientists at a minimum of 90 days after randomisation. All 385 patients included in the CounterCOVID study consented to pseudonymized data collection for the 90-day follow up. The general physician or next of kin was contacted in case any of the above information was missing. In case of a transfer or readmission elsewhere, the concerning hospital or health care facility was requested to provide additional information. The trial was conducted in compliance with the International Conference on Harmonisation Good Clinical Practice guidelines and the Declaration of Helsinki. The study protocol of this follow-up study was approved by the ethics committee and institutional review board of the coordinating hospital.

The number of ventilator-free days was defined as the total number of days free from invasive ventilation, patients who died had zero ventilator free days. The number of additional organ support free days was defined as the total number of days free from cardiovascular support, renal replacement therapy (RRT), and extracorporeal mechanical oxygenation (ECMO) . Patients that died before day 90 were assigned -1 additional organ support free days.

A modified version of the WHO ordinal scale was used to categorise the clinical status of a patient at baseline (day 0), day 9, day 28 and day 90. This ordinal scale ranges from 'discharged without oxygen supplementation' to 'death'. Category 1 indicates that the patient was not hospitalised, and received no oxygen supplementation; 2. was not hospitalised, but received supplemental oxygen; 3. was hospitalised, without the use of supplemental oxygen; 4. was hospitalised and received supplemental oxygen using a nasal cannula or mask ; 5. was hospitalised and received oxygen through non-invasive ventilation or high-flow devices; 6. was hospitalised and received invasive ventilation with no extra organ support ; 7. was hospitalised and received invasive ventilation plus additional organ support: vasopressors, renal replacement therapy (RRT), or extra corporal membrane oxygenation (ECMO); and 8. died.

During invasive ventilation, the courses of oxygenation parameters, ventilatory conditions, and the sequential organ failure assessment (SOFA) were assessed daily for the first 14 days. This was done in a subgroup of patients that received invasive ventilation and consisted of 30 (15%) patients who received imatinib and 26 (14%) who received placebo. These parameters were assessed at the time point with the lowest PaO2/FiO2, in which PaO2 was measured in arterial blood. Assessment of the total SOFA score was adapted from the method used by Ferreira et al., and was calculated as the sum of the SOFA scores of the 6 organ domains: respiratory, coagulation, hepatic, cardiovascular, neurological, and renal (1).

Data for this study were collected retrospectively and recorded in an electronic database in the Castor Electronic Data Capture System (Castor EDC; Amsterdam, Netherlands). Statistical analyses were performed using RStudio version 2021.09.0 (https://cran.r-project.org/). The investigators that performed the statistical analyses were aware of treatment allocation.

Categorical variables were analysed using a chi-square test or Fisher's exact test. Continuous variables were analysed using an unpaired T-test or a Mann-Whitney U test depending on the distribution. Time to mortality was analysed using Cox regression analyses. Linear mixed models were used to assess the effect of imatinib on the eight-point ordinal scale. Recovery was treated as a dichotomous outcome and was defined as either being hospitalised and not requiring supplemental oxygen (category 3), not being hospitalised, but received supplemental oxygen at home (category 2) or not being hospitalised, and received no oxygen supplementation (category 1). Death or respiratory failure was also treated as a dichotomous outcome and was defined as death (category 8), received mechanical ventilation with or without organ support (category 6 or 7), non-invasive ventilation (category 5) or high-flow oxygen devices (category 5). For recovery and respiratory failure or death, the effect of the intervention was derived from logistic generalized estimating equation (GEE) analyses. For the analysis of mortality and the eight-point ordinal scale, baseline characteristics that were imbalanced at baseline (P<0·10) were carried forward as a correction factor (sex, diabetes, obesity (BMI > 30kg/m 2 ), and cardiovascular disease).

Linear mixed models were used to assess the effect of treatment with imatinib on the clinical parameters during invasive ventilation and to test whether treatment with imatinib is an effect modifier for time (interaction). In the linear mixed model, a random intercept on individual patient level was entered to adjust for the high correlation of these repeated measures. For all tests two-sided p-values are provided. P-values below 0·05 were considered significant. (8) 11 (6) 18 (10) Adjusted for obesity (BMI > 30kg/m 2 ) 0·47 (0·25, 0·89) 0·020

Adjusted for diabetes 0·55 (0·31, 0·97) 0·045

Adjusted for cardiovascular disease 0·56 (0·31, 1·00) 0·048

Adjusted for all the above 0·52 (0·28, 0·99) 0·045

Hazard ratios and confidence intervals were calculated using Cox regression analysis BMI = body mass index, CI = Confidence interval 

Through day 90 -unadjusted -0·53 (-0·94, -0·11) 0·014

Through day 90 -adjusted for baseline imbalances † -0·56 (-0·99, -0·13) 0·012

Category at day 9 -0·54 (-0·99, -0·09) 0·018 1 118 (60) 108 ( * The effect of the intervention is expressed as the coefficient with associated confidence interval derived from a linear mixed model. Category 1 indicates that the patient was not hospitalised, and received no oxygen supplementation; 2. was not hospitalised, but received supplemental oxygen; 3. was hospitalised, without the use of supplemental oxygen; 4. was hospitalised and received supplemental oxygen using a nasal cannula or mask ; 5.

was hospitalised and received oxygen through non-invasive ventilation or high-flow devices; 6. was hospitalised and received invasive ventilation with no extra organ support ; 7. was hospitalised and received invasive ventilation plus additional organ support: vasopressors, renal replacement therapy (RRT), or extra corporal membrane oxygenation (ECMO); and 8. died. † Adjusted for sex, diabetes, obesity (BMI > 30kg/m 2 ), and cardiovascular disease. ‡ The effect of the intervention is expressed as an odds ratio derived from logistic generalized estimating equation (GEE) analyses. Death or respiratory failure was defined as death (category 8), receiving mechanical ventilation with or without organ support (category 6 or 7), non-invasive ventilation (category 5) or high-flow oxygen devices (category 5). § The effect of the intervention is expressed as an odds ratio derived from logistic generalized estimating equation (GEE) analyses. Recovery was defined as either being hospitalised and not requiring supplemental oxygen (category 3), not being hospitalised, but received supplemental oxygen at home (category 2) or not being hospitalised, and received no oxygen supplementation (category 1). 

Serial evaluation of the SOFA score to predict outcome in critically ill patients

Collaborators on behalf of the CounterCOVID Study Group (in alphabetical order): Jurjan Aman, Sara Azhang, 

Contributions AV, ED, JS, MR, PS acquired the data and take responsibility for the integrity of the data. ED and JT performed the statistical analyses. AVN, ED, FM, HJB, JA, and JS were involved in data interpretation and drafting of the manuscript. All authors provided critical feedback for the manuscript and approved the final version. All authors had full access to the study data and take responsibility for the decision to submit for publication.

J. Aman and A. Vonk Noordegraaf are inventors on a patent (WO2012150857A1, 2011) covering protection against endothelial barrier dysfunction through inhibition of the tyrosine kinase abl-related gene (arg). All other authors have no competing interests. JA has served as a non-compensated scientific advisor for Exvastat TM .

Qualified researchers can request access to anonymised patient-level data and related study documents after publication. Study documents include the study protocol, structure of the electronic case report form, statistical analysis plan, and dataset specifications.

This project was funded with an unrestricted grant from the Amsterdam Medical Center Foundation, a bottom-up grant from NWO ZonMW (no. 10430 01 201 0007), and from the Innovative Medicines Initiative 2 Joint Undertaking (no. 101005142). The funding sources had no role in the design of the study, data collection, analysis, or in the decision to submit the paper for publication.

This study is registered with the European Union Clinical Trials Register (EudraCT 2020-001236-10, https://www.clinicaltrialsregister.eu/ctr-search/trial/2020-001236-10/NL) and the Netherlands Trial Register (NL8491, https://www.trialregister.nl/trial/8491).