SGLT2 Inhibitor TrEatment iN Patients Awaiting cOronary arTery bYpass Surgery to Reduce Post-opErative AF

Study centers: Up to 7 centers in Sweden, Denmark, and the Czech Republic.

Phase of development: III.

Objectives: In a multicenter, prospective, randomized, controlled clinical trial to compare dapagliflozin 10 mg once daily and placebo in preventing post-operative atrial fibrillation (AF) and acute kidney injury (AKI) in patients with chronic coronary syndrome undergoing coronary artery bypass surgery (CABG).

Methodology: Following informed consent, 800 participants are randomized in a 1:1 fashion to the intervention arm receiving a 10 mg tablet of dapagliflozin once daily and the comparator arm receiving a placebo tablet once daily for a minimum of 1 week while awaiting scheduled CABG surgery. Treatment will be terminated at the time of discharge from hospital after CABG surgery. Aside from a follow-up telephone call one week after discharge to collect safety data-and an on-site visit for type 2 diabetes patients-no extra monitoring or visits are included in the trial. Patients will be followed up 30 days and 12 months through national registries.

Primary endpoint: Incidence of new onset post-operative AF during hospitalization following CABG surgery.

Secondary endpoints:

Key secondary endpoint: AKI (as defined by at least a 1.5-fold increase from baseline in serum creatinine within the prior seven days, an absolute increase in serum creatinine of 26.5 µmol/L or more within 48 hours, or a urine volume of less than 0.5 mL/kg/h for at least six hours, following Kidney Disease Improving Global Outcomes grade 1 or above) in-hospital after CABG surgery.

Safety endpoint: The safety and tolerability of dapagliflozin will be evaluated by reporting adverse events (AEs), serious adverse events (SAEs), and suspected unexpected serious adverse events (SUSARs) from the start of treatment until one week post-discharge.

Other secondary variables during hospitalization after CABG surgery include:

1. All-cause mortality; 2. Stroke; 3. New onset heart failure; 4. Ventricular arrhythmia; 5. Electrical cardioversion for AF; 6. Use of amiodarone for AF; 7. Change in inflammatory and cardiac biomarkers from baseline to three days after CABG surgery; 8. Change in HbA1c from baseline to three days after CABG surgery; 9. Length of intensive care unit and hospital stay.

Secondary variables after hospital discharge include:

1. 30-day all-cause mortality; 2. 12-month all-cause mortality; 3. 12-month myocardial infarction; 4. 12-month stroke; 5. 12-month hospital admission for AF or heart failure; 6.

12-month dialysis treatment.

All events will be adjudicated by an independent clinical endpoint committee.

Quality assurance: Sites will be monitored according to a specific study monitoring plan.

Before starting the clinical trial all centers will have a telephone/web-based start meeting with presentation of the study, study procedures and documentation. The first visit at site will be when the center has included some patients into the study. During the study period, monitors will have regular contact with the participating departments to ensure that the trial is conducted in compliance with the protocol and applicable regulatory requirements. The monitors will also provide information and support to the investigator(s). The monitors will review source documents for verification of consistency with the study data recorded in case report forms according to risk based monitoring. Investigators and other responsible personnel must be available during the monitoring visits, possible audits and inspections and should devote sufficient time to these processes. Patient recruitment status is continuous (updated every 24 hours on weekdays).

Reporting for adverse events: Registration of adverse events will start after informed consent and when treatment with study medication has been given and continue until the patient leaves the hospital after the CABG. The patients will be informed to contact the investigator or study nurse if any adverse event should occur during this timeframe.

Reporting procedures for Adverse Events and Serious Adverse Events: Only adverse events and serious adverse events that are not considered as signs and symptoms expected and related to CABG or known side effects from the study drug will be reported in this study. Events defined as endpoints in the study (e.g. all-cause death, stroke, new onset heart failure) will not be reported as adverse events. This means that other clinical signs and symptoms, which are reported by the patient and observed by the investigator, and in the opinion of the investigator are unexpected in relation to actual diagnosis, will be reported.

SUSAR reporting procedure: If the responsible investigator judges the SAE as being drug-related and unexpected the event must be reported to the sponsor within one working day. The documentation will be on a CIOMS form (http://www.cioms.ch/index.php/cioms-form-i). The sponsor is then responsible for reporting SUSAR to the regulatory authorities and ethics committee. The sponsor is also responsible for information to all involved investigators in the study.

A SUSAR resulting in death or judged as life threatening must be reported to regulatory authorities and ethics committee within 7 days after the sponsor has been notified about the event. A full report must be sent to the authorities within 15 days.

A SUSAR which is not resulting in death or is life threatening has to be reported to regulatory authorities and ethics committee within 15 days after the sponsor has been notified about the event. A full report has to be sent to the authorities as soon as possible.

Annual report: A safety report, including assessment of overall safety and all reported SUSARs will be submitted yearly to the Regulatory Authorities and if requested to the Ethics Committee.

Sample size: We assumed an incidence of post-operative AF of 22% in the control group and of 14% in the SGLT2 inhibitor group (a 36% relative decrease). With a power of 80% and alpha of 5%, a total of 361 subjects completing the trial in each of the two groups are required. To account for drop-outs, 400 subjects will be randomly assigned to each of the two groups.

Statistical analyses: All results will be reported and analyzed according to the intention-to-treat principle. Baseline characteristics will be reported by randomized treatment and overall. For continuous variables, we will report the frequency, mean and standard deviation, or median and interquartile range, as appropriate. For binary or categorical variables, the frequency and percentage will be reported. The number of missing values, if any, will be reported. The disposition of subjects in the trial, including the number randomized, the number lost to follow-up/withdrawn, and the number of subjects at each follow-up visit will be summarized in a CONSORT figure. For binary outcomes, we will report the number and percentage of subjects with the outcome by randomized group. A risk ratio and 95% confidence interval (CI) will be calculated and reported along with the corresponding p value from a chi-squared test or Fisher´s exact test, as appropriate. These include: incidence of AF before hospital discharge (primary endpoint); AKI before hospital discharge (key secondary endpoint); in-hospital all-cause mortality, and stroke, new onset heart failure, ventricular arrhythmia, electrical cardioversion for AF, and use of amiodarone for AF; 12-month dialysis treatment; reported AEs, SAEs, and SUSARs at one week post-discharge (safety endpoint). For continuous outcomes, we will report the frequency, mean, and standard error of the change from baseline by randomized group. The between-group difference in mean change from baseline and 95% CI will be estimated using a linear regression model including the baseline value and randomized treatment as covariates in the model. If the change from baseline is found to seriously violate the required assumption of normality, an appropriate transformation may be applied. These include: change in cardiac and inflammatory biomarkers from baseline to three days after CABG surgery, and change in HbA1c from baseline to three days after CABG surgery. For time-to-event outcomes, the number of subjects with the event and the cumulative percentage (estimated using the Kaplan-Meier method) will be reported by randomized group. Hazard ratios and 95% CIs will be estimated using a Cox proportional hazards model with randomized treatment included as the only covariate in the model, and a p value estimated using a log-rank test. These include: 30-day and 12-month all-cause mortality; 12-month myocardial infarction; 12-month stroke; 12-month hospital admission for AF or heart failure. The length of intensive care unit and hospital stay endpoint will be summarized by randomized group using the median and interquartile range. Geometric mean differences and 95% CIs will be estimated using Poisson or negative binomial regression models, as appropriate, including randomized treatment as a covariate in the model. Two-sided statistical significance levels of 5% will be used and all estimates will be presented with 95% confidence intervals. To account for the stratified randomization, the primary and key secondary endpoint analyses will be stratified by site.

Full detail of the statistical analysis will be described in a statistical analysis plan, which will be completed and signed before the final trial database is locked and unblinded. Any changes to the statistical analysis plan will be documented in amendments to the statistical analysis plan and will be signed off before the database is finalized and unblinded.

Interim Safety Analysis: A maximum of three months following inclusion of the first 100 patients an independent endpoint committee will monitor study endpoints. Variables to be assessed are all-cause death, AKI, dialysis, new onset AF, heart failure, and diabetic ketoacidosis. Premature termination of the study will be mandated in the event that one of the treatment strategies shows statistically significance at the 0.001 alpha level for the composite of the above variables.