Effect of ciclosporin on safety, lymphocyte kinetics and left ventricular remodelling in acute myocardial infarction

Aims Following a favourable pilot trial using a single bolus of ciclosporin, it has been unclear why 2 large studies (CYCLE and CIRCUS) failed to prevent reperfusion injury and reduce infarct size in STEMI (ST elevation myocardial infarction). The purpose of this study was to assess the effect of ciclosporin on myocardial injury, left ventricular remodelling and lymphocyte kinetics in patients with acute STEMI undergoing primary percutaneous coronary intervention. Methods In this double‐blind, single centre trial, we randomly assigned 52 acute STEMI patients with an onset of pain of <6 hours and blocked culprit artery to a single bolus of ciclosporin (n = 26) or placebo (n = 26, control group) prior to reperfusion by stent percutaneous coronary intervention. The primary endpoint was infarct size at 12 weeks. Results Mean infarct size at 12 weeks was identical in both groups (9.1% [standard deviation= 7.0] vs 9.1% [standard deviation = 7.0], P = .99; 95% confidence interval for difference: −4.0 to 4.1). CD3 T‐lymphocytes dropped to similar levels at 90 minutes (867 vs 852 cells/μL, control vs ciclosporin) and increased to 1454 vs 1650 cells/μL at 24 hours. Conclusion In our pilot trial, a single ciclosporin bolus did not affect infarct size or left ventricular remodelling, matching the results from CYCLE and CIRCUS. Our study suggests that ciclosporin does either not reach ischaemic cardiomyocytes, or requires earlier application during first medical contact. Finally, 1 bolus of ciclosporin is not sufficient to inhibit CD4 T‐lymphocyte proliferation during remodelling. We therefore believe that further studies are warranted. (Evaluating the effectiveness of intravenous Ciclosporin on reducing reperfusion injury in pAtients undergoing PRImary percutaneous coronary intervention [CAPRI]; NCT02390674)

its known immunosuppressive function, ciclosporin is also a potent inhibitor of the MPTP, 7 which has demonstrated a beneficial effect on ischaemia/reperfusion injury under experimental conditions. 8 However, 2 recent multicentre trials (published after the start of the CAPRI trial) using ciclosporin prior to PPCI have failed to demonstrate a beneficial effect. 9,10 It has been debated whether either preclinical models do not resemble the clinical situation, or whether there have been inadequacies in the trial design and, more importantly, whether cardioprotective therapies should continue to be pursued at all. 11 A significant drawback of the negative clinical trials targeting IRI has been the lack of mechanistic insight into their failure. Ciclosporin is widely used as an immunosuppressive drug to prevent organ transplant rejection. [12][13][14] Through its intracellular binding to cyclophylin it inhibits calcineurin, a serine/threonine phosphatase, and thereby inhibits various T-cell responses, 14 most importantly T-lymphocyte proliferation. 15 Our recently published data also suggests depletion of circulating Tlymphocytes in STEMI patients immediately following PPCI, consistent with an early participation of T-lymphocytes in human myocardial I-R injury, 16 making lymphocytes another pharmacological target in the treatment of STEMI.
The aim of our pilot trial was to investigate the effect of ciclosporin on IRI and concomitantly achieve mechanistic insight by studying its biological effects on the adaptive immune response as a proximal readout. To avoid design-specific flaws, the CAPRI trial was set up as a single-centre, double-blinded and randomized trial using state of the art cardiac magnetic resonance imaging (MRI) to assess infarct size, microvascular obstruction and cardiac remodelling, confirm ciclosporin blood level kinetics, and finally quantify lymphocyte subset dynamics over 2 weeks from fresh blood samples.

| Experimental protocol
Following angiography, participants were randomised in a 1:1 ratio to either ciclosporin or control (saline) using a blocked allocation (permuted random blocks of variable length) system. Randomisation

| Cardiac MRI
Cardiac MRI scans were obtained at 2-7 days as well as 12 weeks post-myocardial infarction with a Siemens Avanto 1.5 Tesla MRI scanner, using a phased array body coil combined with a spine coil. Intravenous gadobutrol contrast (Gadovist, Bayer Schering Pharma AG, Berlin, Germany) was administered at a dose of 0.1 mmol/kg and, after 10 minutes, short axis end-diastolic late gadolinium enhancement images were obtained. All analysis was performed using validated cardiac MRI analysis software (cvi42, Circle Cardiovascular Imaging Inc., Calgary, Canada) as previously described. 16

| Statistical analysis
The primary endpoint, percentage infarct size post-PPCI (as measured by cardiac MRI) was compared at 12 weeks between groups using the

| Patient population
Out of 199 screened patients (Figure 1), 54 participants were recruited into the trial but only 52 were randomised to 1 of the 2 arms.
One participant was not included in the trial due to a problem encountered with viewing the allocation arm from the randomisation website and the second was due to a technical problem with opening the prescription. Participants were predominantly male, past or nonsmokers with little previous serious medical history or ongoing diabetes (

| Peri-interventional parameters
There were also no notable or apparent clinically important differences in peri-interventional characteristics ( Table 2) between arms.
Time from symptom onset to reperfusion was around 3 hours in both arms, and door-to-balloon time remained under 30 minutes despite randomisation and trial drug infusion prior to reperfusion.
Periprocedural antiplatelet therapy (ticagrelor, prasugrel) was similar in both arms, and TIMI 3 flow was achieved in 50 out of 52 patients.
Target vessel diameter and stented length were similar in both arms.

| Clinical endpoints
Reperfusion arrhythmias (y/n) 10   While the 95% CIs are wide, they do not extend beyond the clinically important difference of 4.6% used to establish the sample size.

| Effect of ciclosporin on infarct size and LV remodelling
The absolute reduction of 5% deemed to be of clinically important significance 19 was also not observed. Finally, we looked at LV remodelling, comparing the 12-week MRI scans to the baseline MRI scans (Table 3). There was a nonsignificant reduction in infarct size in both treatment arms (2.4 vs 1.8%), probably owing to discrete overestimation at baseline secondary to myocardial oedema. 20,21 LVEF appeared to improve in both arms at 12 weeks from baseline, and may have been greater improvement in control arm than intervention arm (6.8% (SD = 7.6) vs 2.1% (SD = 8.8), Table 3 and Table S4). While a change in EDV is important for remodelling, only changes in ESV appeared to correlate with ΔLVEF (r = −0.77, Figure 2C,D). Hence, LV volumes appeared to be more reduced in the control arm, especially difference for ESV (ΔESV −12.2 vs +1.4 mL; ΔEDV −2.0 vs +9.1 mL, Table S3).     (Table S6)

| T-cell response and ciclosporin in STEMI
The role of T-lymphocytes in infarct healing post STEMI has not been investigated before in clinical trials 26,27 ; however, Weirather and colleagues have previously demonstrated the role of regulatory T-cells in the remodelling process in a mouse model. 28 The Frangogiannis group has also identified in the mouse that CCR5 signalling suppresses inflammation and reduces adverse remodelling of the infarcted heart, mediating recruitment of regulatory T cells. 29 We have previously shown that the magnitude of the proinflammatory Th1 response from CD4 T-cells in acute STEMI patients correlates with total ischaemic time. 30 We also identified a proliferative response among CD4 T-cells during the first 24 hours postreperfusion, 31 which has now been confirmed in the CAPRI trial (Online Table 4a

| Alternative mechanism and timing for ciclosporin therapy in STEMI
One of the reasons why ciclosporin did not demonstrate additional protection could have been the co-medication with ticagrelor, [34][35][36] prasugrel or ACE inhibitors, 37,38 which have all demonstrated favourable effects in patient outcome by themselves. Many of our patients have call to balloon times of over an hour despite short door to balloon times. Especially when patients are admitted to an emergency room in a noninterventional hospital, interhospital transfer adds a significant delay to successful reperfusion. In this trial all patients had ciclosporin been given immediately (5 min) before successful reperfusion. It is possible that ciclosporin is not so effective for reperfusion injury (which is not a proven phenomenon in human patients), but rather in ischaemic cardiomyocytes and endothelial cells, preventing or delaying ongoing apoptosis in those cell types.
Ciclosporin (10 μmol/L) completely inhibited the oxLDL-induced release of cytochrome C in a study by Walter and colleagues. 39 Moreover, tumour necrosis factor-α-and angiotensin II-induced cytochrome C release was also prevented by ciclosporin treatment. There is a clinical window of at least 60 minutes in most STEMI patients where, other than pain medication and aspirin, basically no therapy is present. Application of ciclosporin 60 minutes before reperfusion could have a significant effect on infarct size, if it prevents cardiomyocyte or endothelial cell apoptosis in vivo at the given concentration.

| CONCLUSIONS
We confirm in this clinical trial that intravenous administration of the mitochondrial pore inhibitor ciclosporin before reperfusion by primary coronary angioplasty is safe, but does not affect myocardial injury.
Serum levels of ciclosporin during 90 minutes of reperfusion were 10-fold higher than established therapeutic levels for immunosuppression, suggesting that either (i) ciclosporin does not reach ischaemic cardiomyocytes, or (ii) ciclosporin needs to be given at an earlier timepoint during myocardial ischaemia (e.g. before transfer for primary PCI). Also, 1 bolus of ciclosporin is not sufficient to inhibit CD4 Tlymphocyte proliferation during remodelling. Given that ciclosporin is safe in STEMI patients, we suggest conducting a new randomised controlled trial where ciclosporin is given at the first point of contact (e.g. ambulance) and also test prolonged therapy in order to inhibit Tcell proliferation during remodelling.