Efficacy and safety of colchicine and spironolactone after myocardial infarction: the CLEAR-SYNERGY trial in perspective

Evidence before the study

Inflammation reduction with low-dose colchicine and canakinumab has been shown to lower the risk of cardiovascular (CV) events in three large trials, both in patients with chronic coronary disease and after myocardial infarction (MI).^1–3 The effects of colchicine in a CV trial conducted during the COVID-19 pandemic are not known.

Activation of the renin–angiotensin–aldosterone system has been associated with vascular inflammation, myocardial apoptosis, and fluid retention in patients with acute MI. While aldosterone blockade prolongs survival in post-MI patients with clinical heart failure (HF) and decreased systolic function,^4–6 limited data exist regarding the benefits of routine aldosterone blockade in patients with ST-elevation MI (STEMI) with or without HF.

Aim of the study

The main objective of CLEAR-SYNERGY^7,8 was to assess whether colchicine reduces recurrent atherosclerotic events and whether spironolactone reduces a combination of HF and atherosclerotic events in patients who presented with acute STEMI and were managed with percutaneous coronary intervention (PCI).

The primary outcome for colchicine was the composite of CV death, recurrent MI, stroke, or unplanned ischemia-driven revascularization. The co-primary outcomes for spironolactone were (i) the composite of the total numbers of CV deaths or new or worsening HF and (ii) the composite of CV death, new or worsening HF, recurrent MI, or stroke.

Study design

CLEAR-SYNERGY was designed as a multinational factorial randomized controlled trial of low-dose colchicine 0.5 mg daily vs. placebo and spironolactone 25 mg daily vs. placebo. Eligible patients presented with acute STEMI [<5% of patients had non-STEMI (NSTEMI) with extensive myocardial necrosis] and were randomized as soon as possible, but no more than 72 h after PCI for the qualifying MI. Key exclusion criteria included systolic blood pressure <90 mm Hg, creatinine clearance <30 mL/min/1.73 m², serum potassium >5.0 mEq/L, active diarrhoea, or indication for colchicine or a mineralocorticoid inhibitor.

Other salient features/characteristics

Randomization was 1:1:1:1 between the four groups and stratified by MI type (STEMI or NSTEMI). The colchicine dosage employed in the trial was initially weight-based, with patients weighing ≥70 kg receiving twice-a-day dosing for the first 3 months. After a blinded interim analysis showing higher than expected discontinuation rates, the dosing regimen was changed to once daily colchicine 0.5 mg throughout the treatment period irrespective of body weight. If patients developed gastrointestinal symptoms, the managing physician could reduce the colchicine/placebo dose by half. In the case of hyperkalaemia, the spironolactone/placebo dose could be reduced to half a tablet daily. Because of a lower-than-expected aggregate event rate, the sample size was increased from 4000 to 7000 patients to maintain study power of 80%, with an estimated 546 events needed to detect a 25% relative risk reduction, and the follow-up was extended to an estimated median follow-up of 3.5 years for necessary accrual of events.

Principal findings

Between February 2018 and November 2022, the trial recruited 7062 participants from 104 centres in 14 countries (∼2500 patients in four sites in Macedonia and 500 patients in Serbia). The mean patient age was 61 years and 20% were women. The vast majority of patients were enrolled after STEMI (95%) and 49% had multivessel coronary disease.

The composite of CV death, MI, stroke, or ischemia-driven revascularization occurred by 5 years in 9.1% of patients allocated to colchicine and 9.3% allocated to placebo [hazard ratio (HR) 0.99, 95% confidence interval (CI) 0.85–1.16), P = 0.93]. There was a reduction in the incidence of non-CV deaths in the colchicine arm compared with placebo (HR 0.68, 95% CI 0.46–0.99). High-sensitivity C-reactive protein (hs-CRP) concentration at 3 months was 3.0 mg/L (95% CI 2.6–3.5) in the colchicine group and 4.3 (95% CI 3.9–4.6) in the placebo group. Diarrhoea occurred in 10.2% vs. 6.6% (P < 0.001) and serious infection in 2.5% vs. 2.9% (P = 0.85).

The total numbers of CV death or new or worsening HF were 183 patients (1.7 per 100 patient-years) in the spironolactone group and 220 events (2.1 per 100 patient-years) in the placebo group (HR 0.91, 95% CI 0.69–1.21, P = 0.51). The composite of CV death, new or worsening HF, recurrent MI, or stroke occurred in 7.9% and 8.3% of patients in the spironolactone and placebo groups, respectively (HR 0.96, 95% CI 0.81–1.13, P = 0.60).

Serious adverse events were reported in 255 patients (7.2%) in the spironolactone group and 241 (6.8%) in the placebo group.

In perspective

The neutral results of CLEAR-SYNERGY⁷ are discordant with those of COLCOT, LoDoCo-2, and a recent meta-analysis (including these two trials and CONVINCE) of 14 738 patients and 1198 CV events which showed a reduction of 27% in the composite endpoint of CV death, MI, ischaemic stroke, and coronary revascularization with colchicine.^1,2,9

In CLEAR-SYNERGY, there was a reduction of 22% in the incidence of a primary endpoint event with colchicine vs. placebo before the pandemic, which was lost during the pandemic, with an interaction between the COVID-19 phase and the treatment effect (P < 0.10). The inverted relationship between the incidence of non-fatal MI and all-cause deaths in CLEAR-SYNERGY is consistent with the published impact of COVID-19 on CV clinical care and under-reporting of non-fatal events.^10,11 The same phenomenon was observed in Dal-GenE-1, in which the MI/all-death ratio decreased from 1.74 before the COVID-19 pandemic to 0.96 during the pandemic.¹² In CLEAR-SYNERGY, the MI/all-cause death ratio was even much more markedly inverted at 0.62, with rates of MI and death at 3.1% and 5.1% for a median follow-up of 3 years.⁷ In contrast, the COMPLETE trial conducted by the same group before the COVID era reported rates of MI and death of 7.9% and 5.2% with the same median follow-up of 3 years.¹³ These data suggest major under-reporting of MI, by ≥60%, during the pandemic in CLEAR-SYNERGY. Dal-GenE-1, IRONMAN, and GUIDE-HF are other published trial results that were affected by COVID-19, with a loss of significance during the pandemic.^12,14,15

Furthermore, inflammation was not controlled in the colchicine arm of CLEAR with a least-squares mean hs-CRP of 3.0 mg/L (95% CI 2.6–3.5), in contrast to the on-treatment median values of 1.12 mg/L [interquartile range (IQR): 0.77–2.10] in COLCOT and 0.94 mg/L (IQR: 0.53–1.93) in LoDoCo2. These data are important, given that CANTOS has previously shown that inflammation reduction therapy lowers the incidence of CV events only in those with an on-treatment hs-CRP <2.0 mg/L (as in COLCOT and LoDoCo2), but not in those with hs-CRP ≥ 2.0 mg/L (as in CLEAR-SYNERGY).¹⁶ This might also explain at least in part the strange lack of benefit of colchicine, and even a trend for worsening, on the occurrence of pericarditis (HR 1.53, 95% CI 0.88–2.65) in CLEAR-SYNERGY.

The neutral results of CLEAR-SYNERGY as they pertain to spironolactone also differ greatly from the significant benefits of mineralocorticoid receptor antagonists in EPHESUS, RALES, and EMPHASIS.^4–6 This discordance with these three previous trials further demonstrates the effect of COVID on CLEAR-SYNERGY, which makes its results uncertain and uninterpretable. The ongoing COLCOT-T2D trial testing colchicine on ischaemic CV events in 10 000 patients with Type 2 diabetes and no previous coronary disease will help to determine whether its benefits extend to primary prevention.

Funding

There was no funding associated with this editorial.

Data availability

No new data were generated or analysed in support of this research.