NEJMoa2405922

The n e w e ng l a n d j o u r na l of m e dic i n e
Original Article
Colchicine in Acute Myocardial Infarction

S.S. Jolly, M.-A. d’Entremont, S.F. Lee, R. Mian, J. Tyrwhitt, S. Kedev,
G. Montalescot, J.H. Cornel, G. Stanković, R. Moreno, R.F. Storey, T.D. Henry,
S.R. Mehta, M. Bossard, P. Kala, J. Layland, B. Zafirovska, P.J. Devereaux,
J. Eikelboom, J.A. Cairns, B. Shah, T. Sheth, S.K. Sharma, W. Tarhuni, D. Conen,
S. Tawadros, S. Lavi, and S. Yusuf, for the CLEAR Investigators*
A BS T R AC T
BACKGROUND
Inflammation is associated with adverse cardiovascular events. Data from recent The authors’ full names, academic degrees,
trials suggest that colchicine reduces the risk of cardiovascular events. and affiliations are listed in the Appendix.
Dr. Jolly can be contacted at s anjit.jolly@
phri.ca or at Hamilton General Hospital,
METHODS Rm. C3-118, DBCVSRI Bldg., 237 Barton St.
In this multicenter trial with a 2-by-2 factorial design, we randomly assigned patients East, Hamilton, ON L8L 2X2, Canada.
who had myocardial infarction to receive either colchicine or placebo and either *A complete list of the CLEAR investiga-
spironolactone or placebo. The results of the colchicine trial are reported here. The tors is provided in the Supplementary
primary efficacy outcome was a composite of death from cardiovascular causes, Appendix, available at NEJM.org.
recurrent myocardial infarction, stroke, or unplanned ischemia-driven coronary This article was published on November 17,
revascularization, evaluated in a time-to-event analysis. C-reactive protein was 2024, at NEJM.org.
measured at 3 months in a subgroup of patients, and safety was also assessed. DOI: 10.1056/NEJMoa2405922
Copyright © 2024 Massachusetts Medical Society.
RESULTS
A total of 7062 patients at 104 centers in 14 countries underwent randomization;
at the time of analysis, the vital status was unknown for 45 patients (0.6%), and
this information was most likely missing at random. A primary-outcome event
occurred in 322 of 3528 patients (9.1%) in the colchicine group and 327 of 3534
patients (9.3%) in the placebo group over a median follow-up period of 3 years
(hazard ratio, 0.99; 95% confidence interval [CI], 0.85 to 1.16; P = 0.93). The inci-
dence of individual components of the primary outcome appeared to be similar in
the two groups. The least-squares mean difference in C-reactive protein levels be-
tween the colchicine group and the placebo group at 3 months, adjusted according
to the baseline values, was −1.28 mg per liter (95% CI, −1.81 to −0.75). Diarrhea
occurred in a higher percentage of patients with colchicine than with placebo (10.2%
vs. 6.6%; P<0.001), but the incidence of serious infections did not differ between
groups.
CONCLUSIONS
Among patients who had myocardial infarction, treatment with colchicine, when
started soon after myocardial infarction and continued for a median of 3 years,
did not reduce the incidence of the composite primary outcome (death from cardio-
vascular causes, recurrent myocardial infarction, stroke, or unplanned ischemia-
driven coronary revascularization). (Funded by the Canadian Institutes of Health
Research and others; CLEAR ClinicalTrials.gov number, NCT03048825.)
n engl j med nejm.org 1

The New England Journal of Medicine is produced by NEJM Group, a division of the Massachusetts Medical Society.
Downloaded from nejm.org at Sun Pharma on November 17, 2024. For personal use only.
No other uses without permission. Copyright © 2024 Massachusetts Medical Society. All rights reserved.
I
nflammation is thought to be an with ST-segment elevation myocardial infarction
important mechanism for atherosclerosis in (STEMI) was embedded within the larger trial of
both the acute and the chronic phases. In- colchicine and spironolactone, and the results of
creased levels of circulating inflammatory mark- the registry-based trial have been published pre-
ers are also associated with a worse prognosis in viously.11
patients with acute coronary syndromes.1 Can Initially, patients were eligible to participate
akinumab, an interleukin-1β inhibitor, reduced in the trial if they had STEMI and underwent
ischemic events in patients with previous myo- percutaneous coronary intervention. To increase
cardial infarction but increased fatal infections.2 recruitment numbers, the steering committee
Therefore, more data about the effects of antiin- modified the protocol on April 5, 2020, to enroll
flammatory agents on cardiovascular events are patients with large non–ST-segment elevation
needed. myocardial infarction (NSTEMI) who had under-
Colchicine inhibits the actions of neutrophils gone percutaneous coronary intervention and had
and the release of inflammatory chemokines, in- one or more of the following risk factors: a left
cluding interleukin-1 and interleukin-6.3 A trial ventricular ejection fraction of no more than 45%;
involving 4745 patients in which treatment with diabetes mellitus; multivessel coronary artery
colchicine was initiated within 30 days after disease, defined by at least 50% stenosis of a
myocardial infarction and a trial involving 5522 second major epicardial vessel; previous myocar-
patients with stable coronary artery disease dial infarction; or age greater than 60 years. The
showed beneficial cardiovascular effects of col- detailed eligibility criteria are provided in Table
chicine; however, two recent trials involving pa- S1 in the Supplementary Appendix, available at
tients with ischemic stroke showed no reductions NEJM.org.
in cardiovascular events with colchicine treatment, The ethics committee of each participating
although in one trial the treatment duration was center and the relevant national regulatory au-
only 3 months.4-7 thorities approved the trial. All patients provided
The European Society of Cardiology recently written informed consent. The Population Health
provided a class IIa recommendation for the use Research Institute at McMaster University and
of colchicine in patients with atherosclerotic coro-Hamilton Health Sciences in Hamilton, Canada,
nary artery disease.8 However, the use of colchi- coordinated the trial and collected and held all
cine in such patients is not widespread. Given the trial data; the third and fourth authors conducted
biologic rationale for a benefit from colchicine all analyses. The steering committee developed the
and the encouraging evidence that colchicine may trial protocol, and the members of the committee
improve cardiovascular outcomes, we conducted (listed in the Supplementary Appendix) vouch for
the CLEAR trial to examine the effects of colchi- the completeness and accuracy of the data and
cine in patients after myocardial infarction. for the fidelity of the trial to the protocol. The
first author wrote the manuscript with the input
of the coauthors, and the steering committee made
Me thods
the decision to submit the manuscript for publi-
Trial Design cation. The trial funders had no role in the design
We used a 2-by-2 factorial design in this interna- and conduct of the trial. An independent data
tional, investigator-initiated, multicenter, pro- and safety monitoring committee monitored the
spective, randomized, placebo-controlled trial of accumulating safety and efficacy data.
colchicine and spironolactone in patients with
acute myocardial infarction. Information about Randomization
the trial design was published previously and is Patients were randomly assigned in a factorial
provided in the protocol, available with the full 1:1:1:1 allocation to receive colchicine and spirono-
text of this article at NEJM.org.9 The results of lactone, colchicine and placebo, spironolactone
the spironolactone trial are reported elsewhere.10 and placebo, or placebo only as soon as possible
All patients, investigators, health care providers, after the index percutaneous coronary interven-
data collectors, and outcome adjudicators were tion. Randomization was performed with the use
unaware of trial-group assignments. A registry- of permuted blocks within a 24-hour computer-
based trial of SYNERGY stents in 733 patients ized central system at the Population Health Re-
2 n engl j med nejm.org
search Institute. Randomization was stratified der of the treatment period, beginning in Sep-
according to trial center and the type of myocar- tember 2020.4
dial infarction (STEMI or NSTEMI).
Statistical Analysis
Outcomes The initial calculation of sample size to provide
The primary efficacy outcome was a composite of the trial with 80% power to detect a 25% relative
death from cardiovascular causes, recurrent myo- risk reduction was based on a time-to-event analy-
cardial infarction, stroke, or unplanned ischemia- sis of the primary outcome, a composite of death
driven coronary revascularization, evaluated in a from cardiovascular causes, recurrent myocardial
time-to-event analysis. Key secondary outcomes infarction, or stroke; we anticipated a cumulative
were a composite of death from cardiovascular incidence of events in the placebo group of 15%
causes, recurrent myocardial infarction, or stroke, at 3 years, a two-sided type I error level of 5%, a
evaluated in a time-to-event analysis; and the to- loss to follow-up of 2% of patients in both the
tal number of primary-outcome events. C-reactive colchicine group and the placebo group, discon-
protein levels and safety were also assessed. Mea- tinuation of the trial regimen by 12.5% of pa-
surement of C-reactive protein was not mandated tients, and no interaction with spironolactone.
by the trial protocol, but measurements taken in On the basis of these assumptions, we estimated
the course of clinical care were recorded on case that 4000 patients and 512 primary-outcome
report forms. Detailed, prespecified definitions events were needed to detect a 25% relative risk
of outcomes are provided in Table S2, and a full reduction with a log-rank test. In April 2020, a
list of outcomes is provided in Table S3. blinded interim analysis of the incidence of events
A committee of clinicians who were unaware found an incidence of 3% per patient-year, with an
of trial-group assignments adjudicated all prima- estimated cumulative incidence of 9% at 3 years,
ry-outcome events, episodes of major bleeding, which was consistent with the data from other tri-
and stent thrombosis events. Staff at an angio- als.4,12 As a result, the sample size was increased
graphic core laboratory at the Population Health from 4000 to 7000 patients to maintain a power of
Research Institute who were unaware of trial- 80%, and we estimated that 546 primary-outcome
group assignments reviewed all ischemia-driven events would be needed to detect a 25% relative
revascularization and stent thrombosis events. risk reduction. The sample size was increased
without knowledge of any treatment effects.
Trial Interventions For the primary analysis, patients were evalu-
The trial products were colchicine tablets of 0.5 ated according to the groups to which they were
mg, spironolactone tablets of 25 mg, and place- randomly assigned. A two-sided log-rank test was
bos matching the colchicine and spironolactone used to compare the colchicine and placebo
tablets. Tiofarma provided both trial drugs and groups. A P value of less than 0.05 was considered
placebos, which were manufactured with raw to indicate statistical significance. The hazard
materials produced by Indena. At the beginning ratios and 95% confidence intervals were esti-
of the trial, colchicine dosage was based on mated with the use of a Cox proportional-haz-
weight for the first 90 days of treatment; pa- ards regression model in which the trial group
tients weighing 70 kg or more received a dose of was the independent variable and patients were
0.5 mg of colchicine or matching placebo twice stratified according to whether they received
a day, and patients weighing less than 70 kg spironolactone or spironolactone-matched placebo
received a dose of 0.5 mg or matching placebo and whether they had STEMI or NSTEMI. At the
once a day. After the first 90 days of treatment, request of the Journal, the Fine–Gray subdistribu-
all patients received the trial product once a day. tion hazard model was used to account for com-
However, after blinded interim analyses showed peting risks: death from noncardiovascular causes
higher-than-expected rates of discontinuation for outcomes that include death from cardiovas-
and the Colchicine Cardiovascular Outcomes cular causes, death from cardiovascular causes
Trial (COLCOT) showed efficacy with once-daily for outcomes that include death from noncardio-
colchicine, the steering committee adopted the vascular causes, and death from any cause for
regimen of once-daily colchicine at a dose of 0.5 other, nonfatal outcomes. A total event analysis
mg or matching placebo throughout the remain- was performed with the Prentice–Williams–Peter-

Table 1. Demographic and Clinical Characteristics at Baseline.*
Colchicine Placebo
Characteristic (N = 3528) (N = 3534)
Demographic characteristics
Mean age — yr 60.6±10.3 60.7±10.3
Age >75 yr — no. (%) 301 (8.5) 270 (7.6)
Female sex — no. (%) 725 (20.5) 713 (20.2)
Race or ethnic group — no. (%)†
American Indian or Alaskan Native 7 (0.2) 3 (0.1)
Asian 95 (2.7) 89 (2.5)
Black 24 (0.7) 23 (0.7)
Native Hawaiian or other Pacific Islander 9 (0.3) 9 (0.3)
White 3233 (91.6) 3249 (91.9)
Other 153 (4.3) 159 (4.5)
Geographic region — no. (%)
North America 1010 (28.6) 1012 (28.6)
Europe 2356 (66.8) 2359 (66.8)
Other 162 (4.6) 163 (4.6)
Clinical characteristics
Killip class ≥II — no. (%)‡ 25 (0.7) 24 (0.7)
NSTEMI at presentation — no. (%) 165 (4.7) 184 (5.2)
STEMI at presentation — no. (%) 3363 (95.3) 3350 (94.8)
Myocardial area affected by STEMI — no./total no. (%)
Anterior 1304/3363 (38.8) 1326/3350 (39.6)
Inferior 1940/3363 (57.7) 1892/3350 (56.5)
Lateral 423/3363 (12.6) 434/3350 (13.0)
Posterior 341/3363 (10.1) 319/3350 (9.5)
Multivessel coronary disease — no. (%) 1735 (49.2) 1742 (49.3)
Medical history — no. (%)
Current smoker 1461 (41.4) 1423 (40.3)
Hypertension 1620 (45.9) 1613 (45.6)
Diabetes mellitus 658 (18.7) 645 (18.3)
Previous myocardial infarction 309 (8.8) 324 (9.2)
Previous percutaneous coronary intervention 345 (9.8) 364 (10.3)
Medications at discharge — no. (%)
Aspirin 3428 (97.2) 3405 (96.3)
Clopidogrel 1478 (41.9) 1497 (42.4)
Ticagrelor 1611 (45.7) 1571 (44.5)
Prasugrel 381 (10.8) 413 (11.7)
Angiotensin-converting–enzyme inhibitor 2750 (77.9) 2768 (78.3)
or angiotensin-receptor blocker
Statin 3408 (96.6) 3416 (96.7)
Sodium–glucose cotransporter 2 inhibitor 110 (3.1) 101 (2.9)
Table 1. (Continued.)
Colchicine Placebo
Characteristic (N = 3528) (N = 3534)
Initial percutaneous coronary intervention§
Placement of bare-metal stent — no. of stents/total no. (%) 12/4797 (0.3) 8/4898 (0.2)
Placement of ≥1 drug-eluting stent — no. of stents/total no. (%) 4619/4797 (96.3) 4694/4898 (95.8)
Angioplasty only — no. of stents/total no. (%) 146/4797 (3.0) 165/4898 (3.4)
Median no. of stents per patient (IQR) 1.0 (1.0–2.0) 1.0 (1.0–2.0)
Mean stent length — mm 23.8±8.6 23.8±8.8
Mean stent diameter — mm 3.2±0.5 3.1±0.5
Coronary artery bypass grafting — no. of stents/total no. (%) 137/4797 (2.9) 139/4898 (2.8)
Placement of intraaortic balloon pump — no. of patients (%) 45 (1.3) 49 (1.4)
* Plus–minus values are means ±SD. IQR denotes interquartile range, NSTEMI non–ST-segment elevation myocardial
infarction, and STEMI ST-segment elevation myocardial infarction.
† Race or ethnic group was reported by the patients.
‡ The Killip classification system is a tool to assess the risk of death based on the severity of heart failure in patients with
acute myocardial infarction. The scale ranges from I to IV, with higher numbers indicating greater risk.
§ The total number of stents placed was 9695, with 4797 in the colchicine group and 4898 in the placebo group.
son model with the gap-time approach and with data analysis] the pandemic). Post hoc subgroups
the Lin–Wei–Yang–Ying model. Secondary out- according to geographic region (North America vs.
comes were analyzed with the same approach. Europe vs. other) were added to show consistency.
The widths of the confidence intervals have not We undertook on-treatment analyses with the
been adjusted for multiplicity, so the intervals exclusion of patients who discontinued the trial
may not be used in place of hypothesis testing. regimen on the day of randomization and censored
An interaction among the assigned trial regimens patients 7 days after permanent discontinuation
was not expected. of the trial regimen.
The prespecified subgroups were analyzed with C-reactive protein values were analyzed with
the use of the Cox regression model with an inter- the use of a linear mixed model with repeated
action term for the subgroup. Patients were divided measures and adjusted according to the base-
into subgroups according to the prespecified char- line value; the mean difference between groups
acteristics: age (<65 years vs. ≥65 years), sex at 3 months, along with the 95% confidence
(female vs. male), diabetes versus no diabetes, interval, is reported. Additional information re-
multivessel disease versus single-vessel disease, garding the statistical analyses is provided in the
STEMI versus NSTEMI, estimated glomerular Supplementary Appendix.
filtration rate (<60 ml per minute per 1.73 m2 of
body-surface area vs. ≥60 ml per minute per R e sult s
1.73 m2), initial trial dose and weight (patients
weighing <70 kg who received once-daily colchi- Patients
cine during the first 90 days, patients weighing Between February 1, 2018, and November 8, 2022,
≥70 kg who received twice-daily colchicine during we enrolled 7062 patients from 104 centers in 14
the first 90 days, and patients weighing ≥70 kg countries; 3528 patients were assigned to receive
who received once-daily colchicine during the first colchicine and 3534 to receive placebo (Fig. S1).
90 days), and timing of trial enrollment with re- At the time of our analyses, the vital status was
spect to the coronavirus disease 2019 (Covid-19) unknown for 45 of the 7062 patients (0.6%); a
pandemic (before [February 1, 2018, to January 30, primary-outcome event was known to have oc-
2020], during [January 31, 2020, to January 31, curred in 2 of these patients. The remaining 43
2022], or after [February 1, 2022, to the time of patients had no recorded outcome or final visit.

Table 2. Primary and Secondary Outcomes.
Colchicine Placebo Hazard Ratio

Outcome (N = 3528) (N = 3534) (95% CI)*
number (percent)
Primary outcome
Death from cardiovascular causes, recurrent myocardial 322 (9.1) 327 (9.3) 0.99 (0.85–1.16)
infarction, stroke, or ischemia-driven coronary
revascularization†
Components of the primary outcome
Death from cardiovascular causes 117 (3.3) 113 (3.2) 1.03 (0.80–1.34)
Recurrent myocardial infarction 102 (2.9) 111 (3.1) 0.88 (0.66–1.17)
Stroke 50 (1.4) 43 (1.2) 1.15 (0.72–1.84)
Ischemia-driven coronary revascularization 164 (4.6) 166 (4.7) 1.01 (0.81–1.26)
Other outcomes
Death from cardiovascular causes, recurrent myocardial 241 (6.8) 250 (7.1) 0.98 (0.82–1.17)
infarction, or stroke
Death from all causes 162 (4.6) 179 (5.1) 0.90 (0.73–1.12)
Death from noncardiovascular causes 45 (1.3) 66 (1.9) 0.68 (0.46–0.99)
Pericarditis 33 (0.9) 22 (0.6) 1.53 (0.88–2.65)
Atrial fibrillation 91 (2.6) 89 (2.5) 0.98 (0.72–1.33)
* The widths of the confidence intervals have not been adjusted for multiplicity, and the intervals may not be used in
place of hypothesis testing.
† The P value for the primary outcome is 0.93.
Given that the missing data were rare and evenly 3 months, adjusted according to the value at base-
distributed between the colchicine and placebo line, was 2.98±0.19 mg per liter among 1384
groups (Table S4), the data are most likely miss- patients in the colchicine group and 4.27±0.19
ing at random. mg per liter among 1419 patients in the placebo
Baseline characteristics of the patients ap- group (difference, −1.28 mg per liter; 95% con-
peared to be well balanced between the groups; fidence interval [CI], −1.81 to −0.75).
the mean age of patients was 61 years, and
20.4% of patients were women (Table 1). A total Efficacy
of 9.0% of patients had previous myocardial in- A primary-outcome event occurred in 322 of
farction, 10.0% had previous percutaneous coro- 3528 patients (9.1%) in the colchicine group as
nary intervention, and 18.5% had diabetes mel- compared with 327 of 3534 patients (9.3%) in
litus; 95.1% of patients had STEMI, and 4.9% the placebo group (hazard ratio, 0.99; 95% CI,
had NSTEMI. The median time from symptom 0.85 to 1.16; P = 0.93) (Table 2 and Fig. 1), with a
onset to randomization was 26.8 hours (inter- median duration of follow-up of 1089 days (2.98
quartile range, 15.9 to 42.4), and the median years) in the colchicine group and 1090 days
time from randomization to the first dose of the (2.98 years) in the placebo group. The spirono-
trial product was 1.6 hours (interquartile range, lactone factorial had no significant effect on the
0.6 to 7.4). The medications provided to patients results of the comparison of colchicine with pla-
at discharge from the hospital appeared to be cebo for the primary outcome (P = 0.96 for inter-
similar in the two groups (Table 1). The median action). Death from cardiovascular causes, recur-
duration of follow-up was 3.00 years (interquar- rent myocardial infarction, or stroke occurred
tile range, 2.14 to 3.71); 25.9% of patients in the in 241 patients (6.8%) in the colchicine group
colchicine group and 25.5% in the placebo and 250 patients (7.1%) in the placebo group
group discontinued the trial regimen. The least- (hazard ratio, 0.98; 95% CI, 0.82 to 1.17). An
squares mean level (±SE) of C-reactive protein at analysis of total events that included recurrent
primary-outcome events showed that there were

100
376 events (3.53% per patient-year) in the colchi- 15 Hazard ratio, 0.99 (95% CI, 0.85–1.16)
14 P=0.93
cine group as compared with 389 events (3.67% 90 13 Placebo
per patient-year) in the placebo group (hazard 12
80 11
ratio, 0.98; 95% CI, 0.85 to 1.13), with a median 10 Colchicine
Cumulative Incidence (%)

70 9
duration of follow-up of 1078 days (2.95 years) in 8
7
the colchicine group and 1092 days (2.99 years) 60 6
in the placebo group. 5
50 4
Death from cardiovascular causes occurred 3
40 2
in 117 patients (3.3%) in the colchicine group 1
0
and 113 (3.2%) in the placebo group (hazard 30 0 1 2 3 4 5
ratio, 1.03; 95% CI, 0.80 to 1.34). Other compo- 20
nents of the primary and secondary outcomes
10
are summarized in Table 2. Death from noncar-
diovascular causes occurred in 45 patients (1.3%) 0
0 1 2 3 4 5
in the colchicine group as compared with 66 pa-
Years of Follow-up
tients (1.9%) in the placebo group (hazard ratio,
0.68; 95% CI, 0.46 to 0.99). No. at Risk
Colchicine 3528 3329 2688 1686 697 183
The results of on-treatment analyses were Placebo 3534 3349 2683 1674 659 163
consistent with those of the primary analysis
and are shown in Table S5. The incidence of pri- Figure 1. Kaplan–Meier Event Curves for Death from Cardiovascular
mary-outcome events appeared to be consistent Causes, Recurrent Myocardial Infarction, Stroke, or Ischemia-Driven
Revascularization.
across all prespecified subgroups, with the excep-
The inset shows a magnified version of the graph.
tion of patients weighing at least 70 kg who re-
ceived twice-daily colchicine for the first 3 months,
in whom the number of events was lower (Fig. 2).
The results seemed to be similar during the dif- underwent randomization within 30 days after
ferent phases of the Covid-19 pandemic. acute myocardial infarction to receive colchicine
at a dose of 0.5 mg daily or placebo; a total of
Safety 301 primary-outcome events were reported, and
The incidence of serious adverse events and ad- colchicine treatment was associated with a 23%
verse events did not differ between groups (Ta- relative reduction in death from cardiovascular
ble 3). Diarrhea occurred in a total of 361 patients causes, recurrent myocardial infarction, resusci-
(10.2%) in the colchicine group and 233 patients tation after cardiac arrest, stroke, or urgent
(6.6%) in the placebo group (P<0.001). There was hospitalization for angina that led to revascular-
no apparent difference in the number of serious ization.4 The Low Dose Colchicine 2 (LODOCO 2)
infections between the groups. trial assigned patients with stable coronary artery
disease to receive colchicine at a dose of 0.5 mg
daily or placebo; 451 primary-outcome events were
Discussion
reported, and colchicine was associated with a 31%
Among patients with acute myocardial infarction, relative reduction in the composite primary out-
the antiinflammatory agent colchicine did not come (death from cardiovascular causes, myo-
reduce the incidence of the composite primary cardial infarction, ischemic stroke, or ischemia-
outcome (death from cardiovascular causes, re- driven coronary revascularization).5 One of the
current myocardial infarction, stroke, or unplanned largest trials of colchicine, the CHANCE-3 (Col-
ischemia-driven coronary revascularization) over chicine in High-Risk Patients with Acute Minor-
a median treatment duration of 3 years as com- to-Moderate Ischemic Stroke or Transient Isch-
pared with placebo. As expected, colchicine in- emic Attack) trial, involved 8345 patients who
creased the incidence of diarrhea as compared were randomly assigned to receive colchicine at
with that among patients who received placebo. a dose of 0.5 mg twice daily on days 1 through
The previous trial most comparable to CLEAR 3 and 0.5 mg once daily on days 4 through 90 or
is COLCOT, which involved 4745 patients who to receive placebo. A total of 534 primary-out-

Hazard Ratio for Death from Cardiovascular Causes,

Myocardial Infarction, Stroke, or Revascularization
Subgroup Colchicine Placebo (95% CI)
no. of events/total no. of patients (%)
All patients 322/3528 (9.1) 327/3534 (9.3) 0.99 (0.85−1.16)
Age
<65 yr 182/2343 (7.8) 192/2320 (8.3) 0.93 (0.76−1.14)
≥65 yr 140/1185 (11.8) 135/1214 (11.1) 1.06 (0.84−1.34)
Sex
Female 72/725 (9.9) 64/713 (9.0) 1.12 (0.80−1.57)
Male 250/2803 (8.9) 263/2821 (9.3) 0.95 (0.80−1.13)
Diabetes mellitus
Yes 79/658 (12.0) 85/645 (13.2) 0.88 (0.65−1.20)
No 243/2870 (8.5) 242/2889 (8.4) 1.01 (0.85−1.21)
Single-vessel or multivessel disease
Multivessel 192/1735 (11.1) 200/1742 (11.5) 0.97 (0.79−1.18)
Single vessel 130/1793 (7.3) 127/1792 (7.1) 1.02 (0.80−1.30)
Type of myocardial infarction
STEMI 310/3363 (9.2) 315/3350 (9.4) 0.98 (0.84−1.15)
NSTEMI 12/165 (7.3) 12/184 (6.5) 1.13 (0.51−2.52)
Estimated GFR level
<60 ml/min/1.73 m2 56/291 (19.2) 48/278 (17.3) 1.10 (0.75−1.62)
≥60 ml/min/1.73 m2 266/3237 (8.2) 279/3256 (8.6) 0.96 (0.81−1.13)
Dosing
Once daily (patient weight <70 kg) 74/721 (10.3) 72/697 (10.3) 1.01 (0.73−1.40)
Twice daily (patient weight ≥70 kg) 110/1161 (9.5) 136/1137 (12.0) 0.78 (0.61−1.00)
Once daily (patient weight ≥70 kg) 138/1646 (8.4) 119/1700 (7.0) 1.20 (0.94−1.54)
Covid-19 phase
Before pandemic 100/998 (10.0) 125/991 (12.6) 0.78 (0.60−1.02)
During pandemic 170/1773 (9.6) 159/1799 (8.8) 1.09 (0.88−1.35)
After pandemic 52/757 (6.9) 43/744 (5.8) 1.19 (0.79−1.78)
Geographic region
North America 96/1010 (9.5) 95/1012 (9.4) 0.93 (0.11−7.66)
Europe 216/2356 (9.2) 221/2359 (9.4) 0.97 (0.81−1.17)
Other 10/162 (6.2) 11/163 (6.7) 0.90 (0.38−2.13)
0.5 1.0 1.5 2.0 2.5
Colchicine Better Placebo Better
Figure 2. Forest Plot of the Primary Outcome According to Prespecified Subgroups.

The hazard ratio for the North American subgroup at 1 year of follow-up was calculated with the use of the Cox
model with a time-dependent covariate to account for the violation of the proportional-hazards assumption. GFR
denotes glomerular filtration rate (expressed as milliliters per minute per 1.73 m2 of body-surface area), NSTEMI
non–ST-segment elevation myocardial infarction, and STEMI ST-segment elevation myocardial infarction.
come events occurred, and colchicine was found ischemic stroke were randomly assigned to re-
to have no effect on the incidence of stroke (haz- ceive colchicine at a dose of 0.5 mg daily or
ard ratio, 0.98; 95% CI, 0.83 to 1.16) or on the usual care for a median of 34 months; with 338
secondary outcome, a composite of death from primary-outcome events reported, no effect of
cardiovascular causes, myocardial infarction, tran- colchicine on the primary outcome — a compos-
sient ischemic attack, or stroke (a total of 607 ite of ischemic stroke, myocardial infarction, car-
events; hazard ratio, 0.96; 95% CI, 0.82 to 1.13).7 diac arrest, hospitalization for unstable angina,
However, the CHANCE-3 trial was a short-term or death from vascular causes — was found
treatment trial, and the duration of treatment (hazard ratio, 0.84; 95% CI, 0.68 to 1.05).6
may not have been long enough to show a treat- In our trial, 649 first primary-outcome events
ment effect. In the CONVINCE (Colchicine for occurred; an analysis of previous trials showed
Prevention of Vascular Inflammation in Non- that those with more than 600 outcome events
cardioembolic Stroke) trial, 3154 patients with rarely produced spurious results disproven by
subsequent trials.13 The results of our on-treat- Table 3. Adverse Events.

ment analysis were consistent with those of our
intention-to-treat analysis. Furthermore, our sub- Colchicine Placebo
Event (N = 3528) (N = 3534) P Value
group analyses did not suggest that our results
were affected by the Covid-19 pandemic. The ef- number (percent)
fects estimates appeared to be consistent through-
Any serious adverse event 235 (6.7) 261 (7.4) 0.22
out all the individual components of the primary
Serious adverse gastrointestinal 35 (1.0) 33 (0.9) 0.81
and secondary outcomes and on-treatment anal- event
ysis. The increase in the incidence of diarrhea
Serious adverse hematologic 0 (0) 8 (0.2) 0.005
and the reduction in C-reactive protein were as event*
expected and provide support for the biologic Serious infection 87 (2.5) 101 (2.9) 0.85
effects of colchicine in this trial.
Any adverse event 1124 (31.9) 1119 (31.7) 0.86
The reasons for the divergence between the
results of our trial and those of previous trials Diarrhea 361 (10.2) 233 (6.6) <0.001
of colchicine in patients with cardiovascular * Anemia occurred in 3 patients, febrile neutropenia in 2 patients, pancytopenia
diseases are not immediately evident; however, in 3 patients, and thrombocytopenia in 2 patients.
three of the latest trials, CLEAR, CHANCE, and
CONVINCE, provide what is likely to be the
most recent evidence to date of the effects of trexate did not reduce inflammatory markers.
colchicine in patients with vascular disease. Re- The ongoing ARTEMIS (Effects of Ziltivekimab
cent meta-analyses have shown a nominal excess vs. Placebo on Cardiovascular Outcomes in Pa-
in death from noncardiovascular causes with tients with Acute Myocardial Infarction) trial
colchicine.14 We found the opposite: a lower rate involving 10,000 patients is evaluating the ef-
of death from noncardiovascular causes in the fects of ziltivekimab on the primary outcome, a
colchicine group than in the placebo group. composite of death from cardiovascular causes,
COLCOT showed an excess of pneumonia among myocardial infarction, or stroke. The divergent
patients who received colchicine, whereas the results of the trials of canakinumab, methotrex-
LODOCO 2 trial did not show any increase in ate, and colchicine highlight the need for large
serious infection.4,5 We also found no excess in trials that target different parts of the inflam-
serious infection with colchicine as compared matory pathways.
with placebo. Our trial has limitations. Women and mem-
In 2024, the European Society of Cardiology bers of diverse racial and ethnic groups were
upgraded its recommendation for colchicine from underrepresented in the trial relative to the inci-
class IIb to class IIa for patients with atheroscle- dence of cardiovascular disease in these groups
rotic coronary artery disease; however, this change worldwide (Table S6). The percentage of patients
was made before the current data were available.8 who discontinued the trial regimen, 25%, was
Similarly, the U.S. Food and Drug Administration higher than anticipated; however, our on-treat-
had approved colchicine to treat coronary artery ment sensitivity analyses produced results con-
disease before the current data were available. sistent with those of our primary analyses. Our
Canakinumab, an interleukin-1β inhibitor, trial was not designed to evaluate the effect of
was associated with a 15% reduction in ischemic twice-daily colchicine, and this dosing will need
events in patients after myocardial infarction in to be tested in future clinical trials. Adherence
CANTOS (the Canakinumab Antiinflammatory to the trial regimen was assessed only through
Thrombosis Outcome Study).2 Another trial involv- patient reports, because pill counts were not
ing 4786 patients who had myocardial infarction possible during the pandemic. The incidence of
or multivessel coronary artery disease showed no gout was not assessed as an outcome in the trial.
benefit of methotrexate with respect to the pri- Among patients with acute myocardial infarc-
mary outcome, a composite of death from cardio- tion, treatment with colchicine that began soon
vascular causes, nonfatal myocardial infarction, after myocardial infarction and continued for a
stroke, and hospitalization for unstable angina median of 3 years did not reduce the incidence
that led to urgent revascularization (hazard ra- of the composite primary outcome (death from
tio, 0.96; 95% CI, 0.79 to 1.16); however, metho- cardiovascular causes, recurrent myocardial infarc-

tion, stroke, or unplanned ischemia-driven revas- Marlborough, MA. Trial drugs were partially donated by Tiofarma,
Oud-Beijerland, Netherlands.
cularization) but was associated with an increase Disclosure forms provided by the authors are available with
in the incidence of diarrhea. the full text of this article at NEJM.org.
Supported by the Canadian Institutes of Health Research, the A data sharing statement provided by the authors is available
Population Health Research Institute, and Boston Scientific, with the full text of this article at NEJM.org.
Appendix
The authors’ full names and academic degrees are as follows: Sanjit S. Jolly, M.D., Marc‑André d’Entremont, M.D., M.P.H., Shun Fu
Lee, Ph.D., Rajibul Mian, Ph.D., Jessica Tyrwhitt, B.Sc., Sasko Kedev, M.D., Ph.D., Gilles Montalescot, M.D., Ph.D., Jan H. Cornel, M.D.,
Ph.D., Goran Stanković, M.D., Ph.D., Raul Moreno, M.D., Ph.D., Robert F. Storey, M.D., Timothy D. Henry, M.D., Shamir R. Mehta, M.D.,
Matthias Bossard, M.D., Petr Kala, M.D., Ph.D., Jamie Layland, M.D., Ph.D., Biljana Zafirovska, M.D., Ph.D., P.J. Devereaux, M.D., Ph.D.,
John Eikelboom, M.B., B.S., M.Sc., John A. Cairns, M.D., Binita Shah, M.D., Tej Sheth, M.D., Sanjib K. Sharma, M.D., Wadea Tarhuni,
M.D., David Conen, M.D., M.P.H., Sarah Tawadros, M.B., B.S., Shahar Lavi, M.D., and Salim Yusuf, M.D., D.Phil.
The authors’ affiliations are as follows: the Population Health Research Institute, McMaster University (S.S.J., M.-A.E., S.F.L., R.
Mian, J.T., S.R.M., P.J.D., J.E., T.S., D.C., S.T., S.Y.), and Hamilton Health Sciences (S.S.J., M.-A.E., S.F.L., R. Mian, S.R.M., P.J.D., J.E.,
T.S., D.C.), Hamilton, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke (M.-A.E.), the University of British Columbia and
Centre for Cardiovascular Innovation, Vancouver Coastal Health, Vancouver (J.A.C.), the Department of Medicine, University of Sas-
katchewan, Saskatoon (W.T.), and London Health Sciences, University of Western Ontario, London (S.L.) — all in Canada; the Univer-
sity Clinic of Cardiology, Medical Faculty, University Ss. Cyril and Methodius, Skopje, North Macedonia (S.K., B.Z.); Sorbonne Univer-
sity, ACTION Study Group, Centre Hospitalier Universitaire Pitié–Salpêtrière Assistance Publique–Hopitaux de Paris, Paris (G.M.); the
Dutch Network for Cardiovascular Research, Utrecht, Radboud University Medical Center, Nijmegen, and Northwest Clinics, Alkmaar
— all in the Netherlands (J.H.C.); the University Clinical Center of Serbia and the Faculty of Medicine, University of Belgrade, Belgrade
(G.S.); the Cardiology Department, University Hospital La Paz, Universidad Autónoma de Madrid, Madrid (R. Moreno); NIHR Sheffield
Biomedical Research Centre, Sheffield Teaching Hospitals NHS Foundation Trust, and the Division of Clinical Medicine, University of
Sheffield — both in Sheffield, United Kingdom (R.F.S.); the Caril and Edyth Lindner Center for Research and Education, Christ Hospi-
tal Health Network, Cincinnati (T.D.H.); the Cardiology Division, Heart Center, Luzerner Kantonsspital, and the Faculty of Health
Sciences and Medicine, University of Lucerne, Lucerne, Switzerland (M.B.); University Hospital Brno, Brno, Czech Republic (P.K.); the
Department of Cardiology, Peninsula Health, Frankston, VIC, and Peninsula Clinical School, Central Clinical School, Monash Univer-
sity, Melbourne, VIC — both in Australia (J.L.); the Division of Cardiology, Department of Medicine, NYU Grossman School of Medi-
cine, and the Section of Cardiology, Department of Medicine, VA New York Harbor Healthcare System, New York (B.S.); and B.P.
Koirala Institute of Health Sciences, Dharan, Nepal (S.K.S.).
References
1. Barrett TD, Hennan JK, Marks RM, of vascular recurrent events in non-car- cardial infarction. N Engl J Med. DOI:
Lucchesi BR. C-reactive-protein-associat- dioembolic stroke (CONVINCE): a ran- 10.1056/NEJMoa2405923.
ed increase in myocardial infarct size af- domised controlled trial. Lancet 2024; 11. Jolly SS, Lee SF, Mian R, et al. SYNER-
ter ischemia/reperfusion. J Pharmacol Exp 404:125-33. GY-everolimus-eluting stent with a bioab-
Ther 2002;303:1007-13. 7. Li J, Meng X, Shi F-D, et al. Colchicine sorbable polymer in ST-elevation myocar-
2. Ridker PM, Everett BM, Thuren T, et al. in patients with acute ischaemic stroke or dial infarction: CLEAR SYNERGY OASIS-9
Antiinflammatory therapy with cana kin transient ischaemic attack (CHANCE-3): Registry. Am J Cardiol 2024;220:111-7.
umab for atherosclerotic disease. N Engl multicentre, double blind, randomised, 12. Jolly SS, Cairns JA, Yusuf S, et al. Out-
J Med 2017;377:1119-31. placebo controlled trial. BMJ 2024;385: comes after thrombus aspiration for ST
3. Leung YY, Yao Hui LL, Kraus VB. Col- e079061. elevation myocardial infarction: 1-year
chicine — update on mechanisms of ac- 8. Vrints C, Andreotti F, Koskinas KC, et follow-up of the prospective randomised
tion and therapeutic uses. Semin Arthritis al. 2024 ESC guidelines for the manage- TOTAL trial. Lancet 2016;387:127-35.
Rheum 2015;45:341-50. ment of chronic coronary syndromes. Eur 13. Montori VM, Devereaux PJ, Adhikari
4. Tardif J-C, Kouz S, Waters DD, et al. Heart J 2024;45:3415-537. NKJ, et al. Randomized trials stopped early
Efficacy and safety of low-dose colchicine 9. d’Entremont M-A, Lee SF, Mian R, et for benefit: a systematic review. JAMA 2005;
after myocardial infarction. N Engl J Med al. Design and rationale of the CLEAR 294:2203-9.
2019;381:2497-505. SYNERGY (OASIS 9) trial: a 2x2 factorial 14. Fiolet ATL, Opstal TSJ, Mosterd A, et al.
5. Nidorf SM, Fiolet ATL, Mosterd A, et randomized controlled trial of colchicine Efficacy and safety of low-dose colchicine in
al. Colchicine in patients with chronic versus placebo and spironolactone vs pla- patients with coronary disease: a systematic
coronary disease. N Engl J Med 2020;383: cebo in patients with myocardial infarc- review and meta-analysis of randomized
1838-47. tion. Am Heart J 2024;275:173-82. trials. Eur Heart J 2021;42:2765-75.
6. Kelly P, Lemmens R, Weimar C, et al. 10. Jolly SS, d’Entremont M-A, Pitt B, et Copyright © 2024 Massachusetts Medical Society.
Long-term colchicine for the prevention al. Routine spironolactone in acute myo-

NEJMoa2405922

Uploaded by

Copyright:

Available Formats

NEJMoa2405922

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NEJMoa2405922

Uploaded by

Copyright:

Available Formats

The n e w e ng l a n d j o u r na l of m e dic i n e

Colchicine in Acute Myocardial Infarction

n engl j med﻿﻿ nejm.org﻿ 1

2 n engl j med﻿﻿ nejm.org﻿

n engl j med﻿﻿ nejm.org﻿ 3

Table 1. Demographic and Clinical Characteristics at Baseline.*

4 n engl j med﻿﻿ nejm.org﻿

n engl j med﻿﻿ nejm.org﻿ 5

Table 2. Primary and Secondary Outcomes.

Colchicine Placebo Hazard Ratio

6 n engl j med﻿﻿ nejm.org﻿

primary-outcome events showed that there were

Cumulative Incidence (%)

n engl j med﻿﻿ nejm.org﻿ 7

Hazard Ratio for Death from Cardiovascular Causes,

Colchicine Better Placebo Better

Figure 2. Forest Plot of the Primary Outcome According to Prespecified Subgroups.

8 n engl j med﻿﻿ nejm.org﻿

subsequent trials.13 The results of our on-treat- Table 3. Adverse Events.

n engl j med﻿﻿ nejm.org﻿ 9

10 n engl j med﻿﻿ nejm.org﻿

You might also like

n engl j med nejm.org 1

2 n engl j med nejm.org

n engl j med nejm.org 3

4 n engl j med nejm.org

n engl j med nejm.org 5

6 n engl j med nejm.org

n engl j med nejm.org 7

8 n engl j med nejm.org

n engl j med nejm.org 9

10 n engl j med nejm.org