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 Patients
with Chronic Coronary Disease
S.M. Nidorf, A.T.L. Fiolet, A. Mosterd, J.W. Eikelboom, A. Schut, T.S.J. Opstal,
S.H.K. The, X.-F. Xu, M.A. Ireland, T. Lenderink, D. Latchem, P. Hoogslag,
A. Jerzewski, P. Nierop, A. Whelan, R. Hendriks, H. Swart, J. Schaap, A.F.M. Kuijper,
M.W.J. van Hessen, P. Saklani, I. Tan, A.G. Thompson, A. Morton, C. Judkins,
W.A. Bax, M. Dirksen, M.M.W. Alings, G.J. Hankey, C.A. Budgeon, J.G.P. Tijssen,
J.H. Cornel, and P.L. Thompson, for the LoDoCo2 Trial Investigators*​​

A BS T R AC T

BACKGROUND
Evidence from a recent trial has shown that the antiinflammatory effects of col- The authors’ full names, academic de-
chicine reduce the risk of cardiovascular events in patients with recent myocardial grees, and affiliations are listed in the Ap-
pendix. Address reprint requests to Dr.
infarction, but evidence of such a risk reduction in patients with chronic coronary Nidorf at GenesisCare, 3/140 Mounts
disease is limited. Bay Rd., Perth 6000, WA, Australia, or at
­smnidorf@​­gmail​.­com; or to Dr. Mosterd
METHODS at the Department of Cardiology, Mean-
In a randomized, controlled, double-blind trial, we assigned patients with chronic der Medical Center, P.O. Box 1502, 3800
BM Amersfoort, the Netherlands, or at a­ ​
coronary disease to receive 0.5 mg of colchicine once daily or matching placebo. .­mosterd@​­meandermc​.­nl.
The primary end point was a composite of cardiovascular death, spontaneous (non-
*A complete list of the investigators in
procedural) myocardial infarction, ischemic stroke, or ischemia-driven coronary the LoDoCo2 trial is provided in the
revascularization. The key secondary end point was a composite of cardiovascular Supplementary Appendix, available at
death, spontaneous myocardial infarction, or ischemic stroke. NEJM.org.

RESULTS Drs. Nidorf and Fiolet and Drs. Mosterd,

Cornel, and Thompson contributed equal-
A total of 5522 patients underwent randomization; 2762 were assigned to the col- ly to this article.
chicine group and 2760 to the placebo group. The median duration of follow-up
This article was published on August 31,
was 28.6 months. A primary end-point event occurred in 187 patients (6.8%) in the 2020, at NEJM.org.
colchicine group and in 264 patients (9.6%) in the placebo group (incidence, 2.5 vs.
DOI: 10.1056/NEJMoa2021372
3.6 events per 100 person-years; hazard ratio, 0.69; 95% confidence interval [CI], Copyright © 2020 Massachusetts Medical Society.
0.57 to 0.83; P<0.001). A key secondary end-point event occurred in 115 patients (4.2%)
in the colchicine group and in 157 patients (5.7%) in the placebo group (incidence,
1.5 vs. 2.1 events per 100 person-years; hazard ratio, 0.72; 95% CI, 0.57 to 0.92;
P = 0.007). The incidence rates of spontaneous myocardial infarction or ischemia-
driven coronary revascularization (composite end point), cardiovascular death or
spontaneous myocardial infarction (composite end point), ischemia-driven coro-
nary revascularization, and spontaneous myocardial infarction were also signifi-
cantly lower with colchicine than with placebo. The incidence of death from noncar-
diovascular causes was higher in the colchicine group than in the placebo group
(incidence, 0.7 vs. 0.5 events per 100 person-years; hazard ratio, 1.51; 95% CI, 0.99
to 2.31).
CONCLUSIONS
In a randomized trial involving patients with chronic coronary disease, the risk of
cardiovascular events was significantly lower among those who received 0.5 mg of
colchicine once daily than among those who received placebo. (Funded by the Na-
tional Health Medical Research Council of Australia and others; LoDoCo2 Austra-
lian New Zealand Clinical Trials Registry number, ACTRN12614000093684.)

n engl j med  nejm.org 1

The New England Journal of Medicine
Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

D
espite lifestyle changes and risk- Me thods
factor reduction, patients with chronic
coronary disease remain at high risk for Trial Design and Oversight
acute cardiovascular events.1-3 The central role of Patient recruitment in the LoDoCo2 trial com-
inflammation in the progression of coronary dis- menced on August 4, 2014, at 13 centers affili-
ease is well recognized.4,5 The possibility that ated with GenesisCare and the Heart and Vascu-
antiinflammatory therapy may improve cardio- lar Research Institute of Sir Charles Gairdner
vascular outcomes was first highlighted in the Hospital in Western Australia. On October 27,
Canakinumab Antiinflammatory Thrombosis Out- 2016, patient recruitment was expanded with the
come Study (CANTOS) involving patients with a inclusion of 30 centers of the Dutch Network for
history of myocardial infarction and an elevated Cardiovascular Research in the Netherlands.
baseline level of C-reactive protein; the results Enrollment was completed by December 3, 2018.
showed that the risk of recurrent cardiovascular The design of the trial has been published previ-
events was lower among those who received ously.13 The trial protocol, available with the full
canakinumab than among those who received text of this article at NEJM.org, was approved by
placebo.6 However, in another trial, a clinical a centralized institutional review board in each
benefit with methotrexate was not observed in participating country. An independent data and
patients with chronic coronary disease.7 safety monitoring board reviewed cumulative safe-
Colchicine is an antiinflammatory drug orig- ty data to safeguard the well-being of the pa-
inally extracted from the autumn crocus (Colchi- tients. Full details of the trial organization and a
cum autumnale) and was used by the ancient Greeks list of the trial sites and investigators are provided
and Egyptians. In contrast to selective inhibition in the Supplementary Appendix, also available at
of interleukin-1β by canakinumab, colchicine NEJM.org.
has broad cellular effects that include inhibition The academic and clinical investigators de-
of tubulin polymerization and alteration of leu- signed the study, collected and managed the data,
kocyte responsiveness.8-10 In the Colchicine Car- performed the statistical analyses, and drafted
diovascular Outcomes Trial (COLCOT) involving the manuscript. The funders had no role in the
patients who had a myocardial infarction within design or writing of the protocol and statistical
30 days before enrollment, the percentage of those analysis plan; in the selection or monitoring of
who had the composite end point of cardiovascu- the participating sites; in the enrollment or fol-
lar death, resuscitated cardiac arrest, myocardial low-up of patients; in the distribution or admin-
infarction, stroke, or urgent hospitalization for istration of the trial drug or placebo; in the col-
angina leading to coronary revascularization was lection, storage, analysis, and interpretation of
lower among those who received 0.5 mg of col- the data; in the drafting of the manuscript; or in
chicine once daily than among those who received the decision to submit the manuscript for publi-
placebo.11 cation. The trial drug and matching placebo were
In an earlier trial of low-dose colchicine donated by Aspen Pharmacare in Australia and by
(LoDoCo) involving patients with chronic coro- Tiofarma in the Netherlands. The members of
nary disease, we found that the risk of acute the steering committee and the trial statisticians
cardiovascular events was lower among those who had unrestricted access to the data and vouch for
received 0.5 mg of colchicine once daily than the completeness and accuracy of the data and
among those who did not receive colchicine.12 analyses and for the fidelity of the trial to the
This was an open-label trial involving only 532 protocol.
patients, and the results required confirmation.
Accordingly, we conducted an investigator-initi- Trial Population
ated, randomized, controlled, double-blind, event- Patients 35 to 82 years of age were eligible if they
driven trial of low-dose colchicine (LoDoCo2) to had any evidence of coronary disease on invasive
determine whether 0.5 mg of colchicine once coronary angiography or computed tomography
daily, as compared with placebo, prevents cardio- angiography or a coronary-artery calcium score
vascular events in patients with chronic coronary of at least 400 Agatston units on a coronary-artery
disease. calcium scan. Patients were required to have been

2 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 Colchicine in Patients with Chronic Coronary Disease

in a clinically stable condition for at least 6 months farction; ischemia-driven coronary revasculariza-
before enrollment. Patients were not eligible if tion; spontaneous myocardial infarction; ischemic
they had moderate-to-severe renal impairment, stroke; death from any cause; and cardiovascular
severe heart failure, severe valvular heart disease, death. The list of end points, including the pri-
or known side effects from colchicine. Renal mary end point, was revised several times during
function was defined on the basis of the Kidney the trial; the latest and final revision took place
Disease: Improving Global Outcomes (KDIGO) in January 2020 before the data were unblinded.
Clinical Practice Guideline for Acute Kidney In- End points were adjudicated by a committee whose
jury.14 A full list of the inclusion and exclusion members were unaware of the trial-group assign-
criteria is provided in Table S1 in the Supplemen- ments. Additional end points and definitions are
tary Appendix. All the patients provided written provided in Table S2.
informed consent to participate.
Statistical Analysis
Run-in, Randomization, and Follow-up The trial was designed to accrue a minimum of
After signing the informed-consent form, eligible 331 primary end-point events and to have a mini-
patients entered an open-label run-in phase for mum follow-up of 1 year. On the basis of a tar-
1 month, during which time they received 0.5 mg get enrollment of 6053 patients in the open-label
of colchicine once daily. At the end of the open- run-in phase, with 5447 undergoing randomiza-
label run-in phase, the patients who were in sta- tion after screening, we estimated that the trial
ble condition and had no unacceptable side effects, would have more than 90% power, at a two-sided
had adhered to the open-label colchicine regimen, alpha level of 0.05, to detect a 30% lower rate
and remained willing to continue participation (i.e., a hazard ratio of 0.70) of a primary com-
were randomly assigned in a 1:1 ratio to receive posite end-point event in the colchicine group
0.5 mg of colchicine once daily or matching pla- than in the placebo group, assuming a 10% rate
cebo. Randomization was performed in a double- of discontinuation of colchicine or placebo and an
blind manner with the use of a computerized annual rate of the primary end point in the con-
algorithm, with stratification according to coun- trol group of 2.6%. Details of the statistical meth-
try. Clinical evaluations were scheduled before the ods are provided in the Supplementary Appendix.
run-in phase, at the time of randomization, and The main analysis was based on the time from
at 6-month intervals until the completion of the randomization to the first occurrence of any com-
trial. All follow-up assessments were performed ponent of the primary composite end point. If the
in person, if possible, or by telephone. The trial incidence of the primary end point was signifi-
regimens were continued until the completion of cantly lower in the colchicine group than in the
the trial. Moreover, clinical follow-up was con- placebo group (P<0.05), then the ranked second-
tinued until the date of trial completion regard- ary end points were tested in a hierarchical fashion
less of premature discontinuation of colchicine at a significance level of 0.05 in order to preserve
or placebo. the alpha level. The original protocol did not in-
clude a plan to adjust for multiple testing; hierar-
End Points chical testing was included in the protocol in
The primary end point was a composite of car- January 2020 before the data were unblinded to
diovascular death, spontaneous (nonprocedural) be consistent with the new guidelines for statis-
myocardial infarction, ischemic stroke, or ische­ tical reporting in the Journal.15
mia-driven coronary revascularization. Second- The main analysis was performed according
ary end points, which were tested in hierarchical to the intention-to-treat principle and included
fashion, were ranked in the following order: the all adjudicated end-point events that occurred be-
composite of cardiovascular death, spontaneous tween randomization and the end-of-trial date in
myocardial infarction, or ischemic stroke (key all patients who had undergone randomization,
secondary end point); the composite of sponta- regardless of whether they adhered to their as-
neous myocardial infarction or ischemia-driven signed regimen. Cause-specific hazard ratios in
coronary revascularization; the composite of car- the colchicine group, as compared with the pla-
diovascular death or spontaneous myocardial in- cebo group, and 95% confidence intervals were

n engl j med  nejm.org 3

The New England Journal of Medicine
Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

An exploratory sensitivity analysis of the pri-

6528 Patients were enrolled mary end point in the on-treatment data set was
in the open-label run-in phase also performed. The on-treatment analysis was
performed in patients who had received at least
1006 (15.4%) Did not undergo one dose of colchicine or placebo, with additional
randomization
611 Had perceived side censoring of data 30 days after the last dose was
effects received (in addition to the data that were cen-
395 Had other reasons
sored according to the rules for the main inten-
tion-to-treat analysis). Analyses of the primary
5522 Underwent randomization and secondary end points were also performed
with the use of Fine and Gray subdistribution
hazard models to account for competing risks.

2762 Were assigned to receive 2760 Were assigned to receive

R e sult s
colchicine and were included placebo and were included
in the main analysis in the main analysis Patients
Among the 6528 patients who provided written
informed consent and started the open-label run-
44 (1.6%) Had data that were censored 34 (1.2%) Had data that were censored in period, 5522 underwent randomization and
early in the intention-to-treat early in the intention-to-treat
end-point analysis owing to death end-point analysis: 33 died
5478 received at least one dose of colchicine or
from noncardiovascular causes from noncardiovascular causes placebo (Fig. 1 and Table S3). Among the 1006
and 1 was lost to follow-up
Median duration of follow-up, 29.0 mo
patients (15.4%) who had started the run-in period
(IQR, 20.7–45.7) Median duration of follow-up, 28.1 mo but did not undergo randomization, the most
(IQR, 20.3–43.5)
common reason was gastrointestinal upset (in
437 patients).
The baseline characteristics of the patients
289 (10.5%) Permanently discontinued 291 (10.5%) Permanently discontinued
colchicine prematurely placebo prematurely were well balanced between the trial groups (Ta-
22 Never received a dose 22 Never received a dose ble 1). The mean (±SD) age of the patients was
95 Had perceived side effects 93 Had perceived side effects
124 Withdrew from the trial 125 Withdrew from the trial 66±8.6 years, and 846 (15.3%) were female; 11.7%
48 Were withdrawn by physician 51 Were withdrawn by physician of the patients were current smokers, and 18.2%
or had intercurrent illness or had intercurrent illness
had diabetes. Most patients (4658 [84.4%]) had a
history of acute coronary syndrome; in 68.2% of
Figure 1. Enrollment, Randomization, and Follow-up. the patients, the acute event had occurred more
Premature permanent discontinuation of colchicine or placebo was deter- than 24 months before randomization. At base-
mined to have occurred if colchicine or placebo was permanently discontin- line, the patients were well treated with respect to
ued more than 30 days before the occurrence of a primary end-point event,
the occurrence of noncardiovascular death, or the regular end-of-trial date,
chronic coronary disease, with 99.7% taking an
whichever came first. antiplatelet agent or an anticoagulant, 96.6% a
lipid-lowering agent, 62.1% a beta-blocker, and
71.7% an inhibitor of the renin–angiotensin sys-
determined with the use of Cox proportional- tem. Distribution of baseline characteristics ac-
hazards models, stratified according to country. cording to country is provided in Table S4.
If an end-point event had not occurred, follow-up
data were censored at the time of the competing Adherence and Follow-up
risk event (death from noncardiovascular causes The date of the last follow-up contact with a pa-
or death from any cause, as appropriate) or at the tient was February 17, 2020. The database was
end of the trial. Two-sided P values for superiority locked on May 22, 2020. The primary end-point
were calculated with the use of log-rank tests, as status was available for all but one patient. The
governed by the rules of hierarchical testing. The median duration of follow-up was 28.6 months
prespecified subgroup analyses were performed (interquartile range, 20.5 to 44.4). In each trial
with the use of the Cox proportional-hazards group, 10.5% of the patients permanently discon-
method. tinued colchicine or placebo prematurely (Fig. 1).

4 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 Colchicine in Patients with Chronic Coronary Disease

Table 1. Characteristics of the Trial Patients at Baseline.*

Colchicine Placebo
Characteristic (N = 2762) (N = 2760)

Age — yr 65.8±8.4 65.9±8.7

Female sex — no. (%) 457 (16.5) 389 (14.1)
Country — no. (%)
Australia 951 (34.4) 953 (34.5)
The Netherlands 1811 (65.6) 1807 (65.5)
Current smoker — no. (%)† 318 (11.5) 330 (12.0)
Hypertension — no. (%) 1421 (51.4) 1387 (50.3)
Diabetes — no. (%)
Patients receiving any treatment for diabetes 492 (17.8) 515 (18.7)
Patients dependent on insulin 140 (5.1) 147 (5.3)
Renal function — no. (%)‡
Stage 1 or 2 2614 (94.6) 2602 (94.3)
Stage 3a 148 (5.4) 158 (5.7)
Prior acute coronary syndrome — no. (%) 2323 (84.1) 2335 (84.6)
Time since last acute coronary syndrome — no. (%)
≤24 mo 753 (27.3) 726 (26.3)
>24 mo 1570 (56.8) 1609 (58.3)
Prior coronary revascularization — no. (%) 2301 (83.3) 2320 (84.1)
Coronary-artery bypass grafting 319 (11.5) 391 (14.2)
Percutaneous coronary intervention 2100 (76.0) 2077 (75.3)
History of atrial fibrillation — no. (%) 332 (12.0) 317 (11.5)
History of gout — no. (%) 220 (8.0) 226 (8.2)
Medication use — no. (%)
Single antiplatelet therapy 1849 (66.9) 1852 (67.1)
Dual antiplatelet therapy 638 (23.1) 642 (23.3)
Anticoagulant 342 (12.4) 330 (12.0)
No antiplatelet agent or anticoagulant 4 (0.1) 11 (0.4)
Statin 2594 (93.9) 2594 (94.0)
Ezetimibe 551 (19.9) 522 (18.9)
Any lipid-lowering agent 2670 (96.7) 2665 (96.6)
Renin–angiotensin inhibitor 1995 (72.2) 1965 (71.2)
Beta-blocker 1692 (61.3) 1735 (62.9)
Calcium-channel blocker 633 (22.9) 611 (22.1)

* Plus–minus values are means ±SD.

† Information on smoking was missing for 21 patients.
‡ Stage 1 refers to an estimated glomerular filtration rate of at least 90 ml per minute per 1.73 m2 of body-surface area
(normal or high), stage 2 to a rate of 60 to 89 ml per minute per 1.73 m2 (mildly decreased), and stage 3a to a rate
of 45 to 59 ml per minute per 1.73 m2 (mildly to moderately decreased). Stages are based on the Kidney Disease:
Improving Global Outcomes (KDIGO) Clinical Practice Guideline for Acute Kidney Injury.14

n engl j med  nejm.org 5

The New England Journal of Medicine
Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

A Primary End Point in the placebo group, with incidence rates of 2.5
100 and 3.6 events, respectively, per 100 person-years
20
90 Hazard ratio, 0.69 (95% CI, 0.57–0.83) (hazard ratio, 0.69; 95% confidence interval [CI],
P<0.001
Placebo 0.57 to 0.83; P<0.001) (Figs. 2 and 3). This treat-
80 15
ment effect was consistent in the on-treatment
Cumulative Incidence (%)

70
10
analysis (Fig. S1 and Table S5).
60 A key secondary composite end-point event of
Colchicine
50
5
cardiovascular death, spontaneous myocardial in-
40 farction, or ischemic stroke occurred in 115 pa-
30 0
tients (4.2%) in the colchicine group and in 157
20
0 12 24 36 48 60 patients (5.7%) in the placebo group, with inci-
dence rates of 1.5 and 2.1 events, respectively, per
10
100 person-years (hazard ratio, 0.72; 95% CI,
0
0 12 24 36 48 60 0.57 to 0.92; P = 0.007) (Figs. 2 and 3). In the pre-
Months since Randomization
specified hierarchical testing of the ranked sec-
ondary end points, the rates of the first five
No. at Risk
Placebo 2760 2655 1703 821 590 161 secondary end points, including spontaneous myo-
Colchicine 2762 2685 1761 890 629 166 cardial infarction, were significantly lower in the
colchicine group than in the placebo group
B Key Secondary End Point
(Fig. 3). Colchicine did not result in a lower in-
100
20 cidence of death from any cause than placebo
90 Hazard ratio, 0.72 (95% CI, 0.57–0.92)
P=0.007 (73 vs. 60 fatalities; incidence, 0.9 vs. 0.8 events,
80 15 respectively, per 100 person-years; hazard ratio,
Cumulative Incidence (%)

70 1.21; 95% CI, 0.86 to 1.71). Fine and Gray sub-

60
10 Placebo distribution hazard ratios were virtually identi-
50 cal to the cause-specific hazard ratios (Table S6).
5 Colchicine
40
Additional End Points
30 0
0 12 24 36 48 60
A primary composite end-point event as defined
20 in the first LoDoCo trial (sudden cardiac death,
10 nonfatal out-of-hospital cardiac arrest, acute coro-
0 nary syndrome [myocardial infarction or unstable
0 12 24 36 48 60
angina irrespective of revascularization], or ath-
Months since Randomization erosclerotic ischemic stroke) occurred in 201 pa-
No. at Risk tients in the colchicine group and in 290 patients
Placebo 2760 2694 1760 863 625 174
Colchicine 2762 2714 1787 913 651 176 in the placebo group, with incidence rates of 2.7
and 4.0 events, respectively, per 100 person-years
Figure 2. Cumulative Incidence of the Primary End Point and the Key Sec- (hazard ratio, 0.67; 95% CI, 0.56 to 0.81) (Fig. 3).
ondary End Point. The results with respect to the occurrence of new
Panel A shows the cumulative incidence of the primary composite end onset or first recurrence of atrial fibrillation,
point of cardiovascular death, myocardial infarction, ischemic stroke, or
ischemia-driven coronary revascularization, and Panel B shows the cumula-
deep-venous thrombosis or pulmonary embolism
tive incidence of the key secondary composite end point of cardiovascular or both, and new-onset diabetes did not differ
death, myocardial infarction, or ischemic stroke. The inset in each panel significantly between the trial groups.
shows the same data on an enlarged y axis.
Adverse Events
Noncardiovascular deaths occurred more fre-
Primary and Secondary End Points quently among the patients who received colchi-
The primary composite end-point event of cardio- cine than among those who received placebo,
vascular death, spontaneous myocardial infarc- with incidence rates of 0.7 and 0.5 events, respec-
tion, ischemic stroke, or ischemia-driven coronary tively, per 100 person-years (hazard ratio, 1.51;
revascularization occurred in 187 patients (6.8%) 95% CI, 0.99 to 2.31) (Table 2 and Table S7). We
in the colchicine group and in 264 patients (9.6%) observed similar rates of cancer diagnosis, hos-

6 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 Colchicine in Patients with Chronic Coronary Disease

pitalization for infection, hospitalization for pneu- 95% CI, 1.01 to 1.31). Dysesthesia was reported
monia, and hospitalization for a gastrointestinal in 143 patients (7.9%) in the colchicine group and
reason in the two trial groups, in both the inten- in 150 patients (8.3%) in the placebo group (cumu-
tion-to-treat analysis and the on-treatment anal- lative incidence ratio, 0.95; 95% CI, 0.76 to 1.18).
ysis (Table 2 and Table S8). Gout occurred in 38
patients (1.4%) in the colchicine group and in 95 Subgroup Analyses
patients (3.4%) in the placebo group (cumulative The effects of colchicine, as compared with pla-
incidence ratio, 0.40; 95% CI, 0.28 to 0.58). Neutro- cebo, on the primary end point were consistent in
penia and myotoxic effects were uncommon in the prespecified subgroups defined according to
both trial groups. Among the patients from the sex, age (>65 years vs. ≤65 years), smoking status
Netherlands, myalgia was reported in 384 (21.2%) (current vs. former or never), hypertension (yes vs.
in the colchicine group and 334 (18.5%) in the no), diabetes (yes vs. no), renal function (stage 1
placebo group (cumulative incidence ratio, 1.15; or 2 vs. stage 3a [stages are based on the KDIGO

Colchicine Placebo
End Point (N=2762) (N=2760) Hazard Ratio (95% CI) P Value
no. of no. of no. of no. of
patients events/100 patients events/100
(%) person-yr (%) person-yr
Primary end point
Cardiovascular death, myocardial 187 (6.8) 2.5 264 (9.6) 3.6 0.69 (0.57–0.83) <0.001
infarction, ischemic stroke,
or ischemia-driven
coronary revascularization
Secondary end points in ranked order
Cardiovascular death, myocardial 115 (4.2) 1.5 157 (5.7) 2.1 0.72 (0.57–0.92) 0.007
infarction, or ischemic stroke
Myocardial infarction or ischemia- 155 (5.6) 2.1 224 (8.1) 3.0 0.67 (0.55–0.83) <0.001
driven coronary revascularization
Cardiovascular death or myocardial 100 (3.6) 1.3 138 (5.0) 1.8 0.71 (0.55–0.92) 0.01
infarction
Ischemia-driven coronary revas- 135 (4.9) 1.8 177 (6.4) 2.4 0.75 (0.60–0.94) 0.01
cularization
Myocardial infarction 83 (3.0) 1.1 116 (4.2) 1.5 0.70 (0.53–0.93) 0.01
Ischemic stroke 16 (0.6) 0.2 24 (0.9) 0.3 0.66 (0.35–1.25) 0.20
Death from any cause 73 (2.6) 0.9 60 (2.2) 0.8 1.21 (0.86–1.71)
Cardiovascular death 20 (0.7) 0.3 25 (0.9) 0.3 0.80 (0.44–1.44)
Additional end points
The primary end point in the first 201 (7.3) 2.7 290 (10.5) 4.0 0.67 (0.56–0.81)
LoDoCo trial
New onset or first recurrence in atrial 126 (4.6) 1.7 148 (5.4) 2.0 0.84 (0.66–1.07)
fibrillation or atrial flutter
Deep-vein thrombosis or pulmonary 17 (0.6) 0.2 16 (0.6) 0.2 1.06 (0.53–2.10)
embolism or both
Any myocardial infarctions 85 (3.1) 1.1 117 (4.2) 1.5 0.72 (0.54–0.95)
New-onset diabetes 34 (1.2) — 49 (1.8) — 0.69 (0.44–1.06)
0.3 0.5 1.0 2.0

Colchicine Better Placebo Better

Figure 3. Key End Points and their Components.

The cause-specific hazard ratios were estimated from Cox proportional-hazard regression analysis with death from noncardiovascular
causes or death from any cause as a competing risk event. Myocardial infarction refers to spontaneous (nonprocedural) myocardial in-
farction. The primary end point in the first low-dose colchicine (LoDoCo) trial was a composite of sudden cardiac death, nonfatal out-of-
hospital cardiac arrest, acute coronary syndrome (myocardial infarction or unstable angina irrespective of revascularization), or athero-
sclerotic ischemic stroke. Any myocardial infarctions refers to either spontaneous or procedural myocardial infarctions. The ratio for
new-onset diabetes is presented as a cumulative incidence ratio because time-to-event data were not collected. The size of the data
points is inversely proportional to the precision (the standard error of the log of the hazard ratios or cumulative incidence ratio) of the
estimates, with larger data points representing more precise estimates. The testing hierarchy for statistical significance was broken at
the end point of ischemic stroke.

n engl j med  nejm.org 7

The New England Journal of Medicine
Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 2. Adverse Events in the Intention-to-Treat Population.*

Hazard Ratio or
Colchicine Placebo Cumulative Incidence
Event (N = 2762) (N = 2760) Ratio (95% CI)

no. of patients/ no. of events/100 no. of patients/ no. of events/100

total no. (%) person-yrs total no. (%) person-yrs
Noncardiovascular death 53/2762 (1.9) 0.7 35/2760 (1.3) 0.5 1.51 (0.99–2.31)
Diagnosis of cancer 120/2762 (4.3) 1.6 122/2760 (4.4) 1.6 0.98 (0.76–1.26)
Hospitalization for infection 137/2762 (5.0) 1.8 144/2760 (5.2) 1.9 0.95 (0.75–1.20)
Hospitalization for pneumonia 46/2762 (1.7) 0.6 55/2760 (2.0) 0.7 0.84 (0.56–1.24)
Hospitalization for gastrointestinal reason 53/2762 (1.9) 0.7 50/2760 (1.8) 0.7 1.06 (0.72–1.56)
Gout 38/2762 (1.4) — 95/2760 (3.4) — 0.40 (0.28–0.58)
Neutropenia 4/2762 (0.1) — 3/2760 (0.1) — NR
Myotoxic effects† 3/2762 (0.1) — 3/2760 (0.1) — NR
Myalgia‡ 384/1811 (21.2) — 334/1807 (18.5) — 1.15 (1.01–1.31)
Dysesthesia: numbness or tingling‡ 143/1811 (7.9) — 150/1807 (8.3) — 0.95 (0.76–1.18)

* Hazard ratios are reported for noncardiovascular death, diagnosis of cancer, hospitalization for infection, hospitalization for pneumonia,
and hospitalization for gastrointestinal reason; cumulative incidence ratios are reported for gout, myalgia, and dysesthesia because time-to-
event data were not collected for these end points. Cumulative incidence ratios are not reported (NR) for neutropenia and myotoxic effects
because of the low numbers of events.
† Rhabdomyolysis occurred in one patient in the colchicine group, who had a full recovery.
‡ Data were collected for the Netherlands cohort only. Reporting of these adverse events was requested by the Medicines Evaluation Board in
the Netherlands when the trial was expanded to include patients from that country.

Clinical Practice Guideline for Acute Kidney In- cine than with placebo. The observed between-
jury14]), prior acute coronary syndrome (yes vs. no), group difference in the incidence of noncardio-
prior coronary revascularization (yes vs. no), atrial vascular death was not significant, as shown by
fibrillation (yes vs. no), statin dose (high dose vs. the 95% confidence interval, and could have been
low or moderate dose), and ezetimibe use (yes vs. due to chance, although the hazard ratio of 1.51
no). When examined according to country, the is of potential concern. The individual causes of
effect of colchicine, as compared with placebo, on death (Table S7) do not permit a clear interpreta-
the primary end point was directionally consis- tion of this finding. In the COLCOT trial, non-
tent but appeared to be quantitatively larger in cardiovascular death occurred in 23 patients who
Australia than in the Netherlands (Fig. S2). received colchicine and in 20 patients who re-
ceived placebo.11
Among the patients who were enrolled in the
Discussion
run-in phase, 15.4% did not undergo random-
Among patients with chronic coronary disease, ization; the most common reason was gastro-
most of whom were already receiving proven sec- intestinal upset. Among the patients who had
ondary prevention therapies, 0.5 mg of colchicine successfully completed the run-in phase and had
once daily resulted in a 31% lower relative risk undergone randomization, 10.5% in each trial
of cardiovascular death, spontaneous myocardi- group permanently discontinued colchicine or
al infarction, ischemic stroke, or ischemia-driven placebo prematurely. Our results provide no evi-
coronary revascularization (the primary end point) dence for a clinically important interaction be-
than placebo, with a hazard ratio of 0.69. The tween low-dose colchicine and high-dose statins,
effects of colchicine appeared to be consistent which were used by 3413 patients (61.8%) in the
across each component of the primary end point trial. Myalgia, which was assessed only in the
and all secondary composite end points. Netherlands cohort, was common in both trial
The incidence rates of death from any cause and groups, although it was reported more frequently
noncardiovascular death were higher with colchi- in the colchicine group.

8 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 Colchicine in Patients with Chronic Coronary Disease

The CANTOS trial provided evidence suggest- factor control. We did not routinely measure
ing that inflammation plays a causal role in the C-reactive protein levels or other laboratory indi-
pathogenesis of cardiovascular disease and relat- cators of inflammation at baseline, and we can-
ed complications and that interventions to miti- not explore the effects of treatment according to
gate inflammation may reduce the risk of cardio- inflammatory state at baseline. However, the ef-
vascular events.6 Our results with colchicine are fects of treatments were consistent across the
consistent with those obtained in the first majority of clinical subgroups examined.
LoDoCo trial and the COLCOT trial and provide The results of our trial show that among pa-
further support for the potential benefits of anti- tients with chronic coronary disease, most of
inflammatory therapy in patients with coronary whom were already receiving proven secondary
disease.11,12 The magnitude of benefit of low-dose prevention therapies, the occurrence of cardio-
colchicine in the LoDoCo2 trial is consistent with vascular events was significantly lower with low-
that shown in previous trials of antiinflammatory dose colchicine than with placebo.
therapy and in previous trials of other secondary Supported by the National Health Medical Research Council
prevention therapies, including lipid-lowering, of Australia, a grant from the Sir Charles Gairdner Research Ad-
visory Committee, the Withering Foundation the Netherlands,
blood pressure–lowering, and antithrombotic the Netherlands Heart Foundation, the Netherlands Organiza-
therapies, and a benefit was observed in the tion for Health Research and Development, and a consortium of
patients who were already receiving therapy with Teva, Disphar, and Tiofarma in the Netherlands.
Disclosure forms provided by the authors are available with
these agents.1,3,16-18 Furthermore, the benefits the full text of this article at NEJM.org.
emerged early and continued to accrue through- A data sharing statement provided by the authors is available
out the trial, with no suggestion of attenuation with the full text of this article at NEJM.org.
We thank all the patients for their participation in the trial; the
of benefit during up to 5 years of treatment. trial investigators and coordinators at all the centers; and the trial
The LoDoCo2 trial has several limitations. monitors and staff from GenesisCare, including Penny Buczec,
The percentage of women in the trial was lower Denny Craig, Karen Doherty, Louise Ferguson, Louise Nidorf, and
Karen Youl, from the Heart and Vascular Research Institute of
than would be expected given the percentage of Sir Charles Gairdner Hospital, including Louise Ferguson, and
women with chronic coronary disease in the gen- from the Dutch Network for Cardiovascular Research, including
eral population. We did not collect blood-pressure Marjelle van Leeuwen (project manager), Ingrid Groenenberg and
Glentino Rodriguez for data management, Erik Stroes, Max Silvis,
or lipid levels at baseline or during the trial, and and Tim de Vries for medical review, and Petra Bunschoten and
we cannot report outcomes according to risk- Wendy Tousain for site monitoring.

Appendix
The authors’ full names and academic degrees are as follows: Stefan M. Nidorf, M.D., Aernoud T.L. Fiolet, M.D., Arend Mosterd, M.D.,
John W. Eikelboom, M.D., Astrid Schut, M.Sc., Tjerk S.J. Opstal, M.D., Salem H.K. The, M.D., Xiao‑Fang Xu, M.D., Mark A. Ireland,
M.D., Timo Lenderink, M.D., Donald Latchem, M.D., Pieter Hoogslag, M.D., Anastazia Jerzewski, M.D., Peter Nierop, M.D., Alan
Whelan, M.D., Randall Hendriks, M.D., Henk Swart, M.D., Jeroen Schaap, M.D., Aaf F.M. Kuijper, M.D., Maarten W.J. van Hessen,
M.D., Pradyot Saklani, M.D., Isabel Tan, M.D., Angus G. Thompson, M.D., Allison Morton, M.D., Chris Judkins, M.D., Willem A. Bax,
M.D., Maurits Dirksen, M.D., Marco M.W. Alings, M.D., Graeme J. Hankey, M.D., Charley A. Budgeon, Ph.D., Jan G.P. Tijssen, Ph.D.,
Jan H. Cornel, M.D., and Peter L. Thompson, M.D.
The authors’ affiliations are as follows: GenesisCare Western Australia (S.M.N., X.-F.X., M.A.I., D.L., A.W., R.H., P.S., I.T., A.G.T.,
A. Morton, P.L.T.), the Heart and Vascular Research Institute (S.M.N., P.L.T.) and the Department of Neurology (G.J.H.), Sir Charles
Gairdner Hospital, and the Faculty of Health and Medical Sciences (G.J.H., P.L.T.) and the School of Population and Global Health
(C.A.B.), University of Western Australia, Perth, the Department of Cardiology, Fiona Stanley Hospital, Murdoch, WA (C.J.), and the
Harry Perkins Institute of Medical Research, Nedlands, WA (P.L.T.) — all in Australia; the Dutch Network for Cardiovascular Research
(A.T.L.F., A. Mosterd, A.S., S.H.K.T., T.L., P.H., A.J., P.N., H.S., J.S., A.F.M.K., M.W.J.H., M.D., M.M.W.A., J.H.C.), the Netherlands
Heart Institute (A.T.L.F.), and the Department of Cardiology (A.T.L.F.) and the Julius Center for Health Sciences and Primary Care (A.
Mosterd, M.M.W.A.), University Medical Center Utrecht, Utrecht, the Department of Cardiology, Meander Medical Center, Amersfoort
(A. Mosterd), the Departments of Cardiology (T.S.J.O., M.D., J.H.C.) and Internal Medicine (W.A.B.), Northwest Clinics, Alkmaar, the
Department of Cardiology, Radboud University Medical Center, Nijmegen (T.S.J.O., J.H.C.), the Department of Cardiology, Treant Zorg-
groep, Hoogeveen, Emmen, and Stadskanaal (S.H.K.T.), the Department of Cardiology, Zuyderland Medical Center, Heerlen and Sittard
(T.L.), the Department of Cardiology, Isala Diaconessenhuis, Meppel (P.H.), the Department of Cardiology, Gelre Hospitals, Apeldoorn
(A.J.), the Department of Cardiology, Franciscus Hospital (P.N.), and Cardialysis (J.G.P.T.), Rotterdam, the Department of Cardiology,
D&A Research and Genetics, Sneek (H.S.), the Department of Cardiology, Amphia and Breda (J.S., M.M.W.A.), the Department of
Cardiology, Spaarne Hospital, Haarlem and Hoofddorp (A.F.M.K.), the Department of Cardiology, Green Heart Hospital, Gouda
(M.W.J.H.), and the Department of Cardiology, Amsterdam UMC, Amsterdam (J.G.P.T.) — all in the Netherlands; and the Department
of Medicine, McMaster University, Hamilton, ON, Canada (J.W.E.).

n engl j med  nejm.org 9

The New England Journal of Medicine
Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

References
1. Sabatine MS, Giugliano RP, Keech AC, 8. Leung YY, Hui LLY, Kraus VB. Colchi- 14. Kellum JA, Lameire N, Aspelin P, et al.
et al. Evolocumab and clinical outcomes cine — update on mechanisms of action Kidney Disease: Improving Global Out-
in patients with cardiovascular disease. and therapeutic uses. Semin Arthritis comes (KDIGO) clinical practice guide-
N Engl J Med 2017;​376:​1713-22. Rheum 2015;​45:​341-50. line for acute kidney injury. Kidney Int
2. Bhatt DL, Steg PG, Miller M, et al. 9. Dalbeth N, Lauterio TJ, Wolfe HR. Suppl 2012;​2:​1-138.
Cardiovascular risk reduction with icosa- Mechanism of action of colchicine in the 15. Harrington D, D’Agostino RB Sr, Gat-
pent ethyl for hypertriglyceridemia. N Engl treatment of gout. Clin Ther 2014;​ 36:​ sonis C, et al. New guidelines for statisti-
J Med 2019;​380:​11-22. 1465-79. cal reporting in the Journal. N Engl J Med
3. Eikelboom JW, Connolly SJ, Bosch J, 10. Cronstein BN, Molad Y, Reibman J, 2019;​381:​285-6.
et al. Rivaroxaban with or without aspirin Balakhane E, Levin RI, Weissmann G. 16. Cholesterol Treatment Trialists’
in stable cardiovascular disease. N Engl J Colchicine alters the quantitative and (CTT) Collaboration. Efficacy and safety
Med 2017;​377:​1319-30. qualitative display of selectins on endo- of more intensive lowering of LDL choles-
4. Ross R. Atherosclerosis — an inflam- thelial cells and neutrophils. J Clin Invest terol: a meta-analysis of data from
matory disease. N Engl J Med 1999;​340:​ 1995;​96:​994-1002. 170,000 participants in 26 randomised
115-26. 11. Tardif J-C, Kouz S, Waters DD, et al. trials. Lancet 2010;​376:​1670-81.
5. Libby P, Loscalzo J, Ridker PM, et al. Efficacy and safety of low-dose colchicine 17. Ettehad D, Emdin CA, Kiran A, et al.
Inflammation, immunity, and infection after myocardial infarction. N Engl J Med Blood pressure lowering for prevention of
in atherothrombosis: JACC review topic of 2019;​381:​2497-505. cardiovascular disease and death: a sys-
the week. J Am Coll Cardiol 2018;​72:​2071- 12. Nidorf SM, Eikelboom JW, Budgeon tematic review and meta-analysis. Lancet
81. CA, Thompson PL. Low-dose colchicine 2016;​387:​957-67.
6. Ridker PM, Everett BM, Thuren T, for secondary prevention of cardiovascular 18. Antithrombotic Trialists’ (ATT) Col-
et al. Antiinflammatory therapy with disease. J Am Coll Cardiol 2013;​61:​404-10. laboration. Aspirin in the primary and
canakinumab for atherosclerotic dis- 13. Nidorf SM, Fiolet ATL, Eikelboom JW, secondary prevention of vascular disease:
ease. N Engl J Med 2017;​377:​1119-31. et al. The effect of low-dose colchicine in collaborative meta-analysis of individual
7. Ridker PM, Everett BM, Pradhan A, et patients with stable coronary artery dis- participant data from randomised trials.
al. Low-dose methotrexate for the preven- ease: the LoDoCo2 trial rationale, design, Lancet 2009;​373:​1849-60.
tion of atherosclerotic events. N Engl J and baseline characteristics. Am Heart J Copyright © 2020 Massachusetts Medical Society.
Med 2019;​380:​752-62. 2019;​218:​46-56.

10 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at WEILL CORNELL MEDICAL COLLEGE LIBRARY on August 31, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.