Low dose chloroquine decreases insulin resistance in human metabolic syndrome but does not reduce carotid intima-media thickness

Abstract

Background: Metabolic syndrome, an obesity-related condition associated with insulin resistance and low-grade inflammation, leads to diabetes, cardiovascular diseases, cancer, osteoarthritis, and other disorders. Optimal therapy is unknown. The antimalarial drug chloroquine activates the kinase ataxia telangiectasia mutated (ATM), improves metabolic syndrome and reduces atherosclerosis in mice. To translate this observation to humans, we conducted two clinical trials of chloroquine in people with the metabolic syndrome.

Methods: Eligibility included adults with at least 3 criteria of metabolic syndrome but who did not have diabetes. Subjects were studied in the setting of a single academic health center. The specific hypothesis: chloroquine improves insulin sensitivity and decreases atherosclerosis. In Trial 1, the intervention was chloroquine dose escalations in 3-week intervals followed by hyperinsulinemic euglycemic clamps. Trial 2 was a parallel design randomized clinical trial, and the intervention was chloroquine, 80 mg/day, or placebo for 1 year. The primary outcomes were clamp determined-insulin sensitivity for Trial 1, and carotid intima-media thickness (CIMT) for Trial 2. For Trial 2, subjects were allocated based on a randomization sequence using a protocol in blocks of 8. Participants, care givers, and those assessing outcomes were blinded to group assignment.

Results: For Trial 1, 25 patients were studied. Chloroquine increased hepatic insulin sensitivity without affecting glucose disposal, and improved serum lipids. For Trial 2, 116 patients were randomized, 59 to chloroquine (56 analyzed) and 57 to placebo (51 analyzed). Chloroquine had no effect on CIMT or carotid contrast enhancement by MRI, a pre-specified secondary outcome. The pre-specified secondary outcomes of blood pressure, lipids, and activation of JNK (a stress kinase implicated in diabetes and atherosclerosis) were decreased by chloroquine. Adverse events were similar between groups.

Conclusions: These findings suggest that low dose chloroquine, which improves the metabolic syndrome through ATM-dependent mechanisms in mice, modestly improves components of the metabolic syndrome in humans but is unlikely to be clinically useful in this setting.Trial registration ClinicalTrials.gov (NCT00455325, NCT00455403), both posted 03 April 2007.

Keywords: Atheroma; Blood pressure; Carotid intima-media thickness; Chloroquine; Glucose disposal; Insulin sensitivity; JNK; Lipids; Metabolic syndrome.

Conflict of interest statement

Competing interestsThe authors declare that they have no competing interests.

Figures

Fig. 1

CONSORT statement flow diagram for subjects in dose escalation study

Fig. 2

Chloroquine dose escalation trial design and results for hyperinsulinemic-euglycemic clamping and lipids. a Diagram of the dose escalation protocol. A 5–7 week washout period was required between limbs to allow for hematologic recovery following the blood drawing associated with clamps. b Glucose disposal rates expressed as μmol glucose/kg body weight/min. Insulin infusion rates were 0, 56, 181, or 486 pmol/m2/min. c Hepatic glucose production expressed as (μmol glucose/kg body weight/min) per pmol/L insulin. d Hepatic insulin sensitivity expressed as percent suppression of glucose production at 56 pmol/m2/min. e Total, non-HDL, and LDL cholesterol at the end of each limb. f Triglycerides and HDL cholesterol at the end of each limb. Data represent mean ± SE. *P < 0.05 by Tukey’s test for multiple comparisons after repeated measures ANOVA

Fig. 3

CONSORT statement flow diagram for subjects in the yearlong randomized clinical trial

Fig. 4

Yearlong chloroquine trial design and results for carotid imaging. a Diagram of the vascular endpoint trial. After 12 months of placebo or chloroquine, both were stopped, then subjects returned at 24 months for limited additional studies. b CIMT results. Standard deviations (not included to simplify data presentation) and n values for these data follow. Chloroquine: 0, 0.76625 ± 0.17270 (n = 56); 6 months, 0.75827 ± 0.17634 (n = 56); 12 months, 0.75769 ± 0.16061 (n = 54); 24 months, 0.77450 ± 0.16540 (n = 53). Placebo: 0, 0.76900 ± 0.11653 (n = 53); 6 months, 0.76500 ± 0.13087 (n = 51); 12 months, 0.76833 ± 0.11487 (n = 50); 24 months, 0.77280 ± 0.13290 (n = 49). c MRI-determined contrast enhancement. P = 0.0697. d Representative images at baseline and at 12 months for the same vessels. e MRI-determined common carotid lumen area *P = 0.0033. f MRI-determined common carotid artery diameter. *P = 0.0019. Data represent mean ± SD in c, e, f. Analysis by mixed model testing in b, paired t tests in c, e, f

Fig. 5

Blood pressure and lipid responses in the yearlong chloroquine trial. a Diastolic blood pressure over 24 months determined by conventional testing. b–d Blood pressures over 12 months, determined by an ambulatory monitoring device. e–i Lipids and lipoproteins over 24 months. P values were determined by mixed model treatment of repeated measures

Fig. 6

Effects of placebo and chloroquine on potential mechanistic mediators. a Activated JNK in circulating monocytes at baseline and 12 months. *P = 0.0164; P = 0.3343 for placebo; paired t tests. b Plasma concentrations of the oxysterol cholestane-3β, 5α, 6β-triol at 12 months. *P = 0.0031; unpaired t test