Effect of Metformin on Carotid Atherosclerosis for Non-diabetic Patients With PsA (CAMPA Study)

Effect of Metformin on Carotid Atherosclerosis for Non-diabetic Patients With Psoriatic Arthritis - a Pilot Randomized Controlled Trial

Background: Psoriatic arthritis (PsA) is a common chronic inflammatory disease with a prevalence up to 670 every 100,000 subjects. Patients with PsA has an increased risk of cardiovascular disease (CVD) which is one of the major causes of death. The investigators hypothesize that metformin in combination of a treat-to-target (T2T) strategy aiming at tight disease control is more effective in preventing progression of subclinical arthrosclerosis than T2T strategy alone in non-diabetic PsA patients. Objective: To investigate the vascular effects of metformin in PsA patients without diabetes mellitus. The metabolic and anti-inflammatory roles of metformin will also be explored. Study design: This is a 1-year, single-centered, pilot, open-labelled, randomized controlled trial. A total of 24 enrolled patients with PsA being followed at the Prince of Wales Hospital rheumatology clinics will be recruited and randomized to either metformin group or control group in a 1:1 ratio. Participants randomized to the metformin group will be instructed to take 500 mg metformin daily for 1 week before titrating up to twice a day (one with the morning meal, one with the evening meal) to reduce gastro-intestinal adverse events. Expected outcomes: The data from this study will support that there will be significant difference in the proportion of subjects with carotid plaque progression between the metformin group and control group over a period of 1 year.

Psoriatic arthritis (PsA) is a common chronic inflammatory disease with a prevalence up to 670 every 100,000 subjects. Patients with PsA has an increased risk of cardiovascular disease (CVD) which is one of the major causes of death. The investigators hypothesize that metformin in combination of a treat-to-target (T2T) strategy aiming at tight disease control is more effective in preventing progression of subclinical arthrosclerosis than T2T strategy alone in non-diabetic PsA patients. All participants will be followed up at the Prince of Wales Hospital and receive tight-control treatment aiming at achieving minimal disease activity (MDA). It will be adjusted according to a standardized protocol based on the European League Against Rheumatism (EULAR) recommendation and the Hong Kong guideline on the use of biologics. Patients will be treated with methotrexate monotherapy and then other conventional synthetic disease modifying anti-rheumatic drugs (csDMARDs) or biologic DMARDs if they fail to achieve the pre-determined target. Assessment of efficacy will be performed every 3-6 months. If a patient does not achieve the desired treatment goal within 3-6 months of therapy, therapy will be stepped up, unless the patient declines or a toxic effect precludes this approach. Prednisolone at a dose ≤10mg/day will be allowed during the study period. Intra-articular steroid or local steroid injection to enthesitis and dactylitis are allowed during the study but will be forbidden in the 4 weeks before assessment. The details of the treatments received will be recorded. Participants randomized to the metformin group will be instructed to take 500 mg metformin daily for 1 week before titrating up to twice a day (one with the morning meal, one with the evening meal) to reduce gastro-intestinal adverse events. Although the minimum effective dose for cardiovascular protection with metformin has not been determined, relatively low dose metformin has been shown to be effective for blood glucose control with reduced risk of gastrointestinal side-effects in the Asian population. Patients will be given the necessary number of metformin at each visit. At the following visits, surplus medication will be returned to the investigator. Compliance is calculated as a percentage, based on the number of tablets returned. Anti-hypertensive, anti-platelet and lipid-lowering medications can be initiated or titrated at the discretion of the treating physicians. The details of their usage will be documented. The following clinical and laboratory variables will be assessed at each visit: erythrocyte sedimentation rate, C-reactive protein, bone profile, number of swollen joints (0-66), number of tender joints (0-68), visual analogue scale (VAS) for pain (0-100 mm, 100 mm=most pain), VAS for patient's global assessment (0-100 mm, 100 mm=worst score) and VAS for physician's global assessment (0-100 mm, 100 mm=worst score). The Maastricht Ankylosing Spondylitis Enthesitis Score, the Leeds Enthesitis Index, the Spondyloarthritis Research Consortium of Canada Enthesitis Index, number of digits with dactylitis, Bath ankylosing spondylitis disease activity index (for patients with axial involvement), body surface area and Psoriasis Activity and Severity Index are used to measure joint and skin disease activity, and will be recorded every visit. The number of damaged joints will be assessed yearly. Functional disability is assessed by the Health Assessment Questionnaire (HAQ) Disability Index (0-3=most functional disability) every visit. Minimal disease activity (MDA) will be employed as the treatment target of PsA. The MDA criteria assess 7 domains, with the following cutoffs: tender joint count ≤1, swollen joint count ≤1, enthesitis count ≤1, skin score ≤1 (or body surface area ≤3%), function score ≤0.5 (measured by the HAQ), patient's global assessment ≤20 on a 100-mm VAS, and patient-reported pain ≤15 on a 100-mm VAS. If 5 of the 7 cutoffs for these domains are met, then the patient is classified as having MDA. Achieving MDA results in significant benefits in articular disease and can distinguish clearly active treatment from placebo. The investigators have previously shown that achieving sustained MDA, which is believed to be more stringent, but not other composite target, was associated with a protective effect in subclinical atherosclerosis and arterial stiffness progression.

Patients randomized to metformin group will be instructed to take 500 mg metformin daily for 1 week before titrating up to twice a day

Difference in the proportion of subjects with carotid plaque progression between the metformin group and control group

Carotid plaques will be assessed at baseline and 12 months using a high-resolution B mode ultrasound machine (Philips EPIQ7) by a single experienced sonographer blinded to all clinical information using a 30-MHz linear vascular probe (Philips L12-3 broadband linear array transducer). Plaque progression is defined as an incident plaque in a segment without plaque before or an increase in number of plaque. The primary outcome will be analyzed using multivariate logistic regression model. In univariate analysis, potential covariates including age, sex, traditional cardiovascular risk factor, disease activity and use of statin or biologic or targeted synthetic disease modifying anti-rheumatic drugs would be included. Variables which are significantly different between the two groups in the univariate analyses and are biologically plausible will be adjusted in the multivariate logistic regression model to confirm the metformin treatment effect.

The CIMT will be measured offline in the distal common carotid artery, bulb, and proximal internal carotid artery using dedicated software (Philips Xcelera Cardiology Enterprise Viewer Client 4). The mean and maximal IMT values of 6 arterial segments will be calculated for further analysis. The secondary outcomes on other carotid and coronary atherosclerosis parameters, multivariate linear regression or logistic regression model similar to the primary outcome analysis would be used with adjustment for covariates.

For TPA, the plane for measurement of each plaque will be chosen by reviewing the video of the scan to find the largest extent of plaque as seen on the longitudinal view. TPA will be recorded as the sum of the areas of all plaques in the right and left carotid arteries. Reading of the ultrasound scans obtained at baseline and follow-up will be performed concurrently by a single reader who is aware of the temporal order of the images but is blinded to all clinical data. The secondary outcomes on other carotid and coronary atherosclerosis parameters, multivariate linear regression or logistic regression model similar to the primary outcome analysis would be used with adjustment for covariates.

Coronary computer tomography angiogram (CCTA) scans will be performed at baseline and 12 months with 256-rows dual source multi-detector CT (Siemens Somatom Drive). CAC will be quantified by the Agatston method (normal = 0, higher = worse score). The images will be analyzed by an experienced radiologist blinded to all clinical data. The secondary outcomes on other carotid and coronary atherosclerosis parameters, multivariate linear regression or logistic regression model similar to the primary outcome analysis would be used with adjustment for covariates.

CCTA scans will be performed at baseline and 12 months with 256-rows dual source multi-detector computer tomography (CT) (Siemens Somatom Drive). Coronary arteries will be standardized to the American Heart Association 15-segment model. Non-calcified plaque volumes will be measured (CardIQ Xpress 2.0 Reveal, General Electic). The images will be analyzed by an experienced radiologist blinded to all clinical data. The secondary outcomes on other carotid and coronary atherosclerosis parameters, multivariate linear regression or logistic regression model similar to the primary outcome analysis would be used with adjustment for covariates.

CCTA scans will be performed at baseline and 12 months with 256-rows dual source multi-detector computer tomography (CT) (Siemens Somatom Drive). Coronary arteries will be standardized to the American Heart Association 15-segment model. Lesions rendering over 50% stenosis of the lumen will be considered as obstructive. For multiple plaques, the most stenotic one will be recorded. The images will be analyzed by an experienced radiologist blinded to all clinical data. The secondary outcomes on other carotid and coronary atherosclerosis parameters, multivariate linear regression or logistic regression model similar to the primary outcome analysis would be used with adjustment for covariates.

The patients' body mass index (BMI) will be checked at baseline and after 1 year. Mann-Whitney U test will be used to evaluate the differences between the metformin and the placebo group at baseline as well as after 1 year.

The patients' fasting blood glucose will be checked at baseline and after 1 year. Mann-Whitney U test will be used to evaluate the differences between the metformin and the placebo group at baseline as well as after 1 year.

The patients' Hba1c will be checked at baseline and after 1 year. Mann-Whitney U test will be used to evaluate the differences between the metformin and the placebo group at baseline as well as after 1 year.

Changes in cholesterol level in terms of cholesterol ratio (total cholesterol/high density lipoprotein)

The patients' lipid profile will be checked at baseline and after 1 year. Mann-Whitney U test will be used to evaluate the differences between the metformin and the placebo group at baseline as well as after 1 year.

All adverse events will be reported and recorded. Occurrence of adverse effects will be compared by Chi-square test or fisher exact test.

The MDA criteria assess 7 domains: number of tender joints (0-68), number of swollen joints (0-66), visual analogue scale (VAS) for pain (0-100 mm, 100 mm=most pain), VAS for patient's global assessment (0-100 mm, 100 mm=worst score), tender entheseal point and Psoriasis Area and Severity Index(PASI)/body surface area(BSA). Functional disability is assessed by the Health Assessment Questionnaire (HAQ) (0-3=most functional disability). The MDA criteria have the following cutoffs: tender joint count ≤1, swollen joint count ≤1, enthesitis count ≤1, PASI ≤1 (or BSA ≤3%), function score ≤0.5 (measured by the HAQ), patient's global assessment ≤20 and patient-reported pain ≤15 on a 100-mm VAS. If 5 of the 7 cutoffs for these domains are met, then the patient is classified as having MDA. Chi-square test or fisher exact test will be used to evaluate the differences in proportion of patients with MDA between the metformin and the placebo group at 1 year.

Inclusion Criteria: fulfill the ClASsification criteria for Psoriatic ARthritis (CASPAR), are over 18 years old, have a Framingham Risk Score ≥ 10% (i.e. moderate to high CV risk) and have carotid plague on previous ultrasound. Exclusion Criteria: have prior therapy with metformin during the last 6 months; have pre-existing diabetes as defined by the World Health Organization (WHO) criteria: fasting plasma glucose values of ≥ 7.0 mmol/L, 2-h post-load plasma glucose ≥ 11.1 mmol/L, HbA1c ≥ 6.5% or a random blood glucose ≥ 11.1 mmol/L); have liver impairment: ALT more than or equal to 2.5×upper limits of normal; have renal impairment: serum creatinine levels more than or equal to 135µmol/L in males and more than or equal to 110µmol/L in females; have had ACS within the previous 3 months; have New York Heart Association functional class 3 or 4 heart failure; have uncontrolled angina; have 70% or more stenosis on initial CCTA; have clinically relevant current malignancy; are pregnant; breastfeeding or of childbearing potential but unwilling to use adequate contraception; are unable to give written informed consent; patients with moderate to heavy alcohol intake.