This study aims to evaluate the impact of simvastatin on the bone density of postmenopausal women with type 2 diabetes over a duration of 18 months, using a randomized controlled trial design. Aiming to recruit 240 patients, half of them will be randomly assigned to receive simvastatin treatment, while the other half will receive ezetimibe, also a lipid-lowering agent with no known effect on bone. Bone density will be measured at the baseline and the end of the study for comparison of the changes between the simvastatin and the ezetimibe groups.

This is an investigator-initiated study. The principal investigator and the study team will be responsible for ensuring that the study is conducted in compliance with this protocol and the study data collected are verified against the relevant source documents.

All participants will undergo clinical and biochemical assessments at baseline of the trial. Participants will be seen by an endocrinologist at baseline and subsequent follow-up visits at 3, 6, 12 and 18 months respectively.

Study Procedures and Assessments Screening Procedures

Patients will be recruited from medical out-patient clinics, mainly from the primary health care clinics, in the Hong Kong West Cluster of the Hong Kong Hospital Authority. Consecutive participants who fulfil the inclusion and exclusion criteria are invited to participate in this randomized controlled trial after obtaining informed consent.

Clinical and biochemical assessments

Participants will attend a clinical assessment session at baseline after an overnight fast for at least 8 hours. Demographic data and medical history will be obtained using a standardized questionnaire. Personal and family history of fragility fractures (spine, hip, humerus, wrist and ankle) will be recorded. Important clinical risk factors of osteoporosis will be evaluated, including smoking, drinking, family history of fragility fractures, parental history of hip fractures, prior use and duration of hormonal replacement therapy, and levels of physical activity. The levels of physical activity will be assessed through the International Physical Activity Questionnaire (IPAQ). Daily calcium intake will be assessed using a semi-quantitative questionnaire. Body weight, body height and blood pressure (BP) will be measured. Hypertension is defined as BP ≥140/90 mmHg or the use of antihypertensive medications.

Fasting blood will be drawn for plasma glucose, HbA1c, insulin, lipid profile, albumin, calcium, phosphate, creatinine levels and eGFR. Patients with 25OHD levels <50 nmol/L will be given additional cholecalciferol 1000 units/day for 8 weeks followed by reassessment of 25-hydroxyvitamin D (25OHD) to ensure repletion.

Follow-up visits (3, 6, 12, 18 months after baseline visit)

During each follow-up visits, participants will have fasting blood tests checked for liver and renal function tests, creatine kinase, calcium and phosphate, fasting glucose, HbA1c and lipid profile.

Glycaemic control will be managed according to the standard of care. During follow-up visits, tolerance to lipid-lowering therapy will be assessed. Persistent elevation of liver enzymes (alanine aminotransferase and aspartate aminotransferase) to >3 times the upper limit of normal, increase in creatine kinase with clinical signs and symptoms of muscle involvement, or drug intolerance will be excluded from the trial.

During follow-up visits, participants may have changes in clinical conditions and thus be indicated for anti-osteoporosis therapy, or higher-intensity statin. Those who develop incident hip or vertebral fractures, or osteoporosis, and thus require anti-osteoporosis therapy, will be excluded from the trial. Those who develop incident atherosclerotic cardiovascular disease, and thus require higher intensity statin, will also be excluded from the trial.

Withdrawal from study

Patients are informed orally and in writing that they are free to withdraw their participation in the study without bias or prejudice. A patient who decides to stop all therapeutic interventions will be followed by the clinical team for the usual medical care. The date of withdrawal and reason for discontinuation, if known, will be recorded in the medical record.

Statistical Analyses

All statistical analyses will be performed with using R package or IBM® SPSS® version 26. Data will be presented as mean± standard deviation (SD), 95% confidence interval (CI), median with 25th-75th percentile, and number with percentage as appropriate.

All analyses described in this section will be performed based on intention-to-treat analysis and repeated for the per-protocol analysis. Data not normally distributed will be logarithmically transformed before analysis. Between-group comparisons are done with t-test or Mann-Whitney U test for continuous data, and chi-square or Fisher exact test for categorical variables. Two-sided p-value <0.05 will be considered statistically significant.

Baseline demographics and anthropometric parameters (age, BMI), laboratory parameters (HbA1c, LDL-cholesterol, serum creatinine), bone health assessment, including bone mineral density (BMD), trabecular bone score (TBS), bone turnover markers and fracture risk assessment (FRAX) score, clinical risk factors of osteoporosis, family history of fragility fractures (partly reflecting involvement of genetics in bone health), and concomitant medications (various classes of anti-hypertensives and anti-diabetic agents) will be compared between simvastatin and ezetimibe arms.

If applicable, adjustments for covariates may be employed in Analysis of Covariance (ANCOVA). Any adjustments will be pre-specified in the description of the analysis.

BMD is measured with a dual-energy X-ray absorptiometry (DXA) machine (Hologic QDR 4500, Waltham, MA, USA). BMD change over the 18-month period will be calculated for each participant. The BMD change over the lumbar spine (LS), femoral neck (FN), total hip (TH) and distal radius from baseline will be the dependent variable in the ANCOVA model.

Sample size calculation

Sample size calculation is performed with R package (1). According to the study on the impact of lovastatin on TH BMD by Safaei et al. (2), a difference in BMD change over 18-month study period between lovastatin-treated and control groups was 0.098 g/cm2 for TH, with a pooled standard deviation of 0.13 g/cm2. The estimated effect size of TH BMD change over the 18-month study period is 0.75 (95% CI: 0.2-1.3). Assuming the lower CI for the estimated effect size of 0.2, the study would require a sample size of 96 in each group (i.e. total sample size of 192), to achieve 90% power and a level of significance of 5%, for declaring that simvastatin is superior to the active comparator at a 0.02 g/cm2 margin of superiority. To ensure an analysis size of 96 individuals, an overall sample size of 120 individuals per treatment arm will be recruited, anticipating a drop-out rate of approximately 20%.

Protocol Deviations

Protocol deviations are defined as any instance in which the patient or clinical team does not perform scheduled activities/tasks such as missed visits and missed tests and evaluations. Protocols deviations will be reported on the patient's chart.

Inclusion Criteria:

• Chinese women
• Aged 50 to 74 years (inclusive);
• Type 2 diabetes Mellitus;
• Postmenopausal: confirmed with the last menstrual period >12 months by the time of recruitment into the study

Exclusion Criteria:

• Entry HbA1c >8.5%;
• On thiazolidinedione;
• Baseline LDL-cholesterol >3.0 mmol/L, triglyceride >5.0 mmol/L, or known familial hypercholesterolaemia;
• History of hip and/or clinical vertebral fractures;
• Osteoporosis by BMD criteria on DXA;
• On anti-osteoporosis therapy within the prior 2 years;
• Evidence of secondary causes of osteoporosis including Cushing's syndrome, acromegaly, thyrotoxicosis, primary hyperparathyroidism, metabolic bone diseases (e.g. osteomalacia), and systemic glucocorticoid treatment;
• Evidence of documented ASCVD, which includes previous acute coronary syndrome, stable angina, coronary revascularization, stroke and transient ischaemic attack and peripheral arterial disease;
• On lipid-lowering therapy within the prior 2 years;
• Known contraindications to statin therapy including allergy, intolerance and significant liver function abnormality (alanine aminotransferase level >3 times upper limit of normal);
• Significant diabetic complication(s): pre-proliferative / proliferative diabetic retinopathy, diabetic maculopathy, overt proteinuria, estimated glomerular filtration rate (eGFR) <30 mL/min;
• Inability to give an informed consent

The computer-generated sequence will be supplied by the study statistician, independent of the investigators. To facilitate double-blinding, placebo tablets with the same shapes as simvastatin 10mg and ezetimibe 10mg, respectively.