Relationship Between Insulin Resistance and Statin Induced Type 2 Diabetes, and Integrative Personal Omics Profiling

1. SIGNIFICANCE Statins and the risk of T2D: Statin treatment is associated with an increase in incident T2D.1-4 5 Mechanism of statin-induced T2D: It is unclear whether statins increase the risk of T2D by decreasing insulin action, secretion, or both. Several manuscripts have been published that substantially increase understanding of the link between statin use and incident T2D. Swerdlow, et al.2 based on evidence from genetic analysis and randomized trials, concluded that the increased risk of T2D noted with statins is at least "partially explained by HMG-coenzyme A reductase (HMGCR) inhibition." They also noted an association of weight gain with HMGCR variants in statin-treated patients, leading to the notion that decreases in insulin sensitivity contribute to statin-induced diabetes. In that context, Cederberg, at al.9 have shown in a large prospective study (n=8749 men) that participants treated with statins (n=2142) had a 46% increase in incident T2D, associated with a 24% decrease in insulin sensitivity and a 12% decrease in insulin secretion assessed by surrogate measures. Identifying those with at enhanced risk of statin-induced T2D: Studies of 3 randomized clinical trials with atorvastatin by David Waters' group 1,3,4 have demonstrated that "baseline fasting glucose, body mass index, hypertension, and fasting triglycerides were independent predictors of T2D." These abnormalities form a cluster attributed to insulin resistance.10 Since insulin resistance is a predictor of developing T2D, it seems likely that the more insulin resistant the individuals are before treatment, the greater is their risk to for statin-induced T2D. In that context, relatively little attention has been given to the role that metabolic heterogeneity in patients with elevated LDL-C concentrations might play in statin-induced T2D. Specifically, subjects with elevated LDL-C concentrations, whose plasma triglyceride (TG) concentrations are also elevated, are insulin resistant, hyperinsulinemic, and glucose intolerant as compared to those with isolated LDL-C levels. As such, this subset of patients with elevated LDL-C concentrations can be viewed as being at a "tipping point," and any adverse effect of statins on insulin action and/or secretion, irrespective of how mediated, places them at enhanced risk to develop statin-induced diabetes. Indeed, we have shown (Kohli et al)1 that patients with both insulin resistance (as estimated by high TGs) and prediabetes are at particularly high risk of statin-induced T2D. We seek to address 2 important unanswered questions: Do statins primarily affect insulin resistance or insulin secretion?; Are there subsets of individuals at highest risk of statin-induced T2D?
2. RATIONALE, HYPOTHESIS Rationale: T2D develops when insulin resistant individuals cannot maintain the degree of compensatory hyperinsulinemia needed to maintain normal glucose tolerance. However, significant fundamental questions remain. For example, what is the cellular/molecular link between statin treatment and changes in insulin action and secretion? This proposal is based on the premise that studying the effect of statins on insulin action and insulin secretion using "gold standard" methods will help determine if statins adversely affect the risk of T2D by increasing insulin resistance or decreasing insulin secretion. We use the insulin suppression test (with the read-out of steady-state plasma glucose, SSPG) to ascertain insulin sensitivity and the graded glucose infusion test (with the readout of insulin secretion rate, ISR) to ascertain insulin secretion both before and after statin treatment in non-diabetic individuals. We hypothesize that treatment with atorvastatin 40 mg/day for approximately 10 weeks will impair insulin sensitivity and/or insulin secretion and that this effect may be exacerbated in those with underlying insulin resistance. Thus, we plan to look at the effect of atorvastatin not only in all participants but also in subsets of individuals with baseline insulin resistance (which will be enriched for by recruiting volunteers with elevated plasma TG levels (≥150 mg/dL) at baseline. The rationale for this is that plasma TGs are a surrogate measure for insulin resistance with a modest correlation with the direct measure of insulin resistance (steady-state plasma glucose) measured by the insulin suppression test. Clinically, subjects with elevated TGs prior to statin treatment would have substantial clinical benefit from statins, and one of investigators' secondary goals is to demonstrate that a simple measurement of plasma TG concentration (as a surrogate for insulin resistance) can help identify those most at risk of statin induced derangements in glycemic control. Consequently, we propose to enroll nondiabetic volunteers at high-risk for T2D, free of known atherosclerotic cardiovascular disease (ASCVD), not receiving statins, eligible for statin therapy according to ACC/AHA (American College for Cardiology/American Heart Association) 2013 guidelines.11 We will also target recruitment efforts to enrich for subjects with plasma TG concentration ≥ 150 mg/dL to ensure that we enroll subjects across the range of insulin sensitivity. Hypothesis: We hypothesize that high intensity atorvastatin treatment for approximately 10 weeks will impair insulin sensitivity and/or insulin secretion and that this effect may be exacerbated in those with underlying insulin resistance.
3. STUDY DESIGN Sample Size

• We aim to recruit and retain 75 total participants in this study. Study Location
• Clinical and Translational Research Unit (CTRU) at 800 Welch Road, Palo Alto, CA 94304. Duration
• We anticipate that the entire study will take 4 - 5 years through the end of data analysis. Each eligible candidate who voluntarily consents to participate in the study will be active in the study for a total of 5 months from screening to the end of their last visit.

1. STUDY PROCEDURES Recruitment Preliminary recruitment strategies will include: Volunteers will be recruited from the San Francisco Bay Area through advertisements in newspapers, posted flyers, and the social networking site NextDoor as well as from the Preventive Cardiology Clinics at Stanford Health Care. Our goal is to ensure recruitment across a broad range of insulin sensitivity. Prior work from our group and others has shown that high plasma TG concentrations are associated with increased insulin resistance as assessed by reference measures. Therefore, we will target advertisements to enrich for individuals with high TG levels (> 150 mg/dL) as a surrogate for increased insulin resistance. Participant Visits and Procedures Potential participants will be screened initially when they call or email in response to recruitment ads, or a letter from their MD, describing the study as follows: Preliminary intake will occur over the phone. Visit 1 Screening visit. Visit 2: Oral Glucose Tolerance Test (OGTT): This test will take approximately 3 hours.
2. Glycemic status: Participants will be classified as having normal glucose tolerance (NGT), isolated impaired fasting glucose (IFG), isolated impaired glucose tolerance (IGT), or combined IFG/IGT12. In addition, the total integrated plasma glucose response during the OGTT will be calculated by the trapezoidal method (Glucose-AUC). ii) β-cell function: Calculations of the Insulin Secretion- Sensitivity Index-2 (ISSI-2) will be used to quantify insulin secretory function 13, if for some reason a graded glucose infusion study is not done. The ISSI-2 is a validated OGTT- derived method to measure of β-cell function, analogous to the disposition index obtained from the intravenous glucose tolerance test 14. It is calculated by multiplying the insulin secretory response during the OGTT (Insulin-AUC/Glucose-AUC) by insulin resistance (Matsuda index). The fasting and 2-hour glucose results will be discussed with the study participant and a copy of the results will be given to them. Visit 3. Graded Glucose Infusion, GGIT 15,16: This test will take approximately 6 hours. This test is designed to assess the ability of the pancreas to produce insulin in response to a graded glucose infusion. During this test, subjects will have two small IV catheters placed, one in each arm. One IV will be used for drawing samples and the other for the infusion of glucose. During the GGIT, continuous intravenous infusions of glucose will be given at progressively increasing rates: 1, 2, 3, 4, 5, 6, and 8 kg/min in six infusion periods of 40-min duration. Blood samples will be collected for measurements of glucose, insulin, and C-peptide concentrations at fasting and at 30 min and 40 min into each infusion period. The two values during the last 10 min of each infusion period will be averaged. The amount of blood taken for this test will 47.5 ml. Visit 4.Insulin Suppression test, IST 17,18: This test will take approximately 6 hours. This test is designed to determine whole body insulin sensitivity. Following an overnight fast, subjects will have an IV placed in each arm. One for collection of blood and the other for infusion with octreotide (0.27 μg/m2/min), insulin (32 mU/m2/min), and glucose (267 mg/m2/min) for 180 minutes. During the test, endogenous insulin is suppressed and all individuals are given the same concentration of insulin, based on their body surface area. Blood is drawn every 30 minutes for 150 minutes and then at 10-minute intervals from 150 to 180 minutes of the infusion to measure plasma glucose and insulin concentrations. The mean of the last four values is used as the steady-state plasma insulin (SSPI) and glucose (SSPG) concentrations for each individual. As SSPI concentrations are similar in all subjects during the IST, the SSPG concentration provides a direct measure of the ability of insulin to mediate disposal of an infused glucose load; the higher the SSPG concentration, the more insulin resistant the individual. Labs to check kidney and liver function plus a lipid panel and a urine pregnancy test (if appropriate) will be done at this visit. Blood drawn or IST will be 58.5 ml and 5 ml for SHC labs. Labs for iPOP will be drawn at this time and additional samples will be obtained for transcriptome, microbiome, metabolome, and proteome analysis in blood; nasal, tongue, skin surface swabs; urine; and stool. The study drug, atorvastatin 40 mg will be given to study participants, once all labs have been reviewed and participant qualifies. Visit 5-7: Visits will be every 2 weeks for a total of 10 weeks on study medication (statin). Participants will be assessed for any side effects or adverse events (AE) on the statin. Adherence to study medication will be assessed at each visit. Visit 8: Weight, vital signs, and OGTT described above. Visit 9: Repeat GGIT as described above. Visit 10: Repeat IST and iPOP lab testing and samples as described above. At the end of this visit, the statin will be stopped and the study participant would be scheduled for 4- and 8-week follow up visits. Visit 11: One month off statin study visit - weight, vital signs, and iPOP laboratory testing and samples will be done as described above. Visit 12: Last study visit - weight, vital signs, and iPOP laboratory testing and samples will be done as described above. Participants will be asked to fill out questionnaire about her/his physical activity status, food and eating habits, and stress at the time of each iPOP.
3. STATISTICAL CONSIDERATIONS

Based on our prior work19, we calculated that with 60 subjects, we would be able to detect an 8% change in SSPG concentration and an 8% change in ISRAUC after atorvastatin therapy with 80% power and two-side significance level of 5% using a paired samples t test. Thus, we estimated needing to enroll 75 subjects with an anticipated a dropout rate of 20%.

Summary statistics will be reported as median (interquartile range) or number (percent) of participants unless otherwise specified. Shapiro-Wilk tests will be used to assess normality of data, and variables that are not normally distributed will be log-transformed. Percent changes in variables will be calculated by the formula: [(end-of study value) - (baseline value) / baseline value] x 100. Paired samples t tests will be used to compare baseline and end-of-study means. One sample t tests will be employed to evaluate whether percent changes in variables are significantly different from zero (no change). Pearson correlation coefficients will be calculated to determine the strength of association between variables of interest. Prespecified subgroup analyses will be carried out by stratifying for insulin resistant versus insulin sensitive subjects. The SSPG concentration median will be used to define subjects as being insulin resistant or insulin sensitive. Insulin resistant and insulin sensitive group means will be compared by independent sample t tests and proportions by chi-square tests or Fisher's exact tests. Statistical analyses will be performed by using statistical software IBM SPSS version 26.0.