The objective of the proposed study is to investigate for the first time whether there are age-related changes in hepatic fat accumulation, a hallmark of NAFLD, in marmosets. We hypothesize an age-associated increase in hepatic steatosis and alterations in serum lipid profile in the marmoset model. To test our hypothesis, we propose the following Aims.

Aim 1) To investigate whether hepatic fat accumulation increases with age in marmosets. In this aim, we will measure liver and abdominal fat in young and old male and female marmosets using magnetic resonance imaging (MRI) and spectroscopy (MRS). We will also utilize diffusion-weighted imaging (DWI) which is an emerging tool to evaluate liver fibrosis. 

Aim 2) To elucidate whether there are changes in serum lipid profile with age in marmosets. A serum lipidomic profile will be determined and evaluated to look for significant changes in the lipids with aging. To investigate associations between hepatic fat accumulation and cardiovascular health, blood pressure measurements will also be conducted. 

Inhibitors of the sodium-glucose co-transporter (SGLT2) are FDA-approved for the treatment of type 2 diabetes (T2DM). Their mechanism of action involves lowering of blood glucose concentration secondary to increased glucose excretion of glucose by the kidney. These drugs also cause significant improvements in body weight, blood pressure and cardiac function. Based on these pleiotropic effects, including its calorie restriction-mimetic properties, we hypothesize that SGLT2 drugs will impact several markers related to aging, including reductions in oxidative damage to DNA and proteins, DNA methylation, advanced glycation end products-receptor for AGE (AGE-RAGE), cellular senescence, and improvements in mitochondrial function.

Aim 1: To determine whether SGLT2 inhibitors improve biomarkers of aging in older obese adults with pre-diabetes
Aim 2: To determine whether changes in aging-related biomarkers are linked to changes in glucose metabolism and healthspan.

Study Question: Does rapamycin improve age-related changes in cardiac compliance and reduce fibrosis/collagen?

Preliminary RC2-supported studies using cardiovascular magnetic resonance imaging (CMR) suggest that rapamycin treatment improves diastolic function in healthy older adults (see RC2). Now,  this DP will use CMR with late gadolinium enhancement (LGE) to evaluate the effects of 2 months of rapamycin (vs. placebo) on parameters related to cardiovascular aging in 20 healthy adults over 70 years old. CMR data will include measurements of global and regional ventricular systolic and diastolic function, and LGE measurements of myocardial extracellular volume to assess fibrosis. RC1 will conduct parallel studies in marmosets; from an ongoing study, Dr. Ikeno will quantify collagen and elastin in banked aorta and heart samples from young (<5 yrs.) and older (5-15 yrs.) marmosets previously treated with rapamycin (or placebo) for up to 3.5 years. RC3 will carry out the required statistical analyses. If positive, findings from this study will lay the foundation for a clinical trial in older adults with heart failure with preserved ejection fraction, a growing population with few treatment options.

Study Question: Can changes in the circulating lipidome be developed as a cross-species biomarker of aging, age-related disease, and functional decline?

Diverse lipid signaling pathways can modulate the aging process and systematic analyses of the total lipid structure – the lipidome – in clinically relevant samples can reveal novel mechanisms in aging biology, biomarkers for diagnosis, and targets for therapeutics. As an initial step, using samples provided from generally healthy marmosets (RC1) and humans (RC2) across the normal age range for both species, this DP will assess the effects of age on the plasma lipidome. RC1 will provide plasma from ~20 each young (2-5 yrs.), middle-aged (6-9 yrs.) and old (10+ yrs.) naturally aging marmosets. All animals will be phenotyped by our common battery and resilience assessment. RC3 will assist with statistical comparisons of effects of age on changes and test the extent to which the lipidome reflects health and functional status. Identification of similarities in the aging lipidome across species may elucidate important biomarker targets for geroscience. Reflecting the growing interest in this topic, NIA recently released RFA-AG-20-039, “Lipid Signaling in Healthspan and Longevity Regulation”.

Senolytic/senomorphic drugs hold promise for aging and aging-related diseases. However, clinical trials to evaluate these drugs will require sensitive and specific senescence biomarkers. The goal of this project is to lay the foundation for the development and evaluation of non-invasive measures of cellular senescence. The ongoing repository (STARR) will be leveraged to (i) link known markers of senescence [p16 in CD3+ cells, senescence associated secretory phenotype (SASP) gene expression, and b-gal staining] obtained from tissues (blood, skin, fat) with healthspan outcomes; and to (ii) identify novel senescence biomarkers. This DP will also leverage ongoing and future trials on drugs/interventions with senolytic/senomorphic activity (e.g. dasatinib, polyphenols, metformin, mTOR inhibitors, exercise, weight loss) to determine which biomarkers change with the intervention and can predict functional outcome measures. In the future, this DP will conduct earlyphase precision medicine research on senolytics. For example, it will evaluate whether transcriptomic profiling (by RC3) of adipose tissue obtained in vivo can be used to determine which senolytics are most effective in clearing senescent cells and reducing SASP using in vitro cell functional assays. We could then test if molecular profiling predicts in vitro and in vivo clearance of senescent cells and whether their clearance is linked with changes in putative peripheral (non-invasive) senescence biomarkers and healthspan-related outcomes.

RC3 is focused on designing aging-related trials that are rigorous, efficient, feasible, and based on solid preliminary data. This can be challenging because 1) trials with multimorbidity endpoints are novel; 2) biomarkers related to these endpoints are in development; and 3) treatment effect sizes are unknown.

Goal: Through this DP, we will create a database of aging-related clinical trials involving drug classes related to aging, multimorbidity endpoints, and aging-specific biomarkers. The database and research publication will include trials’ primary clinical endpoints, anticipated/realized effect sizes, sample sizes, inclusion/exclusion criteria, durations of treatments, classes of compounds, secondary endpoints and related effect sizes. 

Methods: RC3 will formally examine translational geroscience-focused trials (completed and in-progress) through a systematic review of the literature and clinicaltrials.gov. This online database of multimorbidity and disease-agnostic healthspan-extending trials will be freely available to all OAICs. The initial trial searches will focus on SA OAIC priority agents such as rapamycin, metformin, senolytics, and other compounds under study by OAIC investigators. This database will also include a meta-analytic perspective on the sensitivity to intervention of the assessed aging biomarkers so that investigators will be informed by empirical evidence in selecting cost-effective assays to measure treatment effects. Initial biomarker searches will focus on SA OAIC priority outcomes, namely frailty, epigenomic aging assays, and senescence markers. Article search criteria will be aided by a research librarian (funded by RC3). Abstraction will be done by Dr. Michalek (Project Lead) and Dr. Gelfond, with quality control and abstraction done in coordination with RC2 lead Dr. Espinoza. RC3 will record trial design consultations and note those consultations that use the meta-analytic database. The web-accessible database will allow for crowdsourcing feedback to evaluate accuracy and adapt search criteria. Reporting will comply with Preferred Reporting Items for Systemic Reviews and Meta-Analyses guidelines. This systematic review and meta-analysis will inform power calculations and primary/secondary outcome selection in future studies supported by RC3 of the SA OAIC as well as other scientists in the field.

Rationale: To help with recruitment of OAIC studies during the current grant cycle, RC3 investigator Dr. Alex Bokov used the i2b2 application and data warehouse containing de-identified electronic medical record (EMR) data for 1.7 million patients to pull data from diverse sources (Epic Clarity, Sunrise, IDX, etc.), seeking potential participants who meet trial inclusion/exclusion criteria. While this uncovered participants who met highly specific criteria, subjects were not always efficiently enrolled. Goal: This DP will use machine-learning methods to leverage information within the EMR and clinical trial operational databases to more efficiently identify eligible participants who are more likely to enroll in the trial. Methods: Dr. Zozus (Project Lead) and Dr. Gelfond will use machine learning to adaptively model the full i2b2 patient profiles to match participant characteristics with those associated with successful trial enrollment. Using machine-learning tools (KNN, SVM, LASSO, etc.), subjects more similar to enrolled participants will be prioritized for screening. The effectiveness of this algorithm will be measured by a changepoint analysis that compares the enrollment rate (proportion who successfully enroll) before and after project implementation and examines the accrual rates in specific randomization strata to minimize sampling bias. Efficiency will also be measured by in-person screenings per enrolled subjects. We hypothesize that adaptive cohort identification will enhance accrual rates. If our hypothesis is supported, this algorithm will be made available to other scientists in the OAIC network and broad scientific community.