Metabotyping in the Postmenopausal Stage: Cross-Sectional Observational Study
Trial Snapshot
- Phase
- Not Applicable
- Status
- Completed
- Sponsor
- Enrollment
- 200
- Locations
- 1
- Primary Endpoint
- Metabolomics in serum
Study Overview
Brief Summary
Menopause is defined as the absence of menstrual periods for twelve consecutive months. Although the onset may vary, natural menopause occurs between the ages of 45 and 55 and is considered a stage in the aging process for women. Menopause is a stage strongly conditioned by hormonal modulations with effects on the cardiovascular system associated with abdominal obesity, insulin resistance, decreased energy expenditure, endothelial dysfunction, hypertension, and dyslipidemia. Furthermore, an increase in the production of proinflammatory cytokines involved in numerous pathologies such as osteoporosis has been observed.
The results of several studies suggest that intestinal microbiota (IM) profile may be related to menopause condition by several means, although the data are stil inconclusive.
Estrogen reduction leads to a progressive loss of bone density, a reduction in the bone formation/resorption balance and an increased risk of bone fractures among postmenopausal women. Recently, the alternative to estrogen therapies to reduce the risk of fractures are nutritional strategies fundamentally based on the use of probiotics, whose effect are associated with beneficial modulations of IM.
SHE-HEALTH is a study in which, in a cohort of postmenopausal women, metabolomics, transcriptomics and metagenomics will be combined with the analysis of usual anthropometric and clinical biomarkers and also with genetic and epigenetic analyses to identify population groups (clusters). This study will allow establishing solid scientific bases to define, in future projects, effective nutritional strategies based on group nutrition in postmenopausal women.
The main objective of the present study is to obtain clusters of postmenopausal women, identifying metabotypes (similar metabolic profiles) and enterotypes (similar IM profiles), and combining complementary variables such as classical anthropometric, biochemical and clinical biomarkers.
The secondary objectives of the study are to characterize: 1) The genetic profile of the study cohort; 2) The epigenetic profile of the study cohort; 3) The gene expression profile of the study cohort.
Detailed Description
Cross-sectional observational study in which samples of blood, faeces, urine, hair and hair follicles will be collected to characterize the metabolic profile, intestinal microbiota (IM), gene expression profile, genetic and epigenetic profile of postmenopausal women. Data on lifestyle habits, anthropometric measurements and nutritional and hormonal status will also be collected.
The study will be conducted in a cohort of 200 postmenopausal women.
Each volunteer will make 2 visits:
- A pre-selection visit (to check inclusion/exclusion criteria) (V0) and, if the inclusion criteria are met,
- A study visit (V1) in which samples will be collected from faeces, urine, blood, hair and hair follicles.
In V1, the participants must present themselves fasting for 8 hours to obtain blood and urine collected during the last 24 hours. In addition, during the visit the sample of hair and hair follicles will be collected. Participants are given a basic guide of healthy eating and lifestyle recommendations suitable for postmenopausal stage.
Study Design
- Study Type
- Observational
- Observational Model
- Cohort
- Time Perspective
- Cross Sectional
Eligibility Criteria
- Ages
- 40 Years to 63 Years (Adult)
- Sex
- Female
- Accepts Healthy Volunteers
- Yes
Inclusion Criteria
- •Women between 40 and 63 years old with amenorrhea for a period of time equal or greater than 12 months.
- •Without hormone replacement therapy.
- •Sign the informed consent.
Exclusion Criteria
- •Women diagnosed with diabetes (or serum glucose ≥ 126 mg/dL) or other chronic pathologies (coronary, cardiovascular, celiac disease, Crohn's disease and chronic kidney diseases (or serum creatinine ≥ 1.5 mg/dL).
- •Women taking medications prescribed for hypertension and dyslipidemia. Women who have consumed during the week prior to start to start of the study anti-inflammatory drugs.
- •women with chronic gastrointestinal problems.
- •Women with a body mass index (in kg/m2) <18 or ≥
- •Women who are participating in another clinical trial or following a prescribed diet for any reason, including weigh loss, during the last month.
- •Women who consume more than 14 alcoholic beverages per week.
- •Women current smokers.
Outcomes
Primary Outcomes
Metabolomics in serum
Time Frame: At day 1
Non-targeted metabolomics of serum samples measured using proton nuclear magnetic resonance. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Metabolomics in erythrocytes
Time Frame: At day 1
Non-targeted metabolomics of erythrocytes samples measured using proton nuclear magnetic resonance. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Metagenomics in faeces
Time Frame: At day 1
Faecal intestinal microbiota analysis will be done by 16sRNA sequencing. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum FSH levels
Time Frame: At day 1
Serum FSH levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum inhibin B levels
Time Frame: At day 1
Serum inhibin B levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum CTX-I levels
Time Frame: At day 1
Serum CTX-I levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum AMH levels
Time Frame: At day 1
Serum AMH levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum triglycerides levels
Time Frame: At day 1
Serum triglycerides levels will be measured by Cobas Mira Plus autoanalyzer (Roche Diagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum AST levels
Time Frame: At day 1
Serum AST levels will be measured by Cobas Mira Plus autoanalyzer (RocheDiagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Metabolomics in urine
Time Frame: At day 1
Non-targeted metabolomics of urine samples measured using proton nuclear magnetic resonance. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum TNFalpha levels
Time Frame: At day 1
Serum TNFalpha levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum TRAP5b levels
Time Frame: At day 1
Serum TRAP5b levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum LDL-cholesterol levels
Time Frame: At day 1
Serum LDL-cholesterol levels will be measured by Cobas Mira Plus autoanalyzer (Roche Diagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum HDL-cholesterol levels
Time Frame: At day 1
Serum HDL-cholesterol levels will be measured by Cobas Mira Plus autoanalyzer (Roche Diagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum ALT levels
Time Frame: At day 1
Serum ALT levels will be measured by Cobas Mira Plus autoanalyzer (RocheDiagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum IL-6 levels
Time Frame: At day 1
Serum IL-6 levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum BALP levels
Time Frame: At day 1
Serum BALP levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum PINP levels
Time Frame: At day 1
Serum PINP levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum 17beta E2 levels
Time Frame: At day 1
Serum 17beta E2 levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum testosterone levels
Time Frame: At day 1
Serum testosterone levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum SHBG levels
Time Frame: At day 1
Serum SHBG levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum glucose levels
Time Frame: At day 1
Serum glucose levels will be measured by Cobas Mira Plus autoanalyzer (Roche Diagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum insulin levels
Time Frame: At day 1
Serum insulin levels will be measured by Cobas Mira Plus autoanalyzer (Roche Diagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Homeostatic Model Assessment from Insulin Resistance index (HOMA-IR)
Time Frame: At day 1
HOMA-IR will be calculated using serum glucose and insulin levels. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum hsCRP levels
Time Frame: At day 1
Serum hsCRP levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum osteocalcin levels
Time Frame: At day 1
Serum osteocalcin levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum total cholesterol levels
Time Frame: At day 1
Serum total cholesterol levels will be measured by Cobas Mira Plus autoanalyzer (Roche Diagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum uric acid levels
Time Frame: At day 1
Serum uric acid levels will be measured by Cobas Mira Plus autoanalyzer (RocheDiagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Urine F2-isoprostanes levels
Time Frame: At day 1
Urine F2-isoprostanes levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum creatinine levels
Time Frame: At day 1
Serum creatinine levels will be measured by Cobas Mira Plus autoanalyzer (RocheDiagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Serum urea levels
Time Frame: At day 1
Serum urea levels will be measured by Cobas Mira Plus autoanalyzer (RocheDiagnostics Systems, Madrid, Spain). Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Urine 8-OHdG levels
Time Frame: At day 1
Urine 8-OHdG levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Urine NTX levels
Time Frame: At day 1
Urine NTX levels will be measured by human ELISA kits. Data will be analysed together with the other primary outcomes for cluster identification. Data will be scaled using unit variance scaling. Principal components analysis, Partial Least-Squares Discriminant Analysis and hierarchical clustering will be used to identify clusters and to detect differences among metabotypes. The quality of the model will be judged by the goodness-of-fit parameter, the predictive ability parameter and cross-validation test.
Secondary Outcomes
- Height(At day 1)
- Waist circumference to height ratio(At day 1)
- Body weight(At day 1)
- Body mass index(At day 1)
- Blood pressure (in mmHg)(At day 1)
- Transcriptomics analysis in total blood.(At day 1)
- DNA methylation analysis in total blood.(At day 1)
- Waist circumference(At day 1)
- Dietary intake(At day 1)
- Body composition(At day 1)
- MicroRNAs analysis in total blood.(At day 1)
- Transcriptomics analysis in hair follicles.(At day 1)