Metabolic Phenotyping for Personalized Obesity Therapy

Establishment of a Specific Metabolic Phenotyping for the Personalization of Obesity Therapy in Morbidly Obese Individuals

This study aims to develop a simple, clinically applicable method for metabolic phenotyping to personalize obesity therapy in morbidly obese individuals. The underlying concept is that the way a person's resting metabolic rate (RMR) responds to a 24-hour fast can help distinguish between two metabolic phenotypes. Individuals with a "thrifty" metabolism show a significant drop in RMR during fasting, which may make them less responsive to conventional weight loss interventions. In contrast, those with a "spendthrift" metabolism exhibit little to no drop-or even a slight increase-in RMR, suggesting they may lose weight more readily.

The trial is designed as a prospective, single-center, longitudinal cohort study involving 20 morbidly obese patients (BMI >40 kg/m²) who are already participating in a multimodal obesity therapy program. The study is divided into three phases. In the baseline phase, participants undergo comprehensive screening, which includes physical examinations, blood tests, and body composition assessments. RMR is measured using indirect calorimetry both before and after a 24-hour fasting period, and a device (Lumen™) is used to assess whether the body is primarily burning carbohydrates or fats.

After the fasting measurements, participants perform a low-protein meal test by consuming a specially calibrated chocolate beverage. Their RMR is then monitored at several time points to determine the energy required for digestion. Following this, the study moves into the very-low-calorie diet (VLCD) phase, where participants consume approximately 800 kcal per day using formula meals tailored to meet their nutritional needs despite the calorie restriction. During this 12-week phase, changes in body weight, composition, and metabolic parameters are closely monitored.

The final phase of the study is a 12-week weight maintenance period, during which the focus is on sustaining the achieved weight loss. In addition to RMR and dietary assessments, advanced techniques such as metabolomics are employed. Blood, urine, and saliva samples are collected to analyze metabolic profiles and identify potential hormonal biomarkers-such as leptin, FGF21, and adrenaline-that could further differentiate the "thrifty" and "spendthrift" phenotypes. Body composition is also assessed using methods like bioimpedance analysis (BIA), quantitative magnetic resonance imaging (qMR), and air displacement plethysmography (BodPod).

By correlating the changes in RMR with metabolic and hormonal markers, the study tests the hypothesis that individuals with a marked RMR decrease during fasting (the "thrifty" phenotype) may experience less weight loss during a hypocaloric diet compared to those with minimal RMR change (the "spendthrift" phenotype). If validated, this approach could allow clinicians to predict weight loss outcomes more accurately and tailor obesity treatments to the individual's unique metabolic profile.

Study Overview

• Status: Recruiting
• Conditions: Obesity and Overweight

Detailed Description

This study is a prospective, single-center, longitudinal cohort trial designed to unravel the metabolic heterogeneity among morbidly obese individuals and to develop a clinically feasible method for metabolic phenotyping. The aim is to identify distinct metabolic responses to a controlled fasting period that may predict weight loss outcomes and guide personalized obesity therapy.

Potential participants are morbidly obese patients (BMI >40 kg/m²) already enrolled in a multimodal obesity therapy program at the University Hospital Schleswig-Holstein, Campus Kiel. The screening process includes a thorough medical history, physical examination, and laboratory testing to exclude confounding conditions that might affect energy expenditure or the interpretation of metabolic measurements.

In the baseline phase, enrolled subjects undergo extensive assessments to characterize their metabolic status. Body composition is evaluated using several complementary techniques: bioimpedance analysis (BIA), quantitative magnetic resonance imaging (qMR), and air displacement plethysmography (BodPod). These methods provide detailed insights into fat mass, lean mass, and overall body composition. Resting metabolic rate (RMR) is measured using indirect calorimetry, with subjects placed under a canopy (haubenkalorimeter) that records oxygen consumption and carbon dioxide production to yield precise energy expenditure data.

A critical component of the study is the evaluation of RMR before and after an extended fasting period. Initially, RMR is measured following a 12-hour overnight fast. Participants then complete a 24-hour fasting period, during which continuous monitoring-via devices such as a continuous glucose monitor-ensures compliance with the fasting protocol. A second RMR measurement is taken after the fasting period, and the percentage change in RMR is calculated. Based on previous research, a significant RMR reduction (≥5.7%) classifies a subject as having a "thrifty" metabolic phenotype, while a minimal reduction or slight increase (≥+3.8%) indicates a "spendthrift" phenotype. Subjects with intermediate changes are not categorized for the primary analysis.

Following the fasting assessments, participants undergo a low-protein meal test. They consume a standardized chocolate beverage calibrated to provide 50% of their baseline RMR in caloric content. Postprandial RMR is then measured at several intervals over a three-hour period to assess the thermic effect of food and the energy cost associated with digestion. This test helps to elucidate differences in nutrient oxidation and metabolic flexibility between the two phenotypes.

After completing the baseline phase, subjects enter a 12-week very-low-calorie diet (VLCD) phase. During this period, they consume approximately 800 kcal per day through nutritionally complete formula meals that are designed to maintain a balanced macronutrient profile despite severe caloric restriction. Weight, body composition, and metabolic parameters are monitored on a weekly basis to capture the effects of the dietary intervention. The hypothesis is that individuals with a "thrifty" metabolic response may lose less weight compared to those with a "spendthrift" response, owing to their reduced energy expenditure during fasting.

Following the VLCD phase, patients transition into a 12-week weight maintenance phase. In this period, the focus shifts to sustaining weight loss and monitoring long-term metabolic changes. RMR, body composition, and dietary intake continue to be assessed periodically. Participants also use an activity tracker to document daily physical activity, ensuring that variations in energy expenditure are accounted for in the analysis.

An innovative aspect of this study is the incorporation of metabolomic analysis to identify potential biomarkers linked to the metabolic phenotypes. Biological samples-including blood, urine, and saliva-are collected at various time points throughout the study. These samples undergo metabolomic profiling using advanced techniques such as liquid chromatography-mass spectrometry (LC/MS) and nuclear magnetic resonance (NMR) spectroscopy. The goal is to identify specific metabolites or patterns-such as variations in leptin, fibroblast growth factor 21 (FGF21), and catecholamines-that correlate with the magnitude of RMR change during fasting. This analysis may reveal novel biomarkers that predict weight loss responsiveness and metabolic health.

Data collection extends beyond metabolic measurements. Participants complete visual analog scales (VAS) to assess subjective feelings of hunger and satiety, providing insights into the behavioral dimensions of appetite regulation. In addition, continuous glucose monitoring and measurements of beta-hydroxybutyrate levels are used to verify fasting compliance and to ensure the accuracy of the metabolic assessments.

The study is managed by a multidisciplinary team led by Dr. Tim Hollstein, with contributions from specialists in endocrinology, diabetology, clinical nutrition, and metabolomics. Collaborations with experts from associated institutions enhance the methodological rigor and analytical capacity of the trial. Statistical power calculations, based on prior studies, indicate that approximately 20 subjects (10 per extreme phenotype group) are required to detect significant differences in weight loss outcomes. To accommodate variability and potential dropouts, an initial screening of around 80 individuals is planned.

Safety and compliance are paramount. The protocol includes rigorous monitoring for adverse events related to fasting, blood sampling, and the use of continuous monitoring devices. Subjects who do not meet fasting criteria-confirmed via continuous glucose readings or ketone measurements-are excluded from the primary analysis but continue to receive standard obesity therapy.

Ultimately, the study seeks to bridge the gap between complex laboratory-based measurements and practical clinical applications. By establishing a reliable and straightforward method for metabolic phenotyping, it aims to enable personalized obesity treatments that are tailored to an individual's unique metabolic profile. This approach could revolutionize obesity management by providing clinicians with predictive tools to optimize weight loss interventions and improve long-term metabolic outcomes.

• Study Type: Observational
• Enrollment (Estimated): 80

Contacts and Locations

• Study Contact: Name: Tim Hollstein, PD Dr. med.; Phone Number: +4943150022000; Email: tim.hollstein@uksh.de

Study Locations

• Germany
  • Schleswig-Holstein
    • Kiel, Schleswig-Holstein, Germany, 24105
      • Recruiting
      • University Hospital Schleswig-Holstein
      • Contact: Tim Hollstein, PD Dr. med.; Phone Number: +4943150022000; Email: tim.hollstein@uksh.de

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

The study population consists of adult men and women (ages 18 to 70) with a body mass index (BMI) above 40 kg/m² who are receiving care at the Department of Internal Medicine I at the University Hospital Schleswig-Holstein (UKSH), Campus Kiel. All participants are enrolled in a multimodal obesity therapy program that includes nutritional counseling, behavioral interventions, and exercise components. Eligible individuals must meet the specified inclusion criteria and have no conditions that could confound study results or compromise patient safety.

Description

Inclusion Criteria:

• Morbid obesity (BMI > 40 kg/m²)
• Currently undergoing a multimodal obesity therapy program at the Department of Internal Medicine I at the University Hospital Schleswig-Holstein (UKSH), Campus Kiel

Exclusion Criteria:

• Type 1 or Type 2 diabetes mellitus, or fasting blood glucose >125 mg/dL, or HbA1c >6.5%
• Conditions affecting appetite or energy expenditure (e.g., Cushing's syndrome, uncontrolled hyper-/hypothyroidism, diabetes mellitus)
• Gastrointestinal diseases that may impair nutrient absorption (e.g., inflammatory bowel disease, malabsorption syndromes, peptic ulcers)
• Psychiatric disorders that influence eating behavior (e.g., active depression, anorexia nervosa, bulimia nervosa, borderline personality disorder)
• Acute, unstable cardiovascular disease requiring hospitalization within the last 6 months (e.g., stent implantation)
• Cancer requiring treatment within the past 5 years
• Chronic kidney disease at Stage IV or worse according to NKF criteria
• Active infectious disease (e.g., HIV, hepatitis)
• Active nicotine use
• Drug use (e.g., amphetamines, cocaine, heroin, marijuana)
• Regular intense physical activity (≥1 hour of sport per day)
• Non-MRI-compatible metallic implants (e.g., joint replacements, metal plates)
• Pregnancy or breastfeeding
• Use of weight-loss medications
• Clinically relevant claustrophobia
• Any other condition not specified above that, in the judgment of the investigator, could interfere with study participation or compromise patient safety

Study Plan

How is the study designed?

Number of groups / cohorts

3

Cohorts and Interventions

Group / Cohort
Extreme Thrifty; Individuals with a fasting-induced decrease of RMR < -5.7%
Extreme Spendthrift; Individuals with a fasting-induced increase of RMR > +3.8%
No extreme phenotype; Individuals with fasting-induced RMR changes greater than -5.7% and lower than +3.8%

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change in Resting Metabolic Rate (RMR) From Baseline to 24 Hours After Fasting	RMR (measured in kcal/day) is assessed via indirect calorimetry at baseline (after an overnight fast) and again following a 24-hour fasting period. The primary metric is the absolute change in RMR (kcal/day) or percentage change from baseline. Participants who experience a significant RMR decrease ("thrifty" phenotype) are hypothesized to lose less weight during a subsequent hypocaloric diet than those with a minimal or increased RMR ("spendthrift" phenotype).	Baseline to Week 12

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change in Plasma FGF21 From Baseline to 24 Hours After Fasting	Fibroblast growth factor 21 (FGF21) is measured in plasma (e.g., pg/mL) at baseline (after an overnight fast) and again after 24 hours of fasting. The difference in FGF21 concentrations serves as a potential biomarker for the thrifty phenotype, correlating with a significant RMR drop. Higher FGF21 changes may indicate greater metabolic adaptation to fasting.	Baseline, after 24-hour fast, and throughout the 6-month study period (up to Week 26)
Meal-Induced Thermogenesis as a Predictor of Weight and Fat Mass Loss	Assesses whether the thermic effect of a standardized low-protein meal (meal-induced thermogenesis, MIT) can predict subsequent weight and fat mass changes. MIT is measured via indirect calorimetry at baseline after a 24-hour fast, and these data are correlated with weight and body composition outcomes during the 12-week very-low-calorie diet (VLCD) and the following 12-week weight maintenance phase.	Day 1 (immediately following 24-hour fast) for the low-protein meal test, with follow-up assessments at Week 12 (end of VLCD) and Week 26 (end of weight maintenance).
Metabolic Flexibility (Respiratory Quotient Changes)	Assesses whether "thrifty" individuals (significant RMR drop) show reduced metabolic flexibility, defined by a smaller change in respiratory quotient (RQ) after the 24-hour fast and after a low-protein meal challenge.	Baseline (Day 0), after 24-hour fast (Day 1), and during the low-protein meal test (Day 1)
Subjective Hunger After 24-Hour Fast	Evaluates whether participants with a significant RMR drop ("thrifty" phenotype) report greater hunger following a 24-hour fasting period compared to those with minimal or increased RMR. Hunger is measured using visual analog scales (VAS) at baseline and immediately after the fast.	Baseline (Day 0) and after 24-hour fast (Day 1)
Changes in Fat Mass During and After a Hypocaloric Diet	Determines whether "thrifty" individuals lose less fat mass during a 12-week very-low-calorie diet (VLCD) and whether they regain weight and fat mass more rapidly in the subsequent 12-week weight maintenance phase compared to "spendthrift" individuals. Body composition is measured via bioimpedance, BodPod, and qMR.	Baseline, Week 12 (end of VLCD), and Week 26 (end of maintenance phase)
Gut Microbiome Differences	Explores whether the gut microbiome composition differs between thrifty and spendthrift individuals at baseline and whether these differences persist or change following the VLCD and maintenance phases. Stool samples are collected for microbiome analysis.	Baseline, Week 12, and Week 26
Thermogenic and Appetite-Regulating Hormone Composite Index	A composite index derived from standardized concentrations of leptin, FGF21, free T3 (FT3), free T4 (FT4), growth hormone (GH), adiponectin, ghrelin, GLP-1, secretin, and insulin. Each hormone is measured in its relevant unit (e.g., pg/mL, ng/mL) at baseline, after a 24-hour fast, and at Weeks 12 and 26. The data are converted into z-scores and averaged to create a single composite value, where higher scores indicate higher overall hormone levels. This index is used to compare changes in thermogenic and appetite-regulating hormone activity between "thrifty" and "spendthrift" metabolic phenotypes over time.	Baseline, after 24-hour fast (Day 1), Week 12, and Week 26
Long-Term Change in Metabolic Phenotype	Evaluates whether the metabolic phenotype (thrifty vs. spendthrift) remains stable or changes following the 12-week VLCD and 12-week weight maintenance period. RMR is measured again at the end of the study to detect any persistent or altered metabolic adaptations.	Baseline, Week 12, and Week 26
Association of RMR Changes With Meal-Induced Thermogenesis	Investigates the correlation between the magnitude of RMR change after fasting and the thermic effect of a low-protein meal. Postprandial energy expenditure is measured via indirect calorimetry at multiple time points following meal ingestion.	Baseline (Day 0) and during low-protein meal test (Day 1)
Association Between CO₂ Production and Energy Expenditure	Pearson's correlation coefficient (dimensionless) is calculated between CO₂ production (e.g., mL/min) measured by a handheld device (Lumen™) and energy expenditure (kcal/day) measured by indirect calorimetry. Measurements are taken at baseline, after a 24-hour fast, and during the postprandial phase of a low-protein meal test. A higher positive coefficient indicates a stronger association between CO₂ production and total energy expenditure.	Baseline (Day 0), after 24-hour fast (Day 1), and throughout the 6-month study period
Correlation Between Resting Metabolic Rate and Respiratory Quotient	Pearson's correlation coefficient (dimensionless) is calculated between resting metabolic rate (kcal/day) and respiratory quotient (unitless) measured at baseline, after a 24-hour fast, and during subsequent diet phases (up to Week 26). A higher positive correlation indicates a stronger link between RMR and RQ changes	Baseline (Day 0), after 24-hour fast (Day 1), and during subsequent diet phases (up to Week 26)
Correlation Between Diet-Induced Thermogenesis (DIT) and Postprandial FGF21 Levels	Pearson's correlation coefficient (dimensionless) is calculated between DIT (measured in kcal or kcal/day) and FGF21 concentrations (measured in pg/mL) at baseline, after a 24-hour fast, and during a low-protein meal test. A higher positive correlation indicates a stronger relationship between the magnitude of meal-induced thermogenesis and postprandial FGF21 secretion.	During the low-protein meal test (Day 1) and optionally at subsequent time points (Week 12, Week 26)

Collaborators and Investigators

• Sponsor: University Hospital Schleswig-Holstein

Study record dates

Study Major Dates

• Study Start (Actual): September 15, 2023
• Primary Completion (Estimated): September 30, 2026
• Study Completion (Estimated): September 30, 2027

Study Registration Dates

• First Submitted: February 27, 2025
• First Submitted That Met QC Criteria: March 7, 2025
• First Posted (Actual): March 25, 2025

Study Record Updates

• Last Update Posted (Actual): March 25, 2025
• Last Update Submitted That Met QC Criteria: March 7, 2025
• Last Verified: March 1, 2025

More Information

Terms related to this study

• Keywords: obesity; VLCD; weight loss; fasting; indirect calorimetry; metabolic phenotype; thrifty; spendthrift; meal test
• Additional Relevant MeSH Terms: Nutrition Disorders; Overnutrition; Body Weight; Overweight; Obesity
• Other Study ID Numbers: D 496/21; 508000993 (Other Identifier: DFG (German Research Foundation))

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No