About

Christos Mantzoros is a professor in the Department of Medicine at Harvard Medical School, affiliated with Beth Israel Deaconess Medical Center. His work focuses on various aspects of nutrition, metabolism, and related health conditions. His research encompasses areas such as body composition, bone health, cancer, cardiovascular health, circadian rhythm, clinical research, community health, counseling, critical care, culinary medicine, diabetes, disease prevention, endocrinology, epidemiology, fats, feeding disorders, food insecurity, gastroenterology, immunology, intestinal failure, iron, lifestyle medicine, malnutrition, metabolic disease, metabolism, microbiome, neonatology, nutrition support, obesity, omics, pediatrics, pulmonary health, translational research, vitamin D, weight management, and women's health. His contributions are integral to advancing understanding in these fields, with a particular emphasis on clinical and translational research aimed at improving health outcomes related to nutrition and metabolic diseases.

Research topics

• Internal medicine
• Medicine
• Bioinformatics
• Pediatrics
• Intensive care medicine
• Nursing
• Endocrinology
• Gastroenterology

Selected publications

• Patient-specific deep offline artificial pancreas for blood glucose regulation in type 1 diabetes

  Smart Health · 2026-01-18

  articleOpen access

  Due to insufficient insulin secretion, patients with type 1 diabetes mellitus (T1DM) are prone to blood glucose fluctuations ranging from hypoglycemia to hyperglycemia. While dangerous hypoglycemia may lead to coma immediately, chronic hyperglycemia increases patients' risks for cardiorenal and vascular diseases in the long run. In principle, an artificial pancreas - a closed-loop insulin delivery system requiring patients to manually input insulin dosage according to the upcoming meals - could supply exogenous insulin to control the glucose levels and hence reduce the risks from hyperglycemia. However, insulin overdosing in some type 1 diabetic patients, who are physically active, can lead to unexpected hypoglycemia beyond the control of the common artificial pancreas. Therefore, it is important to take into account the glucose decrease due to physical exercise when designing the next-generation artificial pancreas. In this work, we develop a framework integrating systems biology-informed neural networks (SBINN), deep reinforcement learning (RL) algorithms, and T1DM data collected from wearable devices, to automate insulin dosing for patients. In particular, we build patient-specific computational models using SBINN to mimic the glucose-insulin dynamics for a few patients from the dataset, by simultaneously considering patient-specific carbohydrate intake and physical exercise intensity. Our patient-specific artificial pancreas, based on two deep RL algorithms, provided better insulin dosage, leading to safer glucose levels compared to those in the original dataset.

  PublisherDOI

• OR09-07 Dynamic Proteomic Profiling in a 5-week Crossover Trial of Liraglutide in Obesity Reveals Weight-Dependent Myostatin Downregulation and Multi-Pathway Effects Beyond Weight Reduction

  Journal of the Endocrine Society · 2025-10-01

  articleOpen accessSenior author

  Abstract Disclosure: K. Stefanakis: None. P. Veeragandham: None. V. Gutierrez de Pineres: None. C.S. Mantzoros: Novo Nordisk. The recent FDA approval of generic Liraglutide necessitates a better understanding of its downstream molecular mechanisms to develop obesity treatments that reduce adiposity while preserving lean body mass, key factors in mitigating sarcopenia associated with GLP-1 agonists. In a 5-week crossover randomized, double-blind, placebo-controlled trial, 20 participants with obesity received Liraglutide (3.0 mg/day) or placebo in random sequence, separated by a minimum 3-week washout period. The SomaScan v4.1 platform was used to quantify approximately 7,000 proteins across serum and plasma samples with robust quality controls. Longitudinal proteomic changes were assessed using FDR-corrected pre-to-post comparisons and adjusted mixed models, whereas enrichment analyses were performed using gene and disease ontology, and data from deCODE GWAS. Liraglutide treatment resulted in a significant weight reduction (-2.99 kg placebo-subtracted) and improved glucose and lipid profiles. Out of 6,249 common proteins analyzed, 338 proteins had a significant Time*Treatment interaction, of which Liraglutide significantly upregulated 84 and downregulated 104 proteins, whereas placebo had minimal effects. Specifically, Liraglutide increased exocrine pancreatic proteins (e.g., PNLIPRP1, PRSS2, CTRB2) and fatty acid–binding proteins (e.g., FABP4), and decreased proteins related to fibrosis, inflammation, collagen deposition, and neural development - potentially related to known effects of Liraglutide on appetite regulation and hypothalamic centers. Most proteomic changes were independent of weight and glucose; however, myostatin (MSTN), a negative regulator of muscle growth, was downregulated in a weight-dependent manner, and its inhibitor WFIKKN2, alongside TGF-β receptor BMPR1a were upregulated, suggesting a compensatory mechanism to preserve lean mass during early weight loss. ELISA validation confirmed the proteomic measurements for MSTN (r=0.78, p<10-14). Pathway enrichment analyses indicated reductions in fibrosis and inflammatory processes, and in pathways associated with cardiovascular, renal, and liver diseases, alongside differential regulation of apoptosis, and enhancements in digestion, lipid metabolism, and humoral immunity. Secondary analyses identified distinct proteomic signatures in seven participants who lost more than 3% of their baseline weight, involving neurotransmitter and insulin-regulating proteins. Unlike longer-term weight loss studies, these results capture early molecular adaptations, with the myostatin axis emerging as a target for combination therapies to minimize lean tissue loss. These insights are critical to better understand the mechanisms of weight loss and to optimize the development of novel compounds aimed towards myostatin antagonism to concomitantly lose fat and preserve lean body mass and metabolic health. Presentation: Saturday, July 12, 2025

  PublisherOA PDFDOI

• 1620-P: GDF15 Correlates with Insulin Sensitivity But Not with Features of Metabolic Dysfunction–Associated Steatotic Liver Disease

  Diabetes · 2025-06-13

  article

  Introduction and Objective: Growth Differentiation Factor 15 (GDF15) is a biomarker for metabolic diseases, including type-2 diabetes, metabolic dysfunction-associated steatotic liver disease (MASLD) and steatohepatitis (MASH). GDF15 secretion is linked to hepatic mitochondrial stress in mice and to markers of insulin resistance in humans. MASLD is associated with both hepatic mitochondrial adaptations and insulin resistance. Here, we investigated whether GDF15 levels correlate with hepatic mitochondrial respiration and/or insulin sensitivity in participants with and without MASLD. Methods: Fasting blood samples were obtained from obese participants with and without (CON) biopsy-confirmed simple steatosis (MASL) or MASH (MASH) (n=20/group). Groups were matched for age (mean±SD: 41±11, 44±9, 42±10 years) and BMI (49±7, 48±6, 49±8 kg/m2). Whole-body insulin sensitivity was assessed using hyperinsulinemic-euglycemic clamps, hepatic mitochondrial respiration via high-resolution respirometry and serum GDF15 by ELISA. Group differences were analyzed with one-way ANOVA or Kruskal-Wallis-Test and correlations with Pearson’s or Spearman’s methods. Results: Fasting GDF15 levels were similar between groups (616±353, 661±252, 635±292 pg/mL in CON, MASL, MASH; p=0.385), and correlated with age (r=0.414, p=0.001), but not with hepatic fatty acid- or NADH+succinate-linked mitochondrial respiration (p>0.05 for all respiration states). GDF15 correlated negatively with whole-body insulin sensitivity (M-value; r=-0.342, p=0.022) and positively with adipose tissue insulin resistance (Adipo-IR; r=0.338, p=0.012). Conclusion: GDF15 associates with whole-body and adipose tissue insulin resistance but not with liver oxidative capacity. In line, GDF15 levels were comparable in participants with and without simple steatosis or MASH. These findings may suggest a potential role for GDF15 in adipose tissue-mediated insulin resistance. Disclosure A. Giannakogeorgou: Research Support; Novo Nordisk. S. Kahl: Research Support; Boehringer-Ingelheim, Novo Nordisk A/S. C. Granata: None. G. Heilmann: None. L. Mastrototaro: None. B. Dewidar: None. I. Esposito: None. S. Trenkamp: None. F.A. Granderath: None. M. Schlensak: None. C.S. Mantzoros: Other Relationship; UpToDate. Research Support; LabCorp, Merck & Co., Inc, ESPERION Therapeutics, Inc., Massachusetts Life Sciences Center, Boehringer-Ingelheim. Other Relationship; TMIOA. Consultant; Novo Nordisk, Amgen Inc, Olympus, Genfit, Lumos, Madrigal Pharmaceuticals, Inc, ESPERION Therapeutics, Inc., ALIGOS Therapeutics, Regeneron Pharmaceuticals, 89bio, Inc, Corcept Therapeutics, Intercept Pharmaceuticals, Inc. Research Support; ANSH Labs. Other Relationship; Elsevier, Cardiometabolic Health Congress. Research Support; Abbott. M. Roden: Research Support; Boehringer-Ingelheim. Advisory Panel; Echosens. Speaker's Bureau; Madrigal Pharmaceuticals, Inc. Advisory Panel; MSD Life Science Foundation. Board Member; Novo Nordisk. Advisory Panel; TARGET PharmaSolutions, Inc. P. Schrauwen: Research Support; AstraZeneca, Pfizer Inc, Antaros Medical.

  PublisherDOI

• OR21-05 A Multinational Study of Serum Metabolomics and Lipidomics in the Pathophysiology and Non-invasive Diagnosis of Metabolic Dysfunction-Associated Steatotic Liver Disease

  Journal of the Endocrine Society · 2025-10-01

  articleOpen accessSenior author

  Abstract Disclosure: K. Stefanakis: None. N. Perakakis: None. O. Verrastro: None. M. Garcovich: None. L. Riccardi: None. M. Eslam: None. G.E. Markakis: None. G. Papatheodoridis: None. G. Mingrone: None. J. George: None. C.S. Mantzoros: None. The growing burden of metabolic dysfunction-associated steatotic liver disease (MASLD) and steatohepatitis (MASH), coupled with novel therapies targeting fibrotic MASH (F2–F3), underscores the need for deeper mechanistic insights and improved non-invasive test (NITs). In a multi-national biobank, we enrolled 443 biopsy-proven individuals (81 healthy controls, 362 patients, including 162 with MASH F≥2) across three countries (Italy, Greece, Australia) and two clinical center types (bariatric, gastroenterology-hepatology). A total of 839 metabolites and 840 lipid species were quantified in the serum of all 443 participants by ultra-high-performance liquid chromatography–tandem mass spectrometry. Beyond absolute changes, we employed supervised/unsupervised classifiers, clustering, and pathway enrichment analyses to map metabolites and lipids across the MASLD histological continuum. Early steatosis and MASLD were characterized by a pronounced surge in triglycerides conjugated with 16:0 (palmitic acid), together with elevated ceramides and inflammatory markers. Transition to MASH was marked by elevated BCAAs, bile acid metabolites, sphingolipids, urea cycle intermediates, and xenobiotic compounds (mannonate, hydroxybenzoate), alongside a further spike in triglycerides. In MASH F2–F3, 108 metabolites were markedly upregulated—notably 3-ureidopropionate (uracil pathway) and kynurenine (tryptophan pathway)—and 137 lipid species (chiefly glycerols with saturated fatty acids), whereas 21 metabolites (purines, n-acetylmethionine, steroidogenic intermediates) and 177 lipids (mainly phospholipids and ceramides) were significantly reduced. With advanced fibrosis (F3–F4), the triglyceride surge attenuated, yet inflammation persisted, coupled with dysregulated xenobiotic metabolism and depletion of >10 fibrinopeptide species, indicative of impaired coagulation. Weighted correlation analyses unveiled distinct modules of functionally and structurally similar metabolites and lipids co-varying with transaminases, GDF-15, activins and follistatins in significant fibrosis; separate from obesity-driven steatosis and early inflammation. Building on these comprehensive profiles, we developed categorical gradient-boosting models (4:1 training–validation split, repeated cross-validation with random resampling) using routine clinical variables (e.g., transaminases) and select metabolites such as 3-ureidopropionate, kynurenine, and sphingadienine. These models achieved cross-validated AUCs exceeding 88%—and up to 94%—in the validation cohort, outperforming over 20 established imaging- and biomarker-based NITs for detecting MASH, at-risk MASH, and F≥3. These findings define a novel paradigm for delineating mechanistic pathways that could reveal future therapeutic targets and streamline the development of accurate NITs. Presentation: Saturday, July 12, 2025

  PublisherOA PDFDOI

• Vitamin D supplementation at low (600 IU/day) or higher dose (3,750 IU/day) does not improve insulin resistance markers at one year: A randomized controlled trial

  Metabolism Open · 2025-03-11 · 1 citations

  articleOpen access

  To compare the performance of newer insulin resistance (IR) indices, triglyceride glucose index (TyG) and metabolic score for IR (METS-IR), with previous markers HOMA-IR and McAuley-IR, and assess the impact of one-year of vitamin D supplementation, at two doses, on these indices in overweight, elderly individuals. Exploratory analyses from a double-blind, multicenter randomized controlled trial involved overweight elderly participants with baseline serum 25-hydroxyvitamin D [25(OH)D] levels of 10-30 ng/ml (clinicaltrial.gov: NCT01315366). Participants received 1000 mg calcium citrate/day and vitamin D supplementation at a low-dose of 600 IU/day, or high-dose of 3750 IU/day. 221 participants received low or high-dose vitamin D supplementation. Mean age was 71 ± 5 years, BMI 30 ± 4 kg/m 2 , 25(OH)D 20 ± 7 ng/ml, with 55% female and 69% with prediabetes. There were no significant baseline differences except for HDL levels (p=0.04). TyG was notably increased in the high-dose group (p=0.02). Mixed linear model analysis showed a greater increase in serum 25(OH)D in the high-dose group compared to the low-dose, with decreases in PTH, cholesterol, and LDL independent of dose. TyG and METS-IR did not differ by dose, time, or dose*time interaction. Subgroup analyses by sex, baseline 25(OH)D cut-off, and glucose tolerance status were null. FokI polymorphism showed a significantly greater METS-IR in the high-dose arm, disappeared after adjusting for fat mass. McAuley-IR was the best IR index compared to TyG and METS-IR, both at baseline and 12 months. Vitamin D supplementation at 3,750 IU/d over one-year did not improve IR markers, including TyG and METS-IR.

  PublisherDOI

• Semaglutide, the first approved GLP-1 receptor agonist for the management of metabolic dysfunction-associated steatohepatitis

  Metabolism · 2025-10-03 · 3 citations

  editorialSenior author
  PublisherDOI

• Best practice recommendations for the diagnosis and management of hypoparathyroidism

  Metabolism · 2025-06-26 · 10 citations

  reviewOpen access

  BACKGROUND: Hypoparathyroidism (HypoPT) is characterized by low serum calcium due to insufficient parathyroid hormone (PTH). This manuscript builds upon the 2022 international HypoPT guidelines and three systematic reviews, which have been further informed by updated narrative reviews and expert consensus. This paper presents current best practice consensus recommendations for the diagnosis and management of HypoPT. METHODS: An International Panel of Experts updated the previous systematic reviews (SR's), conducted narrative reviews, developed, and subsequently approved these best practice recommendations at the Parathyroid Summit, held as a pre-Endocrine Society meeting in May 2024 (Boston, USA). RESULTS: Diagnostic criteria for chronic HypoPT require hypocalcemia with inappropriately normal or low PTH levels. Conventional therapy is recommended as first line therapy and includes calcium supplementation, active vitamin D, correction of vitamin D inadequacy and correction of abnormalities in serum magnesium. Monitoring is required to achieve optimal serum calcium while avoiding hyperphosphatemia, hypercalciuria and declines in renal function. Assessment of HypoPT complications is required including skeletal health assessment in postmenopausal women and men over the age of 50 years. Specific strategies are provided for managing HypoPT during pregnancy and lactation as well as in children. PTH replacement with palopegteriparatide has been approved and is an important therapeutic option, especially when conventional therapy is inadequate or not tolerated. CONCLUSION: These best practice recommendations provide a framework for HypoPT diagnosis and management, emphasizing individualized care, role of DNA analysis in the diagnosis of nonsurgical HypoPT, and role of PTH or PTH analogue therapy as appropriate. They complement the 2022 international guidelines and incorporate updated therapeutic recommendations from the past 3 years including the positioning of the newly approved molecule palopegteriparatide based on recent clinical trial data and expert consensus.

  PublisherOA PDFDOI

• Novel Machine-Learning Models Outperform Noninvasive Tests in Metabolic Dysfunction-Associated Steatohepatitis With Fibrosis

  Clinical Gastroenterology and Hepatology · 2025-11-01 · 1 citations

  articleSenior author
  PublisherDOI

• Association between remnant cholesterol and chronic kidney disease: Systematic review and meta‐analysis

  Diabetes Obesity and Metabolism · 2025-02-14 · 11 citations

  reviewOpen accessSenior author

  Abstract Background and Aims Adequate lipid control has emerged as a key factor in the prevention and management of chronic kidney disease (CKD). Remnant cholesterol (RC), a lipoprotein with an established association with cardiovascular risk, has been investigated in the context of CKD. Given the conflicting results from recent studies, we performed this meta‐analysis to summarize the existing evidence on the association between RC and CKD. Methods Medline, Cochrane Library and Scopus were searched until 16 September 2024. Double‐independent study selection, data extraction and quality assessment were performed. Evidence was pooled using random‐effects meta‐analyses. We set as primary end‐point of interest the association between RC and CKD. Results Twelve studies (4 139 674 participants) were included. Participants with RC values in the highest quantile had significantly greater odds of CKD compared to those in the lowest quantile (Odds Ratio [OR] = 1.46, 95% confidence interval [CI] = 1.26–1.68). In a sensitivity analysis confined to subjects with type 2 diabetes (T2D), those in the higher RC quantile also exhibited significantly increased odds of CKD compared to those in the lowest quantile (OR = 1.46, 95% CI = 1.20–1.78). A significant inverse association was observed between RC and estimated glomerular filtration rate (Mean Difference [MD] = −1.43 mL/min/1.73 m 2 for each 1 mmol/L increase in RC, 95% CI = [−2.67, −0.19]). Additionally, individuals with T2D‐related CKD had a 24% increased risk of progression to end‐stage renal disease for each 1 standard deviation increase in RC (Hazard Ratio [HR] = 1.24, 95% CI = 1.04–1.47). Conclusions RC is directly associated with higher risk for CKD. Beyond traditional lipid markers, greater emphasis should be placed on RC levels in individuals with or at risk for CKD.

  PublisherOA PDFDOI

• Performance of Established and Novel Non-invasive Tests, Including New Clinical and Metabolomics-Based Machine Learning Models, for Detecting MASH with Fibrosis Stages F2-F3: A First-in-Class Approach in the post-Resmetirom approval era

  Metabolism · 2025-06-01

  articleSenior author
  PublisherDOI

Recent grants

• NIH Grant M01RR001032

  NIH · $936k · 2008

• NIH Grant R01DK079929

  NIH · $1.8M · 2014

• Adipokine physiology

  NIH · $1.5M · 2008–2021

• NIH Grant R01DK058785

  NIH · $4.0M · 2012

• NIH Grant R56DK058785

  NIH · $85k · 2006

Frequent coauthors

• John P. Chamberland

  327 shared

• Konstantinos N. Aronis

  Johns Hopkins University

  245 shared

• Nikolaos Perakakis

  German Center for Diabetes Research

  222 shared

• Olivia M. Farr

  Beth Israel Deaconess Medical Center

  168 shared

• Kenneth J. Mukamal

  Harvard University

  147 shared

• Stergios A. Pοlyzos

  Aristotle University of Thessaloniki

  145 shared

• Edward L. Giovannucci

  Harvard University

  137 shared

• Charles S. Fuchs

  137 shared