Trellis

Podcast episode

Best Tools for Gut Health & Weight Loss | Dr. Chris Thompson

basic-science diet-nutrition drug-treatment hormone-health medical-innovation

TL;DR

Harvard gastroenterologist Dr. Chris Thompson walks through how the GI tract regulates hunger, fullness, and metabolic health, then explains what GLP-1 drugs actually do (and don't do), when surgery or endoscopic procedures make more sense, and what early warning signs of metabolic trouble you can catch before diabetes arrives. The episode is dense with actionable detail on fiber, fermented foods, muscle loss on GLP-1 drugs, and emerging treatments including a one-time gene therapy now entering clinical trials.


What was covered

  • GI tract anatomy and function — Thompson describes each segment's role: the esophagus moves food, the stomach grinds and secretes ghrelin, the small bowel absorbs calories (one cell thick, surface area of a pickleball court), and the colon hosts the microbiome and produces butyrate and GLP-1.
  • Fiber, resistant starch, and gut barrier health — Low fiber starves the microbiome, which then digests the mucus layer, reducing butyrate production, weakening tight junctions, and setting off an inflammatory cascade. Two recent studies cited: RS2 resistant starch (raw powdered potato, ~40g) improved fatty liver and insulin sensitivity in humans.
  • GLP-1 drugs: benefits, side effects, and "microdosing" — More than 1 million people per month stop GLP-1 medications; about 30% quit in the first month, ~50% by year one. The main problems patients report are muscle loss (including visible "Ozempic face/butt") and nausea at effective doses. Thompson says lower, personalized doses can preserve benefits and reduce side effects — a practice he supports in his own patients, using it as a maintenance bridge rather than stopping cold.
  • Muscle loss and why cycling on/off GLP-1s is dangerous — Each cycle strips lean mass; when weight returns off the drug, it returns as fat, not muscle, progressively worsening body composition. Resistance training is the primary countermeasure Thompson prescribes for all his weight-management patients.
  • Early metabolic markers worth tracking now — Fasting insulin (not just A1C), continuous glucose monitoring (CGM) for a month or two to map food-specific spikes, waist-to-height ratio, ALT liver enzyme, and a respiratory exchange ratio test for metabolic flexibility. Thompson cites the Whitehall 2 study showing high fasting insulin predicts diabetes 10–15 years in advance; and an NHANES-linked analysis finding fewer than 12% of the general U.S. population is metabolically healthy.
  • Surgical and endoscopic alternatives — Endoscopic sleeve gastroplasty (ESG, developed by Thompson in 2012) folds the stomach to suppress ghrelin and trigger stretch receptors; magnetic anastomosis connecting jejunum to ileum produces GLP-1 spikes without drugs; duodenal mucosal resurfacing (ablating the duodenum lining) drops A1C by more than one point in diabetics; combining procedures targets multiple hormonal levers simultaneously.
  • GLP-1 gene therapy — A viral vector carrying the GLP-1 gene tied to the insulin promoter is injected into the pancreatic tail via endoscopic ultrasound. GLP-1 is then secreted alongside insulin in a nutrient-responsive way — not at super-physiologic constant levels. A clinical trial has begun in the Netherlands.

Notable claims & predictions

  • "Over a million people a month are coming off GLP-1s... about 30% in the first month, 50% by end of year." — Dr. Chris Thompson. The dropout rate mirrors adherence patterns for blood pressure and cholesterol drugs and likely reflects a mix of side effects, needle fatigue, cost, and long-term uncertainty.
  • "When you cycle on and off [GLP-1s], you're not putting the lean mass back on — you're putting the fat back on. You're shifting body composition worse every cycle." — Dr. Chris Thompson. He recommends microdosing as a maintenance strategy rather than stopping.
  • "Less than a third of lean people are metabolically healthy... 12% of the whole population." — Thompson, citing an NHANES-based cross-sectional analysis using criteria including waist circumference, blood glucose, and blood pressure.
  • "GLP-1 drugs are super-physiologic — you're bathing the area postrema in this chemical at levels never seen in human history." — Andrew Huberman, confirmed by Thompson. Standard GLP-1 secretion is small and nutrient-triggered; drug doses are orders of magnitude higher.
  • "Fasting insulin can detect metabolic problems 15 years before diabetes appears." — Thompson, citing the Whitehall 2 longitudinal study of British civil servants, but very few clinicians routinely order it.
  • "A one-time GLP-1 gene therapy injection — the clinical trial has entered the Netherlands." — Thompson, describing work he contributed to early on. If it works in humans, a single injection could produce nutrient-responsive GLP-1 elevation for years.

Fact check

"Colon cancer screening age was moved back to 45." — True for average-risk individuals per updated U.S. guidelines (American Cancer Society moved it to 45 in 2018; the U.S. Preventive Services Task Force followed in 2021). Thompson's advice to start at 40 if a first-degree relative had colon cancer, or 10 years before a relative's diagnosis age, is consistent with standard clinical guidance.

"Ghrelin was discovered over 30 years ago." — Slightly off. Ghrelin was identified in 1999 by Kojima and colleagues — about 25 years ago, not 30+. Huberman guessed "over 30 years" and Thompson agreed without correcting. Not a material error, but worth knowing.

"Retatrutide clinical trial showed 30% reduction in body weight." — The Phase 2 trial (Eli Lilly, published in NEJM 2023) showed up to ~24% total body weight loss at 48 weeks at the highest dose — impressive, but not 30%. Huberman's figure is an overstatement. The distinction matters if you're planning around expectations.

NHANES finding that fewer than 12% of the population is metabolically healthy. — This aligns with a widely cited 2019 Metabolic Syndrome and Related Disorders analysis (Araújo et al.) using NHANES data, which found only 12.2% of U.S. adults had optimal cardiometabolic health. The claim is well-grounded.

"Resistant starch (RS2) at ~40g improved fatty liver." — Thompson describes this as a recent study showing "significant improvement in fatty liver" at roughly 40g of RS2 (raw potato starch), with weight relatively stable. A 2023 Cell Metabolism trial from China used 40g/day of raw potato starch and found significant reduction in liver fat and improved insulin sensitivity. The description is consistent with that research, though the study population was Chinese adults with MAFLD, so generalizability to a Western population deserves a note.

AG1 Pro sponsor read — Thompson is a guest on Huberman Lab, which is sponsored by AG1 (among others). Huberman disclosed sponsors; no factual claims about AG1 were made by Thompson. No issue.


Why this matters for you

  • Check fasting insulin, not just A1C. Thompson argues fasting insulin can flag metabolic trouble a decade or more before standard diabetes screening would catch it. It's an inexpensive blood test. Ask your doctor to add it at your next physical — especially if you have central weight gain, high blood pressure, or a family history of diabetes.
  • If you're on a GLP-1 drug, don't stop cold. Quitting and regaining shifts body composition toward fat and away from muscle with each cycle. If you're considering stopping, Thompson recommends either tapering to a maintenance microdose or having a bridging plan (endoscopic procedure or structured lifestyle intervention) in place first. Discuss this with your prescribing physician before making changes.
  • Resistance training is non-negotiable alongside any weight-loss intervention. Whether you're on a GLP-1 drug, had a bariatric procedure, or are dieting, Thompson prescribes resistance training to all patients. Without it, lean mass loss accelerates. If you're sedentary, this is the

Full analysis

Harvard gastroenterologist Dr. Chris Thompson walks through how the GI tract regulates hunger, fullness, and metabolic health, then explains what GLP-1 drugs actually do (and don't do), when surgery or endoscopic procedures make more sense, and what early warning signs of metabolic trouble you can catch before diabetes arrives. The episode is dense with actionable detail on fiber, fermented foods, muscle loss on GLP-1 drugs, and emerging treatments including a one-time gene therapy now entering clinical trials.


What was covered

  • GI tract anatomy and function — Thompson describes each segment's role: the esophagus moves food, the stomach grinds and secretes ghrelin, the small bowel absorbs calories (one cell thick, surface area of a pickleball court), and the colon hosts the microbiome and produces butyrate and GLP-1.
  • Fiber, resistant starch, and gut barrier health — Low fiber starves the microbiome, which then digests the mucus layer, reducing butyrate production, weakening tight junctions, and setting off an inflammatory cascade. Two recent studies cited: RS2 resistant starch (raw powdered potato, ~40g) improved fatty liver and insulin sensitivity in humans.
  • GLP-1 drugs: benefits, side effects, and "microdosing" — More than 1 million people per month stop GLP-1 medications; about 30% quit in the first month, ~50% by year one. The main problems patients report are muscle loss (including visible "Ozempic face/butt") and nausea at effective doses. Thompson says lower, personalized doses can preserve benefits and reduce side effects — a practice he supports in his own patients, using it as a maintenance bridge rather than stopping cold.
  • Muscle loss and why cycling on/off GLP-1s is dangerous — Each cycle strips lean mass; when weight returns off the drug, it returns as fat, not muscle, progressively worsening body composition. Resistance training is the primary countermeasure Thompson prescribes for all his weight-management patients.
  • Early metabolic markers worth tracking now — Fasting insulin (not just A1C), continuous glucose monitoring (CGM) for a month or two to map food-specific spikes, waist-to-height ratio, ALT liver enzyme, and a respiratory exchange ratio test for metabolic flexibility. Thompson cites the Whitehall 2 study showing high fasting insulin predicts diabetes 10–15 years in advance; and an NHANES-linked analysis finding fewer than 12% of the general U.S. population is metabolically healthy.
  • Surgical and endoscopic alternatives — Endoscopic sleeve gastroplasty (ESG, developed by Thompson in 2012) folds the stomach to suppress ghrelin and trigger stretch receptors; magnetic anastomosis connecting jejunum to ileum produces GLP-1 spikes without drugs; duodenal mucosal resurfacing (ablating the duodenum lining) drops A1C by more than one point in diabetics; combining procedures targets multiple hormonal levers simultaneously.
  • GLP-1 gene therapy — A viral vector carrying the GLP-1 gene tied to the insulin promoter is injected into the pancreatic tail via endoscopic ultrasound. GLP-1 is then secreted alongside insulin in a nutrient-responsive way — not at super-physiologic constant levels. A clinical trial has begun in the Netherlands.

Notable claims & predictions

  • "Over a million people a month are coming off GLP-1s... about 30% in the first month, 50% by end of year." — Dr. Chris Thompson. The dropout rate mirrors adherence patterns for blood pressure and cholesterol drugs and likely reflects a mix of side effects, needle fatigue, cost, and long-term uncertainty.
  • "When you cycle on and off [GLP-1s], you're not putting the lean mass back on — you're putting the fat back on. You're shifting body composition worse every cycle." — Dr. Chris Thompson. He recommends microdosing as a maintenance strategy rather than stopping.
  • "Less than a third of lean people are metabolically healthy... 12% of the whole population." — Thompson, citing an NHANES-based cross-sectional analysis using criteria including waist circumference, blood glucose, and blood pressure.
  • "GLP-1 drugs are super-physiologic — you're bathing the area postrema in this chemical at levels never seen in human history." — Andrew Huberman, confirmed by Thompson. Standard GLP-1 secretion is small and nutrient-triggered; drug doses are orders of magnitude higher.
  • "Fasting insulin can detect metabolic problems 15 years before diabetes appears." — Thompson, citing the Whitehall 2 longitudinal study of British civil servants, but very few clinicians routinely order it.
  • "A one-time GLP-1 gene therapy injection — the clinical trial has entered the Netherlands." — Thompson, describing work he contributed to early on. If it works in humans, a single injection could produce nutrient-responsive GLP-1 elevation for years.

Fact check

"Colon cancer screening age was moved back to 45." — True for average-risk individuals per updated U.S. guidelines (American Cancer Society moved it to 45 in 2018; the U.S. Preventive Services Task Force followed in 2021). Thompson's advice to start at 40 if a first-degree relative had colon cancer, or 10 years before a relative's diagnosis age, is consistent with standard clinical guidance.

"Ghrelin was discovered over 30 years ago." — Slightly off. Ghrelin was identified in 1999 by Kojima and colleagues — about 25 years ago, not 30+. Huberman guessed "over 30 years" and Thompson agreed without correcting. Not a material error, but worth knowing.

"Retatrutide clinical trial showed 30% reduction in body weight." — The Phase 2 trial (Eli Lilly, published in NEJM 2023) showed up to ~24% total body weight loss at 48 weeks at the highest dose — impressive, but not 30%. Huberman's figure is an overstatement. The distinction matters if you're planning around expectations.

NHANES finding that fewer than 12% of the population is metabolically healthy. — This aligns with a widely cited 2019 Metabolic Syndrome and Related Disorders analysis (Araújo et al.) using NHANES data, which found only 12.2% of U.S. adults had optimal cardiometabolic health. The claim is well-grounded.

"Resistant starch (RS2) at ~40g improved fatty liver." — Thompson describes this as a recent study showing "significant improvement in fatty liver" at roughly 40g of RS2 (raw potato starch), with weight relatively stable. A 2023 Cell Metabolism trial from China used 40g/day of raw potato starch and found significant reduction in liver fat and improved insulin sensitivity. The description is consistent with that research, though the study population was Chinese adults with MAFLD, so generalizability to a Western population deserves a note.

AG1 Pro sponsor read — Thompson is a guest on Huberman Lab, which is sponsored by AG1 (among others). Huberman disclosed sponsors; no factual claims about AG1 were made by Thompson. No issue.


Why this matters for you

  • Check fasting insulin, not just A1C. Thompson argues fasting insulin can flag metabolic trouble a decade or more before standard diabetes screening would catch it. It's an inexpensive blood test. Ask your doctor to add it at your next physical — especially if you have central weight gain, high blood pressure, or a family history of diabetes.
  • If you're on a GLP-1 drug, don't stop cold. Quitting and regaining shifts body composition toward fat and away from muscle with each cycle. If you're considering stopping, Thompson recommends either tapering to a maintenance microdose or having a bridging plan (endoscopic procedure or structured lifestyle intervention) in place first. Discuss this with your prescribing physician before making changes.
  • Resistance training is non-negotiable alongside any weight-loss intervention. Whether you're on a GLP-1 drug, had a bariatric procedure, or are dieting, Thompson prescribes resistance training to all patients. Without it, lean mass loss accelerates. If you're sedentary, this is the

Comments