GLP-1 agonists: improve skeletal muscle??

 A recent article evaluated the effects of GLP-1 receptor agonists on free-fat mass loss as associated with the attendant weight loss, in a review of 28 trials (see GLP-1 and weight loss composition DiabObesMetab2024 in dropbox, or DOI: 10.1111/dom.15913) 

   

Introduction: 

-- there are 2 incretins (GLP-1 and GIP) that are secreted in response to nutrients (glucose and other carbohydrates, fats, and proteins)

-- GLP-1 (glucagon-like peptide-1) and GIP (glucose-dependent insulinotropic polypeptide) have somewhat different effects in the body but are synergistic in leading to pancreatic insulin release

-- GLP-1 receptors are located throughout the body (as are GIP receptors, but this blog and discussion will focus on the GLP-1's)

-- here is a graphic which shows the various targets and related relevant systemic effects of these incretins:

 

   -- GLP-1 receptors are an important part of the G protein-coupled receptor family (heterotrimeric guanine nucleotide-binding proteins) that are "signal transducers that communicate signals from many hormones, neurotransmitters, chemokines, and autocrine and paracrine factors" : https://pubmed.ncbi.nlm.nih.gov/12040175/ (ie, they are molecular switches transmitting signals from outside a cell to inside one) and regulate lots of things including glucose levels and lipids

    -- GLP-1 receptors are located predominantly in the gut and pancreas, but also in many other parts of the body including the brain (hypothalamus, hippocampus, hindbrain, caudate, putamen, globus pallidum, cerebellum, and these seem to lead to a lot of the metabolic effects of GLP1 receptor agonists) and also in the spinal cord (https://psychiatryonline.org/doi/full/10.1176/appi.neuropsych.20230226#:~:text=GLP%2D1%20and%20its%20analogs,9%2C%2014%2C%2022).  but there are also receptors in the lungs, heart, kidney, immune system (though not in this diagram, but there are receptors on T lymphocytes, macrophages, and intraepithelial lymphocytes: https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2022.997578/full ), stomach and bowel.

        -- as an aside, there are many known effects of GLP-1RAs on the cardiovascular system, including those produced by weight loss (eg, lowering blood pressure and high lipid levels), but also positive effects on the endothelium, decreasing oxidative stress, reduction in atherosclerotic plaque formation and progression, anti-inflammatory effects, vasodilatory effects that reduce systemic vascular resistance: https://www.frontiersin.org/journals/clinical-diabetes-and-healthcare/articles/10.3389/fcdhc.2023.1293926/full

    -- so, how do the GLP-1RAs get into the brain???

        -- the studies are conflicting, some finding that they cross the blood-brain barrier and some that they may sneak around and penetrate the brain through specialized areas of the blood-brain barrier called "circumventricular organs" as found in the postrema and subfornical organ areas.

        -- GLP-1 is also synthesized in the brain: brain derived GLP-1 is synthesized in a subset of neurons in the NTS (nucleus of the solitary tract) that project into the GLP-1 receptor-expressing regions in the hindbrain and hypothalamus, including the paraventricular nucleus, dorsal medial nucleus of the hypothalamus the arcuate nucleus, as well as the NTS itself; a small amount of GLP-1 is synthesized in the hypothalamus https://pmc.ncbi.nlm.nih.gov/articles/PMC4191040 .

        -- GLP-1 receptor agonists (GLP-1RAs) slow gastrointestinal motility likely through vagal effects emanating from the brain, and presumably cause the nausea that occurs in some recipients of GLP-1 receptor agonists; it is also likely that decreased intestinal peristalsis from the GLP-1RAs is centrally mediated 

        -- the net result of GLP-1RAs is a decrease of food intake of >30% and improved satiety markers; however there also are decreases in individual's preoccupations with food as well as some small changes in taste and food preferences

    --there are such profound effects of GLP-1 RAs alone and especially in combo with the GIP receptor agonist (GIP-RA) tirzepatide in weight reduction that the results are similar to the weight loss from some bariatric metabolic surgery (esp the sleeve gastrectomies)

    --as an overview of the weight reduction effects: 

        -- GLP-1RAs: about 10-15% weight reduction at 1 year (some studies a bit higher)

        -- GLP-1RAs plus a GIP-RA with tirzepatide (the only one now available): 20.9% with max dose at 72 weeks

        -- there are other meds coming out that include GLP-1RAs plus GIP-RAs plus a glucagon receptor agonist (GRA), though trials on GRAs so far do not seem to yield better weight loss outcomes than tirzepatide

    -- but, one big issue/concern is that the weight reduction with GLP-1RAs may lead to lots of skeletal muscle loss

        -- though there are no firm studies suggesting that there are adverse clinical effects of skeletal muscle loss, it is especially a concern for the elderly, those with diabetes, and post-menopausal women who are all at increased risk of protein deficiency

    -- the major way studies have assessed the quality of the change in body composition from weight loss is to quantify changes in fat loss vs fat-free mass loss (FFM), as well as the percentage of FFM loss (%FFML), by using dual-energy X-ray absorptiometry (DXA). DXA does not provide a direct measure of skeletal muscle itself. Actual measurements by MRI or CT scans (the gold standards for direct measurements of skeletal muscle) are expensive and not practical in large studies. As a few examples of study results:

            -- the SUSTAIN 8 trial of patients with Type II diabetes on semaglutide 1mg weekly for 52 weeks found %FFML was 40.4% of their total weight loss

            -- the LEAD-2 trial of 103 people with diabetes on liraglutide 0.6, 1.2, or 1.8 mg daily for 26 weeks found %FFML of 33%, 40.0%, and 46.9% respectively

            -- other studies in patients with obesity and diabetes on dietary interventions found %FFML in the 10%-25% range

 

Here is a review of the main points of the article:

 

GLP-1 hormone and appetite:  

-- as per above, there are pretty profound GI effects of GLP-1RAs, including slowing GI motility, nausea, etc, that are mediated centrally by GLP-1 receptors and lead to vagal effects on appetite and satiety. of note, GLP-1 receptors are located in the hypothalamus, hippocampus, hindbrain (area postrema), and mesolimbic regions

    -- studies suggest that the anorectic effect of GLP-1RAs is likely through the vagus nerve, nodose ganglia, hypothalamic nuclei and the brain stem

-- one clinical effect of this neural input is a change in food preferences, including decreases in preoccupations with/ruminations about foods. it seems from preliminary data that there are changes in taste perception. a mouse study found decreased positive responses to sweets, for example

-- and this all brings up the concerns about changes in food preferences leading to perhaps important changes in bodily function. so far, no vitamin deficiencies have been reported, but it is not clear about changes in macronutrients

 

Skeletal muscle: 

-- skeletal muscle is the most abundant tissue in the body, comprising about 40% of body mass and 25 to 30% of basal energy expenditure 

-- it is composed of 75% water, 20% protein, and 5% inorganic salts and minerals, and it constitutes about 55% of fat-free mass by DXA measurements 

-- skeletal muscle also provides an important part of type II diabetes pathogenesis by increasing insulin sensitivity and postprandial glucose uptake. low skeletal muscle mass is associated with poorer diabetes control in older persons with diabetes. It is also an important site of lipid oxidation 

-- skeletal muscle does atrophy with normal aging on the order 1 to 2% annually after age 50, with loss of muscle fibers, motor neuron input, and muscle contractility (and the associated clinical findings of increased risk of falls and bone fractures, physical frailty, and loss of independence) 

-- very low-calorie diets (800-1000 kcal per day) also cause fat-free mass loss and decreased bone mineral density 

 

Measuring body composition: 

-- the gold standard is either CT or MRI scanning for direct measures of adipose tissue and skeletal muscle 

-- there are several other techniques and markers that have been used including DXA scans, ADP (air displacement plethysmography),  and anthropometric measurements (though these latter measurements are not very sensitive for assessing small changes in body composition)

 

Impact of various weight loss interventions

-- calorie restriction, dietary weight loss and %FFML:

    -- amount of FFM lost is highly variable and depends on lots of factors (baseline body composition, diet composition, age, ethnicity, hormonal status, and the amount and type of exercise)

    -- overall consensus: in absence of structured exercise, men lose 20-25% of FFM; women lose 10-15% of FFM, and about 50% of FFM loss is from skeletal muscle

--significant weight loss, MBS (metabolic bariatric surgery), and %FFML:

  -- after a popular TV show, some participants who lost >30% of their weight (some up to 50%) of their total body mass through at least 2 hours/d of high intensity resistance training, still lost 20% of their weight from FFM at week 30, per DXA measurement

  -- effect of MBS on body composition is also quite variable, but is in the range of 20-30%, with Roux-en-Y gastric bypass having the greatest loss

  -- over 1/2 the  FFM loss is in the first 3 months after the surgery

  -- several studies (and, as i remember, a whole issue of the American Journal of Medicine) dealt with the metabolic effects of bariatric surgery, noting that the benefits of bariatric surgery began almost immediately after the surgery (and before any meaningful weight loss), including decreases in glucose and triglycerides, increasing levels of postprandial adiponectin, insulin, insulin-like growth factor-1, and GLP-1 (and, as per above, GLP1 receptors are located in several areas of the body, though perhaps most relevantly the hypothalamus, which does regulate metabolic and cardiovascular function). Hence, i presume, the name being “metabolic” bariatric surgery

  -- in fact, one study found that the amount of FFM loss with Roux-en-Y bypass was actually 11.1% and that in the first year 96.7% of the FFM loss was skeletal muscle as determined by MRI

-- GLP-1 receptor agonists and %FFML:

   -- much of the data here are not great: substudy analyses and often small clinical trials (and, again, we know that there are pretty big interindividual differences, since many factors affect the weight loss, not just the  meds)

   -- to assess FFM, the current study assessed articles using: air displacement plethysmography (ADP, used in 4 studies below), CT imaging (in 1 study), and dual energy x-ray absorptiometry (DXA used in the other 23 studies), the latter using a 3-compartment model (fat mass, lean soft tissue, and bone mineral), though with no direct measure of skeletal muscle mass (DXA cannot directly assess skeletal muscle). they derived lean soft tissue mass of each extremity and the total appendicular lean mass correlated well with the total muscle mass measured by MRI 

    -- based on studies classifying %FFML by being 25%, less than 25%, and greater than 25%:

            -- those achieving 25% FFML: liraglutide in a Mayo Clinic study on 3.0 mg/d for 16 weeks led to a 5.8kg weight loss, 22.4% attributed to FFM (though no difference from placebo, likely reflecting the fact that this was a relatively short study with a less potent GLP-1RA

            -- those achieving <25% FFML: a double-blind, cross-over RCT with only 15 participants on daily oral semaglutide for 12 weeks, lost 2.7kg with only 4% accounted for by FFP measured by ADP

            -- those achieving >25% FFML: subgroup analysis of the SUSTAIN 8 trial with 53 subjects on semaglutide 1mg weekly had their overall weight loss of 5.3kg, DXA scans finding total fat mass (baseline 33.2kg) reduced by 3.4kg and lean mass (51.3kg) reduced by 2.3kg

            -- tirzepatide studies: FFM reduced by 14.2% of the total weight loss after 28 weeks in 45 subjects, vs 11.6% in 44 subjects on semaglutide, per use of ADP at 72 weeks; pooled data from the SURMOUNT trial of tirzepatide found FFM loss 10.9% from baseline at 72 weeks. another report of 18 weeks of tirzepatide found 32.6% ot toal weight loss was due to FFM

           -- a few studies comparing GLP-1RAs to SGLT-2 inhibitors found no real difference in changes in FFA, with %FFML: in the 20-50% range

 

Endogenous GLP-1 and skeletal muscle/issues of muscle quantity vs quality

-- newer research suggests that endogenous GLP-1 may regulate skeletal muscle function directly through GLP-1 receptors on the muscle itself. animal studies have found that GLP-1RAs actually reduced obesity-induced muscle atrophy and that GLP-1RAs may be helpful in treating skeletal muscle atrophy and inflammation

    -- in mice, exercise increases secretion of GLP1, and overexpression of GLP1 in skeletal muscle enhances endurance capacity: https://pubmed.ncbi.nlm.nih.gov/35636559/

-- human studies have found GLP-1 infusions lead to increases in microvascular blood flow and volume in skeletal muscle, perhaps suggesting that increases in postprandial insulin, amino acids and glucose in these tissues may in fact improve muscle protein synthesis: https://pubmed.ncbi.nlm.nih.gov/32744385/

-- and GLP-1RAs may actually be a treatment for sarcopenia (loss of skeletal muscle, esp in the elderly), perhaps by its suppressing myostatin, which is released by muscle cells and is a potent inhibitor of muscle protein synthesis. of note, bimagrumab, a myostatin inhibitor, has been shown to decrease total body fat mass, but increase lean body mass (skeletal muscle mass), along with decreasing HbA1c and body weight in a trial of 75 participants with type 2 diabetes and who were either overweight or obese (https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2774903). 

-- there is some evidence that the adipose tissue normally within the skeletal muscle is a result of the myocytes storing triglycerides for energy between the muscle bundles, and this intermuscular adipose tissue has been found to be a cardiovascular risk factor at close to the same cardiac risk level as visceral fat [and, is this part of GLP-1RA cardioprotection??]

    -- this intermuscular adipose tissue, IMAT (also referred to as myosteatosis), is increased in those with higher levels of general adiposity, is associated with insulin resistance, aging, poor muscle quality, and reduced mobility function

    -- and  when fat loss occurs with weight loss, it also happens in the fatty portion of skeletal muscle (IMAT): https://pmc.ncbi.nlm.nih.gov/articles/PMC2752367/

-- the SURPASS-3 study found that tirzepatide reduced muscle fat infiltration as compared to insulin, with a 4.4% decrease after 52 weeks: https://link.springer.com/article/10.1007/s00125-023-05969-6#Sec1 . Another study  of 128  women who were obese and had diabetes, found that liraglutide reduced thigh muscle fat by 7.8% and that the combo of high muscle fat and low muscle mass decreased as compared to controls 

-- this all brings up the issue that all skeletal muscle is not the same, and that the DXA quantification of muscle quantity is missing the more important point of the muscle quality (which can only be assessed by MRI)

-- so, GLP1-RAs may actually have a role in improving skeletal muscle mass and functioning....

 

Reasons for heterogeneity in the observed data

-- most studies had small sample sizes and many without sufficient statistical power

-- the major limitation was that of the 28 studies reviewed in the article, 23 used DXA for measurement of fat-free mass, but we know that DXA does not measure skeletal muscle directly nor the fat incorporated in the skeletal muscle

-- the studies were largely of the older GLP-1RAs (17 with liraglutide, 4 exenatide, and only 5 with semaglutide and 2 with tirzepatide)

-- some studies were done with patients having diabetes and some without (and we know that the incretin effect is unfortunately decreased in diabetics on the order of 60-70% vs those without diabetes, this likely being a large component of the glucose intolerance in diabetics and the positive effects of the GLP-1RAs)

 

Potential mitigation strategies to use with GLP-1RAs

-- very limited studies on this issue

-- it makes sense to encourage physical activity to increase muscle mass (an 8-week study of people on very-low-calorie diet but then put on liraglutide 3mg plus randomized to resistance exercise for 1 year, found a lower %FFML in those with the resistance exercise with liraglutide than those only getting liraglutide by 24.7% vs 38.6% . But another study found no benefit from resistance exercise

-- high protein diets may be beneficial, but no trials have assessed this (though it does seem intuitively obvious that this should have been assessed......)

-- there is speculation that slower weight loss than achieved in several studies may be better, though this has not been studies in patients on GLP-1RAs

 

 

Commentary:

-- GLP-1RAs along with glucose-dependent insulinotropic polypeptide (GIP), eg tirzepitide, have become a first-line therapy for weight loss by improving the incretin effect, stimulating pancreatic insulin release and decreasing glucagon release in response to glucose as well as slowing gastric emptying (prolonging a feeling of satiety), since these meds are just about as effective as bariatric surgery with sleeve gastrectomy 

-- these medications typically have between 10 and 20% weight reductions from baseline, with tirzepitide being the most potent 

-- the physiology of GLP-1 receptors and the GLP-1RAs is quite complex and continues to unfold. for an impressive review of the knowledge now, see GLP-1 physiology and receptors NatureMetab 2021 in dropbox, or https://doi.org/10.1038/s42255-020-00327-x

 -- in the past, i did find it particularly confusing that the incretin effect (gut hormones GLP-1 and GIP), leading to the pathway of increased insulin release when carbohydrates etc hit the stomach, could be stimulated by once a week peripheral subcutaneous injections.  this article, reflecting lots of research into these incretins, shows that the effects of both of these incretins are really much more complex and body-wide, and the CNS effects may be extraordinarily important https://pmc.ncbi.nlm.nih.gov/articles/PMC9417299/

 

-- Significant weight loss, metabolic bariatric surgery (MBS), and %FFML: 

    -- the amount of FFM loss after significant weight loss varies from a number of factors including baseline body composition, diet composition, age, ethnicity, hormonal status, and the amount and type of exercise 

    -- the overall percent of FFM lost based on DEXA in most studies is 20-30%, with Roux-En-Y gastric bypass having the most weight loss occurring within the first three months after surgery 

   -- women who had undergone Roux-en-Y bypass surgery at one year had lost 35.2% of the baseline weight, DEXA found that only 16.6% of FFM was lost and whole-body MRI found an 11.1% loss of skeletal muscle: i.e. DEXA reflected almost all of the skeletal muscle loss (96.7%). This of course raises questions as to whether these numbers would apply to later stages of weight loss and did not assess the quality of the skeletal muscle (and the issue of the quality of the skeletal muscle is one of the most important findings in this review)

 

Limitations to these studies:

 -- the authors note a few limitations that contribute to the observed heterogeneity in the data (see the above section on that for some of their concerns):

    -- DXA does not measure muscle directly, and that FFM "used describes all bodily tissues that are not adipose and is a crude surrogate measure for skeletal muscle",  ie, DXA is not a great discriminator of free-fat mass loss, but at best is assessing the quantity and not the quality of what is happening to skeletal muscle. and there are studies cited above that GLP-1RAs may even improve the skeletal muscle quality!!!

    -- more advanced imaging techniques, especially MRI, while more expensive and labor-intensive, would more accurately detect specific changes in skeletal muscle

-- several very important confounding factors that are not currently factored into the studies on GLP-1RAs, including sex, age, hormonal status, degree of weight loss, quantity and quality of exercise and diet (especially protein content of the latter) likely impact the degree of FFM loss relative to total weight loss. 

-- though there is a very real concern about loss of skeletal muscle, there are no clinical data showing that this is in fact clinically relevant.

    -- in this context, the data above finding large differences in the quality of skeletal muscle associated with GLP-1RAs, though intuitively important, needs to be evaluated directly

 

so,

-- i do realize that the above is "too much information", so i will summarize the above in a few bullet points:

    -- if you happen to be one of those perhaps masochistic people who actually want even more information about the physiology of GLP-1, see GLP-1 physiology and receptors NatureMetab 2021 in dropbox, or https://doi.org/10.1038/s42255-020-00327-x

-- all modes of weight loss are associated with high percentages of free-fat mass loss on the order of 20-50%, with perhaps ½ of that being skeletal muscle; a few points:

    -- skeletal muscle loss is a really important finding, given its importance in daily life activities, and ultimately in morbidity (falls, other injuries..) and mortality (hip fractures and their high mortality in the elderly, frailty,...). Though, there are no studies documenting that GLP-1RAs can be accused of that. And, of course, the weight loss by itself may be a life extender for many people (a study of people with obesity found that they had a loss in life expectancy of 2.3  years in 2016: https://www.thelancet.com/journals/eclinm/article/PIIS2589-53702200159-6/fulltext

    -- but the results of these studies above were not so consistent, likely related to large differences in study design, patients recruited, comorbidities, diet, exercise, measures to assess skeletal muscle, etc

        -- this article does delve into the issue of quality of skeletal muscle measured: the GLP-1RAs may actually improve the muscle quality by decreasing the amount of triglycerides around the muscle fibers

        -- using the most common and simplest DXA assessment of muscle quantity (which is actually an indirect and a not-totally-accurate measure of skeletal muscle mass), is missing perhaps the more important point of muscle quality as can be determined by MRI

            -- ie, the issue of weight loss is not the only or perhaps even the major factor in its effect on skeletal muscle; instead, the issue may well be how the weight is lost

        -- it is highly likely that other reversible risk factors are important: food composition and especially intake of adequate protein, exercise, physical functioning (especially in the elderly who might not walk around much: one week of bed rest leads to substantial muscle atrophy: https://diabetesjournals.org/diabetes/article/65/10/2862/35009/One-Week-of-Bed-Rest-Leads-to-Substantial-Muscle ; and in already frail elders, even a couple of days in the hospital without walking can lead to longstanding and sometimes permanent disability

    -- all of this means that we really need studies on the different modes of achieving the very important weight loss for many individuals and minimizing skeletal muscle loss, including an adequate protein intake and exercise program

    -- we do know the importance of these 2 interventions (diet and exercise) in improving skeletal muscle mass in general: https://pmc.ncbi.nlm.nih.gov/articles/PMC7400877/

    -- and, by the way, a meta-analysis of 19 studies of nitrogen balance in older adults found no difference in age in determining protein needs in elderly vs younger individuals: https://pubmed.ncbi.nlm.nih.gov/12499330/ : ie, we all need sufficient protein in our diets

-- bottom line: we also need more information about skeletal muscle and FFML% from studies with MRI, which gives much more information about muscle quality, as well as studies to document that improvement of skeletal muscle function by GLP-1RAs leads to actual clinical benefit. And we need information about whether the combination of GLP-1RAs with GIPs and perhaps with GAs (glucagon agonists) have the same protective effect on skeletal muscles.

-- another incidental issue is that GLP-1RAs and GIPs have myriad effects on the whole body (lungs, heart, kidneys, bones..). what is the clinical relevance of these non-muscle effects??

-- to me, the most important point in this study is that unlike other means of weight loss, GLP-1RAs seem to be protective of and perhaps even beneficial to skeletal muscle

 

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

  

to get access to all of the blogs:

 

 go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them

 

if you would like to see the articles in this blog, please email me. 

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list