The shiny new darlings of Big Pharma are the GLP-1 RAs, drugs such as Ozempic, Wegovy, and Mounjaro that were originally designed to improve blood glucose control in type 2 diabetic patients, but were serendipitously discovered to cause weight loss in diabetics and non-diabetics. Now, it seems every day brings some new malady that *might* be solved by taking a GLP-1RA.

Here’s a partial list, generated by AI, of all the purported *possible* benefits supported by research, currently being investigated, or under consideration to be:

Type 2 diabetes mellitus, obesity and overweight with comorbidities, prediabetes and diabetes prevention, metabolic syndrome, insulin resistance, polycystic ovary disease, fatty liver, renal diseases, kidney stones, cardiovascular disease, heart failure, hypertension, dyslipidemia, chronic kidney disease, obesity‑related liver enlargement and liver enzyme elevation, post‑bariatric surgery weight regain, obstructive sleep apnea, asthma, shortness of breath of severe obesity, infertility associated with obesity/PCOS, low sex hormones linked to obesity and insulin resistance (men and women), osteoporosis, psoriasis and other dermatologic conditions, rheumatoid arthritis, gout, neurodegenerative disorders (Parkinson’s, Alzheimer’s, and other dementias), epilepsy, chronic pain syndrome, migraine, depression, addictive behaviors, substance abuse and eating disorders, and finally cancers of several types.

Some of these benefits are already pretty convincingly validated in RCTs—diabetes improvement and weight loss, for instance—while others are still in the tenuous or preliminary stages.

Case in point: the last mentioned—cancer. A recent article appeared in Cure (a website devoted to issues dealing with cancer) with the following headline: GLP-1 Associated With Colorectal Cancer Prevention, Research Finds.

The article detailed observational research, presented at the 2026 American Society of Clinical Oncology Gastrointestinal Cancers Symposium and published this month in the Journal of Clinical Oncology, comparing the use of GLP-1 RAs to aspirin for prevention of colorectal cancer (CRC), with a primary endpoint of the development of CRC and a secondary endpoint development of adverse events associated with either therapy.

From the jump, this is only an observational study, and as such it cannot on its own impute causality. And to be fair, they say that up front in the title: GLP-1 associated with…prevention. A good weasel word to be sure, but unfortunately one most of the public and journalists will ignore.

So the careful reader will see right away there’s no green light here for pulling out the GLP-1RA syringes to combat colon cancer based on this data, but it might suggest hypotheses that could later be investigated with a more rigorous trial. Let’s see if there’s anything robust here.

The study subjects’ data were drawn from a 150-million-patient multicenter database (TrinetX), and the design excluded those subjects who had taken any other form of NSAID besides aspirin over a given period, any other form of diabetic medication apart from a GLP-1RA, or who had reached a primary or secondary endpoint before the study window, which was the 6 months after their first documented use of the therapy (either aspirin or GLP-1RA). These eliminations resulted in two cohorts of approximately 140,000 subjects each, comprised overwhelmingly of middle-aged participants (mean age 58), who were white (67%) and women (69%), who were matched between the cohorts using propensity scoring.

Their conclusions (and the headline that kicked this discussion off) are prime examples of inflating a minuscule, might-possibly-be-an effect into something that, in the headline, sounds like a major news story about a big medical breakthrough.

First, their conclusions:

“GLP-1RAs were associated with a 26% relative reduction in CRC incidence compared to aspirin. These results along with the favorable safety profile of GLP-1RAs could underscore a potential public health impact and warrant prospective validation in randomized clinical trials.”

Wow! Twenty-six percent! That sounds amazing, doesn’t it? But just a quick look beyond the headline and the conclusion tells a different story.

“CRC incidence was 0.13% (183/140,758) in GLP-1RA users vs 0.176% (247/140,692) in aspirin users, yielding an ARR of 0.0455% and NNT of 2,198.”

So the bottom line is that after they ended their study and crunched the numbers, the difference between the GLP-1RA users’ incidence of developing CRC and that of the aspirin users was a mere 64 cases out of 281,000 patients.

If my math is correct, it’s 0.23 cases per 1,000 more or less. That’s about one-fifth of a case per 1,000 people treated. A very weak association. Truly negligible and practically not worth mentioning. (Unless you make GLP-1RAs, that is.)

In which case, you can massage the numbers with a little sleight of statistical hand to show that the 247 cases of CRC in the aspirin group versus the 183 in the GLP-1RA group represents a 26% relative reduction in cancer incidence, but that’s spurious. It’s a relative risk difference, not an absolute risk difference. It has an average risk reduction of 0.0455% with a large number needed to treat. Which sounds a lot less impressive than a 26% reduction. In other words, not actually a whole lot here to crow about.

There is some very preliminary research that suggests GLP-1RAs may have anti-inflammatory and even anti-proliferative activity in colon (and other) cancers, so I would agree that they might want to look further into them with a well-designed randomized controlled trial (RCT). But even if such a trial showed that giving people these drugs reduced the incidence of colon cancer (or anything else) can you tease out whether it is a newly recognized effect of the drug or merely the effect of the drug to help patients achieve better glucose control, satiety, and weight loss—to make them metabolically fitter in general. Let’s back up to look at the target of this receptor they’re so fond of agonizing.

What is GLP-1 anyway?

Glucagon-Like Peptide-1 is a small hormone of the incretin family, produced locally by enteroendocrine L-cells in the gut in response to foods entering the distal small intestine and first part of the large intestine. Its role is to help regulate insulin and glucagon release, gastric emptying, and appetite. Its release amplifies insulin secretion if the food coming in is carbohydrate-rich and glucose is elevated, but it has minimal effect on insulin if not. It also suppresses glucagon secretion (insulin’s opposing hormone) if glucose levels are above the fasting baseline (but not if there is hypoglycemia). It acts to slow gastric emptying time creating a sense of postprandial fullness, which suppresses appetite.

And all of that action seems to be of physiologic benefit to those who have diabetes, obesity, or some other component of metabolic syndrome.

But if stimulating GLP-1 receptors (which is what GLP-1RAs do) is the biochemical mechanism behind these purportedly positive effects, why not just stimulate the release of GLP-1 with food and let it do its natural thing by binding itself to its receptors?

Dietary fat and protein each roughly double to quadruple circulating GLP‑1 after a meal, with levels peaking within a half hour to an hour and a return toward baseline over roughly two to three hours. Fasting GLP‑1 is typically around 5 to 15 pmol/L. Mixed meals generally raise total GLP‑1 about 2- to 4‑fold, regardless of macronutrient makeup, when energy is reasonably matched.

High‑protein meals can produce some of the highest GLP‑1 responses seen with physiologic feeding. In a crossover study, a 60% protein breakfast produced higher GLP‑1 levels at two hours that remained elevated for the 4‑hour sampling period than did isocaloric high‑fat or high‑carb meals.

But dietary fat is itself a potent GLP‑1 secretagogue in both humans and animals, triggering detectable GLP‑1 release within about 30 minutes that peaks around one hour.

Protein plus calcium (e.g., dairy) appears particularly effective, with human work showing some of the highest postprandial GLP‑1 concentrations reported for normal feeding when the two are combined.

When calories are matched, high‑protein and high‑fat meals both stimulate GLP‑1, but several studies show greater GLP‑1 area under the curve after high‑protein or protein‑enriched drinks compared with carbohydrate‑rich or fat‑rich comparators.

In summary, carbohydrates (especially glucose) trigger GLP‑1 quickly via proximal small intestinal sensing and this action will amplify the signal for insulin release. Fat provides strong and somewhat more prolonged stimulation as it reaches the distal bowel’s L‑cells, and protein (and certain amino acids) can produce robust GLP‑1 release that is enhanced by calcium.

So eating meals that are high in quality protein and fat, and limited in carbohydrate would seem like a recipe for doing what a GLP-1RA does. And clinical research bears out that a well-formulated, protein-rich, fat-rich, low-carb ketogenic diet can indeed control diabetes, reduce excess weight and body fat, improve cardiovascular blood lipid measures, resolve fatty liver, improve sleep apnea, and more, all without the untoward potential GLP-1RA side effects that can include such maladies as nausea, vomiting, gastroparesis, lean body mass loss, Ozempic butt, and Wegovy face.

In fact the list of possible adverse side effects—from the common to the rare—is longer than the list of proposed targets for treatment:

Nausea, early satiety, post‑prandial fullness, vomiting, diarrhea, constipation, abdominal pain, cramping, bloating, dyspepsia/heartburn, fatigue, dizziness, headache, injection‑site reactions (erythema, pruritus, nodules, mild pain), hypoglycemia when combined with insulin or insulin secretagogues (e.g., sulfonylureas), mild resting tachycardia, possible orthostatic symptoms (low blood pressure on standing from volume depletion of GI fluid losses), biliary, pancreatic, and GI motility slowing, gall stones and inflammation, gall bladder colic, acute pancreatitis (rare but clinically important), worsening or unmasking of gastroparesis, rare bowel obstruction or ileus, acute kidney injury from volume depletion, electrolyte disturbances secondary to GI fluid loss, boxed warning for thyroid C‑cell tumors (medullary thyroid carcinoma) in rodents; contraindication in patients with personal/family history of MTC or MEN2, systemic allergic reactions, very rarely anaphylaxis, development of anti–drug antibodies, transient visual blurring related to rapid glycemic shifts, worsening of pre‑existing diabetic retinopathy when glucose control improves very rapidly (not uniformly seen, but enough to be on warning labels for some agents), optic neuropathy and other visual events with some agents (causality not fully established), hair loss (alopecia) reported anecdotally and in some pharmacovigilance datasets, mood changes and rare reports of suicidal ideation are being actively surveilled but not definitively linked.

So, if you want to rid yourself or those you love or care for from the specter of that long list of diseases at the start while avoiding that long list of adverse side effects above, the best nutritional advice comes back to this: eat real food. Eat meat, fish, poultry, eggs, dairy, green leaves and vegetables, nuts and seeds, little fruit, some starch, and no sugar.