GLP-1 Contraindications and 'Stop-and-Ask' Cases

Every FDA-approved GLP-1 receptor agonist carries a prescribing label with two distinct categories of caution. A contraindication is a bright-line case: the label states the drug should not be used at all in that circumstance. A warning or precaution describes a situation where the risk profile shifts and closer monitoring, or a case-by-case medical judgment, is expected. These categories exist because regulators and manufacturers were required to demonstrate, through trial data and post-marketing surveillance, that the listed risks are real and specific rather than generic caution. Understanding the difference matters most for anyone using a GLP-1 pathway compound without a prescriber actively screening for these factors, because in that situation the self-screening burden falls on the individual.

The mechanism behind these precautions traces back to what GLP-1 receptor activation does throughout the body, not only in the gut and pancreas. GLP-1 receptors are present in the thyroid C-cells of rodents, in gastric smooth muscle, in the biliary system indirectly through weight-loss-driven changes in bile composition, and in the retina's vascular supply, which is sensitive to rapid shifts in blood glucose. Each contraindication or warning below maps to one of these physiological touchpoints rather than being an arbitrary regulatory add-on.

Absolute contraindications appear on the labels of Wegovy, Ozempic, and Zepbound, among other approved brands. The first is a personal or family history of medullary thyroid carcinoma (MTC). In rodent studies, semaglutide and related GLP-1 drugs produced thyroid C-cell tumors at clinically relevant exposures; whether this translates to increased human risk has not been established, and long-term surveillance in humans taking these drugs has not shown a clear signal, but regulators have retained the contraindication because the rodent finding was consistent and dose-related. The second is Multiple Endocrine Neoplasia syndrome type 2 (MEN2), a genetic condition that carries elevated baseline MTC risk. The third is a documented serious hypersensitivity reaction, such as anaphylaxis, to the drug itself or its inactive ingredients. [1]

A second tier of 'strong caution' history includes conditions where the label does not flatly prohibit use but where clinicians commonly avoid the drug class when alternatives exist. A personal history of pancreatitis is the clearest example: because acute pancreatitis can be severe and repeated episodes can cause lasting pancreatic damage, many clinicians treat a pancreatitis history as a reason to look elsewhere first. Significant gallbladder disease is a related consideration. GLP-1 drugs are independently associated with new gallstone formation, and rapid weight loss compounds that risk further; someone with an active or recent gallbladder problem carries a different risk calculus than someone with no gallbladder history. Delayed gastric emptying disorders, most notably diabetic gastroparesis, deserve particular caution because GLP-1 receptor activation itself slows gastric emptying as its core mechanism. In someone whose stomach already empties abnormally slowly, adding a drug that further delays emptying can produce severe nausea, dehydration, or bowel-obstruction-like symptoms. [2] [3]

A third category involves conditions that call for monitoring rather than avoidance. Diabetic retinopathy is one: intensive, fast glucose improvement of any kind, not unique to this drug class, has been observed to transiently worsen retinopathy in some people with pre-existing disease, likely through a rapid shift in retinal blood flow and osmotic pressure rather than a direct drug toxicity. People with existing retinopathy who start a GLP-1 drug are generally advised to have ophthalmologic monitoring during the period of most active glucose or weight change. [1] Kidney function is a second monitoring point, largely indirect: the drug class does not require dose adjustment for most levels of kidney impairment, but the gastrointestinal side effects (vomiting, diarrhea, reduced fluid intake from appetite suppression) can cause dehydration that stresses the kidneys, particularly in someone who already has reduced kidney reserve. Pregnancy and active attempts to conceive are a third: GLP-1 drugs are not established as safe in pregnancy, weight loss during pregnancy is not generally recommended regardless of method, and current guidance is to discontinue GLP-1 therapy well before a planned pregnancy. [4]

There is a category that FDA labeling does not fully capture on its own: active eating disorders. GLP-1 drugs work in part by making eating physically less comfortable, through fullness signaling and slowed gastric emptying. For someone in an active anorexia, bulimia, or other restrictive eating pattern, a drug that amplifies the physical discomfort of eating can reinforce the disorder rather than treat an unrelated metabolic condition. This is not something a lab test screens for; it depends on honest personal history, which is one more reason coordinated care with both a physician and a mental health professional matters more here than in most other contraindication categories.

The evidence behind these precautions is a mix of categories worth naming explicitly. The thyroid C-cell finding is Animal Research that has not been confirmed or ruled out in humans; the gallbladder and pancreatitis associations come from Moderate Human Evidence in the randomized trials that support drug approval, where event rates were tracked prospectively across thousands of participants; the retinopathy interaction is Mechanistic Research extrapolated from experience with other rapid-acting glucose-lowering therapies, not a GLP-1-specific finding. Distinguishing these matters because they carry different levels of certainty, even though all of them appear as label language.

Limitations worth naming plainly: these contraindication categories describe population-level regulatory decisions, not an individualized risk assessment for any one person. A family history of thyroid cancer that turns out, on genetic testing, to be unrelated to MEN2 is a different risk situation than an MTC diagnosis in a first-degree relative, but the label language does not distinguish between them; that distinction requires a clinician. Similarly, 'significant gallbladder disease' is not a single defined threshold; some people with old, resolved gallstones tolerate the drug class without incident, and others with a similar history do not.

If a compound is being sourced outside regulated manufacturing and pharmacy channels, an additional layer of uncertainty sits on top of everything above: the active ingredient, its concentration, and the sterility of the preparation may not match what the label on the vial claims, which means published contraindication and monitoring guidance for the approved drug may not fully describe the actual product being used. That gap cannot be closed by reading more carefully; it requires independent verification of the product itself, which is rarely available outside laboratory testing.

A practical way to think about the difference between a contraindication and a warning is in terms of what regulators are asking a prescriber to do. A contraindication asks a prescriber not to write the prescription in that circumstance, full stop, because the potential harm is considered to outweigh any plausible benefit regardless of monitoring. A warning asks a prescriber to make an individualized judgment, factoring in the severity of the condition, the availability of alternatives, and the intensity of monitoring that is realistic for that specific patient. Outside a clinical relationship, there is no equivalent judgment being made by anyone, which is the core reason this category of self-sourced use carries structurally higher risk than the same compound used under active medical supervision, independent of anything about the product's authenticity.

It is also worth naming a pattern that appears repeatedly in adverse event reporting for this drug class: many of the most serious documented complications, severe pancreatitis, gallbladder emergencies requiring surgery, profound dehydration from uncontrolled vomiting, occurred in people who had one or more of the risk factors described above and did not have it identified or monitored before symptoms became severe. This is not evidence that the risk factors are rare; it is evidence that self-identification of risk, without a clinician actively screening history, family history, and current symptoms, is an imperfect substitute for the process these labels were written to support.