A Closer Look at Long-Term Exposure, Side Effects, Nutrition, Skeletal Muscle, Weight Regain, Cancer Risk—and What Happens Downstream
Glucagon-like peptide-1 (GLP-1)-based medications have changed the weight-loss conversation.
People who have struggled with obesity for years are losing substantial amounts of weight. Hunger can become quieter. Food cravings can diminish. People may become full after eating much less than they once did.
For many people, these changes can be life-changing. GLP-1-based medications have demonstrated important benefits for weight loss, blood sugar control, and cardiovascular health in appropriate populations.
But I believe we make a major mistake when we look at the number on the scale and assume it tells us everything we need to know about what is happening inside the body.
Because weight loss is only one outcome.
What happens to nutrition?
What happens to skeletal muscle?
What happens to body composition?
What happens to fiber intake?
What happens to essential fatty acids?
What happens to the gallbladder?
What happens to the brain and systems involved in appetite, reward, and motivation?
What happens when the medication is discontinued?
What happens if substantial weight is regained?
What happens if someone loses 50 or 75 pounds but never learns how to eat in a way that can support that weight loss without pharmacological appetite suppression?
What happens to someone who begins taking one of these medications at age 65 or 70, when preserving skeletal muscle, strength, and independence becomes increasingly important?
And what happens when millions of people potentially remain on these medications for years—or perhaps decades?
That brings me to the central question of this article:
WHAT HAPPENS DOWNSTREAM?
The Question Is Bigger Than “Do GLP-1 Drugs Cause Cancer?”
I am not interested in reducing this entire discussion to one question:
“Do GLP-1 drugs directly cause cancer?”
Because even that question can become misleading if we leave out one critically important variable:
TIME.
Cancer often does not develop overnight.
Depending on the cancer and the exposures involved, the biological processes that eventually lead to detectable disease can unfold over many years.
That means we must be extremely careful about what studies lasting months or a few years can tell us about what might happen after 10, 20, or 30 years of exposure.
Consider cigarette smoking as an example.
Imagine conducting a large study in which people smoked cigarettes for a relatively short period and investigators did not detect a statistically significant increase in cancer during that limited observation window.
Would that establish that cigarettes do not cause cancer?
Of course not.
The obvious problem would be:
WE DIDN’T WAIT LONG ENOUGH TO ANSWER THE QUESTION.
I am not saying GLP-1 medications are equivalent to cigarettes.
They are not.
The comparison illustrates a fundamental principle of cancer research:
THE ABSENCE OF A CANCER SIGNAL DURING A LIMITED PERIOD OF OBSERVATION DOES NOT ESTABLISH THE ABSENCE OF LONG-TERM CANCER RISK.
That distinction is essential when interpreting GLP-1 research.
A major 2026 meta-analysis pooled 148 randomized controlled trials involving 168,875 participants and reported an overall cancer relative risk (RR) of 0.99, with a 95% confidence interval (CI) of 0.92 to 1.05. [1]
Researchers therefore did not detect an increase in overall cancer incidence across the randomized trials they analyzed.
That is worth knowing.
But we must be precise about what it means.
It means an increased overall cancer incidence was not detected during the observation periods represented by those trials.
It does not mean:
GLP-1 DRUGS HAVE BEEN PROVEN INCAPABLE OF CAUSING CANCER.
It does not mean:
10, 20, OR 30 YEARS OF EXPOSURE HAVE BEEN PROVEN CANCER-SAFE.
And it does not answer what might happen indirectly through nutritional, behavioral, physiological, and body-composition changes during long-term treatment.
Those decades of evidence simply do not exist yet for today’s widespread use of long-acting GLP-1-based medications for obesity.
That gives us one of the most important principles in this article:
MORE PEOPLE ≠ MORE TIME.
You could study one million people for a relatively short period and gain tremendous statistical power to identify certain short-term effects.
But you still would not know what happens after 20 years.
SAMPLE SIZE CANNOT SUBSTITUTE FOR LATENCY.
Instead of asking only:
“Does the drug directly cause cancer?”
we should also ask:
WHAT DOES THE DRUG SET IN MOTION?
Medicine Isn’t Always a Straight Line
People often think about medications in a simple cause-and-effect sequence.
A person takes Drug A.
Later, that person develops Disease B.
Then everybody asks:
Did Drug A cause Disease B?
But biology is often more complicated.
A medication produces a biological effect.
That effect can alter appetite, food intake, digestion, hydration, nutrition, metabolism, body composition, activity, hormonal signaling, or behavior.
Those changes can then produce consequences of their own.
The pathway may look more like this:
DRUG → BIOLOGICAL EFFECT → PHYSIOLOGICAL OR BEHAVIORAL CHANGE → DOWNSTREAM CONSEQUENCE
That’s how I want people to think about this issue.
A medication does not necessarily have to directly damage an organ for something occurring during treatment to affect that organ later.
A medication does not necessarily have to be directly carcinogenic for something happening during years of treatment to influence cancer-related risk factors potentially.
Sometimes the better question isn’t:
“Did the drug directly cause the disease?”
Sometimes it is:
“WHAT DID THE DRUG CHANGE, AND WHAT HAPPENED BECAUSE OF THAT CHANGE?”
Don’t stop at the drug.
FOLLOW THE PATHWAY.
Why Does It Feel Like These Drugs Are Being Pushed So Aggressively?
There is another issue I believe deserves attention.
In my own work with clients, I am hearing something with increasing frequency.
People tell me they went to their healthcare provider and were quickly offered a GLP-1-based medication.
When one medication produced side effects, some were encouraged to change doses, try another medication, or keep searching for one they could tolerate.
The conversation, as some clients describe it, sounds something like:
“Let’s try this one.”
“If you don’t tolerate that one, let’s try another.”
“Let’s see which one you can tolerate.”
There can absolutely be legitimate reasons for doing this.
People respond differently to medications. Tolerability varies. Dose titration and changing therapies can be appropriate medical decisions.
Obesity is also increasingly treated medically as a chronic disease, and pharmacotherapy has moved much closer to the center of obesity management.
So I understand why more physicians are prescribing these medications.
What concerns me is something different.
ARE WE PRESCRIBING THE DRUG—OR MANAGING THE WHOLE PERSON?
If we’re willing to spend months figuring out which medication someone tolerates, are we putting the same effort into determining whether that person is consuming adequate protein?
Are we monitoring skeletal muscle?
Are we asking whether the person is performing resistance exercise?
Are we assessing fiber?
Micronutrients?
Essential fatty acids?
Physical function?
Mental health?
Gastrointestinal symptoms?
Gallbladder health?
And perhaps one of the most important questions of all:
ARE WE TEACHING THE PERSON HOW TO EAT?
My concern isn’t simply that doctors prescribe GLP-1 medications.
My concern is what happens if the definition of success becomes:
Can you tolerate the medication?
and
Is the scale going down?
Those cannot be the only measurements of success.
GLP-1 Drugs Have Side Effects—Some of Them Serious
Before we even get into the possible long-term consequences, we should acknowledge something already well established:
GLP-1-BASED MEDICATIONS CAN PRODUCE SIDE EFFECTS.
Many of the most common side effects are gastrointestinal.
Depending on the medication, these can include nausea, vomiting, diarrhea, constipation, abdominal discomfort or pain, indigestion, and other gastrointestinal symptoms. [2,3]
For example, the FDA specifically identifies nausea, diarrhea, vomiting, constipation, abdominal discomfort and pain, fatigue, injection-site reactions, hypersensitivity reactions, belching, hair loss, and gastroesophageal reflux disease among potential side effects of tirzepatide used for chronic weight management. [3]
The prescribing information for these medications also contains warnings and precautions concerning potentially more serious problems, depending upon the specific drug and patient, including pancreatitis, gallbladder disease, kidney injury associated with volume depletion, severe gastrointestinal reactions, and hypoglycemia when used with certain glucose-lowering medications. [2,3]
And then there is a warning that is particularly relevant to the subject of this article:
THYROID C-CELL TUMORS.
Semaglutide and tirzepatide carry FDA boxed warnings concerning thyroid C-cell tumors. [2,3]
In animal studies, these medications caused thyroid C-cell tumors in rats.
The FDA-approved labeling states that it remains unknown whether these medications cause thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), in humans. [2,3]
Because of this concern, these medications are contraindicated in people with a personal or family history of MTC or Multiple Endocrine Neoplasia syndrome type 2 (MEN2). [2,3]
That distinction matters.
I am not saying:
“GLP-1 DRUGS CAUSE THYROID CANCER IN HUMANS.”
That has not been established.
What I am saying is:
THE FDA CONSIDERS THE ANIMAL FINDINGS IMPORTANT ENOUGH TO REQUIRE A BOXED WARNING WHILE ACKNOWLEDGING THAT THE HUMAN RISK REMAINS UNKNOWN.
And isn’t that exactly the type of distinction this entire article is about?
There are things we know.
There are things we don’t know.
And there are things we may not be able to know with confidence until enough people have been exposed for enough time.
That’s why the better question isn’t simply:
“Has thyroid cancer been proven?”
It is:
“WHAT DOES THE EVIDENCE SHOW TODAY, WHAT REMAINS UNKNOWN, AND ARE WE FOLLOWING THESE QUESTIONS LONG ENOUGH TO FIND OUT?”
A Drug Can Suppress Appetite. It Cannot Teach You How to Eat.
This may be one of the saddest parts of what I am seeing.
In my own work, I increasingly encounter people who have successfully lost weight with GLP-1 medications but received very little meaningful nutrition education along the way.
A medication can suppress appetite.
It can make someone feel full sooner.
It can reduce food cravings.
It can make eating dramatically less much easier.
But:
A MEDICATION CANNOT, BY ITSELF, TEACH SOMEONE HOW TO EAT.
It doesn’t teach someone how to build a balanced meal.
It doesn’t teach them how much protein they need.
It doesn’t teach them how to get enough fiber.
It doesn’t teach them how to choose nutrient-dense foods.
It doesn’t teach them how to grocery shop.
It doesn’t teach them how to navigate a restaurant.
It doesn’t teach them how to distinguish physical hunger from boredom, stress, habit, or emotional eating.
It doesn’t teach them how to handle vacations, holidays, celebrations, and stressful periods.
And it doesn’t automatically establish the behaviors someone may need if the medication eventually goes away.
The medication may have changed their appetite.
BUT DID IT CHANGE THEIR SKILLS?
Those are not the same thing.
Most People Aren’t Staying on These Drugs
This is where the real-world evidence becomes particularly interesting.
A large U.S. study examined 125,474 adults with overweight or obesity who initiated a GLP-1 receptor agonist. [4]
By one year:
53.6% HAD DISCONTINUED.
By two years:
72.2% HAD DISCONTINUED.
Among participants without type 2 diabetes, the numbers were even more striking:
64.8% discontinued by one year.
84.4% discontinued by two years.
These aren’t estimates I’m making. They come directly from the study’s analysis.
Think about what that means.
In this large real-world population, more than 8 out of 10 people without type 2 diabetes had discontinued within two years.
There isn’t one reason everyone stops.
Side effects can matter. In this study, moderate or severe gastrointestinal adverse events were associated with a greater likelihood of discontinuation. Cost, access, insurance coverage, and individual preferences can also matter. [4]
Some people restart.
Interestingly, the same study found that greater weight regain after discontinuation was associated with a greater likelihood of restarting a GLP-1 medication. [4]
This creates a glaring disconnect:
We’re increasingly discussing these medications as long-term treatment.
Yet:
Large numbers of real-world patients aren’t remaining on them.
So I believe every person beginning treatment deserves another question:
WHAT IS THE LONG-TERM PLAN?
What Happens When the Medication Stops?
This is where the story becomes even more complicated.
When semaglutide treatment was withdrawn in the STEP 1 extension, participants regained approximately two-thirds of their previous weight loss within one year. [5]
Many of the cardiometabolic improvements achieved during treatment also moved back toward baseline.
Tirzepatide research has demonstrated the same general phenomenon: stopping treatment can result in substantial weight regain, whereas continued treatment helps maintain or extend weight reduction. [6]
I would never say every person gains every pound back.
The evidence doesn’t support that universal claim.
But substantial weight regain following discontinuation is clearly a real concern.
Suppose someone loses 60 pounds because their appetite has been dramatically suppressed.
Then insurance changes.
Or the medication becomes unaffordable.
Or side effects become unacceptable.
Or the person simply decides they don’t want to take it anymore.
Their appetite returns.
What now?
What did they learn during those previous two years?
What habits did they establish?
Did they learn how to structure meals?
Did they learn portion management?
Did they establish adequate protein intake?
Did they develop an exercise routine?
Did they learn strategies for managing hunger?
Did they build behaviors capable of supporting them when pharmacological appetite suppression disappeared?
Because:
A DRUG CAN SUPPRESS APPETITE.
IT CANNOT REPLACE EDUCATION.
If these medications are going to become a major component of obesity treatment, nutrition education and sustainable behavioral skills should not be an afterthought.
They should be part of treatment from day one.
We Don’t Have Lifetime Data—Yet We’re Discussing Long-Term Treatment
And now we arrive at a fascinating tension.
We do not have 20-, 30-, or 40-year data from people continuously using today’s long-acting GLP-1-based medications for obesity.
Those data cannot exist yet.
At the same time, obesity is increasingly treated as a chronic disease, and long-term pharmacotherapy is commonly discussed because discontinuation frequently results in weight regain.
These two realities create an enormous responsibility.
WE DON’T HAVE LIFETIME DATA.
Yet for some patients:
WE’RE TALKING ABOUT POTENTIALLY LIFELONG TREATMENT.
That doesn’t mean long-term treatment is necessarily inappropriate.
Many chronic medications are used before decades of outcome data accumulate.
But it makes continued surveillance more important—not less.
Nutrition becomes more important.
Body composition becomes more important.
Skeletal muscle becomes more important.
Physical function becomes more important.
And long-term disease outcomes become more important.
If the plan is to take the drug for life, we need to keep studying what happens over a lifetime.
What the Great Starvation Experiment Taught Us
Long before GLP-1 medications existed, researchers conducted one of the most famous experiments in nutrition science.
It became known as the Minnesota Starvation Experiment.
The original experiment was conducted during World War II under the direction of Ancel Keys at the University of Minnesota.
Healthy young men underwent prolonged semistarvation followed by nutritional rehabilitation.
The experiment taught us something that remains important today:
WHEN YOU SUBSTANTIALLY REDUCE FOOD INTAKE, THE BODY DOESN’T SIMPLY BECOME A SMALLER VERSION OF ITSELF.
The body adapts.
Researchers later revisited the Minnesota data to examine adaptive thermogenesis—the reduction in resting energy expenditure beyond what would be predicted from changes in body composition.
The research documented reductions in fat mass and fat-free mass (FFM), including skeletal muscle, along with changes in resting energy expenditure and multiple physiological variables. [7]
The original experiment also became famous because the effects extended beyond body weight. Participants experienced substantial physical and psychological changes involving energy, strength, mood, behavior, sexual interest, and their relationship with food.
Now let me make the distinction very clear:
TAKING A GLP-1 DRUG IS NOT THE MINNESOTA STARVATION EXPERIMENT.
I’m not saying people taking semaglutide or tirzepatide are starving.
I’m not saying their degree of caloric restriction is comparable.
The experimental circumstances were very different.
What matters is the biological lesson:
Substantial and sustained reductions in energy intake can produce adaptations extending far beyond body weight.
And reducing energy intake is central to how GLP-1-based medications produce weight loss.
Therefore, the Minnesota experiment should make us more curious, not less.
If food intake is substantially reduced for months or years:
What happens to resting energy expenditure?
What happens to skeletal muscle?
What happens to physical activity?
What happens to nutrient intake?
What happens to hormones?
What happens when normal appetite returns?
What happens during weight regain?
And what happens if someone goes through repeated cycles of:
MEDICATION → WEIGHT LOSS → DISCONTINUATION → WEIGHT REGAIN → RESTARTING?
The Minnesota experiment cannot answer those questions for GLP-1 medications.
But it teaches us why those questions deserve to be asked.
How You Lose the Weight Matters
Imagine two people who both lose 50 pounds.
The scale gives them exactly the same result:
MINUS 50 POUNDS.
But physiologically, those two people might have very different outcomes.
One person might eat adequate protein, consume plenty of nutrient-dense foods, maintain sufficient fiber intake, perform resistance exercise several times a week, and preserve much of their skeletal muscle while primarily losing excess body fat.
The other person may barely eat because their appetite has dramatically declined.
Protein could fall.
Fiber could fall.
Food variety could fall.
Resistance exercise might be absent.
That person could lose body fat—but also meaningful fat-free tissue.
Both scales say:
Minus 50 pounds.
But those are not necessarily the same health outcome.
That’s why:
HOW MUCH WEIGHT YOU LOSE MATTERS.
HOW YOU LOSE IT MATTERS TOO.
When Appetite Goes Down, Nutrition Has to Become More Intentional
One reason these medications help people lose weight is simple:
They reduce food intake.
That can be beneficial.
If someone consumes fewer calories from oversized portions, refined carbohydrates, sugary foods, fried foods, or ultra-processed foods, that could substantially improve health.
But whenever somebody eats much less food, we should ask:
WHAT ARE THEY EATING LESS OF?
Because eating substantially less could also mean less protein, fiber, vegetables, fruit, seafood, calcium, magnesium, zinc, iron, vitamins, eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA).
Those deficiencies do not automatically occur.
A person can reduce calories dramatically while improving nutrient density.
But it requires intention.
That’s why I believe:
DON’T LET APPETITE SUPPRESSION BECOME NUTRITIONAL SUPPRESSION.
If you’re going to eat less:
WHAT YOU EAT MATTERS MORE.
Skeletal Muscle May Be One of the Biggest Missing Pieces
Perhaps no downstream issue concerns me more than skeletal muscle.
Skeletal muscle isn’t just about looking muscular.
It plays important roles in glucose metabolism, insulin sensitivity, physical strength, balance, mobility, physical function, and our ability to remain independent as we age.
Weight loss generally includes more than body fat.
Some fat-free tissue is usually lost as well.
But we have to be precise:
FAT-FREE MASS (FFM) IS NOT THE SAME THING AS SKELETAL MUSCLE MASS (SMM).
William Evans and Steven Cummings made this point in JAMA in 2024. They argued that reassurance about muscle loss during medication-assisted weight reduction was premature because much of the research had measured FFM rather than directly measuring skeletal muscle mass (SMM). They called for better measurement of skeletal muscle, strength, and physical function. [8]
If a study says someone lost ten pounds of FFM, we cannot say they lost ten pounds of skeletal muscle.
But we also cannot assume their skeletal muscle was fully protected.
So I want to know:
How much actual skeletal muscle did they preserve?
What happened to strength?
What happened to physical performance?
Did they resistance train?
Were they eating enough protein?
And what happens downstream if meaningful skeletal muscle is lost?
And What About Someone Who Is 65, 70, or 75?
This may be the population that concerns me most.
Age itself changes the conversation.
Older adults already face an increased risk of losing skeletal muscle mass, strength, and physical function.
Eventually, those losses can contribute to sarcopenia.
Sarcopenia isn’t merely about having smaller muscles.
It can affect strength, balance, walking, mobility, recovery from illness, risk of falls, risk of fractures, and perhaps most importantly:
INDEPENDENCE.
Now put substantial weight loss on top of aging.
We know GLP-1-assisted weight loss can include loss of FFM.
Again:
FFM is not synonymous with skeletal muscle.
And I am not saying every older person taking a GLP-1 medication develops sarcopenia.
But that uncertainty gives me another question:
WHY WOULDN’T MUSCLE PRESERVATION BE A PRIMARY TREATMENT GOAL?
If I’m working with a 35-year-old who loses substantial weight, skeletal-muscle preservation matters.
If I’m working with a 70-year-old, it may matter even more.
That person may already have age-related muscle loss.
They may already have lower strength.
They may already consume inadequate protein.
They may already perform little resistance exercise.
Then we introduce a medication that substantially reduces appetite and produces substantial weight loss.
What happens if protein intake falls?
What happens if resistance exercise isn’t part of treatment?
What happens if meaningful skeletal muscle is lost?
What happens if that 70-year-old remains on treatment until they’re 75?
Or 80?
I don’t want to know only:
“How much weight did you lose?”
I want to know:
How much fat did you lose?
What happened to your skeletal muscle?
What happened to your strength?
Can you get out of a chair as easily as before?
Are you walking better or worse?
Are you eating enough protein?
Are you resistance training?
And ultimately:
ARE WE IMPROVING YOUR HEALTHSPAN—OR MERELY LOWERING YOUR BODY WEIGHT?
Because for an older adult, losing weight while becoming weaker would not necessarily represent the health victory the scale makes it appear to be.
WEIGHT LOSS AND HEALTH IMPROVEMENT ARE NOT ALWAYS SYNONYMOUS.
More Than 414,000 People Give Us a Reason to Care About Muscle
An important United Kingdom Biobank study followed 414,094 participants for a median of 11.7 years.
During that period, 63,379 participants developed cancer.
Researchers examined sarcopenia and subsequent cancer incidence.
Among men, sarcopenia or probable sarcopenia was associated with higher incidence of several cancers, including liver, hematologic, and colorectal cancers. [9]
This was observational research.
It does not prove sarcopenia caused those cancers.
And this wasn’t a GLP-1 study.
THAT’S PRECISELY THE POINT.
It tells us something about a potential downstream variable:
SKELETAL MUSCLE MATTERS.
If substantial medication-assisted weight loss includes meaningful skeletal-muscle loss, we should care about that independently of whether the medication itself directly causes cancer.
And notice something else:
11.7 YEARS OF FOLLOW-UP.
Again:
TIME MATTERS.
Body Composition May Matter to Cardiovascular Health Too
Cancer isn’t the only reason to care about body composition.
A longitudinal study followed 1,048 adults between ages 50 and 80.
Participants who lost more than 8% of their FFM percentage had a hazard ratio (HR) of approximately 3.83 for subsequent cardiovascular events compared with those whose FFM percentage remained stable. [10]
That’s approximately a 283% higher relative hazard.
This was observational research.
It was not a GLP-1 study.
And FFM isn’t synonymous with skeletal muscle.
But the study reinforces a larger principle:
BODY COMPOSITION MATTERS.
If someone loses 60 pounds, don’t just ask:
How much did you lose?
Ask:
WHAT DID YOU LOSE?
Fiber Is Another Downstream Question
Now follow another pathway.
Appetite decreases.
Food intake decreases.
Dietary variety may decrease.
Fiber intake could decrease.
Why should we care?
Because dietary fiber has repeatedly been associated with colorectal health.
A systematic review and dose-response meta-analysis of prospective studies reported that every additional 10 grams per day of dietary fiber was associated with approximately a 10% lower relative risk of colorectal cancer. [11]
Does that prove:
GLP-1 → LOW FIBER → COLORECTAL CANCER?
No.
That’s not what I’m saying.
I’m asking:
IF LONG-TERM APPETITE SUPPRESSION RESULTS IN CHRONICALLY INADEQUATE FIBER INTAKE, COULD WE BE MOVING AN ESTABLISHED COLORECTAL-HEALTH VARIABLE IN THE WRONG DIRECTION?
If that’s possible, why wouldn’t we want to know?
Why wouldn’t we measure it?
Why wouldn’t we correct it?
LOOK DOWNSTREAM.
EPA, DHA, Vitamins, and Minerals Cannot Be Assumed
The same principle applies to essential fatty acids.
Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are long-chain omega-3 fatty acids with important biological functions.
I am not saying someone taking a GLP-1 medication is automatically deficient in EPA or DHA.
But when total food intake is substantially reduced, nutritional adequacy deserves more—not less—attention.
If you have fewer calories available every day, those calories need to deliver the nutrients your body requires.
That means thinking about protein, fiber, EPA and DHA, iron, magnesium, calcium, zinc, vitamin D, B vitamins, vitamin C, polyphenol-rich foods, and overall dietary quality.
Instead of automatically assuming:
“I’m losing weight, therefore everything about my health is improving,”
I would rather ask:
WHAT HAPPENED TO YOUR NUTRITIONAL STATUS WHILE YOU WERE LOSING THE WEIGHT?
Natural GLP-1 and Long-Acting Drug Exposure Are Very Different
Naturally produced active GLP-1 has a circulating half-life of approximately one to two minutes.
It is rapidly broken down, in large part by the enzyme dipeptidyl peptidase-4 (DPP-4). [12]
Modern medications are specifically engineered to last much longer.
Semaglutide has an elimination half-life of approximately one week. [2]
Tirzepatide has an elimination half-life of approximately five days and also acts on the glucose-dependent insulinotropic polypeptide (GIP) receptor. [3]
That extended exposure is part of what makes these medications effective.
I am not saying the longer half-life makes them dangerous.
A longer half-life by itself is not evidence of harm.
But biologically, the exposure pattern is clearly different.
Natural active GLP-1:
MINUTES
Some modern GLP-1-based medications:
DAYS
That gives us another reasonable scientific question:
WHAT ARE THE LONG-TERM CONSEQUENCES—BENEFICIAL, HARMFUL, OR NEUTRAL—OF PHARMACOLOGICALLY STIMULATING THESE SYSTEMS FOR YEARS OR DECADES?
We don’t completely know yet.
Because those decades haven’t happened.
GLP-1 Signaling Reaches Beyond Hunger
GLP-1 biology involves more than the stomach.
GLP-1 signaling interacts with areas of the brain involved in appetite, satiety, motivation, and reward.
Researchers are investigating these medications for conditions involving alcohol and other reward-related behaviors.
That could become another therapeutic benefit.
But it should also make us curious.
If we alter biological systems involved in appetite and reward for years:
What else changes?
Motivation?
Food reward?
Alcohol desire?
Pleasure?
Mood?
Sexual desire?
Compulsive behavior?
There is not evidence that GLP-1 medications universally cause depression or reduced libido.
We shouldn’t claim that.
But I believe this remains a legitimate research question:
WHEN APPETITE GOES DOWN, WHAT ELSE HAPPENS TO DESIRE?
Rimonabant: A Lesson From Weight-Loss Drug History
History provides a useful example of why we should never judge a weight-loss medication solely by how much weight people lose.
Rimonabant was an obesity medication sold in Europe under the brand name Acomplia.
It was not a GLP-1 medication.
It acted through a completely different system by blocking cannabinoid type 1 (CB1) receptors.
The medication produced weight loss.
But psychiatric safety concerns emerged, and the European Medicines Agency (EMA) ultimately concluded that its risks outweighed its benefits and recommended suspension of its marketing authorization in 2008. [13]
Why bring up rimonabant?
Not because GLP-1 medications act like rimonabant.
They don’t.
Not because history tells us GLP-1 drugs will have the same outcome.
It doesn’t.
The lesson is broader:
DON’T ASSUME THE INTENDED EFFECT IS THE ONLY EFFECT THAT MATTERS.
Gallbladder Disease Gives Us a Real Downstream Example
We don’t have to speculate about every downstream consequence.
Some are already documented.
A JAMA Internal Medicine systematic review and meta-analysis included 76 randomized clinical trials involving 103,371 participants.
GLP-1 receptor agonist therapy was associated with increased gallbladder and biliary disease.
The overall RR was 1.37.
In weight-loss trials, the RR was 2.29. [14]
That’s a 129% higher relative risk, not a 129-percentage-point increase in absolute risk.
Rapid and substantial weight loss itself can contribute to gallstone formation.
That makes gallbladder disease an excellent illustration of the larger concept.
Do we need to prove that a GLP-1 molecule physically formed the gallstone?
No.
The better question is:
WHAT HAPPENED DURING TREATMENT THAT INCREASED THE RISK?
That is downstream thinking.
Now Let’s Look Closely at the Cancer Research
This is where I believe people need to slow down.
The 2026 Scientific Reports meta-analysis sounds extraordinarily powerful:
148 RANDOMIZED CONTROLLED TRIALS
168,875 PEOPLE
OVERALL CANCER RR: 0.99
But before anyone says:
“Case closed,”
we need to understand what was actually studied.
The 168,875 participants were not one giant group taking the same GLP-1 medication, at today’s obesity doses, continuously for 20 years.
They were pooled from 148 separate trials involving different populations, medications, doses, indications, comparators, and treatment durations. [1]
The main publication does not provide one single overall average exposure duration for all 168,875 participants.
So we should not pretend that it does.
What we can say is:
THOSE 168,875 PEOPLE DO NOT REPRESENT 168,875 PEOPLE EXPOSED FOR DECADES.
That’s the issue.
RR 0.99: What Does That Number Actually Mean?
The reported RR was:
0.99
with a 95% CI of:
0.92–1.05
That means the pooled randomized evidence did not detect an overall cancer difference between GLP-1 groups and their comparators during the durations represented in those trials. [1]
That’s what the result tells us.
It doesn’t tell us what happens after decades.
When people hear:
168,875 participants and RR 0.99
they may hear:
“Scientists have proven these drugs don’t cause cancer.”
But that goes beyond the evidence.
The more accurate interpretation is:
Across the randomized trials currently available, investigators did not detect an increase in overall cancer incidence during the periods studied.
And then this sentence belongs immediately behind it:
Those data cannot establish the cancer effects, if any, of exposure extending for decades beyond the observation periods currently available.
Those statements belong together.
This Is What I Mean by “Science Washing”
I use the term science washing to describe situations in which legitimate scientific findings are presented in a way that can make the evidence appear more complete or definitive than it actually is.
I’m not accusing researchers of fabricating science.
I’m not saying the statistics are fake.
The number may be perfectly accurate.
The interpretation can still go too far.
For example:
“168,875 participants showed no increase in overall cancer risk.”
Now add:
“Those participants came from 148 different randomized trials and do not represent decades of continuous exposure.”
That changes the context considerably.
Not bad science.
Not useless science.
Properly bounded science.
THE PROBLEM ISN’T NECESSARILY THE STATISTIC.
THE PROBLEM IS WHAT THE STATISTIC IS MADE TO IMPLY.
Cancer Needs Time
Nobody would accept this argument:
“We observed cigarette smokers for a few years and didn’t detect a meaningful increase in cancer, therefore cigarettes don’t cause cancer.”
We understand why that would be inadequate.
Cancer can involve cumulative exposure, multiple biological steps, and long latency.
Again, I am not comparing the danger of GLP-1 medications with cigarettes.
I am comparing the logic of the question.
If an outcome may require many years to become visible, we need sufficient time to observe it.
MORE PEOPLE ≠ MORE TIME.
SAMPLE SIZE CANNOT SUBSTITUTE FOR LATENCY.
The Danish Study Becomes Interesting When You Look at Time
A 2025 Danish nationwide study created matched groups of 19,730 GLP-1 receptor agonist users and 19,730 DPP-4 inhibitor users.
At five years, estimated cancer risks were approximately 11.2 per 100 GLP-1 users and 10.6 per 100 DPP-4 inhibitor users.
The difference was small.
But during years six through ten, researchers reported a cause-specific HR of 1.35, with a 95% CI of 1.05 to 1.73.
At ten years, modeled absolute cancer risks differed by approximately 4.11 diagnoses per 100 sustained users. [15]
Does this prove GLP-1 medications caused those cancers?
NO.
The researchers themselves discussed alternative explanations, including the possibility that GLP-1 users experienced fewer deaths before cancer could be diagnosed.
The combined cancer-or-death outcome did not show a significant increase.
Residual confounding could have affected the findings.
And only 344 sustained users remained at the ten-year point, making those later estimates less certain.
But I also don’t believe the appropriate response is:
“Nothing to see here.”
The responsible interpretation is:
THIS IS A SIGNAL WORTH INVESTIGATING.
An association that appears later doesn’t prove causation.
But neither should a later signal automatically be dismissed because shorter-duration randomized evidence is reassuring.
And We’re Still Focusing Mostly on the Drug
Even when we debate the Danish study, we’re still missing part of the story.
Everybody asks:
“Did the medication itself cause those cancers?”
I want to know something more.
During those years:
What happened to skeletal muscle?
What happened to protein intake?
What happened to fiber?
What happened to EPA and DHA?
What happened to vitamins and minerals?
What happened to physical activity?
What happened to body composition?
What happened to nutritional quality?
Those questions aren’t answered merely by determining whether the drug molecule itself is carcinogenic.
That’s why:
THE DRUG DOESN’T HAVE TO DIRECTLY CAUSE THE PROBLEM FOR WHAT HAPPENS DOWNSTREAM TO MATTER.
FDA Approved Does Not Mean Everything Is Known
Another principle we seem to forget repeatedly is:
FDA APPROVED ≠ EVERYTHING IS KNOWN.
Approval doesn’t mean every rare adverse effect has been discovered.
It doesn’t mean every delayed effect has been discovered.
And it doesn’t give us 20 years of follow-up when 20 years haven’t happened.
That’s why postmarketing surveillance exists.
History gives us examples.
Vioxx.
Rezulin.
Belviq.
Rimonabant in Europe.
These drugs had different mechanisms and different problems.
None predicts what will happen with GLP-1 medications.
The lesson is:
HISTORY DOESN’T TELL US WHAT WILL HAPPEN.
IT TELLS US WHY WE SHOULDN’T ASSUME WE ALREADY KNOW.
What If GLP-1 Drugs Never Directly Cause a Single Cancer?
Let’s take the strongest possible counterargument.
Suppose 20 years from now we have excellent evidence showing that GLP-1 receptor stimulation itself does not directly increase cancer risk.
Fine.
My questions still don’t disappear.
What happened to skeletal muscle over those 20 years?
What happened to protein intake?
What happened to fiber?
What happened to EPA and DHA?
What happened to micronutrient intake?
What happened to physical activity?
What happened to body composition?
What happened to gallbladder health?
What happened to reward and motivation?
What happened to the person who lost 75 pounds but also lost metabolically valuable tissue?
What happened to the person whose appetite was suppressed for a decade but who never received meaningful nutritional counseling?
What happened to the person who discontinued after two years and regained substantial weight?
And what happened to the 70-year-old who lost weight but also became weaker?
Those questions remain.
EVEN IF GLP-1 DRUGS NEVER DIRECTLY CAUSE A SINGLE CANCER, WHAT HAPPENS DOWNSTREAM STILL MATTERS.
We Need a Better Definition of Success
If someone chooses to use a GLP-1-based medication, success should involve much more than watching their weight decrease.
Weight matters.
But body composition matters.
Skeletal muscle matters.
Muscle strength matters.
Physical function matters.
Protein intake matters.
Resistance exercise matters.
Fiber matters.
Food quality matters.
Vitamin and mineral adequacy matters.
Essential fatty acids matter.
Mental health matters.
Gastrointestinal health matters.
Education matters.
And sustainable habits matter.
Because whether someone remains on the medication for 20 years or stops after 20 months, I want that person healthier—not merely lighter.
PROTECT THE REST OF THE PERSON’S HEALTH WHILE THE WEIGHT IS COMING OFF.
If you’re going to eat less:
WHAT YOU EAT MATTERS MORE.
If you’re going to lose weight:
WHAT YOU PRESERVE MATTERS TOO.
If you’re going to suppress appetite:
MONITOR NUTRITION.
If there’s a realistic possibility that you’ll stop:
LEARN HOW TO EAT WHILE YOU’RE STILL TAKING IT.
And if treatment may potentially continue for life:
WE NEED TO KEEP STUDYING WHAT HAPPENS OVER A LIFETIME.
Follow the Pathway
The GLP-1 conversation has become intensely focused on one endpoint:
WEIGHT LOSS.
But medicine doesn’t always work through one straight line.
A medication changes biology.
Biology changes appetite.
Appetite changes food intake.
Food intake changes nutrition.
Weight loss changes body composition.
Body composition affects physiology.
Nutrition affects the body’s ability to function.
Stopping the medication can change appetite again.
Weight may return.
Body composition may change again.
And some consequences may take years to become visible.
That’s why I keep returning to the same word:
DOWNSTREAM.
What happens downstream from appetite suppression?
What happens downstream from eating substantially less food?
What happens downstream from rapid or substantial weight loss?
What happens downstream from losing skeletal muscle?
What happens downstream from inadequate protein?
What happens downstream from chronically inadequate fiber?
What happens downstream if vitamins, minerals, EPA, or DHA become inadequate?
What happens downstream when someone loses substantial weight without learning how to eat differently?
What happens downstream when someone stops the medication?
What happens if they regain much of the weight?
What happens if they restart?
What happens to the 70-year-old whose greatest long-term threat may not simply be excess body fat, but losing the muscle and strength required to remain independent?
And what happens after five years?
Ten years?
Fifteen years?
Twenty years?
We don’t have all those answers.
And saying:
“WE DON’T KNOW.”
is not anti-science.
It’s exactly what science requires when the necessary evidence does not yet exist.
Something harmful may emerge over time.
It may not.
Long-term GLP-1 therapy may ultimately prove remarkably safe.
But we should not pretend that short- and intermediate-duration evidence can tell us everything about decades of exposure.
Especially when we’re discussing cancer.
148 TRIALS CAN GIVE US MORE PEOPLE.
THEY CANNOT GIVE US TIME THAT HASN’T HAPPENED YET.
A successful obesity treatment should ultimately produce more than a smaller number on a scale.
It should help produce a healthier person.
A stronger person.
A better-nourished person.
A person with skills.
A person with sustainable habits.
And especially for an older adult, a person who is still strong enough to walk, climb stairs, get out of a chair, recover from illness, and live independently.
So don’t stop at RR 0.99.
Don’t stop at 168,875 participants.
Don’t stop at pounds lost.
Don’t stop at the drug.
Ask the next question.
Then the next.
Then the next.
Because:
IT’S NOT JUST WHAT THE DRUG DOES.
IT’S WHAT THE DRUG SETS IN MOTION.
DON’T STOP AT THE DRUG. FOLLOW THE PATHWAY.
LOOK DOWNSTREAM.
Because losing weight is only part of the story.
How you lose the weight, what you learn while you’re losing it, what you preserve, what happens if the medication stops, and what happens to the rest of your health along the way matter too.
References
-
- Liang, W., Huang, Z., Long, Y., et al. (2026). Glucagon-like peptide-1 receptor agonists and risk of cancer: A systematic review and meta-analysis of randomized controlled trials. Scientific Reports. https://doi.org/10.1038/s41598-026-63543-7
- United States Food and Drug Administration. (2025). Wegovy (semaglutide) prescribing information.
- United States Food and Drug Administration. (2025). Zepbound (tirzepatide) prescribing information.
- Rodriguez, P. J., Zhang, V., Gratzl, S., Do, D., Goodwin Cartwright, B., Baker, C., Gluckman, T. J., Stucky, N., & Emanuel, E. J. (2025). Discontinuation and reinitiation of dual-labeled glucagon-like peptide-1 receptor agonists among U.S. adults with overweight or obesity. JAMA Network Open, 8(1), e2457349. https://doi.org/10.1001/jamanetworkopen.2024.57349
- Wilding, J. P. H., Batterham, R. L., Davies, M., et al. (2022). Weight regain and cardiometabolic effects after withdrawal of semaglutide: The STEP 1 trial extension. Diabetes, Obesity and Metabolism, 24(8), 1553–1564. https://doi.org/10.1111/dom.14725
- Aronne, L. J., Sattar, N., Horn, D. B., et al. (2024). Continued treatment with tirzepatide for maintenance of weight reduction in adults with obesity: The SURMOUNT-4 randomized clinical trial. JAMA, 331(1), 38–48. https://doi.org/10.1001/jama.2023.24945
- Müller, M. J., Enderle, J., Pourhassan, M., Braun, W., Eggeling, B., Lagerpusch, M., Glüer, C.-C., Kehayias, J. J., Kiosz, D., & Bosy-Westphal, A. (2015). Metabolic adaptation to caloric restriction and subsequent refeeding: The Minnesota Starvation Experiment revisited. The American Journal of Clinical Nutrition, 102(4), 807–819. https://doi.org/10.3945/ajcn.115.109173
- Evans, W. J., & Cummings, S. (2024). Weight loss-induced muscle mass loss. JAMA, 332(16), 1394. https://doi.org/10.1001/jama.2024.17212
- Filis, P., Papagiannopoulos, C. K., Markozannes, G., Chalitsios, C. V., Zerdes, I., Valachis, A., Papandreou, C., Christakoudi, S., & Tsilidis, K. K. (2025). Associations of sarcopenia, sarcopenia components and sarcopenic obesity with cancer incidence: A prospective cohort study of 414,094 participants in UK Biobank. International Journal of Cancer, 157(7), 1316–1332. https://doi.org/10.1002/ijc.35480
- Hu, T., Shen, Y., Cao, W., Xu, Y., Wang, Y., Ma, X., & Bao, Y. (2023). Two-year changes in body composition and future cardiovascular events: A longitudinal community-based study. Nutrition & Metabolism, 20, 4. https://doi.org/10.1186/s12986-023-00727-2
- Aune, D., Chan, D. S. M., Lau, R., Vieira, R., Greenwood, D. C., Kampman, E., & Norat, T. (2011). Dietary fibre, whole grains, and risk of colorectal cancer: Systematic review and dose-response meta-analysis of prospective studies. BMJ, 343, d6617. https://doi.org/10.1136/bmj.d6617
- Müller, T. D., Finan, B., Bloom, S. R., et al. (2019). Glucagon-like peptide 1. Molecular Metabolism, 30, 72–130. https://doi.org/10.1016/j.molmet.2019.09.010
- European Medicines Agency. (2008). Questions and answers on the recommendation to suspend the marketing authorisation of Acomplia (rimonabant).
- He, L., Wang, J., Ping, F., Yang, N., Huang, J., Li, Y., & Xu, L. (2022). Association of glucagon-like peptide-1 receptor agonist use with risk of gallbladder and biliary diseases: A systematic review and meta-analysis of randomized clinical trials. JAMA Internal Medicine, 182(5), 513–519. https://doi.org/10.1001/jamainternmed.2022.0338
- Gamborg, M., Grand, M. K., Grell, K., Rosthøj, S., Pedersen-Bjergaard, U., Torp-Pedersen, C., & Mørch, L. S. (2025). Long-term cancer risk in users of glucagon-like peptide-1 agonists in Denmark: A nationwide emulated trial. The Lancet Regional Health – Europe, 55, 101346. https://doi.org/10.1016/j.lanepe.2025.101346
__________
Robert Ferguson is a California- and Florida-based single father of two daughters, clinical nutritionist, Omega Balancing Coach™, researcher, best-selling author, speaker, podcast and television host, health advisor, NAACP Image Award Nominee, creator of the Diet Free Life methodology, and Chief Nutrition Officer for iCoura Health. He also serves on the Presidential Task Force on Obesity for the National Medical Association and the Health and Product Advisory Board for Zinzino, Inc.
🗓️ Schedule a FREE consultation with Robert Ferguson about becoming a client: SCHEDULE FREE CONSULTATION
👉🏽 To order the BalanceOil+ with the BalanceTEST, > CLICK HERE
👉🏽 Watch a free online presentation on the BalanceOil+ and the BalanceTEST: WATCH NOW.
0 Comments