When Weight-Loss Drugs Don’t Work

[miss-sanne]

Many of my patients show no reaction at all to Wegovy, which is the medication we primarily prescribe here in Denmark.

My own experience matched that: Wegovy did nothing for me.

 

[dickybob]

FortySeven said:

To be more precise, 1 in 10 people who made it into a clinical trial got labeled "non-responders" -- and that label didn't even require zero response, only a smaller one. That's a different thing from the punchy subhead's assertion: 'But one in 10 people are what scientists call “non-responders.”' The implication there is that 1 in 10 people across the entire global population fall into what they'd term "non-responders".

Sloppy science journalism like this keeps bogus claims alive, though I suppose it moves copies.

Consider this: folks enrolled in a clinical trial aren't automatically "average" people. They presumably had to clear various criteria. Those criteria mean the trial group likely contained a larger share than the general public of individuals who were a) obese and b) had attempted weight loss through other means/possibly other medications and not succeeded. (I'm not using 'failed' as a judgment here, to be clear.)

Still, it's a worthwhile article, and it points to other worthwhile articles (such as the one about retatrutide clinical study participants quitting because they were shedding "too much" weight). It's just that science reporting which tosses out easy-breezy claims with seemingly little critical thought really gets under my skin.

The piece starts out by describing someone who spent 15 months on Zepbound and, at best, dropped a pound or two. Forty-Seven skips past that entirely in order to label what the article contains "lazy science." Yes, the New York Times leans liberal, but that is a completely different matter from "lazy science." The New York Times verifies everything. I subscribe to the NY Times and consider its science coverage to be reliable. Personally, I have been a low-responder. When I moved up to roughly 5 mg sub each week, I developed diarhea. To stop that effect, I had to go back down to a smaller dose, 2mg divided into two subs per week. Since then I have been slowly increasing again, by 0.5 mg every four weeks. For roughly five months I dropped no weight whatsoever. Then I became quite ill and, apparently because of the illness combined with tirz, I abruptly shed 4-5 pounds. There are probably two sides to this. At one extreme are those who respond extremely well — for instance, I know a man who maintains his weight on 2.5 mg of tirz every 3-4 weeks — and at the other extreme are those who do not respond at all. Several earlier replies came from people who had little to no reponse. Clearly, the writer is/was too busy writing to bother reading those other responses.

 

[Foggy-Hollow]

sfkid said:

https://share.google/jTyQpIy8C6fxuSXq5

According to this article, it appears that 1 out of 10 people have no response whatsoever to GLP1s.

Is there anyone here who knows a person that experienced absolutely no effect?

At some point — possibly near the end of last year — a person wrote that tirz didn’t produce any weight loss for them until they reached 15mg.

I wanted to know how they prepared their vial, but they never got back to me. A misplaced decimal point is an easy mistake, though it’s also possible they really were a low-responder.

The “normal curve” shows up in so many situations. Given how much people differ, it makes sense that a handful will barely respond or not at all, while a handful will respond extremely strongly. Another possibility is that side effects hit them hard. What’s difficult is that nobody can predict their own outcome in advance, so starting low and increasing slowly is the safest approach.

 

[FortySeven]

dickybob said:


The article itself opens with an example of a person that has been on Zepbound for 15 months and and may have lost a pound or two at most. Forty-Seven jumps right over that in a quest to call the contents of the article "lazy science." New York Times has a liberal bias to be sure, but that is way different than "lazy science." Everything in the New York Times is cross-checked. I take NY Times and find its science articles to be solid. I for one have been a low-responder. I got diarhea when I jumped up to approximately 5 mg sub per week. I had to drop down to a lower dose to prevent this effect, 2mg split into two subs weekly. I have gradually been working up again at the rate of 0.5 mg every four weeks. Didn't lose any weight at all for about five months. I got really sick and evidently between my sickness and tirz, I suddenly lost 4-5 pounds. There are most likely two ends to the stick. On one end, are the people who are extremely responsive, .i.e. I know of someone who takes 2.5 mg of tirz every 3-4 weeks to maintain his weight and the other end of the stick where some people are non-responsive. Some of the previous replies were from folks who had little to no reponse. Evidently, the writer is/was to busy writing to read those other responses.

Click to expand...

What I called out was “lazy science reporting,” not “lazy science.” Those two phrases are verrrrrrrry far apart. My only point: 1 in 10 within a scientific study population isn’t equivalent to 1 in 10 across the general population. That’s all. Yes, it’s a fine semantic distinction, but it’s exactly the kind of thing that keeps misinformation alive. It seems you didn’t read my reply carefully enough to grasp my argument, which never challenged the study’s validity or its results—only how those findings were converted into the article’s subheading.

1:10 clinical trial participants does not equal 1:10 people. It’s not my opinion—it’s statistics.

I subscribe to the NYT. As a rule, their reporting is outstanding.

I skipped nothing, and I’m not on any kind of mission. What I am is a critical thinker—and I urge others to think critically too.

 

[Flexon]

Occasionally, the pharmaceutical industry itself is the one committing what can be called "lazy science reporting":

FDA Warns GLP-1 Maker Over Failure to Properly Report Deaths, Strokes:

https://www.medpagetoday.com/public...Nordisk for,PADE investigation, the FDA noted

 

[Grogu]

FortySeven said:


I didn’t say “lazy science”, I said, “lazy science reporting.” Two verrrrrrrry different things. All I was pointing out is that one in 10 people in a scientific study population is not the same thing as one in 10 people in the general population. That’s it. It’s a subtle issue of semantics, but it’s the sort of misinformation that gets perpetuated. I don’t think you even read my response closely enough to understand what I was saying, which is not questioning the validity of the study—nor its results—but rather questioning the translation of the study’s findings into the subheading of the article.

1:10 clinical trial participants does not equal 1:10 people. It’s not my opinion—it’s statistics.

I am a NYT subscriber. Their reporting is generally exceptional.

I didn’t jump over anything, and I’m certainly not on any sort of quest. I am, however, a critical thinker, and I encourage others to be as well.

Click to expand...

Your remarks got me wondering about how far clinical trial findings can really be stretched to the broader public, especially when the rules for who gets enrolled are tight. I also started questioning whether data from nearly every GLP-1 trial out there — or perhaps all of them — can reasonably be applied to everyone.

So I went back and read the enrollment requirements for SURMOUNT-1. They are quite narrow — tighter, in fact, than I had recalled. Still, the bulk of the design seems aimed at recruiting people who are broadly healthy, satisfy the treatment conditions, and lack other medical issues that might skew outcomes or put them at risk. Take the exclusions for a history of MTC or MEN, or any lifetime suicide attempt — those are likely in place because such cases are uncommon among the general public. Both of those clearly serve to protect participants from harm. That said, I assume the investigators devote a great deal of effort to these choices precisely so the findings will hold up beyond the study group; otherwise the FDA would not sign off. They likely must defend each criterion on scientific grounds.

TL/DR, the mere existence of enrollment rules — and the fact that not everyone can join a trial — does not automatically prevent the results from being generalized to the wider population.

 

[tubby]

FortySeven said:


I didn’t say “lazy science”, I said, “lazy science reporting.” Two verrrrrrrry different things. All I was pointing out is that one in 10 people in a scientific study population is not the same thing as one in 10 people in the general population. That’s it. It’s a subtle issue of semantics, but it’s the sort of misinformation that gets perpetuated. I don’t think you even read my response closely enough to understand what I was saying, which is not questioning the validity of the study—nor its results—but rather questioning the translation of the study’s findings into the subheading of the article.

1:10 clinical trial participants does not equal 1:10 people. It’s not my opinion—it’s statistics.

I am a NYT subscriber. Their reporting is generally exceptional.

I didn’t jump over anything, and I’m certainly not on any sort of quest. I am, however, a critical thinker, and I encourage others to be as well.

Click to expand...

I can't agree with the idea that the NYT excels at much beyond spinning narrow narratives and knocking down simplified versions of complicated debates, all while acting incredibly self-satisfied. That said, I'm with you on every other point, and it's refreshing to find another person here who notices the deceptive tricks pharma companies routinely pull by carefully engineering their trial populations to tilt approval decisions and outcomes in their favor.

Honestly, at this stage it wouldn't shock me to learn of a trial built around subjects who had only just finished prednisone therapy (a treatment recognized for promoting weight gain), with the placebo arm kept on prednisone (maybe through tailored prednisone shots) while the treatment arm got switched to a GLP (serving both to shed pounds and to calm inflammation), all as a way to inflate the weight-loss numbers. And should any drug company representatives be reading these boards, pass this idea along to your trial design people and they'll pay you handsomely for it.

 

[blackkaiser]

sfkid said:

https://share.google/jTyQpIy8C6fxuSXq5

According to this article, it appears that 1 out of 10 people have no response whatsoever to GLP1s.

Is there anyone here who knows a person that experienced absolutely no effect?

Hmm.... So that means 90% of people do respond.... That's actually quite a high number....

 

[RetCurious]

DjJoshua said:

sfkid said:


https://www.nytimes.com/2026/03/12/well/weight-loss-drugs-response-wegovy-zepbound.html

Based on this article it seems 1 of 10 people don't respond the GLP1s at all.

Does anyone know someone that has had zero response?

Click to expand...

Is it only happening on my end, or does that link require signing up for an account before the article can be viewed? Could someone share another link?

I know I'm showing up after everyone else, but I found a saved version of that NYT piece.


https://archive.is/JMxci

 

[RetCurious]

Rolltide61 said:

sfkid said:

https://share.google/jTyQpIy8C6fxuSXq5

According to this article, it appears that 1 out of 10 people have no response whatsoever to GLP1s.

Is there anyone here who knows a person that experienced absolutely no effect?

Reta gave me no results at all. I climbed all the way to 10mg. Tirz did nothing either. Even combining Cagri with Reta, I felt absolutely no reduction in hunger.

But here's the strange part. I started Melanotan ll. No appetite whatsoever.

Go figure.

Genuinely fascinating — thanks.

Here's what Wikipedia says about melanotan II:

"Other effects of melanotan II... loss of appetite (the last via activation of MC4)"

And on the melanocortin 4 receptor:

"The melanocortin 4 receptor (MC4R) is a G protein-coupled receptor involved in regulating energy homeostasis, appetite, and sexual function. It plays a key role in metabolic processes and is predisposes to certain forms of obesity in humans."

Melanocortin 4 receptor - Wikipedia





Image unavailable


en.wikipedia.org

 

[FortySeven]

Grogu said:


I’ve been thinking about your point about the generalizability of clinical study results when the study participant selection criterion is constrained and if the results from most (if not all) the clinical trials on glp-1 medications can be imputed upon the general population.

I looked at the SURMOUNT-1 participant criterion and it’s fairly restrictive. More restrictive than I originally remembered. But most of the protocol appears to be centered on finding generally healthy folks who meet the treatment criteria and don’t have confounding medical conditions which could affect the results or cause harm to the participant. For example, excluding people with a medical history of MTC or MEN or lifetime history of suicide attempt is probably because these are low in the general population. These two are clearly about no harm to the participant. But I’m sure that study researchers spend significant amount of time on these decisions so that the results are generalizable, otherwise the FDA wouldn’t accept the results. They probably have to justify the criterion with science.

TL/DR, just because there is a participant selection criterion and that everyone isn’t eligible to participate in the study doesn’t mean that the results can’t be extrapolated to the population.

Click to expand...

I appreciate the considered reply. Looking at the participant data, was there any indication that any of the test subjects were younger than 48yo? (It’s entirely possible I’m misreading the table, but that appears to be the youngest age.) And assuming I’ve got it right, 100% of those enrolled were “overweight or obese”.

Here in the untamed Wild West of rat experiments, there are people (myself included) whose BMIs have never gone beyond a high “normal”.

I truly wasn’t trying to take a shot at the experiments, since I don’t see myself as a scientist (and I hold no degree that would imply otherwise.) So I welcome being set straight by those who are better at parsing study summaries (you likely fall into that group).

My only point is that the New York Times took a statistic and boiled it down into a false claim by applying the broad brush “people”.

It could absolutely be accurate that 10% of the population Eli Lilly hopes will be future consumers (overweight/obese people) are what the science calls non-responders. I can’t say. (Like I mentioned, were they truly all over 48 years old?) I suspect the FDA would in fact prefer the study to center on the intended audience for the treatment under investigation, which is what happened.

I simply don’t believe the study demonstrated that 10% of “people” are non-responders. That wasn’t what the study set out to do. This isn’t me attacking the study. Honestly, for all we know, if you took the entire global population that gets called “people” (which would have to include babies and children and gym bros looking to cut for a show and folks like me who wanna lose 15 pounds), 30% could be non-responders. And no, I’m not suggesting studies should be run on babies. My point is only that “people” is an extremely, extremely broad word. It’s the whole circle in the Venn diagram of things, right?

I think there’s a troubling pattern of trying to stretch meaning out of data sets from studies that had one purpose (here, “assess the efficacy and safety of retatrutide in obese patients with or without diabetes”), and then drawing secondary conclusions.

If a study were designed to establish the percentage of the population of people as a whole who are non-responders to retatrutide, the study population would need to be widened, wouldn’t it?

Again, I’m not a scientist, but… my issue isn’t really with the science here. It’s with how the English language is being used. Because if the New York Times is willing to be that sloppy with the English language in this case, then what other headlines don’t actually line up with reality? And I think we all know that’s a slippery slope in media and I hate to see it continue. I also feel like I’ve been on my soapbox long enough, and I never intended this to turn into such a heated point of contention.

 

[Grogu]

FortySeven said:


Thanks for your thoughtful response. In the participant data, did you find if any of the test subjects were under 48yo? (I could honestly be reading the table wrong, but it seems like that’s the lowest age.) And if I’m reading it correctly, 100% of the participants were “overweight or obese”.

Out here in the wild Wild West of rat experiments, we have folks (including me) whose BMIs have never been above a high “normal”.

Click to expand...

Your post really made me stop and think.

Looking at the phase 2 retatrutide clinical trial, the mean age shown in the last column is 48.2, and the figure beside it represents the +/- standard deviation. That 12.7 is expressed in years and reflects 1 standard deviation away from the mean (average). Roughly 95% of a population falls within 2 standard deviations. That means about 95% of the participants ranged in age from 22.8 to 73.6. So there were certainly participants younger than 48. The same logic applies to BMI. The average BMI came out to 37.3, and with a standard deviation of 5.7, there were certainly individuals whose BMI sat in the high 20s. Still, keep in mind these medications were developed for people who are clinically obese, or overweight with comorbidities.

If a study were conducted to determine the percentage of the population of people as a whole who are non-responders to retatrutide, the study population would have to be expanded, no?

Click to expand...

It might surprise you how few participants are needed to reach conclusions about a population, provided the statistical methods are sound. The number could be as low as 30. That said, I doubt anyone would deliberately build a sample to test a population the medication was never intended to treat. There's no logic in that whatsoever.

FortySeven said:


All I am saying is that the New York Times took a statistic and reduced it to a false claim by using the broad stroke “people”.

Click to expand...

According to the NYT article, "....But one in 10 people are what scientists call “non-responders."

What that suggests is that 1 in 10 people (who take the medication) will be non-responders. After all, you can't be a non-responder unless you're taking the medication. And the only individuals who are meant to be on the medication are those with the clinical indications. That's exactly why the study was set up to choose a sample from that population.

They never claimed 1 in 10 people worldwide, since the general public as a whole shouldn't be using the medication.

 

[deluge]

miss-sanne said:

Many of my patients show no reaction at all to Wegovy, which is the medication we primarily prescribe here in Denmark.

My own experience matched that: Wegovy did nothing for me.

Same here. I was paying for Sema out of pocket with my own money, and my weight actually went up. I stopped that pretty fast.

Foggy-Hollow said:

sfkid said:

https://share.google/jTyQpIy8C6fxuSXq5

According to this article, it appears that 1 out of 10 people have no response whatsoever to GLP1s.

Is there anyone here who knows a person that experienced absolutely no effect?

At some point — possibly near the end of last year — a person wrote that tirz didn’t produce any weight loss for them until they reached 15mg.

I wanted to know how they prepared their vial, but they never got back to me. A misplaced decimal point is an easy mistake, though it’s also possible they really were a low-responder.

The “normal curve” shows up in so many situations. Given how much people differ, it makes sense that a handful will barely respond or not at all, while a handful will respond extremely strongly. Another possibility is that side effects hit them hard. What’s difficult is that nobody can predict their own outcome in advance, so starting low and increasing slowly is the safest approach.

That’s a smart thing to ask! It’s easy to get it wrong. There are other factors as well. For instance, today I found out that the cheap pen I bought from Amazon was repeatedly delivering 3.5 to 4 units even though the dial was turned to 10. If that had kept happening, I might have ended up disappointed with how that specific peptide worked for me (GHK-Cu, in this instance).

 

[FortySeven]

Grogu said:


Your post gave me a lot to think about.

In the phase 2 retatrutide clinical trial the average age was 48.2 (last column) but the number next to it is the +/- standard deviation. The 12.7 is in years and this is one standard deviation from the mean (average). Two standard deviations capture about 95% of a given population. So, 95% of the participants were between the age of 22.8 and 73.6. So, there were definitely some people younger than 48. Same for BMI. Average BMI was 37.3, but with a 5.7 standard deviation, there were definitely folks with a BMI in the high 20s. But remember that these medication were designed for the clinically obese and those overweight with commorbidities.

View attachment 17561

You'd be surprised how small a sample needs to be to draw conclusions about a population, if proper statistical methods are followed. It could be as low as 30. But I don't think anyone would ever design a sample to test a population that the medication was never designed to treat. That makes no sense at all.

The NYT article says, "....But one in 10 people are what scientists call “non-responders."

The implication is that it's 1 in 10 people (who take the medication) will be non-responders. Because you can't be a non-responder if you don't take the medication. And the only people who are supposed to be taking the medication are those with the clinical indications. And hence why the study was designed to select a sample of that population.

They never said 1 in 10 people in the world, because the entire general public shouldn't be taking the medication.

Click to expand...

Learning something new is something I really enjoy—and your post taught me quite a bit! Thanks!

 

[Grogu]

FortySeven said:


Okay I love learning new stuff, and I learned a lot from your post! Thank you!

Click to expand...

Your point regarding the term “people” is fair—perhaps “users” or “patients” would have fit better.

What I was really trying to get at, I suppose, was the issue of generalizability, the statistics, and whether the study holds up.

Honestly, I’ve gone through the tirzepatide trials quite thoroughly (retatrutide less so), and they are remarkably well constructed and carried out. Over the years I’ve pored over plenty of medical research for my own various conditions , and a fair amount of it is messy, since the majority of doctors doing research haven’t had training in statistical methods. The lead investigator for both the tirzepatide and retatrutide trials (Dr. Jastreboff) holds an MD and a PhD, and she has EL behind her. These studies are extremely solid.

 

[FortySeven]

I’m grateful for the way you broke down the data and how it was presented—and it’s great that you went so thoroughly into how the study was built and its methods! I’m on this forum to pick up knowledge from people such as yourself!

 

[dickybob]

FortySeven said:


I didn’t say “lazy science”, I said, “lazy science reporting.” Two verrrrrrrry different things. All I was pointing out is that one in 10 people in a scientific study population is not the same thing as one in 10 people in the general population. That’s it. It’s a subtle issue of semantics, but it’s the sort of misinformation that gets perpetuated. I don’t think you even read my response closely enough to understand what I was saying, which is not questioning the validity of the study—nor its results—but rather questioning the translation of the study’s findings into the subheading of the article.

1:10 clinical trial participants does not equal 1:10 people. It’s not my opinion—it’s statistics.

I am a NYT subscriber. Their reporting is generally exceptional.

I didn’t jump over anything, and I’m certainly not on any sort of quest. I am, however, a critical thinker, and I encourage others to be as well.

Click to expand...

FortySeven said:


I didn’t say “lazy science”, I said, “lazy science reporting.” Two verrrrrrrry different things. All I was pointing out is that one in 10 people in a scientific study population is not the same thing as one in 10 people in the general population. That’s it. It’s a subtle issue of semantics, but it’s the sort of misinformation that gets perpetuated. I don’t think you even read my response closely enough to understand what I was saying, which is not questioning the validity of the study—nor its results—but rather questioning the translation of the study’s findings into the subheading of the article.

1:10 clinical trial participants does not equal 1:10 people. It’s not my opinion—it’s statistics.

I am a NYT subscriber. Their reporting is generally exceptional.

I didn’t jump over anything, and I’m certainly not on any sort of quest. I am, however, a critical thinker, and I encourage others to be as well.

Click to expand...

I owe you an apology. Last week I misread what you wrote, and you were right about that. I accept the correction.

 

[FortySeven]

dickybob said:


I wanted to apologize. You are correct in that I did misread your response last week. I stand corrected.

Click to expand...

no worries. seems I got a few things incorrect, which @Grogu pointed out.

Group hug?

 

[BBA1969]

I’m in the middle of switching from tirzepatide (Eli Lilly’s Mounjaro) over to grey retatrutide.

My previous approach used split dosing—2 × 3.75 mg each week—and I’ve now replaced the Monday evening injection with 2.5 mg of retatrutide. Come Wednesday, appetite suppression falls away quite sharply for me, and the Friday morning Mounjaro shot only just arrives in time to carry me through the weekend.

The plan: raise the retatrutide gradually and find out whether that gives me stronger and steadier appetite control. After things feel reasonably consistent across the whole week, I’ll complete the changeover to retatrutide.

So far, the good part is that neither compound has produced any negative side effects for me.

Has anyone else gone through a similar switch, and how did it go for you?

 

[Calm Logic]

For me, what caps the dose on any GLP is the side effects, and reta gives me just as many. So even if reta looks stronger for weight loss on paper, I'd say tirz and reta share more similarities than differences.

When people move over to reta, the usual error is pushing reta too hard, or cutting the tirz back before they should.

Given that you've had no side effects so far, your long-term path looks solid.