Semaglutide & weight loss

Executive Summary

Please login to view this content


An obesity drug remains one of the biggest opportunities for pharmaceutical companies, but it has proved to be one of the most elusive to date. Medications approved for weight loss by the Food and Drug Administration (FDA), the European Medicines Agency (EMA), and other regulatory bodies have had a troubled history. Several approved drugs have been withdrawn following serious adverse events. Various amphetamines were withdrawn for addiction, palpitations, and other side-effects. Fenfluramine and dexfenfluramine were withdrawn because of heart valve damage. An increase in major adverse cardiovascular events prompted the withdrawal of sibutramine in Europe and the US. Rimonabant reduced body weight and ameliorated some cardiovascular risk factors, but it was withdrawn because of adverse psychiatric issues. Drugs that have retained approval, such as Orlistat, have severe and debilitating gastrointestinal side-effects, which have limited their use.

I have written about two diet drugs in recent years. In September 2018, I covered a report from the European Society of Cardiology conference (Ref 1). A diet drug was presented at a cardiology conference because it was being claimed to be the first weight-loss drug deemed safe for heart health with long term use. The weight loss difference at one year was 4.2kg in the drug group and 1.4kg in the placebo group. The drug was called Lorcaserin and it was being marketed under the trade name Belviq. In February 2020 it was announced that lorcaserin was being withdrawn from the US market due to increased occurrence of cancer (Ref 2).

In 2015, I reported on a drug called liraglutide (Ref 3). It was approved by the EMA for type 2 diabetes (T2D) in July 2009 and it was approved by the FDA for T2D in January 2010. It was approved by the FDA for obesity on 29th December 2014, where it is marketed as “Saxenda.” It was approved by the EMA on 23rd January 2015 for obesity. The manufacturer is Novo Nordisk.

This week’s note is about semaglutide – also made by Novo Nordisk. Liraglutide and semaglutide are both drugs for T2D, which are now being used/researched for obesity (semaglutide is not yet approved). The drugs are injected into the stomach – liraglutide daily and semaglutide weekly. Both drugs belong to a class of drugs called “incretin mimetics” because these drugs mimic the effects of incretins. Incretins are hormones produced and released into the blood by the intestine in response to food. The incretin of key relevance with liraglutide is GLP-1 (Glucagon-Like-Peptide-1). GLP-1 increases the release of insulin from the pancreas, slows absorption of glucose from the gut and reduces the amount of glucose being produced by the liver. All three actions are intended to reduce levels of glucose in the blood stream – hence their use for T2D.

The liraglutide dose for T2D is typically 1.2mg per day. The dose for obesity is 3mg per day. The liraglutide trial was 56 weeks long (with a 12 week follow-up period). Both the drug and placebo group were given the same diet and exercise regime to follow. The drug group lost an average of 8.40kg; the placebo group lost an average of 2.8kg. Those were significant differences.

I cautioned in that note that liraglutide came with a major warning for thyroid cancer.

Let’s now look at semaglutide…

The study

The ‘semaglutide for obesity’ study was published in the New England Journal of Medicine on February 10th, 2021 (Ref 4). It was a randomised controlled trial (RCT), which was also described as “double-blind” meaning that neither the participants nor the researchers knew who was in which group. Adults with obesity but not diabetes were recruited for the trial. The trial was large – 1,961 people – and 68 weeks long. The recruited people came from 129 different sites in 16 countries in Asia, Europe, north America, and south America. Novo Nordisk sponsored the trial, designed the trial, and oversaw its conduct. There were enough participants such that there were no significant differences between groups, when randomised.

The average (mean) BMI was 38 at the start of the trial. The participants were randomly assigned in a 2:1 ratio to 68 weeks of treatment (once weekly injected semaglutide at a dose of 2.4mg) or placebo injection – plus lifestyle intervention. The main outcomes of interest were the change in body weight and weight reduction of at least 5%.

The intervention

Semaglutide was approved in 2017, at doses up to 1mg administered by injection once weekly, for the treatment of T2D in adults and for reducing the risk of cardiovascular events in people with T2D and cardiovascular disease (Ref 5).

The trial to examine the efficacy of the drug for obesity started with the injection of 0.25mg of semaglutide weekly for the first 4 weeks, with the dose increased every 4 weeks to reach the maintenance dose of 2.4mg weekly by week 16 (lower maintenance doses were permitted if participants had unacceptable side effects with the 2.4mg dose). No details were given about the placebo. I found the original trial protocol and this also did not reveal the placebo. It was described as “semaglutide matching placebo” (Ref 6). An earlier paper on the trial also gave no details about the placebo (Ref 7).

In addition to the drug or placebo injection, participants received individual counselling sessions every 4 weeks to help them adhere to a reduced-calorie diet (500 calories deficit per day relative to their own starting calorie intake) and increased physical activity (150 minutes a week). Both diet and activity were recorded in a diary or smartphone app.

The results

The average (mean) reduction in bodyweight from baseline to week 68 was 14.9% in the semaglutide group as compared with 2.4% in the placebo group. In kg terms, this was an average (mean) weight loss of 15.3kg (2.4 stone) in the drug group as compared with an average (mean) weight loss of 2.6kg (0.4 stone) in the placebo group.

More participants in the semaglutide group than in the placebo group achieved weight reductions of 5% or more (86% of participants vs 32%), 10% or more (69% vs 12%) and 15% or more (51% vs 5%) at week 68.

Those were impressive weight loss results. They were the best I have seen in any drug vs placebo weight loss trial. They were almost unbelievable – with the weight loss being virtually double that reported with the company’s similar drug, liraglutide. Having sponsored, designed, and run the trial, Novo Nordisk is currently seeking approval for the drug for obesity (Ref 8).


The authors offered the following as an explanation for the mechanism by which semaglutide aided weight loss: “Weight loss with semaglutide stems from a reduction in energy intake owing to decreased appetite, which is thought to result from direct and indirect effects on the brain.” Having researched obesity for years, this didn’t make sense at first sight. I have found obesity to be related to over consumption of specific foods (invariably processed foods, with an unnatural carbohydrate/fat combination). I have yet to meet the person who became obese while overeating salmon and green beans. I am skeptical about claims that drugs reduce appetite and thus overeating, as I have not seen people overeating because they were generally hungry (in which case salmon and green beans would suffice). I have seen people overeating because they were specifically ‘hungry’ – specifically hungry for bread, cereals, biscuits, cakes, confectionery, crisps etc. When someone has cravings for specific foods – and shows sign of addiction to such foods – hunger is rarely the issues and conversely appetite control (reducing hunger) rarely helps, in my experience.

I turned to the side-effect’s information in the paper and a more plausible explanation for weight loss could be found. The table made me increasingly suspicious about the placebo. Adverse events were reported in 90% of drug participants and 86% of placebo participants. Even though we know that people can report side effects when taking a placebo, 86% is extraordinarily high.

The adverse events reported in the drug group were primarily gastrointestinal: 44% reported nausea; 32% diarrhea and 25% reported vomiting. Overall 74% reported gastrointestinal disorders. These numbers were all significantly higher than those reported in the placebo group. Perhaps semaglutide reduced appetite because eating became so traumatic that it wasn’t worth the effort. Diarrhea and vomiting would reduce absorption of food and this could have impacted weight loss. (It would also have impaired nutrient absorption). Twice as many people in the drug group withdrew from the trial (compared to the placebo group). Nearly 12 times as many people withdrew from the drug group for gastrointestinal reasons as did in the placebo group.


An editorial accompanying the article noted some limitations of the trial (Ref 9). “First, the demographics in this trial are not reflective of the general U.S. population. Most of the participants were White, with only 6% Black or African American and 12% Latinx, whereas nearly 40% of the U.S. population is non-White. Similarly, males were underrepresented (26%). In addition, more than 40% of the cohort had prediabetes. These factors, taken together, raise additional questions about the efficacy of subcutaneous semaglutide in persons with obesity and normal glucose tolerance. Second, the present trial, although 68 weeks in length, still does not address long-term efficacy, since obesity is a chronic problem requiring constant attention. Third, in the real world, it seems unlikely that once-weekly subcutaneous administration would be a palatable or cost-effective solution in the long run.”


The paper summarised the common side effects reported during the trial. The patient leaflet gives more comprehensive information about other potential serious outcomes. The patient leaflet for the 2.4mg dose is not yet available. The patient leaflet for a 0.5mg dose cautions that serious outcomes include: pancreatitis; hypoglycemia; retinopathy (diabetic eye disease), as well as all the gastrointestinal effects reported in the trial (Ref 10). The retinopathy and eye problems are reported to be common serious side effects (affecting up to 1 in 10 people). Acute pancreatitis is reported as may affect 1 in 100. Nausea and diarrhea are reported as very common and vomiting and hypoglycemia as common.

The patient leaflet doesn’t mention the most serious issue. This is prominently featured in the FDA approval document (Ref 11). The header of page one of the approval document has this prominent warning about thyroid cancer.

As shared in the introduction, Lorcaserin was presented at a cardiology conference in August 2018 and was already withdrawn from the US market by February 2020 due to increased occurrence of cancer. Perhaps semaglutide will achieve approval and then it might also be withdrawn soon afterwards. Although liraglutide has the same thyroid cancer concerns and it is still on the market.

We don’t know the medium, let alone long-term effects of semaglutide. The endocrine system has shown itself to be relatively easy to damage, with approximately 1 in 10 Americans now having diabetes – the vast majority of these cases being type 2 (Ref 12). What damage will incretin mimetics do beyond a year? Can we encourage production of insulin and discourage production of glucose indefinitely without consequence? What happens if semaglutide is stopped? Or does it need to be injected into the stomach weekly forever? Weight is invariably regained following calorie deficit dieting. Will weight be regained with or without continued injections?

The desire to lose weight is so strong that it almost has no limits. There will undoubtedly be many people who see these impressive weight loss results and want to try the drug if it is approved. Obesity itself has its own complications – how much harm is acceptable to balance benefit from losing weight? This drug will probably be approved and one day it might be withdrawn. There are so many questions to which we currently don’t know the answers. I can fully understand why people will want to try this but I’m concerned about what might happen if they do.



Ref 1:
Ref 2:
Ref 3:
Ref 4: Wilding et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity. NEMJ. February 2021.
Ref 5:
Ref 6:
Ref 7: Kushner et al. Semaglutide 2.4 mg for the Treatment of Obesity: Key Elements of the STEP Trials 1 to 5. Obesity. June 2020.
Ref 8:
Ref 9: Ingelfinger & Rosen. STEP 1 for Effective Weight Control — Another First Step? NEMJ. February 2021.
Ref 10:
Ref 11:
Ref 12:

6 thoughts on “Semaglutide & weight loss

  • With my doctor hat on, I will say that of all the drugs that we can use for type 2 diabetes, my preferred ones are metformin and GLP1 analogues, because these can cause weight loss and reduction in Hba1c. I’ve always assumed that it is because they cause anorexia as a side effect. I appreciate that GLP1 causes a prandial insulin rise, but it is not sustained, hence the easier weight control. I would far rather, of course, treat and REVERSE the type 2 diabetes with the appropriate , ie low carbohydrate, diet, but because of the twin myths of ‘saturated fat is bad for you’ and ‘carbohydrates are essential’, backed up by illogical dietary guidelines, this is a hard battleship to turn around, not helped by the massive reduction in face to face consultations since Covid. Most will have forgotten the drug Acarbose, an alpha glucosidase inhibitor, which reduced the digestive conversion of poly and di saccharides to monosaccharide (glucose and fructose) . This was actually addressing type 2 diabetes at its root cause, excessive absorption of carbohydrate, but because the partially digested starch acted as an osmotic laxative, and also was fermented in the large bowel, the side effects of explosive farty diarrhoea meant that the drug never took off, unlike the patients! This drug could have been marketed as a great way to control you carb intake, reducing your weight and Hba1c at one go. However, back then fat was even more of a demon than it is now, hence oblivion for acarbose.
    Instead, we are pushing drugs that either reduce the renal threshold for glucose, so that you pee sugar, and get thrush or urinary infections for your troubles, or we ‘tidy the glucose way’ into glycogen stores, and then fat stores when the glycogen stores are filled, by using sulphonylureas to stimulate the patient’s own insulin secretion, or insulin itself, all the while making the patient fatter and fatter.
    We (by which I mean the medical world) are so fixated on the numbers of blood glucose, and it’s long term equivalent, Hba1c, and getting them down, we have lost sight of the fact that if we only controlled how much gets in to the body in the first place (carbohydrate restriction) we wouldn’t have to spend billions of pounds on drugs keeping the levels down, and also treating the myriad complications of type 2 diabetes as well.
    I often use the analogy of plastic shopping bags. Before we had to pay for them they would accumulate around the house, and because they were seen as too valuable to throw out (the way the body views glucose) they would end up being put in drawers and cupboards, until every available storage space was stuffed with plastic shopping bags, just like the way the body stuffs every available fat cell with fat (from converted glucose) when really what we had to do was stop bringing the plastic bags into the house in the first place.

    • Hi Meddyg
      That’s fascinating – thanks so much for sharing this. I didn’t know all the ‘toys’ you had in the box to play with along the way. I had heard of the unfortunate side effect ones – a bit like that weight loss drug Ali! I had also heard that some caused thrush or urinary infections and didn’t make the connection with peeing sugar – duh – of course!

      You’re totally right that it seems so obvious that we have to stop putting the sugar in, but unfortunately this seems like the last thing on the agenda. The plastic bag analogy is great too – I’ve managed to stop them coming into the house with home deliveries although I’d love to get back to friendly shopping one day.

      Thanks again
      Best wishes – Zoe
      p.s. I remember a chat with Jason Fung about packing stuff away. We were debating if obesity was a healthy response to type 2 diabetes because at least sugar was being turned to fat and then packed away in fat cells. Far better than hanging around toes and eyes where real damage can be done!

        • Wow – it’s as if you know the note I am working on for next week! ;-)

  • Hi, thank you for a good post.

    Just checked the GIP (upper gut predominant) and GLP (lower gut) incretin hormone effects. Obese seem to have an imbalance, and the roux-en-Y eliminating the major part GIP production recreates the balance. This resembles the omega6/omega-3 situation, where it is wiser to first decrease O-6 and possible increase O-3. These medicines actually seem to leave GIP unaffected but increase GLP (have I got it right?).

    Dr. Eades had a very interesting weekmail about development of Orlistat mid 1990’s. They were tasked to recruit volunteer to first loose 4% weight on low-fat diet (nicely designed by drug company) during 6 months, after which the eligible subjects entered the paid test phase.
    Biggest first problem: talking to the normal patients on low-carb hi-protein diet on waiting room; Study subjects had hard time loosing this 4% in months while Eades patients were easily loosing this in one month. They wanted to quit and join… Secondly, the drop-out rate became huge, despite the salary for participating.

    The GiP is largely affected by refined carbs, the finer the particle size the higher it goes. Fat-and-carb combo send GIP levels to stars (and conseqently insulin). So, if one wants to affect GIP/Insulin levels, how to most effectively ameliorate this effect thru diet? In order to match the scalpel treatment or uncontrolled wet farts generating drug. Could something be left uneaten instead…?

    • Hi JR
      Many thanks for your comment. I’m also on Mike Eades’ newsletter and the Orlistat (and Hawaii) one was a corker!

      Semaglutide does activate GLP – I didn’t see anything during the research for this note on GIP, so you might well be right.

      We come to the same conclusion – can’t we just eat differently rather than inject these drugs daily or weekly into our stomachs?! I know how difficult overcoming cravings is, but every day you don’t eat craved stuff, it becomes easier to avoid it. Got to be a better way!

      Best wishes – Zoe

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.