GLP-1 Drugs

Obesity

Why do we care about obesity? Obesity is the largest risk factor for basically every disease and condition. By reducing people’s weight, we reduce and prevent the downstream complications including cancer, heart disease, diabetes, kidney disease, obstructive sleep apnea.

About ~40% of Americans are obese (BMI > 30) in 2024, and ~50% will be obese by 2030. We eat an obesogenic diet and our eating habits as a nation are not going to get better. Same goes for our lifestyle, we will continue to be more sedentary.

More so, the nature of gaining weight is insidious. Specifically, eating an excess of 11 calories per day (i.e., an extra potato chip or two) results in gaining one pound over a year (h/t Randy Seely). Over a few decades from chilhood to middle age, this results in a) ~30 pounds which will place most middle-age adults in the obese category and b) a habit of eating that is hard to break (h/t Power of Habit).

Losing weight is fundamentally an energy expenditure game: energy in (eating) must be less than energy out (exercise, movement, basal metabolic rate). But, this is hard because eating is easy and limiting it is equivalent to breaking an addiction. And, exercise is hard to do consistently when you are already overweight (h\t personal experience) and requires discipline (very hard). These factors are not going to improve in the United States or other developed countries. Drugs for treating the end-cause of this problem are key and going to becomre more and more important.

Weight loss targets

Modest weight loss at any BMI has been associated with improved health measures including but not limited to blood pressure, hemoglobi A1c, and LDL. A 5% decrease in weight loss was associated with significant improvements in CVD risk factors (Wing 2011). 5-10% weight loss is associated with reduced health care costs (Ryan 2017). The 5% level is generally accepted as the minimum clinically meangingful level.

What does weight loss do? Per Ryan 2017, Glycemic improvement in type 2 diabetes, triglyceride reduction, HDL increase, improvement in obsstructive sleep apnea, reduces osteoarthritic knee pain, reduces emergent knee pain, hepatic steatosis reduction, quality of life score improvement, depression improvement, mobility improvement, urinary incontinence improvement, sexual function improves, PCOS and infertility improvements, decrease in health care costs, and decrease in mortality. In general 5-10% is sufficient to see considerable clinical improvement along these metrics.

For a number of these conditions like OSA and NASH (now called MASH), there is a dose-response relationship between weight loss and these parameters: more weight loss means more improvement.

5-10% weight loss had no effect on major cardiovascular outcomes (Wing 2013). But there is a reduction in mortality with weight loss greater than 10% (Sjostrom 2007). Note that the most robust data for these weight loss reductions and outcome measures are in post-MBS patients since these procedures have been done for much longer. Given that the magnitude of weight loss is the same in post-MBS patients and these newer GLP-1RA drugs, there is no reason to suspect that the results would not transfer.

Bariatric Surgery

Bariatric surgery results in a ~20% weight loss (Maciejewski 2016). Poor weight loss or weight regain, resulting in less than 20% weight loss, affects up to 1 in 4 patients who undergo bariatric surgery (Mok 2023). The BARI-OPTIMISE trial showed an 8% weight loss with liraglutide (older generation GLP-1RA) after 24 weeks in post-operative metabolic and bariatric surgery (post-MBS) patients compared to the placebo group who experienced a 0.5% reduction. Retrospective studies with semaglutide in post-MBS patients showed ~15% weight loss at 48 weeks (Lautenbach 2023 , Murvelashvili 2023).

It used to be true that bariatric surgery was more effective than any nonsurgical intervention at promoting weight loss and improvement in co-morbid outcomes (Sudlow 2019). But, this was before the recent GLP-1RA drugs came out.

What are GLP-1 drugs?

Glucagon-like peptide 1 (GLP-1), part of a broader group of naturally occuring molecules called incretins, is a hormone essential for regulating glucose levels in the human body (Anderson 2018). GLP1s act by stimulating GLP-1 receptors.

We have created drugs that resemble GLP-1s to take advantage of these receptors and their downstream signalling. GLP-1 Receptor Agonists (GLP-1RAs) are molecules that activate GLP-1 receptors in the pancreas which lead to enhanced insulin release and reduced glucagon release-responses that are both glucose-dependent and carry a very low risk for hypoglycemia (Shaefer 2015). GLP-1RA also acts on GLP-1R in the brain which cause reduced appetite and increased satiety (Anderson 2018). It is thought that much of the weight-loss effect of GLP-1 drugs is by their action on GLP-1R in the brain, not in peripheral tissue. It also acts in the GI tract which lead to delayed glucose absorpotion due to slower gastric emptying (Masseli 2021). GLP-1 receptors are expressed in a range of tissues - heart, skeletal muscle, pancreas, specific nuclei in the brain, lung, smooth muscle, and kidney - which suggests that GLP-1 has many effects outside of regulating glucose metabolism.

Lixisenatide, exanitide, liraglutide, and other older-generation GLP-1RA’s will not be covered here. They did not work particularly well relative to the modern class of medications listed here and are not in favor with prescribers in the USA. But, data from these older generation GLP-1RAs can motivate new avenues for development. One example is the positive results of the Phase 2 trial of lixasenatide in Parkinson’s disease (post).

For MASH, trizapetide appears to be better at improving liver fibrosis stage compared to semaglutide. These are early data but encouraging for trizapetide. When looking at patients with improving liver fibrosis, 55% and 51% in the 5 mg and 10 mg trizapetide group compared to 30% in placebo whereas 43% in 0.4mg semaglutide compared to 33% in placebo. The semaglutide studies concluded no significant different whereas the trizapetide concluded that there was a significant difference.

This paper in JAMA internal medicine showed that semaglutide reductions in weight were lower than trizapetide reductions in patients obese or overweight. This was not an RCT but a retrospective analysis of EHR data frrom May 2022 to September 2023. Among the matched population at risk undergoing treatment, 82% on tirzepatide vs. 67% on semaglutide achieved 5% of greater weight loss; 62% vs. 37% acheived 10% or greater, and 42% vs. 18% achived 15% or greater highlighting the increased ability of tirzepatide relative to semaglutide in weight reduction. Findings in this study are broadly consistent with existing evidence from RCTs when comparing across RCTs examining semaglutide vs. placebo and tirzepatide vs. placebo. Consistent with clinical trials, larger reductions were seen in those without T2D relative to those with T2D.

Broadly, for both semaglutide and tirzepatide, there was a decreased risk of obesity-associated cancers (12 cancers including kidney, breast, colorectcal, pancreatic, ovarian, and multiple myeloma) for T2DM patients on GLP1-RAs compared to those on insulin (HR for each cancer around 0.7ish on average). Data from comparison of metformin vs. GLP1RA was less clear with hazard ratio 95% confidence intervals overlapping 1. There may be a reduction in the risk of cancer in high risk populations though it is probably modest when compared to insulin and very small compared to metformin. Data from this retrospective study found here

There is some off-label use of GLP-1RAs in treating PCOS-related infertility by OB/GYNs. The typical use is 6 months of GLP-1RA and contraception then getting off both and trying to conceive. There is no clinical trial yet to suggest that GLP1-RAs are effective in approaching infertility.

GLP-1s limit cravings which can be helpful in addiction, specifically alcohol. It has been noted that people report drinking less while on GLP-1s. There is very limited evidence to suggest GLP-1s help addiction but literature is building (here, here, and here).

In fact, Roche anticipates that based on current study data, the unique oral GLP-1 RA CT-966 could be used as maintenance treatment after weight loss induced by injectable medications in the class.

Roche joins other pharma companies with its entry into the oral incretin-based weight loss drug arena, including Structure Therapeutics with its GSBR-1290 delivering a placebo-adjusted weight loss of 6.2% after 12 weeks in a phase 2 trial; Viking Therapeutics with an announcement in March 2024 of a placebo-adjusted weight reduction of 3.3% after 4 weeks of treatment with its once daily tablet VK2375.

Amgen is in phase 3 trials for maritide, a GLP-1RA and GIPR antaongist. This is interesting because tirzepatide is a GLP-1RA and GIPR agonist. Randy Seeley in his presentation from AACR 2024 on anti-obesity medications was not convinced that the GIPR antagonist strategy will be as successful as a GIPR agonist approach. Amgen data for their GIPR antagonist strategy appears to come from adipocytes. Randy suggested that much of the benefit of semaglutide and tirzepatide comes from its effects on the brain rather than peripheral cells.

This paper from Dec 2023, from fellow Yale medical student Ashwin Chetty does a great job covering the clinical trials of GLP1 drugs at the time of publication: (Chetty et al).

There are many targets besides GLP-1, and GIP. These include PYY, glucagon, amylin, and activin. Some of these like amylin are being targeted by current trials.

Side effect profile of GLP-1RAs

Nause and vomiting are the primary side effects of GLP-1RAs and are experienced by most patients. These side effects are elicited from stimulating GLP-1 receptors in the brain. Patients report nausea most frequently which oftentimes in my discussions is more like a queasiness that resolves 24 hours post-injection during the first week of a new dose. Patients describe it as if they had a big thanksgiving meal and express the familiar feeling of “I don’t want to see food or eat it because I’m so full”. Constipation is often reported by patients which if untreated leads to overflow diarrhea especially among the older population. In conversation with physicians, they report that if the constipation is treated well then the pressure will not build up and they do not usually feel as nauseaus. Treating constipation well involves starting the patient on a personalized bowel regimen usually beginning with miralax and fiber supplements. As a note, stool softeners like Senna do not usually help here.

GLP-1s carry a risk of gallstone pancreatitis. About 2% of patients on semaglutide develop a gallbladder related disorder, most of which are cholecystitis. Bariatric surgery carries the same risk. Anytime a person loses weight rapidly, they are predisposed to bilary problems.

Ania Jastreboff, who led the SURMOUNT trizapetide trial, states that these side effects are transient and preventable with proper dose escalation strategies. Specifically, she recommends the provider up-titrates slowly and tailors each program to the individual. Essentially, go slowly to avoid side effects and have a physician who is experienced with this work with you. In talking to other physicians who prescribe a lot of GLP-1s or have used them, they also echo these sentiments.

Of course, a small minority of people will not tolerate the drug. But, the same is true of statins and basically every drug. There is some concern that GLP-1RAs carry an increased risk of suicidal ideation. Data from a recent JAMA internal medicine study and Nature Medicine study find that suicide risk is not increased with GLP1 use.

We do not know the long term safety profile of GLP-1RA drugs and the FDA has asked for approved GLP-1RA drugs to report long term safety data.

Absolute contraindications for GLP1-RAs are a history medullary thyroid cancer history, MEN2 status, and pregnancy. There is a moderate increase (HR 1.78 w/ 95% CI 1.04 to 3.05) in the risk of thyroid cancer following GLP-1RA use (Bezin et al).

2 year follow up data of patients on semalgutide revealed that 2.6% (1.3% in placebo) experienced gallbladder disorders, 8% (11.8% in placebo) experienced severe adverse side effects, and 82% experienced (54% in placebo) nausea/vomiting/diarrhea/constipation (study). Side effects are comparable with placebo in the patient population eligible to take semaglutide suggesting that outside of an increase in mild GI disturbances, the drug is tolerable.

The American Society of Anesthesiology says to avoid GLP-1RAs 7 days prior to date of surgery. This may lead to missed doses and a re-titration of GLP-1 dosing after surgery and in-hospital recovery requiring close coordination with the original GLP-1 prescriber. Also, the pens come in fixed doses that cannot be manually adjusted so new pens with the proper doses need to be re-prescribed and may take more time in being delivered requiring further down-titration followed by up-titration. A recent JAMA paper reports that in a retrospective study there was no worse outcomes for patients who did not stop GLP-1RAs prior to surgery compared to those that did. However, this study only measured ICU admission, aspiration pneumonitis, or respiratory failure within 7 days post-op. This is not a) an RCT, b) a complete list of adverse outcomes measured, or c) a sufficient enough time frame to make assessment.

In speaking with Jenna Klimovich, a physician at the CT VA who prescribes GLP-1RAs, she shared that most side effects occur when a new dose is started. So, in a dose escalation strategy that consists of four up-titrations over a month followed by a stable dose, the first 48 hours of each new dose for the first month will be accompanied by GI distress. Then, symptoms will subside till the next dose escalation. After a month, assuming no interruptions, symptoms should be minimal.

Patients may lose a considerable amount of lean muscle mass (see supplement of SURMOUNT trials for example) if they do not dose escalate slowly. All patients on GLP-1RAs because of the risk of lean muscle mass loss and associated fractures and frailty should focus on strength training and protein intake. Since patients on GLP1-RAs consume less food by volume than they are normally able to, it may be beneficial to shape their food intake to be packed with protein and vegetables and if room is left over to continue to other foods. This may help prevent some of the lean muscle mass long if accompanied by resistance training. These measures are not cheap and are time intensive.

GLP-1 dosing strategies for weight loss and maintenance

GLP-1 drugs are not a one time treatment. They are currently daily oral or injectable medicines. Like pills for hypertension, these medicines will likely have to be used regularly in order for patients to continue to realize their benefits and not regain weight. Strategies for doing this in the least invasive manner possible are important to increase patient adherence.

Currently, patients on injectables continue to inject every week in perpetuity. It is possible that in the future after a period of injectable use, patients can maintain their weight loss on an oral pill.

There is also development of a long acting injectable such that patients may be able to receive a single injection once every thirty days. The long acting injectable strategy already exists for drugs like haloperidol (antipsychotic) and depo-provera (birth control) so this technology may be possible for GLP-1RAs too.

Compounding pharmacies and pricing concerns

GLP-1RAs semaglutide and tirzapetide are sold at high price points ($1000/month) and are the subject of much government scrutiny from Biden, Sanders, and others. Simultaneously, Eli Lilly stock is up 180% in the past 2 years and Novo Nordisk stock is up 142%.

Compounded versions of Wegovy, Ozempic, and Mounjaro exist on the market and are being sold by online companies like HIMS. The FDA allows for compounded versions of patented drugs to be sold if there is a shortage of the drug. Shortage currently refers to inadequate supply to match demand. This is true. Though, given the pricing pressure, the government could look to expand the definition of shortage to include supply-demand mismatch from unaffordability.

How might this discussion change if Trump takes office instead of Harris?

Hesistations that may limit the success and broad penetration of semaglutide and tirzepatide
How to enter the field?

Regardless of which GLP-1RA company comes out ahead, what are the inevitable consequences of a large portion of the US population that is on anti-obesity medications? What behaviors will people do when they lose 20-30% of their body weight?

Obesity is a chronic disease. And GLP-1s are here to stay. They have the potential to be the statins for obesity. But, they require lots of tinkering. It’s an art. There is dose escalation, side effect management, management of the drug during surgeries and hospitalizations, and titration of other weight-based medications.