What is the pipeline for future medications for obesity?

Obesity is a chronic disease characterised by excess adiposity that impairs health and affects about 650 million people worldwide [1, 2]. It increases the risk for multiple metabolic complications including type 2 diabetes (T2D), metabolic-dysfunction associated steatotic liver disease (MASLD, previously known as non-alcoholic fatty liver disease), cardiovascular disease as well as many mechanical complications such as osteoarthritis and obstructive sleep apnoea (OSA) [3].

Lifestyle interventions (including diet, exercise and behavioural changes) are the cornerstone of obesity management with diverse benefits, but even the most intensive lifestyle interventions result in up to 10% mean weight loss (WL) and weight maintenance remains a challenge, as 80% of WL is expected to be regained over the next 5 years [4, 5]. The main drivers for weight regain after significant WL include the persistence of a lower resting metabolic rate and the increased appetite, possibly mediated through long lasting increased orexigenic and decreased anorexigenic signals [2, 6]. Despite that 5–10% WL is clinically beneficial, greater WL may be required for the individual to improve or achieve remission of some obesity-related complications [7, 8]. Bariatric surgery can result in 25–30% mean WL and long-term weight maintenance, however, it is not scalable at the population level and people may be hesitant to this option due to the perceived risk of postoperative complications [9, 10].

However, there is still a significant difference between the WL that can be achieved with bariatric surgery and the currently approved obesity pharmacotherapies and there is heterogeneity in treatment responses with GLP-1 RA (for example people with T2D may achieve less WL compared to those without diabetes in clinical trials, despite similar lifestyle interventions) [13, 14]. Additionally, the currently approved GLP-1 RA for obesity treatment are in injectable form and some people may be reluctant to consider injectable treatments [15].

In this review, we discuss the pipeline of obesity pharmacotherapies with focus on entero-pancreatic hormone-based molecules, and we evaluate the data from early and late phase clinical trials on their safety and efficacy on WL in people with and without T2D. We also consider the effect of these molecules on other obesity-related complications [glycaemia in people with T2D and on liver fat content in people with MASLD/metabolic dysfunction-associated steatohepatitis (MASH, previously known as non-alcoholic steatohepatitis)] and we discuss the potential clinical implications and challenges that the new era in obesity pharmacotherapy will bring.

Numerous entero-pancreatic hormones [GIP, glucagon, amylin and peptide YY (PYY) agonists and GIP antagonists] with diverse metabolic actions are currently under investigation either alone or in combination with GLP-1 RA (Fig. 2), aiming to enhance and/or complement the effect of GLP-1 agonism on weight and metabolism [14]. The notion for combining entero-pancreatic hormone-based therapies for obesity treatment is supported by the efficacy of bariatric surgery on WL, an intervention leading to increased postprandial levels of multiple entero-pancreatic hormones [31, 32]. Moreover, preclinical and early phase clinical trials have confirmed the therapeutic potential of entero-pancreatic hormone combinations in obesity and/or metabolic complications such as T2D [33].

GIP is secreted by K-cells in jejunum in response to food intake and its physiological role includes stimulation of insulin secretion, increased glucagon secretion, increased lipogenesis and enhancement of lipid buffering capacity (Fig. 1) [34]. In the context of T2D, the ability of GIP to stimulate insulin secretion and to ameliorate glycaemia is impaired [35].

Studies in animal models have shown an anorexigenic action of GIP receptor agonism [36] and a recent phase 1 clinical trial with a long-acting GIP RA found that repeated dosing for 4 weeks induced modest WL (−1.9 to −3.1 kg vs. −0.4 kg with placebo) in people with T2D, without delay in gastric emptying or nausea and vomiting [37].

Nevertheless, as the simultaneous activation of GLP-1 and GIP receptors in preclinical models results in greater WL and glucose-lowering efficacy compared to activation of each receptor alone [38–40], the interest for the development of unimolecular GLP-1 and GIP RA was increased, despite that this additive effect in food intake and glycaemia was not observed in some acute clinical studies [41, 42].

Tirzepatide is a once weekly, subcutaneous, unimolecular dual RA (GLP-1 and GIP) which has comparable GIP receptor binding affinity to native GIP and 5 times lower GLP-1 receptor affinity than that of native GLP-1. Tirzepatide has been approved for T2D management - the mean HbA1c reduction in SURPASS programme ranged between 1.9 and 2.6% across the trials and 40–69% of participants achieved ≥10% WL with the higher doses (10 and 15 mg), despite that there was no additional support for lifestyle intervention [16, 43]. In a subanalysis of SURPASS-3, tirzepatide 10 and 15 mg resulted also in a relative reduction in liver fat content by 40–47% compared to 11% with insulin glargine at 52 weeks [44]. A significant reduction in appetite and food intake was also observed with tirzepatide compared to placebo in people with T2D [45].

The phase 3 SURMOUNT programme assessed the safety and efficacy of tirzepatide as obesity treatment and the medication has now been approved (November 2023) for chronic weight management by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) [46, 47]. In SURMOUNT-1, 72 weeks of tirzepatide induced mean WL of 16–22.5% in people without diabetes compared to 2.4% with placebo, with no sign of the weight plateauing, suggesting that there may be further WL with long-term use [46]. In SURMOUNT-2, the mean WL was up to 15.7% with tirzepatide 15 mg compared to 3.3% WL with placebo at 72 weeks for people with obesity and T2D [47]. In both studies, improvements in weight associated with improved quality of life parameters, physical function and cardiometabolic risk factors [47].

The most commonly reported AE with tirzepatide were gastrointestinal, including nausea, diarrhoea and vomiting. Most of them were mild to moderate in severity and improved over time. Only 4-7% of participants in SURMOUNT-1 and −2 discontinued the medication due to AE (Tables 2, 3).

Another two trials from the SURMOUNT programme have recently been published [48, 49]. SURMOUNT-3 evaluated the efficacy and safety of tirzepatide compared to placebo for 72 weeks after a 12-week intensive lifestyle intervention lead-in period that included a low-calorie diet and exercise. The trial randomised adults with obesity who have achieved ≥ 5% WL by the end of the 12-week lead-in period (mean WL at the lead-in period was 6.9%) to placebo or tirzepatide. Those taking tirzepatide, achieved an additional 18.4% WL from randomisation to 72 weeks compared to those taking placebo who experienced weight regain of 2.5% [48].

The SURMOUNT-4 trial assessed the weight maintenance with tirzepatide and had also two periods: a 36-week open-label lead-in period during which all participants took tirzepatide, followed by a 52-week double-blind treatment period during which participants were randomised to either continue on tirzepatide or switch to placebo. At the end of the 36-week tirzepatide lead-in period, participants achieved 21.1% mean WL. Those taking tirzepatide, managed an additional 5.5% WL from randomisation, compared to those taking placebo who experienced mean weight regain of 14% from randomisation [49]. At the end of the SURMOUNT-4 study (88 weeks from baseline), 25.9% of people at the placebo group were able to achieve ≥ 15% WL compared to 84.1% at the tirzepatide group.

Several other trials are assessing the safety and efficacy of tirzepatide in improving different cardiometabolic complications such as OSA, MASH and heart failure with preserved ejection fraction (HFpEF) (NCT05412004↗, NCT04166773↗, NCT 04847557↗) [16]. Moreover, multiple other GLP-1/GIP RA in oral or injectable form are in early phase of development (Fig. 1) – CT 388 has shown a placebo-adjusted WL of 8.5% after 4 weeks in a phase 1 trial [50].

The exact molecular mechanisms leading to WL and glycaemia improvements with tirzepatide are still under investigation, mainly because the role of GIP receptor activation with tirzepatide is unclear. One prevailing theory is that tirzepatide is an imbalanced and biased GLP-1 RA with very low efficacy for recruitment of β-arrestin, whilst is acting as an unbiased GIP agonist with full efficacy for β-arrestin recruitment [51, 52]. Nevertheless, a number of other hypotheses have been suggested and both GIP receptor activation and GIP receptor antagonism have been proposed as potential mechanisms to tirzepatide’s efficacy [53].

GIP antagonism is also a potential obesity treatment as in preclinical studies improved the metabolic profile and reduced food intake [54–56]. A possible explanation for the similar effects on weight of both GIP agonism and antagonism is the potential desensitisation of GIP receptors by GIP agonists exposure [57].

AMG133 (subcutaneous, every 4 weeks) is a monoclonal antibody that was designed to antagonise the GIP receptor and is also linked to two modified GLP-1 peptides that activate the GLP-1 receptor [58]. Results from a phase 1 trial have supported further clinical evaluation with a dose-dependent WL up to 14.5% with AMG133 compared to 1.5% with placebo by day 85 [59]. Of note, the WL in participants to this trial was maintained for few months after the last injection. AE were mainly nausea and vomiting which were transient. A 52-week phase 2 trial on AMG133 in people with overweight and obesity is ongoing (NCT05669599↗).

Glucagon is secreted from the pancreatic a-cells and the main physiological site of action is the liver to increase hepatic glucose production [60]. Glucagon agonism reduces also food intake [61] and increases energy expenditure [62], suggesting that it could promote WL (Fig. 1). In vivo studies with long-acting glucagon analogue suggest that hypoaminoacidaemia could be an important contributing mechanism to the observed increase in energy expenditure with glucagon agonism, but it also leads to enhanced lean mass loss [63]. Moreover, the combination of glucagon with GLP-1 actions could improve WL while protecting against the risk of hyperglycaemia [64]. Further glucagon actions include the improvement of the whole body lipid metabolism and the promotion of hepatic fatty acid oxidation which may provide therapeutic actions for MASLD and/or MASH [60].

Initial results from studies in rodents suggested a synergistic role of dual GLP-1 and glucagon agonism on reducing food intake and led to the development of numerous unimolecular GLP-1/glucagon co-agonists. Despite the encouraging results of GLP-1/glucagon co-agonism in experimental models, different GLP-1 and glucagon co-agonists have shown various levels of efficacy and tolerability in people with obesity and/or T2D, in early phase clinical trials, which may be explained by the different ratio of GLP-1 to glucagon activity between the different molecules [65].

Additionally, clinical trials with long-acting dual or triple agonists targeting the glucagon receptor have demonstrated a marked suppressive effect on circulating amino-acids in both humans and animals [66, 67]. This could be due to enhancement of hepatic amino-acid catabolism with hepatic glucagon receptor activation. Over the next years, it will be important to understand the effect of the reduced circulating amino-acids with the novel multi-agonists targeting the glucagon receptor on lean muscle mass (especially in populations with obesity and high risk of sarcopenia) as well as on energy expenditure [66].

Survodutide (BI 456906) is a GLP-1/glucagon co-agonist which has progressed to phase 3 clinical trials as treatment for obesity (SYNCHRONIZE programme). In a recently completed phase 2 trial (abstract data), 46 weeks of survodutide once weekly (0.6 to 4.8 mg) resulted in a dose-dependent mean WL up to 18.7% vs. 2% WL with placebo in people with obesity [68]. The percentage of participants who discontinued the medication due to AE ranged between 20 and 29%. Most treatment discontinuations were due to gastrointestinal AE and occurred during the rapid escalation phase – this may be mitigated with more gradual dose escalation at phase 3 trials.

For people with T2D, survodutide 1.8 mg twice weekly resulted in 9% WL vs. 5.4% WL with semaglutide 1 mg and 1.2% WL with placebo at 16 weeks (phase 2 trial) [69]. Mean HbA1c reduction was superior with the higher doses of survodutide compared to semaglutide 1 mg (1.9% vs. 1.5%, respectively). Despite that the serious adverse events (SAE) were similar to semaglutide 1 mg, the AE leading to drug discontinuation were 10–30% compared to 4% with semaglutide 1 mg (Table 3). Survodutide has also received FDA fast-track designation for adults with MASH (NCT04771273↗).

Mazdutide (IBI362 or LY3305677) is a once-weekly oxyntomodulin analogue which acts both on GLP-1 and glucagon receptors [65]. A phase 2 trial assessing mazdutide 3, 4.5 and 6 mg in a Chinese population with overweight/obesity demonstrates up to 11.3% WL after 24 weeks of treatment compared to 1% weight gain with placebo [70]. Another phase 2 trial using a higher dose of mazdutide for obesity (9 mg) is currently ongoing with a press release revealing a placebo-adjusted WL of 15.4% after 24 weeks [71]. Phase 3 studies using mazdutide as obesity treatment in a Chinese population (doses between 4 and 9 mg) are also ongoing (NCT05607680↗, NCT06164873↗).

For people with T2D, a phase 2 trial with mazdutide (doses 3, 4.5 and 6 mg) has shown reduction in HbA1c (up to −1.7%) and up to 7.1% WL at 20 weeks (Table 3), while early phase trials for MASH are in progress [72]. Regarding safety profile, in phase 2 trials for obesity and diabetes, mazdutide was discontinued due to AE by 0–2% of participants and SAE experienced by up to 7% of participants with mazdutide 6 mg (Tables 2 and 3).

Pemvidutide (ALT-801) is another unimolecular once weekly GLP-1/glucagon agonist undergoing a phase 2 trial as obesity treatment, following a phase 1 trial demonstrating up to 10.3% WL at 12 weeks [73]. A press release of the phase 2 obesity trial (MOMENTUM-1) reports a mean WL up to 15.6% (placebo-adjusted 13.4%) at 48 weeks with the 2.4 mg dose. However, 20% of participants discontinued the pemvidutide 2.4 mg due to AE with the most common being nausea and vomiting - the discontinuation rate was 5–19% with lower doses [74]. In people with T2D, a phase 1b study did not demonstrate improvement in HbA1c, however there was a placebo-adjusted WL of 8.5% after 12 weeks (Table 3). Furthermore, in a phase 2 trial in people with MASLD, 24 weeks of pemvidutide resulted in a relative reduction in liver fat content up to 76% with the 2.4 mg dose vs. 14% with placebo [75, 76]. Pemvidutide is also undergoing phase 2 studies as MASH treatment.

Efinopegdutide (JNJ-64565111; HM12525A) is also a unimolecular once weekly GLP-1 and glucagon agonist that has been investigated in people with obesity, T2D and MASLD/MASH. In a phase 2 clinical trial for obesity, 26 weeks of efinopegdutide (doses 5 to 10mg) resulted in up to 11.8% WL in people with obesity (without diabetes) compared to 7.5% WL with liraglutide 3 mg and 1.8% WL with placebo [77, 78]. For people with T2D, a phase 2 trial showed that 12 weeks of efinopegdutide achieve up to 7.9% WL (compared to 0.7% WL with placebo) without actual improvement in HbA1c [78]. However, in people with T2D and MASLD, 24 weeks of efinopegdutide 10 mg reduced liver fat content by 72.7% compared to 42.3% reduction with semaglutide 1 mg despite similar WL achieved with both interventions [79]. Efinopegdutide has been granted fast track designation by the FDA for the treatment of MASH (NCT04944992↗, NCT05877547↗) and this is the indication that the clinical development programme of this molecule will focus. Most common side effects were gastrointestinal, with 24.5% of participants in the phase 2 obesity trial discontinuing efinopegdutide 10 mg due to AE – however when the same dose was used with gradual escalation at the MASLD trial (starting dose 2.4 mg for 4 weeks, then 5 mg for 4 weeks and then 10 mg), only 5.6% of participants discontinued the medication due to AE.

Other GLP-1 and glucagon co-agonists in early phase clinical trials as treatments for obesity include AZD9550 and LY3305677 (NCT05623839↗).

Given the efficacy and benefits of the dual GLP-1/GIP RA tirzepatide and the dual GLP-1/glucagon RA, a triple agonist targeting all three receptors (GLP-1/GIP/glucagon) may result in superior WL and glycaemic control than dual agonists. Indeed, in pre-clinical models, retatrutide, a triple agonist (GLP-1/GIP/glucagon), resulted in greater WL and improved glucose profile compared to tirzepatide through increased energy expenditure and reduced calorie intake [80, 81].

Retatrutide is administered once weekly and is more potent at human GIP receptors and less potent at GLP-1 and glucagon receptors [80]. In a phase 2 study in people with obesity (without T2D), retatrutide (doses 1 to 12 mg) has led to a dose-dependent WL up to 24.2% at the end of the 48-week treatment period compared to 2.1% with placebo [81]. At 48 weeks, WL of ≥25% had occurred in 36–48% of people with the 8 mg and 12 mg doses compared to no one who received placebo. Greater % WL was attained with retatrutide among participants with BMI≥35 kg/m² and among female participants. Marked improvements in lipid profile and blood pressure were also observed compared to placebo [81].

In people with T2D, retatrutide (0.5 to 12 mg) led to substantial reductions in bodyweight and HbA1c compared to both placebo and dulaglutide 1.5 mg in a phase 2 trial [67]. After 36 weeks, retatrutide resulted in HbA1c reduction up to –2.2% vs. −1.4% with dulaglutide and −0.3% with placebo and a bodyweight reduction of up to 16.9% with retatrutide vs. 2% with dulaglutide and 3% with placebo [67]. The benefits of triple agonism extended also to MASLD – in a subgroup of 98 people, fat content normalised in approximately 90% of people receiving the highest retatrutide doses [82].

The safety profile of retatrutide was consistent with that of other incretin-based therapies. Transient and mostly mild-to-moderate gastrointestinal symptoms were the most frequently reported AE and occurred more frequently with the 4 mg starting dose groups rather than the 2 mg starting dose groups [67]. AE leading to drug discontinuation were observed in 16% of participants with the 12 mg dose, but the overall incidence of SAE was low (up to 6-8%).

A programme of phase 3 trials with retatrutide (TRIUMPH) is ongoing with an aim to assess the safety and efficacy of retatrutide in different populations living with obesity (OSA, T2D, established cardiovascular disease, osteoarthritis). Another triple agonist (HM15211) is undergoing phase 1 clinical trials with no published data so far.

Amylin is co-secreted with insulin from the β-cells of the pancreas and plays a key role in postprandial satiety regulation (Fig. 1). Amylin acts on amylin receptors in the brainstem to reduce food intake [83] and improves glucose metabolism by delaying gastric emptying and inhibiting glucagon secretion [84, 85]. The reduction of food intake with amylin is not accompanied by the expected concomitant decrease in energy expenditure through actions at the sympathetic nervous system.

Pramlintide is the first synthetic amylin analogue that approved for diabetes management with WL up to 7.9% [86]. Cagrilintide is a newer long-acting amylin analogue - in a phase 2 trial, once weekly cagrilintide resulted in a dose-dependent WL between 6 and 10.8% compared to 9% with liraglutide 3 mg and 3% with placebo [87].

Other amylin-based molecules which are in early stage clinical trials include long-acting amylin agonists and a dual amylin and calcitonin RA (Table 1).

As WL with GLP-1 RA and amylin analogues results both from distinct and overlapping pathways, a combination of these entero-pancreatic hormones may induce a synergistic WL effect [84].

In a phase 1b trial, 20 weeks of cagrilintide 2.4 mg once weekly in combination with semaglutide 2.4 mg once weekly (cagrisema) resulted in up to 17.1% WL compared to 9.5% WL with semaglutide 2.4 mg plus placebo in people with obesity [88]. In a phase 2 trial, people with T2D and overweight/obesity achieved greater mean WL with cagrisema 2.4 mg compared to cagrilintide 2.4 mg or semaglutide 2.4 mg alone after 32 weeks of treatment (15.6% WL vs. 8.1% WL vs. 5.1% WL, respectively) [89]. Mean HbA1c reduction with cagrisema was also greater compared to semaglutide 2.4 mg and cagrilintide 2.4 mg alone (−2.2% vs. −1.8% vs. −0.9% respectively).

Gastrointestinal AE with cagrisema were more than with semagutide or cagrilintide monotherapy, however the SAE and AE leading to medication discontinuation were minimal and similar between the groups [87, 90]

A programme of phase 3 clinical trials (REDEFINE) assessing the safety and efficacy of cagrisema in people with obesity (NCT05567796↗, NCT05394519↗, NCT05813925↗) is currently ongoing. Moreover, an oral GLP-1 and amylin co-agonist (amycretin) is also in early phase clinical trials (Table 1).

PYY is co-secreted from the intestinal L cells together with GLP-1 following food intake. Following secretion, PYY is rapidly cleaved by dipeptidyl peptidase-4 (DPP-4) to its active form (PYY 3-36) acting on neuropeptide Y receptor type 2 (Y2R). Y2R receptor is present in the brain and its agonism results in a reduction in food intake and increased satiety [91]. Studies with PYY agonists administered intravenously have shown reduction in food intake with increased satiety [92, 93], however, a nasal PYY agonist showed minimal efficacy with poor tolerability [94]. Long-acting, subcutaneously administered, PYY RA are undergoing early phase clinical trials as obesity treatments either alone or in combination with GLP-1 RA (Table 1, NCT02568306↗ and NCT03574584↗).

In a phase 1 trial, a PYY analogue (Y14 peptide) administered subcutaneously at 7- to 14-days intervals was safe and potentially efficacious (WL between 2.9 and 3.6 kg at 31 days, with a 38–55% reduction in food intake vs. placebo) [95]. NNC0165-1875 has recently completed a phase 2 trial in combination with semaglutide 2.4 mg once weekly (NCT04969939↗), but results are not currently available.

Several therapies that are not based on entero-pancreatic hormones are currently under evaluation for obesity and they represent further therapeutic options, considering their distinct mechanism of WL to the entero-pancreatic hormone therapies.

A new era for obesity treatment has commenced where pharmacotherapy with combinations of entero-pancreatic hormones approach the WL efficacy of bariatric surgery. Tirzepatide is the first dual agonist which has been approved for chronic weight management, but numerous other dual and/or triple agonists (cagrisema, retatrutide, mazdutide and survodutide) are also in phase 3 trials as potential treatments for obesity and its metabolic complications. Moreover, oral GLP-1 RA are also under development and will provide an alternative option.

The plethora of efficacious obesity pharmacotherapies with different mechanisms of action will allow tailored treatment plans based on individual’s preference, comorbidities and treatment response. A percentage of people will not be able to tolerate the new pharmacotherapies or achieve the individualised goals and others may not have long-term access to these treatments. Combining different treatment modalities (including lifestyle interventions, pharmacotherapies and bariatric surgery) may support people to achieve individualised long-term goals, maximise health benefits and improve quality of life.

Obesity pharmacotherapy is a rapidly moving field and further research on long-term clinical efficacy, safety and cost-effectiveness will inform better their place in the treatment algorithms for obesity and obesity-related complications over the next years.

Supplementary material