___________________

¹ This list is the rapporteur’s summary of points made by the individual speakers identified, and the statements have not been endorsed or verified by the National Academies of Sciences, Engineering, and Medicine. They are not intended to reflect a consensus among workshop participants.

The first session explored (1) medications that can affect body weight and composition, (2) the altered pharmacokinetics and clinical effects seen with some drugs in patients with obesity, (3) current practices and considerations for the inclusion of individuals with obesity in clinical trials, and (4) medications in clinical trials or recently approved to treat obesity. Christina R. Chow, Emerald Lake Safety, and Laura Higginbotham, U.S. Food and Drug Administration (FDA), moderated the session.

MEDICATIONS THAT AFFECT WEIGHT STATUS

Donna Ryan, Pennington Biomedical Research Center, provided an overview of the biology of body weight and body composition regulation. She discussed medications approved for chronic weight management, the influence of medications for various indications on weight and body composition, and potential future approaches to obesity treatment with medications toward the goal of optimizing body composition and function.

Biology of Body Weight and Body Composition Regulation

To illustrate the importance of genetics to body weight and composition, Ryan described the similarity in body habitus seen in monozygotic or identical twins—who share 100 percent of their genes—versus the comparative difference in dizygotic twins, who share 50 percent of their genes (Börjeson, 1976). She clarified that in discussing genetics in relation to body weight, she refers to single-nucleotide polymorphisms and not single-nucleotide single-gene defects sometimes associated with obesity (e.g., leptin receptor deficiency, proopiomelanocortin mutations). Single-nucleotide polymorphisms are common; over 200 of them are associated with body mass index (BMI), and more than 100 are associated with waist circumference. However, she noted that genetic susceptibility to obesity is not the sole determinant; the environment also influences body weight and composition. Obesity has increased in modern times due to a host of social and behavioral factors that foster the expression of genes that heighten the risk.

Numerous factors, lifespan events, and biological aspects of a complex regulatory system influence body weight regulation and the expression of obesity, said Ryan. Growth charts developed to track BMI-for-age percentiles indicate that BMI typically increases from birth through the first year of life, when it decreases (CDC, 2000). Around 6 years of age, an adiposity rebound leads to an increase in BMI until adulthood. During puberty, both boys and girls see a steady increase in lean mass, but this trend is much more prominent in boys (Barnes, 1975). At the onset of puberty, girls experience an increase in body fat, whereas fat tends to decrease in boys. By sexual maturation, girls have twice the body fat of boys. Ryan emphasized the critical importance of body fat to reproduction, given that the onset and maintenance of menses requires 17 and 22 percent body fat, respectively. Most people continue to experience increases in body weight and BMI through adulthood until approximately 65—with some variance associated with race and ethnicity—at which point, BMI decreases with age (see Figure 2-1) (Yang et al., 2021). However, a decrease in lean mass relative to total body mass occurs much sooner, beginning in the 30s, and accelerates in the mid-60s. The weight reduction in older life is generally due to reduced lean mass rather than fat mass, which sometimes increases throughout the lifespan, Ryan noted.

Menopause has not been directly associated with weight gain, but dramatic changes in body fat distribution occur, Ryan explained. Although the general population gains an average of 0.5 kg per year during midlife, women in menopause also experience changes in body fat composition and distribution related to ovarian hormones (Davis et al., 2012). A study found that women in peri- or postmenopause have more fat mass and visceral fat than age-matched premenopausal women (Genazzani and Gambacciani, 2006).

Furthermore, Ryan continued, women on hormone replacement therapy have a similar amount of body fat as premenopausal women, which is significantly less than that of peers in peri- or postmenopause. Changes also occur in the amount of body fat within certain regions. In comparison to women in premenopause or on hormone replacement therapy, women in peri- or postmenopause see increases in percentage of total fat in their trunk region and decreases in the legs. Thus, hormonal changes are accompanied by body fat redistribution, decreasing in the hips and thighs and increasing in the central areas, Ryan stated.

The complex system regulating body composition, weight, and mass involves integrating peripheral signals into the central nervous system (CNS), including the homeostatic and reward systems in the brain, Ryan explained (Grannell et al., 2022; Mendieta-Zerón et al., 2008; Yu and Kim, 2012). Researchers in obesity medicine tend to focus on long-term regulators of body weight and energy expenditure. Leptin, a protein produced by fat, is of particular focus. Proportional to body fat mass, leptin levels increase with higher body fat mass and decrease with weight-loss-related reduction in fat mass. Decreased leptin levels trigger the brain to perceive starvation, which drives CNS responses to increase food intake and decrease energy expenditure. Ryan described a lepto-centric view of body mass and weight regulation in obesity research, but she noted an increasing interest in understanding how the body regulates compartments besides body fat. For example, when part of the liver is surgically removed, it will regrow to a certain limit. Ryan stated that obesity research is exploring the role of muscles and the products of muscles, such as myokines, in signaling the brain to regulate weight and muscle mass.

Among the varied categories of drugs associated with increases in weight, antipsychotics likely produce the most, Ryan noted, particularly second-generation antipsychotics. She illustrated this point with a heat map comparing the effects of specific antipsychotics on weight, BMI, glucose, cholesterol, and triglycerides (Pillinger et al., 2020). Older antipsychotics, such as haloperidol and ziprasidone, are not as effective in controlling mental health symptoms, but they seem to produce lesser effects on weight and metabolic end points, she explained. Whereas first-generation antipsychotics block dopamine, norepinephrine or histamine, many second-generation antipsychotics block dopamine and serotonin 5-hydroxytryptamine receptor 2A (5-HT2A) and 5-HT2C, the serotonin receptor subtypes most involved in food intake. The 5-HT2C antagonists increase food intake, whereas the 5-HT2C agonists reduce it. Dopamine antagonists and histaminergic blockers also drive weight gain. Ryan described that the mechanisms of action for some AOMs are the opposite of those for many antipsychotics. For instance, lorcaserin is a 5-HT2C agonist, and phentermine releases norepinephrine.

Most antidepressants are associated with weight gain, said Ryan, with the notable exceptions of fluoxetine and bupropion. In large doses, fluoxetine can produce significant weight loss and was therefore evaluated as a potential AOM. However, counterregulatory mechanisms cause weight regain, and the net effect was insufficient for it to be considered an AOM. Naltrexone/bupropion is approved for chronic weight management, and bupropion is a norepinephrine reuptake inhibitor. Medications for human immunodeficiency virus (HIV) can increase weight and alter body composition, she continued. Some highly active antiretroviral therapies, such as protease inhibitors, cause redistribution of body fat, resulting in reduced subcutaneous fat and increased abdominal and visceral fat (Tondt et al., 2024). This can cause undesirable metabolic consequences, Ryan noted. Hormones can also create weight changes. For example, glucocorticoids can cause weight gain and truncal fat redistribution. Progestin contraceptives can cause weight gain, with injectable or implantable progestins carrying the highest risk. She noted that intrauterine devices coated with progestin typically do not cause weight gain. Testosterone, when used to treat testosterone deficiency in men, can reduce body fat and increase lean mass.

Potential Future Approaches to Obesity Treatment with Medications

Ryan emphasized that treating obesity should extend beyond the goal of weight loss to optimizing body composition and function. Studies of obesity treatments including surgery, lifestyle interventions, and medications indicate that weight loss involves decreases in both fat and lean mass. She said that 75 percent of weight loss should come from fat mass and 25 percent from lean mass and decreased organ size (Christoffersen et al., 2022).

This goal reduces abnormal excess body fat, preserves lean mass, and optimizes organ function.

Loss of muscle mass can lead to consequences, such as frailty fractures in the shoulder, arm, hip, or pelvis, said Ryan. To illustrate, she cited longitudinal data from the Look AHEAD Trial—which included people aged 55 at the start of the study—which indicate that participants lost approximately 8.5 percent body weight during the first year of an intensive lifestyle intervention, but they regained 4 percent of their baseline weight over 4 years, most of which was fat (Johnson et al., 2017). They therefore lost lean mass, and their resulting body composition was worse than that of a control group. A higher proportion of the intervention group experienced frailty fractures, with these steadily increasing with age. Ryan underscored that falls and fractures could be the event initiating decline and death in older adults; therefore, the increased risk of fractures associated with loss of lean muscle is a serious consequence of weight loss. She highlighted that despite this risk of fracture, the intensive lifestyle intervention group saw benefits, such as improved metabolic health and reduced health care costs.

Ryan noted that emerging approaches to obesity focus on improving body composition and weight loss. For example, activin and myostatin inhibitors are being researched for their associations with reduction in fat mass and modest increase in lean mass. Bimagrumab is a monoclonal antibody that blocks activin receptor type-2B (ActRIIB) signaling. Taldefgrobep alfa is an adnectin, similar to a monoclonal antibody, that blocks active myostatin and inhibits ActRIIB signaling. Apitegromab is a monoclonal antibody that targets pro and latent myostatin. Ryan added that selective androgen receptor modulators are sometimes used inappropriately by weightlifters to build muscle mass but are being researched as a mechanism to preserve lean mass during weight loss.

Ryan reiterated that medications can promote weight loss or gain; they can also improve or deteriorate body composition and metabolic function. Thus, better understanding of these processes is needed to work toward the obesity management goal of improving body function and health.

ALTERED PHARMACOKINETICS AND CLINICAL EFFECTS OF DRUGS IN PATIENTS WITH OBESITY

David J. Greenblatt, Tufts University School of Medicine (TUSM), discussed the altered pharmacokinetics and clinical effects of drugs in patients with obesity. He began by introducing the concept of the “70-kilogram man,” which originated during World War II for use in U.S. Armed Forces recruitment efforts and established the mean weight of a healthy young man as 70 kg (154 pounds). Much has changed in the decades since, he observed, with both BMI and the prevalence of overweight and obesity

among people aged 20–39 having increased since at least the 1960s—when the National Health and Nutrition Examination Survey began tracking these data—with marked increases over the past 30 years and trends holding for both men and women (Greenblatt, 2013). He noted that obesity carries the risk of comorbid conditions (obesity-related diseases), such as hypertension, type 2 diabetes, cardiovascular disease, metabolic disorders, depression, osteoarthritis, and sleep apnea.

Greenblatt highlighted the foundational research on drug disposition in obesity conducted by Darrell Abernethy while on faculty at Tufts University. From 1979 to 1984, he ran a series of studies that established a foundation for current knowledge about drug effects in individuals who have overweight and obesity. This research involved (1) characterizing body habitus both numerically and in terms of pharmacokinetics, pharmacodynamics, and drug disposition; (2) observing changes in treatment effects in individuals with obesity and the implications of such changes; and (3) understanding the intrinsic changes in body habitus associated with gender and age.

Greenblatt explained that when Abernethy began in 1979, measures of obesity were limited. Dual-energy X-ray absorptiometry (DEXA) technology for measuring bone density was not yet widely available, and the BMI calculation existed but had not been fully validated. Therefore, Abernethy indexed obesity in quantitative terms using percent ideal body weight (IBW). This measure was based on actuarial data used by insurance companies to partition premiums based on deviations from IBW. It multiplies five by height minus 60 inches and adds 100 pounds for women and 110 pounds for men. Percent IBW is actual weight divided by IBW and multiplied by 100. Greenblatt highlighted close intercorrelation between percent IBW and BMI, noting that BMI reasonably correlates with total body fat measured by DEXA.

Drug Effects and Distribution in Relation to Lipid Solubility

Abernethy correctly hypothesized that a drug’s effects and distribution are in relationship with its lipid solubility, said Greenblatt. Abernethy studied drug distribution corrected for protein binding for a variety of drugs ranging from relatively water soluble to highly lipid soluble. He found that approximately 80 percent of the variance in healthy-weight control individuals is related to lipid solubility. Abernethy then evaluated numerous water-soluble drugs (e.g., caffeine, cimetidine, salicylate), highly lipid-soluble drugs (e.g., diazepam, midazolam), and drugs with intermediate lipid solubility (Bruno et al., 2021). Greenblatt noted that most of the highly lipid-soluble drugs were psychotropics. In total, 19 drugs were tested on a cadre of study participants that included healthy-weight controls, with

a mean weight of 142 pounds and 99 percent IBW, and those with overweight and obesity, with a mean weight of 251 pounds and 179 percent IBW. Greenblatt emphasized that the study was in 1979, when a weight of 251 pounds was more uncommon than it is today, and included slightly and very overweight individuals. The study design consisted of single intravenous doses, if preparations were available, and oral doses when absolute availability was known. Using pharmacokinetic methods, Abernethy and colleagues conducted 19 studies with 20–40 individuals per study in over 600 trials.

The first study measured distribution of intravenous diazepam—a highly lipid-soluble psychotropic drug—and yielded striking results, Greenblatt stated (Greenblatt et al., 2022). Comparison of age- and gender-matched individuals revealed that the volume of distribution in an individual with obesity was sevenfold that of a control, the half-life showed a fivefold increase, and drug clearance was similar. Abernethy repeatedly observed this pattern of results as he conducted more studies, with it most pronounced with highly lipid-soluble drugs. For example, intravenous midazolam, another lipid-soluble psychotropic drug, had prolonged half-life in individuals with obesity compared to controls. Abernethy found a tight relationship between its volume of distribution and degree of obesity, with percent IBW accounting for approximately 85 percent of the variability. In contrast, the pharmacokinetic profiles for intravenous cimetidine, a water-soluble drug, were similar for the controls and individuals with obesity, with only slight increases in the latter group. Greenblatt said that volume of distribution, half-life, and clearance of water-soluble cimetidine were similar for the controls and individuals with obesity, but the volume of distribution and half-life of lipid-soluble midazolam were both threefold higher in the latter group. He added that after correcting volume of distribution for total body weight—calculated by dividing liters distributed by kilograms of total body weight—the increase for midazolam in people with obesity held. Thus, a disproportionate distribution of midazolam into body weight over ideal weight is seen in individuals with obesity, Greenblatt emphasized.

Greenblatt reviewed a model of drug pharmacokinetic activity in healthy-weight people and in those with obesity (Figure 2-2). The model represents the central compartment (C) with a shaded square, the peripheral compartment pharmacokinetic (P) with a white square, and the peripheral compartment anatomic (A) with a dashed circle. The model of a nonlipophilic drug in a healthy-weight individual shows that the C square and A circle are approximately the same size, and the P square is slightly smaller than C and A. However, in a person with obesity, the C and P squares are the same size, but the A circle is much larger than both. This represents that the pharmacokinetic peripheral compartment of a water-soluble drug increases only slightly because it is not soluble in the additional adipose

tissue, Greenblatt explained. However, the activity of lipophilic drugs is quite different. As with a nonlipophilic drug, the C square and A circle are approximately the same size in a healthy-weight person, but the P square is larger than both and probably consists mainly of adipose tissue. In an individual with obesity, the A circle is larger than the C square, and the P square is much larger still. Greenblatt noted that this represents that the pharmacokinetic peripheral compartment in an individual with obesity increases disproportionately and becomes larger than the anatomic increase in body size; this is due to the selective solubility of lipophilic drugs in excess adipose tissue.

Drug Clearance in Individuals with Obesity

Drug clearance is the volume of blood completely cleared of drug per unit of time, explained Greenblatt. He clarified that it is not the elimination half-life, elimination rate constant, or rate of drug disappearance or elimination. Furthermore, drug clearance is independent of distribution and a major determinant of half-life and of steady-state concentration during multiple doses. Calculating it involves dividing the dose by the “area under the curve,” a term that represents total drug exposure integrated over time. Greenblatt explained that drug clearance and volume of distribution are independent of one another and combine to determine half-life, so half-life is the dependent variable. In those with obesity, drug clearance has no evident relation to volume of distribution or lipid solubility and no consistent relation to degree of obesity. He described that in some studies, clearance changes with obesity, and in many studies, it does not.

Greenblatt summarized that volume of distribution and drug clearance are independent and combine to produce half-life. Lipid solubility of a drug and obesity in an individual combine to determine volume of distribution, which might modify a loading dose following a single dose, as with phenytoin. When a loading infusion or loading treatment initiation regimen is used to attain a steady state more quickly, the volume of distribution might influence such a regimen for long-half-life drugs, as with brexpiprazole. He stated that half-life is important due to its ability to determine the duration of action after single doses and the rates of accumulation and washout after termination. Clearance is a principal determinant of steady-state concentration during maintenance therapy. A study of chronic diazepam use provided individuals with obesity and a control group with a constant daily dose for a month followed by discontinuation, said Greenblatt (Abernethy et al., 1983). The rates of state attainment and drug washout were prolonged in individuals with obesity despite similar concentrations in both groups once steady state was obtained. He explained that the steady-state concentrations were similar because drug clearance rates were the same in both groups.

labeling examples pertaining to obesity. She highlighted the general expectation that pivotal clinical trials of a therapeutic product should include its intended patient population. Noting that the U.S. population is becoming increasingly diverse, Vaidyanathan stated that a diverse clinical trial population allows for better assessing the risk–benefit profile in specific groups and aids in providing recommendations for use. However, because premarket clinical trials aim to limit heterogeneity of treatment effects, they often use a population with a homogeneous disease background to limit variability and maximize the chances of demonstrating treatment effects. Vaidyanathan emphasized that this practice often leads to underrepresentation or exclusion of certain subpopulations and gaps in prescribing recommendations for medications.

Clinical Pharmacology and Clinical Trials

Understanding the drivers of response to therapy and its variability is an important goal in clinical pharmacology, said Vaidyanathan. Variability in response may be attributable to intrinsic factors, such as age and body weight, or extrinsic factors, including drug–drug interactions, alcohol or drug use, and the environment (Huang et al., 2007). The influence of some of these factors on treatment effects in certain patient populations is evaluated and addressed during drug development. Regulatory review of clinical pharmacology data in both investigational and new drug applications examines how intrinsic and extrinsic factors influence dose or drug exposure, safety, and efficacy. The typical drug development program consists of several clinical trials, including dedicated clinical pharmacology studies to characterize pharmacokinetics and pharmacodynamics, drug–drug interactions, and effects of various intrinsic factors and other covariates. She noted that these studies are often conducted in healthy individuals. In addition, clinical trials on the intended patient population include dose-finding studies and pivotal efficacy and safety trials. All such studies yield pharmacokinetic data, Vaidyanathan explained.

Intrinsic and extrinsic factors that commonly influence drug exposures are often investigated in dedicated clinical pharmacology studies, Vaidyanathan stated. These stand-alone studies are well controlled and can provide a robust assessment of these interactions. They typically feature a limited number of participants, and their design is frequently a single dose in healthy individuals. She described that population pharmacokinetics and exposure response can be an appropriate methodology to explore the effects of obesity on drugs, especially when an adequate number of individuals with obesity is included in the study and, if possible, in all categories of

overweight and obesity. Testing body weight as a covariate in population pharmacokinetics is a general approach and often includes various weight metrics. Vaidyanathan said that other quantitative systems pharmacology models could be developed to describe the effects of physiological changes in obesity on pharmacodynamics, safety, and efficacy.

Investigational product review is a continuous process, explained Vaidyanathan. The critical stages—such as at the end of phase 2 and during phase 3R planning—entail questions as to whether the development program will provide adequate data regarding drug effects in a broad population that includes subpopulations of interest. Manuals of policies and procedures guide the FDA process of regulatory and scientific evaluations during a new drug application or investigational product review stages (Office of Clinical Pharmacology, 2023; Office of New Drugs, 2013). Additionally, FDA review staff use templates to promote consistency in review format and content. For example, the clinical pharmacology review template enables review staff to conduct a structured evaluation of whether intrinsic factors require an alternative dosing regimen and/or management strategy for any particular subgroup, thus ensuring therapeutic optimization and individualization. Vaidyanathan stated that obesity has known implications related to physiological changes and corresponding effects on overall drug disposition. Moreover, changes in pharmacokinetics of drugs in populations with obesity are highly variable and depend on multiple factors such as drug characteristics, degree of obesity, and specific organ function, she added (Jain et al., 2011).

Current State of Dosing Information for Patients with Obesity

Vaidyanathan highlighted a study of drug labels approved in 2004–2010 that reported that obesity was listed as a risk factor for adverse events or other comorbidities on several labels (Jain et al., 2011). However, the labels provided very limited information on pharmacokinetics or pharmacodynamics, and only one drug included specific dosing instructions for patients with obesity. A recent study found that 32 drug products approved before September 2022 provided information specific to obesity but with a high level of variability in type and amount (Madabushi, 2022). Similarly, two literature reviews on pediatric drug product labels for new molecular entities—for drugs approved 2007–2016 and 2016–2021—found very limited information specific to pediatric obesity (Samuels et al., 2023; Vaughns et al., 2020). Vaidyanathan noted that the information also featured varying definitions and variability on how much pediatric drug development addressed obesity, and data most often originated from adults.

___________________

² For more information about this workshop, visit https://www.fda.gov/drugs/news-events-human-drugs/bridging-efficacy-and-safety-obese-considerations-and-scientific-approaches-11092022 (accessed March 19, 2024).

and safety trials of liraglutide in type 2 diabetes weight management programs included patients with obesity (FDA, 2014). Data from these trials can be used with population pharmacokinetics and exposure response to determine the adequacy of the dose. Vaidyanathan highlighted that several initiatives are underway to promote diversity in clinical trials, including statutes and regulations that require drug developers to include various demographic information and patient subgroup analyses and present subgroup analyses for efficacy and safety. Examples of obesity-specific regulatory activities include engaging stakeholders, collaborating in regulatory research and publications, reviewing drug regulatory submissions, and publishing guidance on obesity-specific recommendations (Ramamoorthy et al., 2022; Vaidyanathan et al., 2023).

Guidance recommendations specify including participants who have obesity, either as a prespecified subgroup with a prospective analysis plan or during drug development, noted Vaidyanathan (Vaidyanathan et al., 2023). This specification is included in recommended approaches for sponsors to take to increase enrollment of underrepresented populations in clinical trials, and it mentions both demographic and nondemographic characteristics, such as patients with extreme body weight or organ impairment (CDER and CBER, 2020). Additionally, guidance recommends evaluation of the influence of obesity on pharmacokinetics, dose, and efficacy. A pediatric pharmacology guidance states that drug development should address the effects of disease state and obesity on drug disposition (CDER, 2022). Vaidyanathan stated that increased body weight can reduce the effectiveness of some contraceptives. Therefore, FDA guidance for establishing the effectiveness and safety of hormonal drug products intended to prevent pregnancy recommends that the trial population should include women who have obesity with a BMI greater than or equal to 30 kg/m² (CDER, 2019) and that the analysis plan for them should include prespecified subgroup efficacy analyses for women who have obesity. Given the overlap of nonalcoholic steatohepatitis (NASH) and metabolic conditions, such as obesity and type 2 diabetes, guidance on NASH with liver fibrosis specifies that the proportion of patients with these comorbidities in clinical trials should reflect the target population (CDER, 2018a). Moreover, the NASH guidance recommends that sponsors engage with FDA before initiating phase 2 and 3 trials. It provides recommendations on developing new drug products for chronic weight management (CDER, 2007). Other therapeutic areas with obesity-specific recommendations include developing drugs for prophylaxis of anthrax (CDER, 2018b) and anti-infective drugs (CDER and CBER, 2021). Vaidyanathan noted that recommendations address excluding individuals with obesity when warranted for safety. Guidance is also available regarding the presentation of relevant information in labeling recommendations for health care providers.

variability in exposure for patients with obesity, but no dose cap was projected to raise concentrations to a level that posed safety concerns for adverse effects. The analyses demonstrated that a 1 mg/kg dosing scheme with a dose cap of 125 mg in persons weighing 100 kg or more was within the therapeutic window for patients with and without obesity and therefore recommended. Vaidyanathan described a two-part decision tree created to prioritize drugs for dedicated pharmacokinetic studies in individuals with obesity (Langevin et al., 2023). It uses drug models such as exposure-response, drug characteristics, physical chemical properties, and elimination pathway to provide and predict exposure changes. The authors recommend that stakeholders use the decision tree to prioritize resources toward the drugs most likely to cause exposure changes.

Vaidyanathan reiterated that clinical trials should reflect the intended patient population and, given the prevalence of obesity, include individuals who have obesity. Sponsors should address drug disposition in those who have obesity early in drug development and determine and discuss with regulators the need to enroll them in phase 2 and 3 trials, she emphasized. Modeling tools, such as exposure response in population pharmacokinetics, can enable early identification of whether obesity is likely to have an appreciable effect on pharmacokinetics. Sponsors can use modeling to consider including those with obesity in pivotal trials. FDA guidances highlight the need for including diverse trial participants in drug development, and regulatory agencies have engaged with stakeholders to discuss including those who have obesity. Vaidyanathan underscored that consistent, collective efforts from all stakeholders—including regulatory agencies, sponsors, clinicians, scientists, and trial participants—are needed to reduce the gap in prescribing information regarding patients with obesity.

TRANSFORMATION: WHAT IS TO COME IN THE PIPELINE OF ANTI-OBESITY MEDICATIONS

Ania Jastreboff, Yale School of Medicine, outlined medications in clinical trials or recently approved by FDA for obesity treatment. She described this moment as transformative and of critical importance to medical history. The potential to effectively treat obesity could unlock the ability to mitigate, treat, or even prevent hundreds of related diseases, she stated. Jastreboff added that progress toward this goal needs to be made in a thoughtful, safe manner.

Case Study Comparison

To illustrate the transformative nature of the new generation of AOMs, Jastreboff shared case studies of two patients. The first was an 18-year-old woman with a BMI of 57 who had several obesity-related diseases and was

fearful of type 2 diabetes, which both of her parents had. Over 3 years, Jastreboff sequentially added four medications: metformin, liraglutide, bupropion, and naltrexone. In less than 4 years, the patient lost 139 pounds, approximately 45 percent of her body weight, and her hemoglobin A1C normalized. The second patient was a 49-year-old woman with a BMI of 34 and several obesity-related diseases. She had tried numerous diets that led to weight loss; however, she repeatedly regained the weight, as would be expected. Jastreboff said that the human body works to regain fat. This patient began a clinical trial of tirzepatide and lost >90 pounds within 1 year, 45 percent of her body weight, and her BMI normalized. Whereas both patients lost the same percentage of body weight, the treatment for the first involved four agents over more than 3 years, compared with one agent over 1 year. Jastreboff described these cases as illustrating the potential of new AOMs. Although medications prescribed to the first patient are effective, especially in combination, the recent introduction of semaglutide and tirzepatide constitutes a leap forward in AOMs that has brought the field to a watershed moment in treatment capabilities, Jastreboff said.

Nutrient-Stimulated Hormone-Based Therapies

Although research on AOMs focuses on numerous mechanisms, including myostatin activin pathway inhibitors and melanocortin-4 receptor agonists for monogenic obesity, Jastreboff focused on nutrient-stimulated hormone-based (NuSH) therapies that have been approved, such as semaglutide, tirzepatide, and liraglutide, and several in phase 3 trials (Jastreboff and Kushner, 2023). These hormones, such as amylin, glucagon, peptide YY, gastric inhibitory polypeptide (GIP), and glucagon-like peptide 1 (GLP-1), are stimulated when food is consumed and signal to the brain and various body tissues about energy homeostasis. They can be grouped to create dual or triple agonist NuSH-based therapies that may increase the therapeutic effect on body weight and other health benefits, Jastreboff explained. She noted that GLP-1 is the most heavily researched NuSH, given that GLP-1 receptor agonists have been used to treat type 2 diabetes for nearly 20 years.

The NuSH pipeline contains several AOMs, including four in phase 3 trials and three that have been approved by FDA for obesity treatment (chronic weight management), said Jastreboff. Semaglutide is a long-acting once-weekly self-injectable GLP-1 receptor agonist. FDA approved it initially for diabetes and then for chronic weight management in adults in 2021 and in adolescents aged ≥12 years in 2022. As with many NuSH-based therapies, it is a once-weekly self-injectable. In the STEP-1 trial, the average weight reduction for adults is 16.9 percent body weight, or 38 pounds, at 68 weeks (Wilding et al., 2021). The STEP TEENS trial found a reduction in BMI of 16.1 percent in adolescents aged 12–17 (Weghuber et al., 2022);

a STEP Young trial is assessing safety and efficacy in children aged 6+ years (NLM, 2019). Furthermore, cardiometabolic measures of adults on semaglutide demonstrated improvements in hemoglobin A1C, blood pressure, and lipid panel, foreshadowing that it may translate to better health outcomes in terms of cardioprotection, she emphasized. The SELECT trial of semaglutide in over 17,000 individuals with cardiovascular disease and obesity without type 2 diabetes demonstrated a 20 percent reduction in composite end point of cardiovascular death, nonfatal myocardial infarction, and nonfatal stroke (Lincoff et al., 2023). Jastreboff underscored that this therapeutic impact was found in a population already optimized in terms of evidence-based treatment. The STEP-HFpEF trial investigating individuals with obesity and heart failure with preserved ejection fraction and without type 2 diabetes found a significant reduction in heart failure symptoms and physical limitations (Kosiborond et al., 2023). Moreover, the magnitude of benefit observed was directly related to the extent of weight reduction. Top-line results of a semaglutide trial of individuals with chronic kidney disease and type 2 diabetes, the FLOW trial, found a 24 percent reduction in risk of kidney disease progression and kidney and cardiovascular death. Jastreboff highlighted that in October 2023, this trial was terminated early based on efficacy, meaning that it was deemed unethical to continue the placebo group when semaglutide treatment appeared to be so effective.

Tirzepatide, a once-weekly, self-injectable, long-acting dual GIP/GLP-1 receptor agonist, is the first dual hormone receptor agonist to be approved by FDA for treatment of diabetes (in 2022) and obesity (in 2023), Jastreboff stated. She led the SURMOUNT-1 trial that demonstrated that participants randomized to the highest dose (15 mg) had an average reduction of 22.5 percent total body weight at 72 weeks, with nearly 40 percent of participants losing at least one quarter of their body weight (Jastreboff et al., 2022). The trial also demonstrated marked improvement in various cardiometabolic measures, and numerous trials are assessing how this may translate into health benefits. The SURMOUNT-2 trial, which looked at tirzepatide for weight management in individuals with type 2 diabetes, demonstrated similar weight reduction in addition to improved hemoglobin A1C (Garvey et al., 2023). SURMOUNT-4 placed all participants on tirzepatide for 36 weeks and then randomized them to continue tirzepatide for a total of 88 weeks or switch to placebo (Aronne et al., 2024). It demonstrated that continuing the treatment resulted in an average reduction of 26 percent of total body weight, which translated to approximately 62 pounds. Moreover, a majority of participants lost at least one quarter of their body weight. Highly effective agents, such as tirzepatide, may require longer to reach a weight plateau (76 weeks in this case). Additionally, participants on the placebo experienced weight regain. Tirzepatide has also been associated with improved cardiometabolic measures and benefit

in diabetes control, she added, with additional trials evaluating its effects on obesity-related conditions and pediatric patients with obesity. Another trial studied it after intensive lifestyle intervention (Wadden et al., 2023). Jastreboff emphasized that lifestyle intervention is important for optimizing health, and that care providers should encourage patients to eat nutritious food and maximize physical activity. Given the many different types of obesity, variability is observed in both weight reduction and regain with all treatments, including these novel AOMs, and is being further explored. Tirzepatide trials are ongoing for individuals who had prediabetes and obesity, primary and secondary cardiovascular prevention and renal outcomes, and pediatric patients.

In addition to the AOMs with FDA approval, Jastreboff noted that several NuSH-based therapies are in phase 3 trials. CagriSema, a weekly injectable, contains cagrilintide (an amylin analog) paired with semaglutide. A study of semaglutide with various doses of cagrilintide found that treatment of 2.4 mg of each drug resulted in average reduction of 17.1 percent body weight in 20 weeks (Enebo et al., 2021). She underscored the rapid nature of this weight loss and noted that, on average, regain occurred if treatment was discontinued. Survodutide is a weekly injectable dual GLP-1 and glucagon receptor agonist. Phase 2 trial results found an 18.7 percent reduction in total body weight at 46 weeks of treatment, with up to 40 percent of participants having lost 20 percent of body weight (le Roux et al., 2024). Retatrutide is a weekly injectable triple hormone receptor agonist targeting GIP, GLP-1, and glucagon receptors. A phase 2 trial led by Jastreboff demonstrated that in 48 weeks of treatment, participants lost an average of 24.2 percent of total body weight, which translated to approximately 58 pounds of absolute reduction within 11 months (Jastreboff et al., 2023). At 48 weeks, participants had not yet plateaued, indicating that additional reduction is likely with extended treatment. The drug is in a phase 3 trial to investigate its full efficacy. Jastreboff highlighted that retatrutide exceeded the 5 percent body weight reduction threshold for efficacy set by FDA for AOMs, as 100 percent of participants reached that target at the two highest doses, 8 mg and 12 mg. She underscored the rare nature of achieving 100 percent efficacy in a scientific study. At the highest dose, 90 percent of participants lost at least 10 percent of body weight, 63 percent lost at least 20 percent, and 26 percent lost 30 percent or more. Jastreboff noted that loss of 30 percent body weight is akin to the results seen with bariatric surgery. The study found variability in weight reduction, which has been demonstrated with all forms of obesity treatments, including bariatric surgery and all AOMs. Furthermore, women experienced a higher mean percent change in body weight, 29 percent, than did men, 22 percent. More research is needed to understand whether and how variability in efficacy is related to genetics,

phenotypic factors, and other considerations, as there are many different types of obesity, she added.

Oral GLP-1 receptor agonists are also in development, Jastreboff stated. Oral semaglutide is FDA-approved to treat type 2 diabetes, and higher doses completed trials for chronic weight management. A phase 3 trial of daily treatment with 50 mg demonstrated an average weight reduction of 17.4 percent at 68 weeks. It is a peptide, and researchers are developing small molecules that do not see the degradation by digestive enzymes that peptides experience. Orforglipron, which is a small-molecule NuSH that has demonstrated an average weight reduction of 14.7 percent at 36 weeks, is also in phase 3 trials. Jastreboff noted that in addition to weekly injectable and daily oral medications, MariTide is a GIP receptor antagonist and GLP-1 receptor agonist in a monthly injectable form that is in phase 2 of development.

Obesity Treatment That Optimizes Health

Now that medications have made total body weight reductions of 15, 20, or 25 percent attainable, the research and clinical focus is extending beyond “just” weight reduction; the goal of treating obesity is to improve health outcomes, Jastreboff said. This includes mitigating related diseases (e.g., cardiovascular disease, type 2 diabetes, sleep apnea), minimizing risk for developing them, and considering the quality of weight reduction, including body composition and muscle function. She described that myostatin activin pathway inhibitors (Tisdale, 2010), such as bimagrumab, taldefgrobep, and SRK-439, are in development and may improve body composition and result in some weight reduction. Therefore, potential future combination therapy may pair new NuSH-based therapies with myostatin activin pathway inhibitors and hold potential to maximize fat loss while minimizing loss of muscle mass.

A focus on treatment to optimize health includes numerous factors and considerations, said Jastreboff. Treatment components include targeting the neurometabolic pathophysiology of obesity to address its underlying biology of the disease. Targeting different mechanisms for various types of obesity involves considering its heterogeneity. Assessment of obesity severity is also needed, including related conditions and the overall health and metabolic profile of each patient. Moreover, providers should consider the quality of weight reduction and determine how to optimize individualized combination therapy, she said. Jastreboff added that treatment should involve assessing the rate of weight reduction and titrating medication to prevent patients from potentially losing weight too quickly. She emphasized that starting with the lowest dose and up-titrating based on how the patient is responding (weight reduction and side effects) is the best approach to

mitigating potential gastrointestinal side effects and tailoring treatment to the patient’s health goals. Jastreboff noted that with bariatric surgery, providers consider potential negative effects of weight reduction—such as bone loss, vitamin deficiencies, muscle loss and function, and side effects—which should also be done for AOMs. Jastreboff added that providers should help support patients in lifestyle measures to improve nutritious food consumption (prioritizing protein, given that food intake will decrease during the weight reduction phase); increase physical activity, including resistance exercise; reduce stress; and improve sleep quality. Furthermore, she continued, providers should always remember and be thoughtful of the bias and stigma patients face and the psychological and psychosocial implications of obesity and even of seeking treatment. Moreover, she underscored that access and affordability are critical considerations, as these new tools are of no use if patients do not have access to them.

DISCUSSION

Chow and Higginbotham moderated a panel discussion that addressed topics including metrics for improvements in body composition, treatment assessment benchmarks, provider resources for potential effects of medications in patients with obesity, in silico modeling, and the expected length of treatment.

Treatment Assessment Metrics

Higginbotham highlighted a challenge in identifying a clinically meaningful end point for body composition and preservation of lean mass. Noting that FDA only accepts laboratory- or imaging-based end points when they are validated surrogates—and does not consider changes in DEXA imaging to be clinically meaningful to patients—she asked about potential end points in a development program for treatments for improved body composition. Ryan said that body composition is a moving target that features sexual differences that change across the lifespan. In addition to sex and age differences, racial differences may also be at play. These dynamics make it challenging to identify a target, she said. Jastreboff commented that assessing body composition and muscle function should be included in medication trials to enable data collection across ages and sexes; metrics for preservation of muscle mass and function could be determined from these data. Higginbotham replied that FDA collects body composition data, but this is performed as a subset in the trial, and data are not collected on every participant.

Given that AOMs can reach near 100 percent success for 5 percent weight loss, Chow asked whether new measures of success would be appropriate.

Ryan responded that many benefits accompany modest weight reduction. However, given that robust reduction is now possible, she suggested that a range of additional metrics would be helpful, including a BMI lower than 25; a waist circumference lower than the cutoff point for age, sex, and race; normal glycemia and other cardiometabolic markers; and patient-reported outcomes for quality-of-life benchmarks. Jastreboff added that the 5 percent weight reduction is an important benchmark due to the related metabolic benefit and that targets that reflect optimized health should also be incorporated. Greenblatt said that the Tufts University Jean Mayer Human Nutrition Research Center on Aging has robust data on the benefits of exercise and resistance training in older adults for changes in body habitus and general health. He maintained that weight loss should be accompanied by efforts to mitigate lean mass loss via resistance training and appropriate diet.

Clinician Resources

Chow asked about the availability of resources that instruct clinicians on how best to administer drugs known to cause weight gain or with altered pharmacokinetics in people with obesity. Greenblatt emphasized the value of referring to the original data on distribution, clearance, and half-life for a specific drug rather than a pharmacy compendium or website, as such sources filter original data. Ryan replied that she refers clinicians to the Obesity Medicine Association’s Obesity Algorithm, a resource that provides information on associated weight gain or loss with different medications and is updated annually (Tondt et al., 2024). To address the lack of guidance regarding altered dosing for patients with obesity, she suggested forming a working group from professional societies to develop recommendations. Vaidyanathan said that physiological-based pharmacokinetic modeling and other modeling tools could be helpful in building dosing information based on pharmacokinetic data available in the literature.

In Silico Modeling

Chow asked whether in silico modeling could increase the speed with which a potential drug is approved. Vaidyanathan replied that FDA has various avenues for supporting modeling-based methodologies, depending on the type of data used to develop them. She noted that many case examples feature sponsor use of in silico approaches to input data in trial design or other questions related to clinical trial drug development. Vaidyanathan underscored that early interaction with FDA is recommended for sponsors proposing in silico approaches.

This page intentionally left blank.