Differentiate next-generation GLP-1s by quality of weight loss
The rise of highly effective glucagon-like peptide-1 receptor agonists (GLP-1s) and multi-agonists has transformed obesity research into a multi-billion-dollar industry with strong momentum and fierce competition. Eli Lilly reported 2025 sales of tirzepatide through September 30th of $24.8 billion, making it the world’s best-selling drug. In November, Pfizer won its bidding war against Novo Nordisk to acquire the metabolic startup Metsera for $10 billion.1
With multiple agents demonstrating 15% to 20% reductions in baseline body weight, weight loss alone is now a prerequisite. Next-generation products face a rising threshold for market entry and differentiation. Products that deliver higher-quality weight loss—improving body composition by reducing fat mass (FM), maintaining (or increasing) lean body mass (LBM), and preserving bone mass density (BMD)—could gain a significant competitive edge.
At Parexel, over the past five years, we have completed more than 60 Phase I to IV studies testing GLP-1s, amycretins, and multi-agonists in obesity, type 2 diabetes, cardiovascular diseases, and metabolic dysfunction-associated steatohepatitis (MASH). We work with an increasing number of obesity drug sponsors who wish to incorporate LBM and other endpoints into their trials. Many ask the same questions: Should we start collecting body composition data in Phase 1 or Phase 2? What is the minimum duration for a trial to capture these data? Which measurement tools and imaging modalities work best? And critically, how can we demonstrate that better body composition in a clinical trial translates into functional benefits in the real world? Our experience suggests that linking improved body composition with better physical function will be critical for differentiating next-generation products.
Body composition is key to quality weight loss
High-quality weight loss requires reducing FM, particularly visceral fat, while actively preserving, or even increasing, LBM. LBM preservation is associated with improved metabolic function, greater physical capacity, and reduced frailty. Sponsors must align with regulators on endpoints if they intend to make label claims about body composition. To convince payers, they must champion the clinical value of preserving lean mass over simple total weight loss—and provide data to support the benefits.
The body composition argument may be particularly compelling for vulnerable patient segments. For instance, older patients are susceptible to sarcopenia and fragility: Losing additional LBM due to pharmacotherapy is counterproductive to functional independence. For patients with MASH, improving body composition often translates directly to metabolic health.
Prioritizing body composition will be crucial to the adoption of the next generation of weight-loss therapies, many of which have potential in multiple treatment settings, including as monotherapies, add-ons to GLP-1s, and maintenance therapies to prevent weight regain after GLP-1 treatment is discontinued.
Body composition measurement strategies
We recently conducted a review of obesity trials and found that studies increasingly use multiple complementary measures of body composition.2 FM is a primary endpoint in 28% of recent obesity trials; LBM in 21%; and BMD in 7% (n=124; trials reviewed from clinicaltrials.gov).
Although no single assessment method is 100% accurate, two measures are currently the most common in the studies we reviewed:
- Dual-energy X-ray absorptiometry (DXA): Used in 35% (44/124) of trials, DXA provides a practical and reliable measure of total body composition and BMD. Standardization is good, but cost and logistics at scale remain challenges. For example, we recently worked on a Phase 1 study that incorporated DXA scans to track changes from baseline in body fat, visceral fat, and fat-free mass. Technicians were unable to complete scans for several patients due to their physiques and had to repeat them because of improper positioning. This slowed down the randomization process for the trial. We worked to mitigate these challenges with site training and advance preparation.
- Bioelectrical impedance analysis (BIA) scales: Used in 36% (45/124) of trials, BIA is highly accessible and low-cost. However, even medical-grade BIA scales are less accurate than site-based methods and highly dependent on hydration status. Both alcohol and caffeine intake can impact readings, so patients must avoid them for 24 to 48 hours before a test.
Computed tomography (CT) and magnetic resonance imaging (MRI) are the gold standards for body composition analysis because they produce high-resolution, cross-sectional images that can distinguish fat mass, lean mass, and muscle quality. However, they are costly, and MRI scanners may not be as widely available as DXA scanners. In addition, CT exposes patients to ionizing radiation, and MRI has weight limits and requires longer scanning times. At Parexel, we recommend using BIA scales in conjunction with DXA scans (or hydrostatic weighing) to capture longitudinal trends, often as a secondary or exploratory endpoint. In clinical practice, these are the easiest for patients and providers to access and will likely be used to gather real-world evidence once drugs are on the market.
Increasing use of MRI in early studies
Although DXA scans are less costly and easier to perform than MRIs, MRI scans allow for more precise and detailed fat distribution analysis as well as a wider array of endpoints pertaining to fat distribution, lean tissue mass, and muscle volume and composition. For these reasons, an increasing number of Parexel’s sponsors are incorporating MRI scans into Phase 1 trials. For example, MRI scans were recently reinstated into a Phase 1 protocol due to the value of obtaining preliminary body composition data that could guide endpoint assessments for later pivotal trials. When MRI scans are incorporated in early studies, we plan ahead for operational complexities. For example, we recommend that sites schedule scans well in advance in a prespecified window, rather than at a single study visit. Scanners are not always available due to site-specific demands. Given the challenges of performing an MRI in extremely obese patients, vendors and site staff may require extra training and practice runs. It is also essential to clarify for sites that central reviewers must read participant MRI data before randomization.
We recently worked on an early-phase energy expenditure study that estimated proton density fat fraction via MRI (MRI-PDFF) to assess liver fat and body composition. This required sites to have MRI scanners from major manufacturers, such as GE, Philips, or Siemens, that included a specific PDFF package. The scanners must be calibrated and maintained regularly to ensure accuracy. Also, incidental findings that arose after the scans were reviewed required the principal investigators and sites to follow up with referrals for further evaluation. Sponsors must plan for those realities.
The choice of measurement strategy will depend on the investigational compound and the aims of the study. If a sponsor wants to focus on muscle volume assessments, MRI offers more muscle-related endpoints, while DXA can assess fat and lean mass parameters.
Integrating body composition into early trial design
Body composition measures should be strategically integrated early in development to maximize their impact on regulatory and commercial positioning. Incorporating LBM preservation, FM reduction, and BMD protection as primary or supportive secondary endpoints as early as an expanded Phase 1 multiple ascending dose (MAD) cohort or a Phase 2 is crucial for gaining mechanistic insight. However, it takes about three to six months to see significant changes in body composition, so this may not provide as much value for shorter MAD studies.
"Switch" studies represent a powerful design option for demonstrating that initial weight loss can be sustained or that further weight loss can be achieved. They involve recruiting patients who have achieved an initial therapeutic goal on a standard GLP-1 agent (for instance, 10% weight loss) and then switching them to the novel test agent. This could establish a durable, differentiated effect such as a superior metabolic profile or LBM protection.
In early-phase trials, Bayesian statistical methods can be used to efficiently detect early signals of improved body composition. Focusing on posterior probabilities provides sponsors with a clear, statistical basis to decide whether to pursue the LBM preservation claim.
Linking body composition to functional outcomes
Sponsors must ensure a clear association between imaging modalities (such as MRI, DXA, and BIA) and performance-based tests that demonstrate sustained or improved muscle function. The following validated clinical and performance-based tests provide objective evidence of muscle strength and function, critical for supporting LBM preservation:
- Short physical performance battery (SPPB): A standardized, multi-faceted measure combining tests for balance, gait speed, and lower body strength.
- Six-minute walk test (6MWT): A measure of endurance and functional capacity.
- Sit-to-stand test: A practical, rapid assessment of lower-body strength and mobility.
- Hand grip and leg strength: Handheld dynamometry (HHD) devices can estimate isometric muscular strength of the flexor muscles of the hand and forearm, or of specific leg muscle groups, such as the quadriceps or hamstrings.
- Stair climb power (SCP) Test: A standardized method of measuring the ability to climb stairs, which is an essential daily function, especially for older adults or individuals with mobility limitations, which can lead to loss of independence.3
Based on our experience, hand grip strength, the sit-to-stand test, and the SPPB are easy to perform and clinically useful, whereas the 6MWT follows a standard protocol and takes longer to perform.
Similar to other metabolic diseases, connected devices can gather objective, continuous data in real-world environments and support clinic-based functional assessments for patients participating in weight-loss trials. Potential devices include:
- Actigraphs: Measure movement, sleep, and gait, capturing changes in physical activity and capacity outside the clinic.
- Dynamometers: Enable physicians to test hand grip and leg strength when the patient exerts maximal force.
- Vital sign monitors: Collect key data points related to comorbidities, such as heart disease.
- BIA-enabled weight scales: Valuable as a secondary or exploratory measure for monitoring longitudinal trends in muscle and fat mass.
Sensor data should be evaluated with subjective data collected via clinical outcome assessments (COAs) to provide context for a patient's exertion and overall quality of life. For instance, heart rate data is important in obesity trials to account for subjective differences in exertion relative to absolute accelerometry values. At Parexel, we advise sponsors to collect patient experience to establish a connection between physical function performance outcome assessments and quality of life (QoL). For example, is a good 6MWT score relevant to patients? How can results be interpreted? Direct patient input can help sponsors understand clinical endpoints that impact QoL.
Securing long-term value with quality outcomes
The highly competitive GLP-1 environment demands a strategic reframing of obesity clinical development beyond simple weight loss. The integrated clinical development plan must be designed to build a superior evidence package that directly links the mechanism of action, body composition benefits, functional outcomes, and long-term adherence.
Sponsors can protect their compounds against competing therapies by accurately measuring body composition, validating functional benefits, and defining what is most valuable to patients. By taking these steps, developers can move beyond weight loss to capture the quality of weight loss, ensuring better patient outcomes and a stronger market position.
Resources
- Eli Lilly’s weight loss and diabetes drug tops Keytruda as world’s best-selling medicine, STAT News (October 30, 2025); Pfizer to Acquire Obesity Drug Start-Up Metsera in $10 Billion Deal, The New York Times (November 8, 2025).
- Parexel review of obesity trials planned, ongoing, or completed as of 2025 (unpublished data). Queried clinicaltrials.gov for 2025 trials containing key search terms (including “obesity,” “overweight,” “bariatric surgery,” and “body composition”) and identified 124 clinical trials.
- The Stair Climb Power Test as an Efficacy Outcome in Randomized Trials of Function Promoting Therapies in Older Men, The Journals of Gerontology: Series A (July 6, 2019).