GLP-1 Changed Obesity Treatment. The Next Challenge May Be Clinical Trial Capacity

Obesity pharmacotherapy crossed a threshold that few anticipated at the start of this decade. Semaglutide and tirzepatide produced weight reductions that bariatric specialists had previously associated only with surgery, and the clinical evidence anchoring those results continues to accumulate. In November 2023, Lincoff and colleagues published the SELECT trial results in The New England Journal of Medicine: 17,604 adults with established cardiovascular disease and obesity but without diabetes, enrolled across 804 clinical sites in 41 countries, with once-weekly subcutaneous semaglutide 2.4 mg reducing the risk of major adverse cardiovascular events by 20% over a median follow-up of nearly four years compared with placebo [1]. That finding, more than any commercial milestone, reframed obesity from a chronic condition to a cardiovascular risk factor requiring active pharmacologic intervention.

The commercial and scientific response has been proportionate. Novo Nordisk's 2024 annual financial report documented Ozempic sales of DKr120.34 billion (approximately USD 17.46 billion) for the year, making it the company's highest-revenue product by a wide margin [2]. Across the industry, dozens of sponsors are now advancing obesity programs from early-phase proof-of-concept to pivotal development. IQVIA, in a February 2026 obesity clinical trials analysis published on its own platform, noted that more than 120 anti-obesity drugs were in active development and that obesity clinical trials had nearly doubled over the prior five years [3]. The conditions under investigation have expanded well beyond weight reduction alone. GLP-1 receptor agonists have now been evaluated in dedicated pivotal trials across a range of adjacent indications: semaglutide improved symptoms and functional status in completed trials in heart failure with preserved ejection fraction (STEP-HFpEF, NEJM 2023) [25] and reduced major kidney outcomes in patients with type 2 diabetes and chronic kidney disease (FLOW, NEJM 2024) [26], with ongoing programs in metabolic dysfunction-associated steatohepatitis, obstructive sleep apnea, and knee osteoarthritis [3].

But clinical research infrastructure does not scale the way manufacturing capacity does. The question the industry has not fully answered is whether the sites, investigators, patients, and operational systems needed to run this volume of obesity trials actually exist at the scale the pipeline now demands.

Why the Trial Volume Has No Historical Parallel

According to GlobalData's clinical trials intelligence platform, 2024 was a record year for obesity clinical trials, with the number of trials rising year-on-year since 2020 [4]. That same analysis found that 2025 was tracking to surpass the 2024 record, with 71% of newly announced obesity trials involving innovator drugs, compared with 64% in 2024, the first year innovator trials had comprised more than half the total [4]. (Note: the trial-volume figures above draw on GlobalData reporting; IQVIA's separately published platform analysis supports the pipeline growth claim. The site-activation and protocol-complexity evidence cited throughout this article reflects general clinical trial operations data, not obesity-specific measurements, except where indicated.) IQVIA's own published analysis noted that obesity clinical trial volume had nearly doubled over the prior five years, reflecting the pace at which new entrants continued to enter the space [3].

These active programs require clinical sites, trained investigators, eligible patients, protocol-specific equipment, and sustained monitoring capacity. Many of them are competing for the same finite pool.

The conditions under active investigation have also diversified in ways that amplify the demand on research infrastructure. Retatrutide, Eli Lilly's GLP-1/GIP/glucagon triple agonist, is being evaluated in Phase 3 programs spanning obesity, obstructive sleep apnea, and knee osteoarthritis. Orforglipron, Lilly's oral small-molecule GLP-1 agonist, was evaluated in the ATTAIN-1 trial, where 3,127 adults with obesity were enrolled across a 72-week treatment period, with the highest dose producing 12.4% mean weight loss using the efficacy estimand and 11.2% using the treatment regimen estimand, per results published in The New England Journal of Medicine in September 2025 [5]. The FDA approved orforglipron (now branded Foundayo) for chronic weight management on April 1, 2026, based on the ATTAIN clinical development program [5a]. A Phase 3 cardiovascular outcomes trial, ATTAIN-Outcomes (NCT07241390), evaluating orforglipron's effect on major adverse cardiovascular events in adults with established atherosclerotic cardiovascular disease and/or chronic kidney disease, was recruiting as of June 2026 [5b]. Each indication, each next-generation compound, and each new Phase 3 program requires sites, investigators, and patients that were not previously committed to obesity research.

The Site Activation Problem Is Not New, but the Scale Is

Research from the Tufts Center for the Study of Drug Development (Tufts CSDD) has documented the site activation gap for years. At the 2024 SCOPE Summit, Tufts CSDD findings showed that site activation rates in North America had dropped to approximately 62%, meaning roughly 40% of sites that sponsors attempt to engage either fail to complete contracting and budgeting, fail to receive IRB approval, or fail to enroll a single patient after activation [6]. That figure, troubling in any therapeutic area, carries particular operational weight in obesity research, where per-protocol infrastructure requirements, including DEXA scanning, lifestyle intervention co-interventions, and body composition monitoring, go beyond what many standard Phase 3 trials in endocrinology or diabetes require.

Obesity trials frequently require on-site dual-energy X-ray absorptiometry (DEXA) scanning to characterize body composition and separate fat loss from lean mass loss. Debate continues over whether FDA should require DEXA as a standard endpoint: a 2024 JAMA Internal Medicine commentary argued that while mandatory DEXA would improve compositional accountability in obesity drug trials, such a requirement could also introduce barriers to drug approval that do not apply to treatments for other chronic diseases [7]. Whether required or not, many obesity protocols include DEXA as either a primary or exploratory endpoint, and site eligibility depends on equipment availability, technician competency, and scanner-specific calibration. Not every high-volume endocrinology practice has that infrastructure, and not every hospital-affiliated research center can absorb the operational overhead of running multiple simultaneous obesity protocols in parallel.

Protocol complexity in obesity trials has also risen in ways that mirror broader industry trends. A Tufts CSDD survey of clinical research sites worldwide, published in Therapeutic Innovation and Regulatory Science in 2024, found that increases in protocol complexity and the resulting effort required of investigative sites to implement protocols have been well documented, with DCT approaches further reshaping the burden distribution [8]. Separately, Tufts CSDD protocol benchmarking data reported through 2025 found that the average Phase 3 trial now averages 3.5 amendments per study, up more than 50% from five years prior, with eligibility criteria changes among the most common triggers [8a]. Obesity trials add further layers on top of that general trend: washout periods for participants already receiving commercial semaglutide or tirzepatide, lifestyle intervention co-interventions requiring dietitian or behavioral coach support, long follow-up windows to capture cardiovascular or durability endpoints, and comorbidity-stratified enrollment targets that demand more granular patient identification than a BMI cutoff alone can support.

The SELECT trial enrolled across 804 sites in 41 countries and ran for nearly four years [1]. Its extension study, SELECT-LIFE (NCT04972721), began in September 2023 and was completed in August 2025, gathering long-term outcome data on SELECT participants who continued or discontinued semaglutide [9]. Asking those same sites to now absorb next-generation GLP-1 trials, oral formulation programs, combination therapy studies, and indication-expansion protocols simultaneously is not a planning assumption; it is a staffing and infrastructure stress test.

Patient Supply Is Not as Abundant as the Prevalence Figures Suggest

The epidemiology of obesity presents an apparent paradox for clinical trialists. The World Obesity Atlas 2025, published by the World Obesity Federation in March 2025, projected that the total number of adults living with obesity would reach 1.13 billion by 2030, representing a 115% increase from 2010 [10]. In the United States, National Center for Health Statistics data from the 2021-2023 NHANES survey found that 72.4% of adults age 20 and older had overweight or obesity, with obesity alone affecting 40.3% [11]. On paper, the eligible patient population is enormous.

In practice, the eligible trial population is considerably smaller, and commercial GLP-1 availability is likely changing the accessible pool faster than many sponsors anticipated.

Several dynamics converge here. First, the commercial success of semaglutide and tirzepatide has created an enrollment constraint that did not exist five years ago. A substantial portion of patients who would otherwise qualify for GLP-1 obesity trials are already receiving commercial therapy. Trials with placebo comparator arms or washout requirements face friction when recruiting patients who are currently achieving meaningful weight loss on approved products. Real-world persistence data illustrates how strongly patients who achieve weight loss remain on therapy: a 2025 academic obesity clinic cohort study published in Diabetes, Obesity and Metabolism found that GLP-1 RA persistence significantly exceeded prior real-world benchmarks when patients had structured clinical support and consistent medication access [12]. That attachment to effective therapy creates a recruitment dynamic that did not exist before semaglutide and tirzepatide reached broad commercial availability.

Second, BMI-based eligibility criteria have drawn sustained criticism for producing cohorts that do not reflect the demographic distribution of obesity in the target treatment population. A systematic review published in the Journal of Racial and Ethnic Health Disparities examined phase 3 and phase 4 obesity clinical trials and found that White non-Hispanic individuals made up the majority of participants, while Black and Hispanic patients, who bear disproportionate disease burden, remained underrepresented despite the NIH Revitalization Act of 1993 mandate for inclusive enrollment [13]. A 2023 AMA Journal of Ethics analysis noted that larger-bodied patients remain underrepresented not only in obesity trials but across clinical research broadly, partly because other therapeutic area trials exclude participants above BMI thresholds [14].

In response to persistent representation gaps, the FDA issued a draft guidance in June 2024 under the Food and Drug Omnibus Reform Act of 2022 (FDORA) describing the format, content, and manner of Diversity Action Plans that sponsors of most clinical studies would be required to submit, specifying enrollment goals and recruitment strategies for underrepresented populations [15]. That draft added planning and operational burden to already complex obesity protocols. Following executive branch directives issued in January 2025, FDA removed the draft guidance from its guidance database; the Federal Register notice from June 2024 remains publicly accessible, the statutory FDORA requirement remains in place, but the implementing draft guidance was no longer listed in FDA's searchable guidance database as of mid-2026 [16].

Third, protocol-level eligibility criteria in obesity trials have evolved in ways that narrow the qualifying pool beyond what BMI alone would predict. Trials layering cardiovascular endpoint requirements, specific comorbidity inclusion rules, prior therapy restrictions, and geographic site density naturally exclude large segments of the obese population. The cumulative effect is that sponsors are designing trials for a patient population that is simultaneously smaller, more geographically constrained, and increasingly already engaged with commercial therapies.

Operational Implications for Sponsors and Sites

The site-level burden of obesity trials extends beyond DEXA requirements and washout logistics. A global Tufts CSDD survey of 387 investigative site professionals published in Applied Clinical Trials in March 2026 found that 22% of sites reported financial losses tied to technology coordination, troubleshooting, and training burdens associated with digital and decentralized trial tools [17]. Sites running multiple concurrent obesity protocols face compounded versions of those burdens: separate training requirements per protocol, different eClinical platforms per sponsor, and patient education demands that scale with participant complexity rather than headcount.

Decentralized clinical trial (DCT) models offer partial relief. Tufts CSDD data showed that sites participating in hybrid or full DCT formats reported nearly a 15% reduction in patient recruitment and retention burden, but the same dataset documented a 25% increase in patient education burden and nearly a 20% increase in adverse event monitoring and reporting load [6]. For obesity trials specifically, the tradeoffs are not straightforward. Weight assessments, body composition measurement, vital sign collection, and injection training are difficult to decentralize without sacrificing data quality. FDA's September 2024 final guidance, "Conducting Clinical Trials With Decentralized Elements," encourages sponsors to use risk-proportionate approaches when determining which procedures can be conducted remotely, but body composition measurement remains a largely in-person endpoint [18].

Protocol amendment patterns compound startup risk. Tufts CSDD protocol benchmarking data found the average Phase 3 trial now averages 3.5 amendments, with eligibility criteria among the most common triggers [8a]. For obesity programs where the enrolled patient population is already constrained, mid-study amendments to inclusion or exclusion criteria create cascading operational problems: previously screened patients must be reassessed, IRB submissions must be updated, and site staff require retraining. Each amendment can add weeks or months to the timeline at sites already running at capacity.

The competitive pressure on sites also has a direct effect on how smaller sponsors plan their programs. When multiple sponsors simultaneously seek to activate sites for GLP-1 or next-generation anti-obesity medication trials in the same geographic markets, site selection decisions are increasingly based on prior trial experience, investigator availability, and infrastructure readiness rather than patient prevalence alone. Programs from established sponsors with brand recognition among site coordinators, higher per-patient budgets, and existing investigator relationships from prior trials face less friction in activation. Smaller sponsors or those conducting first-in-class programs face a more competitive site access environment as a direct consequence.

Regulatory and Documentation Considerations

FDA's current framework for obesity and overweight drug development is a draft guidance published in January 2025 (Revision 2), which revised and replaced the prior 2007 draft guidance entitled "Developing Products for Weight Management" [19]. Published in the Federal Register on January 8, 2025 with a comment deadline of April 8, 2025, the draft guidance identifies mean percentage change from baseline body weight versus control as the primary efficacy endpoint for adult weight-reduction trials, with long-term weight reduction described as assessment over at least one year on the maintenance dose [19]. The guidance explicitly recognizes obesity as a chronic disease, a change from the 2007 framing [19].

Cardiovascular outcomes trials carry additional documentation demands that the 2025 draft guidance addresses in sections on clinical outcome claims and long-term safety requirements. The SELECT trial provides the operational template for this category: an event-driven design with a time-to-first MACE primary endpoint, cardiovascular event adjudication by an independent committee, long follow-up duration, and pre-specified sensitivity analyses consistent with FDA's estimand framework under ICH E9(R1) [1], [20]. Sponsors planning similar programs for next-generation obesity agents will need to replicate the organizational and monitoring apparatus that the SELECT investigators built across 804 sites and four-plus years.

From a GCP standpoint, the updated regulatory framework is ICH E6(R3), which was adopted by the ICH on January 6, 2025, and published in the Federal Register by FDA on September 9, 2025 [21]. ICH E6(R3) introduces a risk-proportionate framework for GCP that has direct relevance to obesity trial operations. The guideline supports more flexible, proportionate oversight approaches and places greater emphasis on quality management systems and critical data identification from the earliest stages of trial design, which theoretically should benefit high-volume obesity programs running across large site networks [21]. In practice, operationalizing risk-based monitoring across 200-plus sites in a cardiovascular outcomes trial requires rigorous data management infrastructure and site training that not all programs have in place at activation.

Informed consent documentation in obesity trials faces specific complexity around lifestyle intervention requirements, potential weight regain upon discontinuation of therapy, and the commercial availability of alternatives to trial participation. FDA's December 2025 final guidance, "Enhancing Participation in Clinical Trials: Eligibility Criteria, Enrollment Practices, and Trial Designs," which superseded the prior November 2020 guidance on the same topic, recommends that sponsors review eligibility criteria for unnecessary restrictions and consider consent language and participant communication materials for accessibility, with particular attention to characteristics including weight range, comorbid conditions, and patient populations with low representation [22].

Where AI and Automation Can Realistically Help

The operational challenges in obesity trial execution map to categories where AI-assisted tools have produced credible evidence of value in clinical research: site identification, patient pre-screening, and regulatory document generation across multi-year, multi-site programs.

Site selection for obesity trials requires matching investigator experience and site infrastructure characteristics against protocol-specific needs, including DEXA availability, endocrinology subspecialty access, and prior enrollment history in metabolic disease studies. AI-driven site intelligence tools that aggregate ClinicalTrials.gov registry data, prior enrollment performance, investigator publication records, and EHR-connected population data can surface site candidates that traditional feasibility questionnaire approaches would not identify. A Tufts CSDD impact report found that AI use in clinical trial activity yields time savings of approximately 18% [23]. Kitsa's KScout product is designed to support this kind of data-informed site selection, providing sponsors with enrollment potential estimates grounded in actual patient population characteristics rather than self-reported site estimates.

Patient pre-screening presents a similar opportunity. Obesity trials with layered eligibility criteria, including BMI ranges, comorbidity specifications, prior therapy restrictions, and cardiovascular risk stratification, require detailed chart review to identify candidates. Manual screening at high-volume sites is time-consuming and depends on coordinator experience. FHIR-connected pre-screening tools, such as KScreener, are designed to systematically apply inclusion and exclusion criteria against structured EHR data before coordinator review begins, reducing false referrals and accelerating identification of genuinely eligible patients. In therapeutic areas where the enrolled-to-screened ratio is under pressure from commercial therapy availability, that efficiency is operationally meaningful.

Regulatory document automation is relevant to obesity programs for a different reason. Multi-site cardiovascular outcomes trials generate substantial documentation across their full duration, including protocols, investigator brochures, DSURs, and eventually CSRs, all of which must maintain consistency across amendments and across the full length of the study. AI-assisted regulatory writing tools, such as KScribe, are designed to support cross-document consistency and reduce the manual burden associated with amendment-driven revisions, while keeping protocol language aligned with FDA and ICH guidance requirements.

None of these tools address the underlying shortage of experienced investigative sites or resolve the enrolled patient pool constraints created by commercial GLP-1 availability. In every case, the AI layer is designed to support human review rather than substitute for investigator judgment or automated eligibility decisions. But they do reduce friction at each step where human capacity is currently being consumed by tasks that do not require expert judgment.

Key Takeaways

• The GLP-1 obesity pipeline expanded at a rate with no recent historical parallel: 2024 was a record year for obesity clinical trials, and 2025 was tracking to surpass it, with innovator drug trials comprising 71% of newly announced programs according to GlobalData's clinical trials database.
• Site activation rates in North America dropped to approximately 62% by 2024, according to Tufts CSDD findings, meaning nearly 40% of engaged sites fail to activate. In obesity research, where per-protocol infrastructure demands are higher than in most therapeutic areas, that gap creates real operational consequences for sponsors that do not account for it in feasibility planning.
• Commercial GLP-1 availability is likely changing the accessible trial population. Patients achieving meaningful weight loss on semaglutide or tirzepatide have reduced incentive to join trials requiring washout or placebo assignment, which may narrow the accessible research population despite obesity's high global prevalence.
• Diversity gaps in obesity clinical trials remain wide. Black and Hispanic patients, who carry disproportionate obesity burden, are persistently underrepresented in phase 3 and phase 4 obesity studies. FDORA created a statutory obligation for sponsors to submit Diversity Action Plans; FDA's June 2024 draft guidance describing the format of those plans was removed from FDA's guidance database following executive branch directives issued in January 2025, though the Federal Register notice remains publicly accessible.
• ICH E6(R3), adopted January 6, 2025 and published by FDA in the Federal Register on September 9, 2025, introduces a risk-proportionate GCP framework directly relevant to obesity trial operations. Its practical application across large, multi-site cardiovascular outcomes programs requires quality-by-design investment at the protocol planning stage, not at activation.
• AI-assisted site selection, patient pre-screening, and regulatory document automation address specific operational bottlenecks in obesity trial execution. They do not solve the underlying infrastructure gap but can improve efficiency at each friction point where human capacity is being consumed by workflow rather than judgment.