Beyond Weight Loss: The Role of DXA in the GLP-1 Era

GLP-1 and related therapies have redefined what is possible in obesity treatment, delivering weight loss that was previously unattainable with pharmacologic intervention. As these therapies scale across broader populations, the central question is evolving from how much weight is lost to what type of weight is lost.

That shift places body composition at the center of clinical, regulatory, and commercial decision-making. Dual-energy X-ray absorptiometry (DXA) is the leading modality for measuring changes in fat mass, lean mass, and bone mineral density, enabling a more precise understanding of therapeutic impact. Yet despite its growing adoption in weight-loss trials, DXA remains operationally fragile. Variability in acquisition, analysis, and quality control can undermine data integrity, distort longitudinal outcomes, and introduce risk at the regulatory stage. This paper examines the intersection of GLP-1 innovation and measurement science, arguing that while DXA is widely accessible, its value depends entirely on execution.

The GLP-1 landscape: why measurement now matters

GLP-1 receptor agonists and dual agonists have rapidly become foundational therapies in obesity and metabolic disease. Current agents routinely deliver 10–25%+ reductions in total body weight, setting a new clinical benchmark and expanding treatment eligibility across diverse patient populations. At the same time, the pipeline is evolving toward preserving lean mass during weight loss, combining mechanisms to enhance metabolic outcomes, and supporting long-term maintenance of weight reduction, a broader strategic shift from short-term weight loss to sustained metabolic health.

As efficacy improves, the limits of traditional endpoints become more apparent. Total body weight, while simple and intuitive, does not distinguish between fat loss and loss of lean tissue. That distinction is increasingly critical: loss of muscle mass is associated with functional decline, increased morbidity, and reduced long-term benefit, particularly in older or metabolically vulnerable populations. The result is a widening measurement gap. Weight alone is no longer sufficient to characterize therapeutic impact, and body composition must fill that gap.

Why DXA is central in the GLP-1 era

DXA is uniquely positioned to address this need, providing quantitative assessment of fat mass, lean mass, bone mineral density, and the regional distribution of tissue. These capabilities allow for a more complete evaluation of both efficacy and safety, enabling differentiation between therapies that may appear similar when assessed by weight alone. Its adoption is also driven by practicality: DXA is widely available across clinical sites, relatively low cost, and associated with minimal radiation exposure, making it suitable for large, multi-site trials and increasingly relevant for real-world evidence generation.

However, the same characteristics that make DXA accessible also introduce risk to the integrity of trial results. The installed base of scanners is heterogeneous, protocols vary, and operator technique is inconsistent. Without rigorous standardization, variability across sites can exceed the underlying biological signal, compromising interpretability. Body composition is therefore not only a valuable endpoint but a differentiating one: sponsors who can generate credible, reproducible DXA data will be better positioned to support regulatory submissions, demonstrate safety, and distinguish their therapies in a competitive market.

Operational reality: why "doing DXA right" is hard

While DXA is often perceived as routine imaging, its use in body composition analysis requires a higher level of precision and control. Several well-established sources of variability can materially affect results: inconsistent patient positioning, differences in scan modes and acquisition settings, calibration variability across devices, drift in region-of-interest definitions, and errors in file handling and visit mapping. These factors can lead to incorrect estimates of lean and fat mass, misinterpretation of longitudinal trends, and conflicting data across sites.

Structural limitations of equipment matter too. Many legacy DXA systems were not designed for higher-BMI populations; constraints such as table width and field of view can result in incomplete body capture, requiring adaptations such as hemi-scanning. Patient movement is a further challenge. Where DXA is treated as a low-risk, routine procedure and quality control is minimal, movement artifacts may go undetected, introducing substantial errors that are often systematic rather than random and therefore difficult to identify without structured oversight.

Real-world longitudinal evidence

Analysis of multi-site longitudinal datasets highlights the practical consequences of these challenges. In one representative dataset spanning 54 participants across five sites over multiple timepoints (W0–W26), expected trends in weight and total mass reduction were observed. Deeper analysis, however, revealed missing or incomplete data at key timepoints, variability in acquisition and analysis across sites, and inconsistencies in regional measurements, particularly in appendicular lean mass. Even when study design is robust, inconsistent execution can compromise data quality; without standardization and quality control, longitudinal interpretation becomes unreliable.

A practical multi-site DXA framework

Effective implementation of DXA in multi-site trials requires a coordinated operational framework. Four components work together to produce defensible, reproducible endpoints suitable for regulatory and publication purposes.

Credibility and differentiation

DXA is often perceived as a commoditized capability. In practice, successful implementation requires specialized expertise: experience in multi-site standardization, a deep understanding of body composition measurement, and validated approaches to calibration and quality control. Advanced tools and methodologies, including proprietary phantoms such as the Bona Fide Phantom (BFP) and Body Composition Phantom (BCP), further support accurate and reproducible measurement. Ultimately, the value of DXA is determined not by access to scanners, but by the rigor of execution and interpretation.

Strategic implications for sponsors

As GLP-1 therapies expand into broader and more complex use cases, including older populations, combination regimens, and long-term maintenance, body composition will play an increasingly central role in defining clinical value. High-quality DXA data supports differentiated labeling strategies, more robust safety narratives, and credible real-world evidence generation. Conversely, poor-quality data introduces risk across all three domains.

Conclusion

GLP-1 therapies have fundamentally changed the evaluation of obesity treatments. The relevant question is no longer how much weight is lost, but what tissue is lost. DXA is uniquely capable of answering it, but its effectiveness depends entirely on execution. Programs that approach DXA as routine imaging risk generating inconsistent or misleading data; those that apply disciplined, standardized methodologies will produce insights that are clinically meaningful, regulatorily defensible, and commercially differentiating. In the GLP-1 era, the challenge is not access to measurement, but the ability to measure correctly.