Going beyond total skeletal muscle area at L3 (tSMA) to accurately assess pancreatic cancer cachexia

Introduction

Pancreatic cancer cachexia (PCC) is characterized by progressive wasting of skeletal muscle and adipose tissue. PCC has been recognized as a systemic disorder in which tumor-derived cachexia-inducing factors result in systemic metabolic and inflammatory disruptions (1). The systemic disturbances contribute to disease progression and complicate clinical management. As a result, measurement of PCC is the subject of active investigation to provide important insights into the dismal clinical outcomes associated with pancreatic cancer.

The recently published study by Davis et al. (2) aimed to determine the impact of edema on assessment of total skeletal muscle area (tSMA) at the third lumbar vertebra (L3) in patients with pancreatic ductal adenocarcinoma (PDAC). Unlike tSMA assessments, which can be confounded by edema within skeletal muscle and subcutaneous fat, psoas muscles, which are used in assessment of psoas muscle skeletal muscle area (pSMA), are more distant from the subcutaneous compartment and may be less prone to the confounding effects of edema. Using a multi-institutional, prospective cohort, they describe the confounding impact of edema in accurately measuring muscle mass using tSMA and reveal that assessing changes in psoas muscle (pSMA) at L3 is independent of edema status and may be a more accurate surrogate of total body skeletal muscle mass and perhaps more prognostic compared to tSMA in patients with PDAC. Should this data be used as justification to replace tSMA with pSMA in this patient population? Additional confirmatory studies are certainly necessary prior to routine use of pSMA, but this study highlights yet another limitation of tSMA-based analysis and more broadly the heterogeneity by which muscle mass and cancer-cachexia assessments are being utilized in the field. In this commentary, we aim to contextualize the study by Davis et al. and highlight the important potential therapeutic implications of accurately assessing cancer-associated cachexia and sarcopenia.

Defining the problem: clinical impact and assessment of cancer cachexia

Although cancer cachexia is a well-established, clinically meaningful syndrome, a uniform definition of this syndrome is yet to be established. Most studies utilize a combination of percent weight loss over time in combination with targeted assessment of body composition and an instrument to assess functionality/quality of life. Whole body fat-free mass (FFM) is a body composition estimate, and is comprised of skeletal muscle, viscera, skin, and bone. It is a measure utilized to describe metabolic health. In cancer, it is considered a critical prognostic tool for assessing treatment toxicity and survival. Prior tools have proven inadequate to identify those at risk for adverse health outcomes. Frequently used measures, include body mass index (BMI) and dual X-ray absorptiometry (DXA). BMI is a calculated measure involving height and weight which limits characterization of body composition (3). DXA, which measures whole-body distribution of lean body mass is a quick and relatively accurate modality to assess body composition, but it is not routinely used as a longitudinal imaging modality in cancer patients, which limits its use to assess cancer associated cachexia (4). Since abdominopelvic computed tomography (CT) imaging is routinely used in cancer care, assessment of skeletal muscle area (SMA), visceral/subcutaneous adipose tissue (VAT/SAT), and skeletal muscle density (SMD) at L3 has become a gold standard for body composition assessment and thus provides an opportunity for longitudinal assessment of cancer associated cachexia with better accuracy than standard DXA (5). Trials have demonstrated a correlation between low abdominal muscle cross-sectional area measurements and worse overall survival, cancer-specific survival, and disease-free survival in patients with a cancer diagnosis (6). Despite associations between changes in body composition and treatment-related toxicities across a wide range of malignancies (7), it has not yet been established if absolute muscle mass or the relative proportion of muscle mass compared with total adiposity is the superior predictor of clinical outcomes and limits clinical applicability. Therefore, research is ongoing to optimize our understanding of radiologic artificial intelligence (AI) measures so that it can help characterize prognosis and guide treatment plans for newly diagnosed patients.

Is pSMA a more reliable assessment of total body skeletal muscle mass than tSMA?

Edema becomes a clinical issue for ~50% of patients with PDAC (8). The authors present the argument that edema may affect CT findings used to estimate tSMA and therefore limit our understanding or assessment of muscle changes during treatment. Therefore, the authors conducted a body composition analysis of 95 patients with PDAC enrolled in the Florida Pancreas Collaborative (FPC) in order to evaluate the impact of edema on tSMA. The primary objectives included comparison of tSMA and pSMA calculations in the presence or absence of edema, as well as to explore the utility of pSMA in body composition assessment.

The FPC cohort study is a multi-institutional prospective observational cohort study (9). The study enrolled 500 individuals, 18 years of age or older, who presented with a pancreatic mass to a participating site between 2018 and 2021. Patients were eligible if a PDAC diagnosis was made after presentation, and CT were available from the time of diagnosis and ~6 months later. Eligibility applied to 95 of the 500 individuals enrolled.

Body composition assessment involved assessment by CT images in the axial view in accordance with standard operating procedures of the FPC. Edema was assessed using a semiquantitative approach utilizing Hounsfield unit (HU) density, since prior analyses confirmed edema has a mean HU density of 0. Patients were classified as edematous or non-edematous based on this assessment. Muscle area was assessed using sliceOmatic v5rev16 with the ABCS plug-in for tSMA and manual segmentation for pSMA by trained raters blinded to edema status.

Calculations of percent changes and monthly rate of change of tSMA and pSMA between diagnosis and follow-up were calculated. The authors examined the following from these measurements: (I) association with edema status; (II) associations with clinically relevant covariates using the Fisher exact test.

The authors demonstrated that in the overall study population, there were no differences in tSMA between diagnosis and follow-up (P=0.20). However, tSMA was lower at follow-up in patients without edema (P=0.02) and there was no change in patients with edema. Whereas pSMA was significantly lower at follow-up (P<0.001), and consistently lower at 6-month follow-up regardless of edema status. The authors surmised that their results support the use of pSMA as a surrogate for tSMA, and specifically in patients without edema.

Strengths and weaknesses of the study

Although a small cohort, a major strength of this study was the use of a prospectively annotated data set of 500 patients from multiple institutions as well as the novelty of assessing the confounding impact of edema on assessment of muscle mass. A weakness of this study stems from the automated assessment of edema. AI in image analysis is constrained by a heavy reliance on large, high-quality, and unbiased data. Key limitations of AI include a lack of contextual understanding and reasoning, high susceptibility to noise or distortion, and inability to handle complex, nuanced, or abstract scenes. In this study, the inability to reliably differentiate edema in the muscle and the subcutaneous space limited the analysis, particularly accurate assessment of tSMA. Since assessment of edema was based upon opacifications in subcutaneous fat, some edematous patients may have been misclassified as non-edematous, particularly those patients who develop muscle edema for which CT assessment is limited. Therefore, in an attempt to assess the impact of edema on body composition measures, the authors managed to highlight the limitations of tSMA as a body composition tool and posit that pSMA may be a superior modality to assess FFM given that it is less influenced by edema If validated in successor studies, utilizing pSMA to assess PCC will theoretically prevent the need for manual assessment of edema in these patients. However, it is important to note that pSMA is still prone to the confounding effects of interstitial fluid within the muscle compartment as well as myosteatosis, both of which can overestimate FFM and represents important limitations of pSMA. Another important limitation of this study is small number of patients with imaging available at baseline and at the 6-month follow up timepoint. The small number influenced the ability to balance demographic and treatment history characteristics, therefore potentially skewing clinical outcomes. The heterogeneity of the patient population likely affected accurate assessment of clinical outcomes. For example, the observation that tSMA increased in patients prior to accounting for edema status is seemingly discordant with prior literature due in part to the fact that previous studies focused on patients with advanced disease and the current study included a significant cohort with earlier stage disease (10,11). Another important limitation in this study is that it does not correlate pSMA with FFM. The correlation between tSMA and FFM is well-documented, however, the correlation between pSMA and FFM is less clear and thus it remains unknown whether pSMA represents a suitable surrogate for FFM (12,13). However, given the concordance between tSMA and pSMA assessments as well as the fact that pSMA may be a more accurate assessment tool in patients with edema, further studies evaluating the clinical utility of pSMA assessment are warranted.

Future directions and clinical implications

Despite its limitations, the study by Davis et al. provides important insights into the limitations of currently accepted tools to assess cancer-cachexia and propose body composition analysis using pSMA as another, perhaps more reliable tool in patients with PDAC.If performance of pSMA is validated in successor studies, this would be an important tool in studying and assessing FFM and PCC in clinical scenarios involving edematous patients. Currently available tools to assess FFM are in part dependent on manual segmentation tools and are thus are subject to limitations such as inter/intra-observer variability and image quality, However, with the rapid advancement of AI, one can easily predict that automated segmentation tools will become more accurate, particularly with determining skeletal muscle vs. edema, and thus improve the accuracy and prognostic capability of tSMA assessments (14). Indeed, tools such as EfficientNetV2-XL have been shown to have superior sensitivity and specificity of assessing sarcopenia via CT compared to conventional models (15). However, these models are still under development and are being refined for broader applicability. These automated segmentation tools will undoubtably have some degree of variability, but improvements in AI algorithms will likely drastically reduce variability compared to currently available manual segmentation tools, including ones used in the current study. Since automated segmentation tools are still being refined evaluation of additional tools, such as pSMA, are necessary to improve the diagnostic and prognostic capability of body composition analysis, particularly in patients with PDAC.

Accurate diagnosis and longitudinal assessment of cancer-cachexia is of paramount importance for prognostic characterization; there are also ongoing efforts to therapeutically target and reverse cancer-cachexia to improve patient outcomes (Table 1). Growth differentiation factor 15 (GDF-15) is a cytokine that has recently been associated with development of cachexia and pre-clinical data suggest inhibition of GDF-15 can reverse cachexia. A randomized, phase II study of the potent GDF-15 inhibitor ponsegromab in patients with cancer cachexia found that at 12 weeks, patients treated with ponsegromab (400 mg group) had on average a 2.81 kg weight gain compared patients treated with placebo as well as improved activity levels (16). RIVER-mPDAC is a phase IIB/III study of ponsegromab in patients with metastatic PDAC and is currently ongoing.

The ghrelin pathway is another promising target in cancer cachexia. When activated, the ghrelin receptor has anabolic and appetite-stimulating effects and anamorelin is a high-affinity, selective ghrelin-receptor agonist. The activity of anamorelin in modulating cancer cachexia was evaluated in two international phase III trials in cachectic patients with non-small cell lung cancer (NSCLC). Between the two studies, anamorelin resulted in a median increased lean body mass of 0.65–0.99 kg (P<0.0001) as assessed by DXA. As a result of these studies, anamorelin has gained regulatory approval in Japan and represents the first approval worldwide of a drug targeting cancer cachexia.

Imaging-based body composition analysis with advanced AI algorithms to facilitate rapid and accurate automated segmentation is the future of body composition analysis. Multi-modal imaging and functional tools incorporated into cancer cachexia assessments will be critical to help develop novel systemic therapies to target and reverse the sequela of cancer cachexia.

Conclusions

The study by Davis et al. provides important contributions to the field of imaging-based assessment of cancer cachexia in patients with PDAC. Further evaluation of pSMA and other imaging-based cachexia assessments are warranted to not only help guide prognosis, but also to aid in the development of novel agents to prevent and treat cancer cachexia.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Journal of Gastrointestinal Oncology. The article has undergone external peer review.

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0221/prf

Funding: None.

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0221/coif). R.S. reports AstraZeneca funding to institution (clinical trial); honoraria for a lecture by Grupo Oncoclinicas; and travel reimbursement (flight, hotel) from DAVA Oncology. K.C.P. reports one-time advisory board member for Exelixis 3/2026, Lily 4/2025 and Novartis 2/2024. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Kordes M, Larsson L, Engstrand L, et al. Pancreatic cancer cachexia: three dimensions of a complex syndrome. Br J Cancer 2021;124:1623-36. [Crossref] [PubMed]
2. Davis EW, Park MA, Basinski TL, et al. The Impact of Edema on Skeletal Muscle Changes among Patients with Pancreatic Ductal Adenocarcinoma. Cancer Epidemiol Biomarkers Prev 2025;34:1609-17. [Crossref] [PubMed]
3. Strulov Shachar S, Williams GR. The Obesity Paradox in Cancer-Moving Beyond BMI. Cancer Epidemiol Biomarkers Prev 2017;26:13-6. [Crossref] [PubMed]
4. Bates DDB, Pickhardt PJ. CT-Derived Body Composition Assessment as a Prognostic Tool in Oncologic Patients: From Opportunistic Research to Artificial Intelligence-Based Clinical Implementation. AJR Am J Roentgenol 2022;219:671-80. [Crossref] [PubMed]
5. Mourtzakis M, Prado CM, Lieffers JR, et al. A practical and precise approach to quantification of body composition in cancer patients using computed tomography images acquired during routine care. Appl Physiol Nutr Metab 2008;33:997-1006. [Crossref] [PubMed]
6. Shachar SS, Williams GR, Muss HB, et al. Prognostic value of sarcopenia in adults with solid tumours: A meta-analysis and systematic review. Eur J Cancer 2016;57:58-67. [Crossref] [PubMed]
7. Bruggeman AR, Kamal AH, LeBlanc TW, et al. Cancer Cachexia: Beyond Weight Loss. J Oncol Pract 2016;12:1163-71. [Crossref] [PubMed]
8. Saito Y, Takekuma Y, Komatsu Y, et al. Impact of baseline hypoalbuminemia on peripheral edema induced by gemcitabine and nab-paclitaxel for pancreatic cancer treatment. Support Care Cancer 2025;33:938. [Crossref] [PubMed]
9. Permuth JB, Dezsi KB, Vyas S, et al. The Florida Pancreas Collaborative Next-Generation Biobank: Infrastructure to Reduce Disparities and Improve Survival for a Diverse Cohort of Patients with Pancreatic Cancer. Cancers (Basel) 2021;13:809. [Crossref] [PubMed]
10. Davis MP, Bader N, Basting J, et al. Are Muscle and Fat Loss Predictive of Clinical Events in Pancreatic Cancer? The Importance of Precision Metrics. J Pain Symptom Manage 2025;69:141-51.
11. Babic A, Rosenthal MH, Bamlet WR, et al. Postdiagnosis Loss of Skeletal Muscle, but Not Adipose Tissue, Is Associated with Shorter Survival of Patients with Advanced Pancreatic Cancer. Cancer Epidemiol Biomarkers Prev 2019;28:2062-9. [Crossref] [PubMed]
12. Van den Broeck J, Sealy MJ, Brussaard C, et al. The correlation of muscle quantity and quality between all vertebra levels and level L3, measured with CT: An exploratory study. Front Nutr 2023;10:1148809. [Crossref] [PubMed]
13. Rollins KE, Gopinath A, Awwad A, et al. Computed tomography-based psoas skeletal muscle area and radiodensity are poor sentinels for whole L3 skeletal muscle values. Clin Nutr 2020;39:2227-32. [Crossref] [PubMed]
14. Lee H, Troschel FM, Tajmir S, et al. Pixel-Level Deep Segmentation: Artificial Intelligence Quantifies Muscle on Computed Tomography for Body Morphometric Analysis. J Digit Imaging 2017;30:487-98. [Crossref] [PubMed]
15. Onishi S, Kuwahara T, Tajika M, et al. Artificial intelligence for body composition assessment focusing on sarcopenia. Sci Rep 2025;15:1324. [Crossref] [PubMed]
16. Groarke JD, Crawford J, Collins SM, et al. Ponsegromab for the Treatment of Cancer Cachexia. N Engl J Med 2024;391:2291-303. [Crossref] [PubMed]