Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Oct 14;8(1):611.
doi: 10.1038/s41746-025-02016-z.

Leveraging Sarcopenia index by automated CT body composition analysis for pan cancer prognostic stratification

Katarzyna Borys  1   2 , Johannes Haubold  3   4 , Julius Keyl  4   5 , Maria A Bali  6 , Riccardo De Angelis  6 , Kévin Brou Boni  7   8 , Nicolas Coquelet  6 , Judith Kohnke  3   4 , Giulia Baldini  3   4 , Lennard Kroll  9 , Sara Schramm  10 , Andreas Stang  10 , Eugen Malamutmann  11 , Jens Kleesiek  4 , Moon Kim  3   4 , Stefan Kasper  5 , Jens T Siveke  5   12   13 , Marcel Wiesweg  5 , Anja Merkel-Jens  10 , Benedikt M Schaarschmidt  3 , Viktor Gruenwald  5 , Sebastian Bauer  5 , Arzu Oezcelik  11 , Servet Bölükbas  14 , Ken Herrmann  15   16 , Rainer Kimmig  17 , Stephan Lang  18 , Jürgen Treckmann  19 , Martin Stuschke  16   20 , Boris Hadaschik  16   21 , Lale Umutlu  3 , Michael Forsting  3   4 , Dirk Schadendorf  16   22 , Christoph M Friedrich  10   23 , Martin Schuler  5   16 , René Hosch #  3   4 , Felix Nensa #  3   4
Affiliations

Leveraging Sarcopenia index by automated CT body composition analysis for pan cancer prognostic stratification

Katarzyna Borys et al. NPJ Digit Med. .

Abstract

This study evaluates the CT-based volumetric sarcopenia index (SI) as a baseline prognostic factor for overall survival (OS) in 10,340 solid tumor patients (40% female). Automated body composition analysis was applied to internal baseline abdomen CTs and to thorax CTs. SI's prognostic value was assessed using multivariable Cox proportional hazards regression, accelerated failure time models, and gradient-boosted machine learning. External validation included 439 patients (40% female). Higher SI was associated with prolonged OS in the internal abdomen (HR 0.56, 95% CI 0.52-0.59; P < 0.001) and thorax cohorts (HR 0.40, 95% CI 0.37-0.43; P < 0.001), as well as in the external validation cohort (HR 0.56, 95% CI 0.41-0.79; P < 0.001). Machine learning models identified SI as the most important factor in survival prediction. Our results demonstrate SI's potential as a fully automated body composition feature for standard oncologic workflows.

PubMed Disclaimer

Conflict of interest statement

Competing interests: K.H. reports personal fees from Bayer, personal fees and other from Sofie Biosciences, personal fees from SIRTEX, non-financial support from ABX, personal fees from Adacap, personal fees from Curium, personal fees from Endocyte, grants and personal fees from BTG, personal fees from IPSEN, personal fees from Siemens Healthineers, personal fees from GE Healthcare, personal fees from Amgen, personal fees from Novartis, personal fees from ymabs, personal fees from Aktis Oncology, personal fees from Theragnostics, personal fees from Pharma15, personal fees from Debiopharm, personal fees from AstraZeneca, personal fees from Janssen, outside the submitted work. MSchuler reports consultant fees from Amgen, AstraZeneca, BIOCAD, Blueprint Medicines, Boehringer Ingelheim, Bristol Myers Squibb, GlaxoSmithKline, Janssen, Merck Serono, Novartis, Roche, Sanofi, Takeda; Honoraries for CME presentations from Amgen, Boehringer Ingelheim, Bristol-Myers Squibb, Janssen, Novartis, Roche, Sanofi; Research funding to institution from AstraZeneca, Bristol Myers Squibb. DS reports grants from Novartis, Amgen, MSD, Roche, BMS and consulting fees from Nektar, Philogen, lnFlarX, Neracare, Merck Sharp & Dohme, Novartis, Bristol Myers Squibb, Pfizer, Pierre Fabre, Replimune, Amgen, SunPharma, Daiichi Sanyo, AstraZeneca, IQVIA, LabCorp, BioAlta, and Sanofi; Honoraria from Merck Sharp & Dohme, Novartis, Bristol Myers Squibb, Merck-Serono, Pierre Fabre, Replimune, SunPharma, LabCorp and Sanofi; Support for attending meetings from Pierre Fabre, BMS, MSD; Leadership or fiduciary role in other board (all unpaid) from EORTC, WTZ, Deutsche Krebshilfe, University Alliance Ruhr. MStuschke reports grants from AstraZeneca, Bristol-Myers-Squibb, Sanofi-Aventis, Janssen-Cilag; Participation on a data safety monitoring or advisory board from Sanofi-Aventis, Bristol-Myers-Squibb, Janssen-Cilag, AstraZeneca, Medupdate GmbH, Bristol-Myers-Squibb. BMS received grants from PharmaCept, Else Kröner-Fresenius-Foundation. JH received support from the German Research Foundation as a member of the clinical scientist program. JK received support from the German Cancer Consortium Joint Funding (DKTK JF RAMTAS). KBB received grants from Elekta. JS received grants from Bristol-Myers-Squibb, Roche/Genentech, Eisbach Bio, Abalos Therapeutics; Consulting fees from Celgene, AstraZeneca, Immunocore, Bayer, Roche, Novartis, SERVIER, MSD Sharpe Dome; Honoraria for lectures from Celgene, Astrazeneca, Immunocore, Bayer, Roche, Novartis, SERVIER, MSD Sharpe Dome; Support for attending meetings from SERVIER; Support for attending meetings from Novartis, Immunocore, Bristol-Myers-Squibb, AstraZeneca; Stock options from Pharma15. MW received grants from Takeda; Consulting fees from GlaxoSmithKline and Novartis; Honoraria from Amgen, AstraZeneca, Roche, and Takeda; Support for attending Meetings from Janssen and GlaxoSmithKline; Participation on a data safety monitoring or advisory board from Amgen, AstraZeneca, Daiichi Sankyo, GlaxoSmithKline, Janssen, Novartis, Pfizer, Roche, Takeda. SK received grants from BMS, Roche, Lilly; Consulting fees from BMS, Lilly, Amgen, Merck Serono, MSD, Novartis, Onkowissen.de, Incyte; Honoraria from BMS, Amgen, MSD, Merck Serono, Lilly, Servier; Support for attending Meetings from Amgen, Pierre Fabre, BMS, Roche, Lilly; Participation on a data safety monitoring or advisory board from Novartis; Leadership or fiduciary role in other board society from DGHO, DKG-AIO; VG received support from Wilhelm-Sander-Foundation; Grants from Pfizer, AstraZeneca, BMS, Ipsen; Consulting fees from AstraZeneca, Apogepha, Astellas, BMS, Novartis, Apogepha, EISAI, MSD, MerckSerono, Roche, EUSAPharm, Nanobiotix, Debiopharm, Oncorena, PCI Biotech; Honoraria from AstraZeneca, Astellas, BMS, Novartis, Ipsen, EISAI, MSD, MerckSerono, Roche, EUSAPharm, Janssen, ONO Pharmaceutical; Support for attending meetings from Merck, Pfizer, Janssen; Leadership or fiduciary role in other board society from German medical oncology working group for cancer treatment, German Cancer Society, German Society of Hematology and Oncology, ESMO, and ASCO; BH reports grants from German Research Foundation; Royalties from Uromed; Consulting fees from AAA/Novartis, ABX, BMS, Bayer, Janssen, Lightpoint, MSD, Pfizer; Honoraria from Astellas, AstraZeneca, Bayer, Janssen, Pfizer; Support for attending meetings from Astellas, Bayer, Janssen; Participation on a data safety monitoring or advisory board from Janssen; Leadership or fiduciary role in other board society from German Cancer Society. All remaining authors declare no conflict of interest.

Figures

Fig. 1
Fig. 1. Kaplan-Meier survival curves and Restricted Mean Survival Time (RMST) differences for body composition features.
Survival curves compare the lowest tertile (T1) with the combined upper tertiles (T2-T3) for the Sarcopenia Index (SI), Skeletal Muscle Index at the L3 vertebra (L3 SMI), and Body Mass Index (BMI) in a the internal abdomen cohort and b the internal thorax cohort. c Shows the corresponding differences in RMST between T1 and T2-T3 for all features across both CT regions.
Fig. 2
Fig. 2. Kaplan-Meier survival curves by Sarcopenia Index (SI) tertiles, stratified by sex.
Survival curves compare the lowest tertile (T1) with the combined upper tertiles (T2-T3) of the SI across three cohorts: a the internal abdomen cohort, b the internal thorax cohort, and c the external cohort.
Fig. 3
Fig. 3. Machine learning results for combined metastatic status groups in the internal cohorts.
Kaplan-Meier survival curves compared the lowest tertile (T1) with the combined upper tertiles (T2-T3) of survival times predicted by gradient-boosted survival trees. Results are shown for two models: a a model trained on the internal abdomen cohort, including both non-metastatic (M0) and metastatic (M1) patients, and b a model trained on the internal thorax cohort, also including both M0 and M1 patients. External validation was performed only for the abdomen-based model. Additionally, impurity-based feature importances are presented.
Fig. 4
Fig. 4. Flowchart of cohort selection.
From an initial pool of 57,620 patients, 11,086 remained after excluding those without eligible CT scans and non-target tumor types. Further exclusion of 746 patients based on mismatched series metadata and scanned body regions yielded a final abdominal CT cohort of 10,340 patients. Additionally, a subcohort of 8760 patients with thorax CT scans was identified. Patients were classified according to their metastatic status into M0 (non-metastatic), M1 (presence of distant metastases), and Mx (staging information unavailable within ±60 days of primary cancer diagnosis).
Fig. 5
Fig. 5. Schematic overview of the study design and analysis.
The internal cohorts include baseline CT scans of both the abdomen and thorax, while the external validation cohort comprises abdominal CT scans only. All cohorts underwent automated body composition analysis (BCA) to extract the Sarcopenia Index (SI). Subsequent statistical and machine learning analyses were performed to evaluate the prognostic value of SI for overall survival.
See this image and copyright information in PMC

References

    1. Tolonen, A. et al. Methodology, clinical applications, and future directions of body composition analysis using computed tomography (CT) images: A review. Eur. J. Radiol.145, 109943 (2021). - PubMed
    1. Gao, M. et al. Associations between body-mass index and COVID-19 severity in 6·9 million people in England: a prospective, community-based, cohort study. Lancet Diab Endocrinol.9, 350–359 (2021). - PMC - PubMed
    1. Del Grande, M. et al. Computed tomography–based body composition in patients with ovarian cancer: Association with chemotoxicity and prognosis. Front. Oncol.11, 718815 (2021). - PMC - PubMed
    1. Shirdel, M. et al. Body composition measured by computed tomography is associated with colorectal cancer survival, also in early-stage disease. Acta Oncol.59, 799–808 (2020). - PubMed
    1. Black, D. et al. Prognostic value of computed tomography: Measured parameters of body composition in primary operable gastrointestinal cancers. Ann. Surg. Oncol.24, 2241–2251 (2017). - PMC - PubMed

LinkOut - more resources