Integration of sarcopenia screening into radiotherapy planning: Validation of a time-efficient SMI measurement method using MIM software in prostate cancer.
Sarcopenia (SP) affects 43.8% of men with prostate cancer (PCa), correlating with higher all-cause mortality and reduced quality of life. Systematic SP screening could reduce treatment complications and improve prognosis. Although several validated software tools exist for CT-based skeletal muscle assessment, their implementation in routine clinical practice remains limited due to time constraints, operator dependency, and lack of integration into radiotherapy planning systems.
We aimed to validate a new SP screening method using MIM® software, a program commonly used in radiotherapy planning, to facilitate its implementation in daily practice.
In 41 patients with PCa, Skeletal Muscle Index (SMI) was retrospectively calculated using MIM® and compared with the validated ImageJ® method. Agreement was assessed via Bland-Altman analysis, while sarcopenic status was classified using literature-based SMI thresholds. Measurement time and reproducibility were evaluated using Wilcoxon tests, Intraclass Correlation Coefficient (ICC), and Fleiss kappa coefficients (κ). Additionally, an automated MIM-workflow was developed and compared to the semi-manual MIM-method in terms of agreement and processing time.
The cohort had a median age of 71 years, with 66% overweight or obese and 22% classified as sarcopenic. The new method demonstrated high concordance with the validated method (mean SMI difference: -0.04 cm2/m2, p = 0.23), while significantly reducing calculation time (3.5 ± 1 min vs. 10 ± 1 min, p < 0.001). Reproducibility was excellent, with an ICC of 0.99 for SMI and perfect agreement on sarcopenic classification (κ = 1). The automated MIM-workflow further decreased processing time (median 3.2 min vs. 3.5 min, p < 0.001) but yielded slightly higher SMI values (mean difference: 2.3 cm2/m2, p < 0.0001).
MIM® software provides a reproducible, reliable, and time-efficient alternative for SP screening, allowing radiation-oncologists to implement this assessment immediately into routine radiotherapy planning. Although the automated MIM-workflow requires further calibration to correct for systematic SMI overestimation, it represents a promising step toward full automation and broader clinical integration of SP assessment within radiation-oncology.
We aimed to validate a new SP screening method using MIM® software, a program commonly used in radiotherapy planning, to facilitate its implementation in daily practice.
In 41 patients with PCa, Skeletal Muscle Index (SMI) was retrospectively calculated using MIM® and compared with the validated ImageJ® method. Agreement was assessed via Bland-Altman analysis, while sarcopenic status was classified using literature-based SMI thresholds. Measurement time and reproducibility were evaluated using Wilcoxon tests, Intraclass Correlation Coefficient (ICC), and Fleiss kappa coefficients (κ). Additionally, an automated MIM-workflow was developed and compared to the semi-manual MIM-method in terms of agreement and processing time.
The cohort had a median age of 71 years, with 66% overweight or obese and 22% classified as sarcopenic. The new method demonstrated high concordance with the validated method (mean SMI difference: -0.04 cm2/m2, p = 0.23), while significantly reducing calculation time (3.5 ± 1 min vs. 10 ± 1 min, p < 0.001). Reproducibility was excellent, with an ICC of 0.99 for SMI and perfect agreement on sarcopenic classification (κ = 1). The automated MIM-workflow further decreased processing time (median 3.2 min vs. 3.5 min, p < 0.001) but yielded slightly higher SMI values (mean difference: 2.3 cm2/m2, p < 0.0001).
MIM® software provides a reproducible, reliable, and time-efficient alternative for SP screening, allowing radiation-oncologists to implement this assessment immediately into routine radiotherapy planning. Although the automated MIM-workflow requires further calibration to correct for systematic SMI overestimation, it represents a promising step toward full automation and broader clinical integration of SP assessment within radiation-oncology.
Authors
De Bruyn De Bruyn, Coquelet Coquelet, Ferreira Ferreira, Michel Michel, Descamps Descamps, Van den Begin Van den Begin, Van Gestel Van Gestel, Preiser Preiser
View on Pubmed