Accepted for/Published in: JMIR Public Health and Surveillance
Date Submitted: Oct 11, 2022
Open Peer Review Period: Oct 11, 2022 - Oct 25, 2022
Date Accepted: Feb 23, 2023
(closed for review but you can still tweet)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Trajectories of Skeletal Muscle Mass Predicts Prognosis of Colorectal Cancer: Deep Learning-Based Automated Determination
ABSTRACT
Background:
Colorectal cancer (CRC) is the third most common cancer, and account for nearly 10% of cancer-related mortality worldwide. Low skeletal muscle profile at the time of diagnosis and muscle depletion are widely known as important prognostic factors in CRC patients. However, previous measurements of muscle and adiposity still comprise manual processes, which hinder the evaluation of continuous prognostic effects, leading to limited clinical application.
Objective:
This study aims to evaluate the prognostic impact of the initial status and trajectories of skeletal muscle, along with body mass index (BMI). A fully automatic UNet-based muscle measurement algorithm was implemented to reduce possible manual procedures at most.
Methods:
We analyzed 4,056 newly diagnosed CRC patients. Skeletal muscle mass index (SMVI) was defined as the muscle volume at 3rd-lumbar spine level, calculated by a deep learning algorithm, then divided by the square of height. Prognostic significances to overall survival (OS) of baseline of BMI and SMVI, and trajectories of BMI and SMVI were calculated and based on the results, predicted prognosis were demonstrated as heatmaps.
Results:
SMVI trajectories were categorized as decreased (20%), steady (50%), or increased (30%), while BMI trajectories were categorized as decreased (20%), steady (56%), or increased (24%). At the time at diagnosis, high SMVI (HR, 0.82; P=.04) showed favorable impact, while decreased SMVI (HR, 1.31; P=.001) was negative prognostic factor for OS. Similarly, baseline status and trajectory of BMI showed significant impact on OS. Considered simultaneously, BMI >30kg/m2 with low SMVI at the time of diagnosis resulted the highest mortality risk. We observed improved survival in patients with increased muscle mass without BMI loss compared to patients with steady muscle mass and BMI.
Conclusions:
Continuous trajectories of body and muscle mass are essential prognostic factors in CRC patients. A noninvasive automatic algorithm provides a unique opportunity to apply longitudinal evaluations of body composition to routine clinical practices and to understand how and when skeletal muscle impacts cancer prognosis.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.