Unsupervised machine learning identified two body composition profiles (High Muscle and Low Muscle) in 189 gastrointestinal cancer patients; the Low Muscle Profile was associated with nutritional risk only in men in an exploratory analysis, while nutritional risk, but not body composition profile, was associated with poorer quality of life.
More than one-third of gastrointestinal cancer patients in the cohort were at nutritional risk according to NRS-2002.
Principal component analysis and k-means clustering identified two body composition profiles: High Muscle Profile (HMP) and Low Muscle Profile (LMP).
No statistically significant association between body composition profile and nutritional risk was observed in the overall cohort.
In gender-stratified multivariable analysis, the Low Muscle Profile was associated with higher odds of nutritional risk among male patients.
Nutritional risk, but not body composition profile, was associated with poorer health-related quality of life (EORTC QLQ-C30) Summary Score.
Nutritional risk was associated with a higher Symptom Score, whereas body composition profiles were not.
The authors conclude results are preliminary and need confirmation in larger, adequately powered longitudinal cohorts, and that added value of BCPs beyond established screening remains to be determined.
This study looked at 189 people with gastrointestinal cancers (stomach, pancreas, rectum, and liver) to see whether grouping patients by their body makeup, such as muscle and fat amounts, could help identify who is at risk of malnutrition. Using a computer-based grouping method, the researchers found two patterns: one with higher muscle and one with lower muscle. About 36.5% of patients were at nutritional risk based on a standard screening tool. Overall, the muscle-based groups did not clearly line up with nutritional risk. Only in men did the low-muscle group appear more likely to be at nutritional risk, and the authors stress this was an exploratory result with considerable uncertainty. In contrast, patients who were at nutritional risk reported worse overall quality of life and more symptoms, while the body composition groups did not predict quality of life. This research suggests that standard nutritional screening captured quality-of-life differences better than these body composition groupings in this sample. Because the study was a single snapshot in time and relatively small, the findings are preliminary, and larger studies that follow patients over time are needed to see whether body composition profiles add anything useful to routine screening.
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Upload Your LabsLuongo A, Di Carlo F, Coluccia S, Rocco P, Russo N, Racca E, et al.. (2026). Exploring Body Composition Profiles in Patients with Gastrointestinal Cancers: A Cross-Sectional Analysis of the NUTRISCREEN Project.. Current oncology (Toronto, Ont.). https://doi.org/10.3390/curroncol33100591