<![CDATA[Background: Chemotherapy remains the major strategy in cancer treatment, although it increases the risk of malnutrition. Evidence identifying hierarchical clusters of malnutrition risk factors beyond chemotherapy remains limited despite their importance for improving nutritional risk stratification.Objectives: This study aims to identify the proximity of malnutrition-related factors among cancer patients, thereby generating phenotypic clusters of cancer patients at risk of malnutrition. Methods: A cross-sectional study was conducted among 148 cancer patients undergoing chemotherapy in Dr. Sardjito Hospital, Yogyakarta. Malnutrition status, body composition, and anthropometric indicators were assessed using Patient Generated-Subjective Global Assessment (PG-SGA), Bioimpedance Analysis (BIA), and standardized measurements, while dietary intake was collected using a semi-quantitative food frequency questionnaire (SQ-FFQ). Data were standardized using z-score and analyzed using k-means clustering. Associations between clusters and clinical characteristics were examined using chi-square tests (p<0.05), and multinomial logistic regression was applied to identify predictors of malnutrition risks.Results: Three clusters (High-fat-high-intake (HFHI), Lean muscular (LM), and Low intake-low body composition (LILB)) were positively associated with gender, smoking status, and cancer type (p<0.05). Compared to cluster LILB, cancer type significantly increased the likelihood of a person being included in both cluster HFHI (OR=7.34; p<0.001) and LM (OR=4.35; p=0.001). Furthermore, gender had a significant effect on being included in cluster HFHI (OR=6.34; p=0.031), but not in cluster LM. Smoking status was not significantly linked with any cluster (p>0.05). Conclusions: Clustering approach can reveal varied nutritional status patterns among cancer patients, necessitating more tailored and multimodal nutritional therapy management]]>
Copyrights © 2025
