P551 - A SIMPLE CLINICAL MODEL USING BMI, WEIGHT LOSS, AND ALT TO PREDICT SARCOPENIA IN LUNG CANCER PATIENTS

P551

A SIMPLE CLINICAL MODEL USING BMI, WEIGHT LOSS, AND ALT TO PREDICT SARCOPENIA IN LUNG CANCER PATIENTS

D. M. Pham^1,*, H. D. Q. Bui¹

¹Nutrition, Military Hospital 103, Hanoi, Viet Nam

Rationale: Sarcopenia is a common comorbidity in lung cancer patients, associated with reduced treatment tolerance and poor prognosis. This study aimed to identify practical, accessible predictors of sarcopenia using routine clinical and laboratory data.

Methods: A cross-sectional analysis was conducted on 89 patients with histologically confirmed lung cancer who were undergoing treatment at the Oncology Center, Military Hospital 103 (Vietnam). Sarcopenia was defined according to AWGS 2019 criteria, incorporating muscle mass, strength, and physical performance. Multivariate logistic regression was performed using BMI categories, recent weight loss, liver enzymes (ALT, AST), albumin, creatinine, electrolytes, hemoglobin, lymphocyte count, and demographic data.

Results: Among lung cancer patients (mean age 62.6 ± 9.8 years; 78.7% male), sarcopenia was diagnosed in 35 individuals (39.3%). Multivariate analysis revealed that low BMI was the strongest independent predictor. Compared to BMI <18.5 kg/m², those with BMI 18.5–23 had significantly reduced risk (OR = 0.038, p = 0.008), and BMI ≥23 further lowered the risk (OR = 0.014, p = 0.001). Weight loss in the previous 6 months was also significant (OR = 1.172 per 1% lost; p = 0.021). Additionally, reduced ALT (GPT) levels were associated with sarcopenia (OR = 0.960 per unit; p = 0.019). Male gender showed a trend toward increased risk (OR = 5.03; p = 0.067).

Conclusion: BMI, recent weight loss, and serum ALT are practical and independent predictors of sarcopenia in lung cancer patients. This clinical model offers a simple and scalable screening approach, particularly valuable in settings lacking advanced facilities such as BIA, DEXA, or CT imaging.

Disclosure of Interest: D. M. Pham Other: The authors declare no conflict of interest., H. Bui Other: The authors declare no conflict of interest.