Predictive validity of the GLIM criteria in treatment outcomes in cancer patients with radiotherapy

Abstract

Background & aims: The Global Leadership Initiative on Malnutrition (GLIM) proposed a scheme for malnutrition diagnosis. However, the validity of the GLIM criteria has not been well-established in adults with cancer. This study aimed to validate the GLIM criteria by comparing its predictive validity in treatment-related outcomes to the Patient-Generated Subjective Global Assessment (PG-SGA), a validated malnutrition assessment tool in cancer patients.

Methods: This is a secondary analysis using data from a multi-center randomized clinical trial. Malnutrition at admission was identified based on the GLIM criteria and PG-SGA. Their predictive validities in treatment-related outcomes were compared, including radiotherapy-induced severe toxicities, radiotherapy interruption, and use of artificial nutrition support during radiotherapy and quality of life (QoL) after radiotherapy.

Results: Data from 468 adult cancer patients (mean age: 59.8 years, gender: 41.9% female) receiving radiotherapy revealed that malnutrition rate was 33.7% and 39.7% based on the GLIM criteria and PG-SGA, respectively. Patients with increased degree of malnutrition at admission had increased rates of severe toxicities, treatment interruption, and use of artificial nutrition support during radiotherapy well as decreased scores of QoL in all domains after radiotherapy. The GLIM criteria performed better in predicting severe toxicities and treatment interruption, while PG-SGA better predicted the use of artificial nutrition support and QoL scores.

Conclusions: This study provided evidence on the predictive validity of the GLIM criteria in treatment-related outcomes in comparison to PG-SGA. The GLIM criteria was found to be a valid tool in diagnosing malnutrition; however, its predictive performances varied in related to different outcomes.

Keywords: Adult; Cancer; Diagnosis criteria; Malnutrition; Validity.