Objective This study aimed to provide evidence-based results on differences in overall survival (OS) rate to guide the diagnosis of cancer cachexia.

Design Data collection and clinical assessment was performed every 3 months (5 visits): baseline data, muscle strength, nutritional and psychosocial status. 2 definitions on cachexia using different diagnostic criteria were applied for the same patient population. Fearon et al's definition is based on weight loss, body mass index (BMI) and sarcopenia. Evans et al nuances the contribution of sarcopenia and attaches additional attention to abnormal biochemistry parameters, fatigue and anorexia. The mean OS rates were compared between patients with and without cachexia for both definitions.

Results Based on the population of 167 patients who enrolled, 70% developed cachexia according to Fearon et al's definition and 40% according to Evans et al's definition. The OS in the cachectic population is 0.97 and 0.55 years, respectively. The difference in OS between patients with and without cachexia is more significant using the diagnostic criteria of Evans et al. The focus of Fearon et al on weight loss and sarcopenia over-rates the assignment of patients to the cachectic group and OS rates have less prognostic value.

Conclusion This study presents a correlation with prognosis in favour of Evans et al’ definition as a tool for cachexia diagnosis. This means that weight loss and BMI decline are both key factors in patients with cancer leading to cachexia but less decisive as stated by Fearon et al. Instead, extra factors gain importance in order to predict survival, such as chronic inflammation, anaemia, protein depletion, reduced food intake, fatigue, decreased muscle strength and lean tissue depletion.

Trial registration number B300201112334.