Cancer-associated anorexia: Validity and performance overtime of different appetite tools among patients at their first cancer diagnosis

Background & aims: Anorexia is a frequent symptom in cancer and we aimed to assess its prevalence among patients at their first cancer diagnosis by different appetite tools and the relationship between each tool with self-reports of food intake. We also tested whether cancer anorexia influences outcomes independently of reduced food intake or body weight loss (BWL) overtime and whether BWL was associated with complications during anticancer-therapy. Methods: Functional Assessment of Anorexia/Cachexia Therapy (FAACT) score, self-assessment of appetite, Anorexia Questionnaire (AQ) and Visual Analog Scale (VAS) were administered. Percent of food intake was used as a criterion measure of anorexia. We registered BWL and anticancer-therapy complications over 3-month-follow-up. Results: 438 cancer patients from 7 cancer-centers worldwide were included. The prevalence of anorexia was 39.9% by FAACT score, 40.2% by VAS, 40.6% by the self-assessment of appetite and 65.4% by AQ. Low food intake (≤50%) was reported in 28% of patients. All appetite tools correlated with food intake percent (P < 0.0001). We documented a correlation between self-assessment of appetite, FAACT score, VAS and BWL overtime (P < 0.04). The self-assessment of appetite (P = 0.0152) and the FAACT score (P = 0.043) were associated with BWL independently of anticancer therapies. Among patients with BWL, the risk to develop complications was greater with respect to those who maintained a stable or gained body weight (P = 0.03). Conclusions: In our sample of cancer patients, FAACT score and self-assessment of appetite performed well when low food intake was used as a criterion measure, and revealed an association of anorexia with BWL, which was, in turn, related to the development of anticancer-therapy complications.