Randomised feasibility study of a more liberal haemoglobin trigger for red blood cell transfusion compared to standard practice in anaemic cancer patients treated with chemotherapy

Objectives: The primary objective of this feasibility study was to identify quality of life (QoL) scores and symptom scales as tools for measuring patient-reported outcomes (PRO) associated with haemoglobin level in chemotherapy-treated cancer patients. Secondary objectives included comparing QoL and symptoms between randomisation arms. Background: Anaemia in cancer patients undergoing chemotherapy is associated with decreased QoL. One treatment option is red blood cell transfusion (RBCT). However, the optimal haemoglobin trigger for transfusion is unknown. Methods: Patients were randomised to a haemoglobin trigger for RBCT of either<9·7gdL-1 (arm A) or<lower normal level, female: 11·5gdL-1, male: 13·1gdL-1 (arm B). Four PROs were used: Functional Assessment of Cancer Therapy-General (FACT-G) and the FACT-Anaemia (FACT-An), a Numeric Rating Scale on symptoms of anaemia and self-reported Performance Status (PS). The association between haemoglobin and PRO variables was assessed using a linear mixed model with random effects. Results: A total of 133 patients were enrolled, of which 86 patients received RBCT (28 in arm A, 58 in arm B). Baseline questionnaires were filled out in 79·7% of cases. Haemoglobin levels were significantly correlated with FACT-An, FACT-An Total Outcome Index (TOI), Functional Well-Being, fatigue and PS. Improvement on several PRO variables was observed in both arms after RBCT, with clinically minimal important differences observed in FACT-G, Physical Well-Being, FACT-An, FACT-An TOI, fatigue and dyspnoea. Conclusions: QoL scores of physical and functional domains as well as self-reported anaemia-related symptoms correlated well with haemoglobin level in chemotherapy-treated cancer patients.