Assessing the Efficacy of RBC Component Transfusion Therapy in Patients with Thalassemia Major

Patients with thalassemia major require RBC components that retain sufficient viability to correct anemia and suppress endogenous erythropoiesis. For novel RBC products prepared with pathogen and leukocyte inactivation (PI) it is also important to meet these goals while having minimal effect on transfused iron burden. Unfortunately, ideal endpoints for the efficacy of RBC transfusion in this patient population have not been extensively defined, and published data characterizing relevant statistical parameters are limited. We studied the statistical properties of three transfusion efficacy endpoints in a sample of thalassemia major patients receiving conventional RBC components. The variability of these parameters is a key consideration in the design of clinical trials assessing RBC transfusion efficacy.

Data from 6 transfusion cycles for adult subjects with thalassemia major were randomly extracted from Webthal, a computerized clinical record for thalassemia patients in use at Italian Thalassemia Centers. Only subjects who received all RBC components during a single transfusion episode for each cycle were included in the analysis. Three potential transfusion endpoints were investigated: (1) cumulative hemoglobin (Hb) use (g/kg/day), (2) 1-h post-transfusion Hb increment (g/dL), and (3) proportional Hb decline per day following transfusion (%/day). Data for Hb increment and Hb consumption were based on volume of RBC components adjusted to a hematocrit of 100%. Cumulative Hb consumption provides a summary assessment of transfusion efficacy and iron burden over the 6 transfusion cycles. For the longitudinal endpoints (2) and (3), mixed effects longitudinal models were used to assess whether these endpoints significantly differed across the 6 transfusion cycles.

When subjects were selected for consistency of transfusion cycles, and all RBC components for a cycle were administered in a single transfusion episode, a relatively homogeneous population may be defined. In this instance, all of the longitudinal outcome measures were consistent and variability in the measures over 6 transfusion cycles was limited. For 18 subjects over 6 transfusion cycles, the mean cumulative Hb consumption was 0.11 g/kg/day (SD=0.033). Values for each endpoint are presented for each cycle (Table).

Based on this analysis, a combination of endpoints related to RBC transfusion for support of thalassemia major can be defined to describe comprehensively the efficacy of novel RBC components. The endpoints presented are clinically relevant to routine care, and thus unlikely to result in missing data. None of the longitudinal endpoints demonstrated substantial differences across the 6 transfusion cycles for patients on a stable transfusion regimen. Thus, these endpoints provide relevant comparative parameters for novel RBC components such as PI RBC. Their limited variability is beneficial for non-inferiority clinical trials. Furthermore, inter-subject variability is much greater than intra-subject variability. A crossover design would further enhance the clinical trial such that a feasible number of subjects may be enrolled to achieve 80–90% power to reject inferiority.

Reed:Cerus Corporation: Employment, Stock and stock options Other. Sherman:Cerus Corporation: Employment, stock and stock options Other. Corash:Cerus Corporation: Employment, stock and stock options Other.