¹Myeloma Institute for Research and Therapy, University of Arkansas for Medical Sciences, Little Rock, AR, USA

Correspondence: Dr EJ Anaissie, Myeloma Institute for Research and Therapy, University of Arkansas for Medical Sciences, 4301 West Markham, Slot 816, Little Rock, AR 72205, USA. E-mail: anaissieeliasj@uams.edu

Received 7 March 2006; Revised 23 June 2006; Accepted 4 July 2006.

Abstract

Melphalan-based autologous stem cell transplant (Mel-ASCT) is a standard therapy for multiple myeloma, but is associated with severe oral mucositis (OM). To identify predictors for severe OM, we studied 381 consecutive newly diagnosed myeloma patients who received Mel-ASCT. Melphalan was given at 200 mg/m² body surface area (BSA), reduced to 140 mg/m² for serum creatinine >3 mg/dl. Potential covariates included demographics, pre-transplant serum albumin and renal and liver function tests, and mg/kg melphalan dose received. The BSA dosing resulted in a wide range of melphalan doses given (2.4–6.2 mg/kg). OM developed in 75% of patients and was severe in 21%. Predictors of severe OM in multiple logistic regression analyses were high serum creatinine (odds ratio (OR)=1.581; 95% confidence interval (CI): 1.080–2.313; P=0.018) and high mg/kg melphalan (OR=1.595; 95% CI: 1.065–2.389; P=0.023). An OM prediction model was developed based on these variables. We concluded that BSA dosing of melphalan results in wide variations in the mg/kg dose, and that patients with renal dysfunction who are scheduled to receive a high mg/kg melphalan dose have the greatest risk for severe OM following Mel-ASCT. Pharmacogenomic and pharmacokinetic studies are needed to better understand interpatient variability of melphalan exposure and toxicity.