The Conventional Body Surface Area (BSA) Method of Calculating the Dose of Melphalan (MEL) Results in Widely Variable MEL Exposure and Mucositis Risk in Myeloma (MM) Patients Undergoing Autologous Stem Cell Transplantation (ASCT).

Background: The importance of MEL dose intensity for patients (pts) with MM is supported by a survival benefit when MEL-ASCT is utilized. This survival advantage is further enhanced when tandem ASCT are applied, suggesting that higher MEL doses would be desirable if gastrointestinal mucositis, the dose limiting toxicity of MEL is not excessive.

Objective: To identify the predictors of severe mucositis in MM pts undergoing MEL - ASCT.

Patients and methods: 382 consecutive MM pts enrolled in our Total Therapy 2 protocol, who received their first MEL - ASCT between 10/1998 and 12/2002. A MEL dose of 200 mg/m² BSA was utilized when serum creatinine was < 3 mg/dl (reduced to 140 mg/m² if > 3 mg/dl). BSA was based on actual weight if < 60 kg or calculated weight if > 60 kg. Mucositis peak was graded 0–4 (NCI Common Toxicity Criteria). Potential covariates included among others age, sex, MM remission status, time period between last chemotherapy and MEL-ASCT, MEL dose in mg/kg, CMV serostatus, DLCO, creatinine clearance (CrCl).

Results: BSA dosing of MEL resulted in a wide range of MEL,when calculated based on actual weight (2.5–6.4 mg/kg). Mucositis developed in 83 % of pts and was severe (Grades 3 & 4) in 23 %. By univariate analysis, the following pre-ASCT risk factors for mucositis were identified: lower BSA (OR 0.31/m²; 95% CI 0.111 – 0.881; p=0.027), lower platelet counts (OR 0.998/1000 platelets / mL; 95% CI 0.996 – 1; p=0.025), lower CrCl (OR 0.991/ml/min; 95% CI 0.985 – 0.998; p= 0.011), higher LDH (OR 1.007U/L, 95% CI 1.002 – 1.011; p=0.005), lower DLCO (OR 0.989/unit; 95% CI 0.978 – 1; p= 0.04) and CMV seropositivity (OR 1.51; 95% CI 1.002 – 2.29; p= 0.04). Multivariate analysis identified higher mg/kg MEL dose as the most important risk factor for mucositis (OR=1.58/mg/Kg; 95% CI 1.148 – 2.177; p= 0.005). Additional pre-ASCT risk factors included higher LDH (OR 1.006; 95% CI 1.001 – 1.012; p= 0.007), lower CrCl (OR= 0.991/ml/mn; 95% CI 0.984 – 0.999; p= 0.02), and lower DLCO (OR 0.989; 95% CI 0.978 – 1; p = 0.04). Using the intended mg/kg MEL dose, pre-ASCT CrCl, LDH and DLCO a predictive model for severe mucositis (grade 3 and 4) was developed based on ordinal logistic regression analysis:

p= exp(−2.1527−0.00248*PLT−0.00894* CrCl + 0.00646*LDH−0.0114*DLCO + 0.4578* Mel) / [1+exp (−2.1527−0.00248*PLT−0.00894*CrCl + 0.00646*LDH − 0.0114* DLCO + 0.4578* Mel)]

This model will be prospectively validated.

Conclusions: Dosing MEL on the basis of BSA results in widely variable MEL exposure and risk for severe mucositis following MEL-ASCT. MM pts scheduled to receive MEL-ASCT using a high mg/kg MEL dose and whose pre-ASCT variables include low CrCl, high LDH and /or low DLCO should be considered at greater risk for severe mucositis following ASCT.