Comparative Evaluation of Single-Dose and Split-Dose Melphalan Conditioning Concerning Toxicity Profiles and...

High-dose melphalan (HDM) remains the gold-standard conditioning regimen for autologous stem cell transplantation (ASCT) in multiple myeloma (MM), significantly improving response depth and long-term survival. However, HDM is associated with substantial toxicities, notably oral mucositis and gastrointestinal complications. To mitigate these, centers vary the administration schedule between a single bolus (200 mg/m²) or a split-dose approach over 2 days. Existing literature comparing these methods is limited and inconsistent regarding both toxicity and engraftment kinetics. This study compares the safety profiles, hematologic recovery kinetics, and long-term clinical outcomes (progression-free survival and overall survival) of single-dose versus split-dose HDM administration in patients with MM undergoing ASCT, utilizing a propensity score-matched analysis to minimize baseline imbalances. This multicenter retrospective cohort study included 262 patients across 3 tertiary centers between 2016 and 2026. To ensure comparability, 202 patients were selected through 1:1 propensity score matching (PSM). PSM was performed using logistic regression including age, sex, ECOG status, ISS stage, and renal function (eGFR <40 versus ≥40 mL/min/1.73 m²). The primary safety endpoint was severe gastrointestinal (GI) toxicity (grade ≥3 mucositis and/or diarrhea). Secondary endpoints included infections, organ toxicities, engraftment timing, and survival. To delineate the independent contributions of dosing schedule and melphalan to cell infusion timing (24 versus 48 hours), multivariable linear regression and Cox proportional hazards models were employed, adjusting for CD34⁺ cell dose and maintenance therapy. Single-dose administration was associated with a significantly lower incidence of severe GI toxicity (19.8% versus 36.6%; odds ratio [OR] .43, 95% confidence interval [CI] .21-.84; P = .012), primarily driven by reduced grade ≥3 diarrhea (OR .40, 95% CI .19 to .81; P = .009). Single-dose dosing also provided significant protection against organ toxicities, specifically reducing creatinine elevation (OR .31; P = .025) and hepatic transaminase elevation (OR .40; P = .043). Although split-dose conditioning initially showed faster neutrophil (11 versus 12 days; P < .001) and platelet engraftment (11 versus 13 days; P < .001), multivariable models revealed this was exclusively driven by the 48-hour melphalan to cell infusion interval used in a subgroup of the split-dose cohort. No significant differences were observed between single-dose and split-dose 24-hour subgroups. At Day +90, overall response rates (94.1% versus 93.8%) and CR rates (71.3% versus 64.4%) were comparable (P = .48). With a median follow-up of 36.5 months, no significant differences were found in median PFS (33.6 versus 34.7 months; P = .552) or OS (P = .598). Single-dose HDM provides a superior safety profile, particularly regarding gastrointestinal, renal, and hepatic toxicities, without compromising long-term disease control. The previously reported faster engraftment with split-dose conditioning appears to be a function of the extended (48-hour) melphalan-to-infusion interval rather than the dosing schedule itself. These findings suggest that a single dose is safer and that infusion timing is a critical, modifiable factor in optimizing ASCT protocols.