Effect of Melphalan Exposure in Patients with Lymphoma Undergoing Autologous Hematopoietic Stem Cell Transplant (AHCT)

Abstract

<h3>Background</h3> Carmustine, etoposide, cytarabine and melphalan (BEAM) is the most common conditioning regimen for autologous hematopoietic cell transplantation (AHCT) recipients with lymphoma. Data in multiple myeloma have shown that dosing melphalan as a fixed dose based on body surface area leads to inter-patient differences with a 5-fold variability in area under the curve (AUC), with higher AUCs associated with improved outcomes but increased toxicity (Shaw BBMT 2012). We evaluated the pharmacokinetics (PK) of Evomela® (propylene glycol free melphalan, PGF-MEL) in lymphoma patients to investigate the effect of melphalan exposure on outcomes and toxicities. <h3>Methods</h3> On D-1 prior to AHCT, 140 mg/m2 of PGF-MEL was given over 30 minutes and serum sample were drawn serially at 5, 15, 30, 40, 75, and 150 min after the dose. AUC PK modeling was done with Cetara Phoenix WinNonlin, Princeton, NJ. Toxicities will be collected using CTCAE v4 through D+100 after AHCT. <h3>Results</h3> Eighty-seven patients received BEAM and had melphalan PK analysis. Among all patients, the median age was 51 (range, 19-74), 62% were male, and 53% had aggressive B-cell lymphoma. The median PK parameters were as follows: AUC 8.1 mg*h/L (range, 3.8-18.8); clearance 31.8 L/h (range 16.4-79.1); volume of distribution 33.2 L (range, 19.7-70.1); and half-life of elimination 0.7 hours (range, 0.5-1.7). The median dose of melphalan given was similar between patients ≤ median AUC and those above (260 mg [range, 200-340 mg) and 270 mg [range, 210-340 mg], respectively.) Median follow-up of survivors was 8.94 months (range 0.72-32.91). Univariate analysis showed no difference in 1-year overall survival (OS) (95.5% [84-100%] vs. 87.4% [76.5-100%], p=0.2) or progression-free survival (PFS) (78.4% [63.6-96.6%] vs. 78% [63.7-95.5%], p>0.99) when evaluating patients ≤ or above median melphalan AUC, respectively. Patients with a melphalan AUC ≤ the median had an average length of stay of 19 days (range, 5-27) vs. 21 days (range, 7-38) for those above the median exposure (OR 1.19 [1.05-1.38], p = 0.004). Toxicity data collection is ongoing. <h3>Conclusion</h3> PGF-MEL demonstrated a 5-fold variability in AUC when used as part of BEAM conditioning in AHCT for lymphoma. Early results do not yet indicate an impact on OS or PFS, but the data requires time to mature. However, melphalan exposure ≤ 8.1 mg*h/L resulted in a shorter length of hospitalization likely driven by toxicities that will be reported and presented.