Introduction: Methotrexate is an antimetabolite chemotherapeutic agent used for over 50 years. The inclusion of high dose methotrexate with leucovorin rescue has enhanced lymphoma and acute leukemia regimens' efficacy. As a weak acid, methotrexate has variable protein binding and about 50% of the drug binds to serum albumin. Animal models suggest that methotrexate conjugated to albumin may improve anti-tumor effects by accumulating in tumor cells through endocytosis, decreasing its clearance. Further, greater methotrexate toxicity has been reported in patients with hypoalbuminemia. Our purpose was to assess serum albumin's impact on methotrexate pharmacokinetics by comparing methotrexate clearance and toxicities between patients with normal serum albumin and hypoalbuminemia.Methods: This IRB approved single-center retrospective study at Memorial Sloan Kettering Cancer Center included patients who were ≥18 years of age, received methotrexate at a dose of ≥1 gram/m2 between January 2007- November 2014, and admitted to the inpatient leukemia or lymphoma service. Patients admitted to other services or who had prior methotrexate exposure were excluded. Patients were identified through the pharmacy database and electronic medical records. Patients were categorized based on their most recent serum albumin level prior to methotrexate administration. Hypoalbuminemia was defined as serum albumin ≤ 3.4 g/dL. The relationship between hypoalbuminemia and clinical variables was assessed with Fisher's exact and Wilcoxon Rank Sum tests. Kaplan Meier methods were used to visualize the relationship between hypoalbuminemia with time to methotrexate clearance. Methotrexate clearance was defined as the first documented time the methotrexate level ≤0.05 micromolar. Univariate and multivariate associations between predictors and clearance were assessed with Cox proportional hazards regression. Clinical variables considered for the Cox model included hypoalbuminemia, urine acidity, renal function, fluid retention, presence of drug interactions, methotrexate dose, and nephrotoxic agents. Any variables significant at p<0.05 were considered for multivariate analysis.Results: Of 523 patients identified, 167 patients were evaluable. Of them 126 had lymphoma (75.4%) and 41 patients had leukemia (24.6%). The most common chemotherapy regimens used were R-CODOX/M-IVAC (21%), pediatric inspired acute lymphoblastic leukemia regimens (15%), HyperCVAD (12%), SMILE (11.4%), and MR-CHOP (10%). 135 patients had normal serum albumin and 32 had hypoalbuminemia. Baseline characteristics were similar between cohorts. Hypoalbuminemia was associated with a higher proportion of patients with fluid retention (34% vs. 12%, p=0.006) and concomitant use of nephrotoxic agents (41% vs. 20%, p=0.02). Hypoalbuminemia was associated with a significantly longer time to methotrexate clearance (median: 96 hours, 95% CI: 72-132 hours) vs normal serum patients (median 72 hours, 95% CI: 63-72 hours), p=0.004) (Figure 1). In univariate analysis, hypoalbuminemia, methotrexate dose, and nephrotoxic agents were significantly associated with time to clearance. In our multivariate model, hypoalbuminemia (p=0.009), methotrexate dose (p=0.002) and nephrotoxic agents (p=0.02) remained significantly associated with methotrexate clearance time. In addition, patients with hypoalbuminemia had a higher proportion of hyperbilirubinemia (28% vs. 13%, p=0.001) and significantly longer hospitalization (median 14 days vs. 5 days, p<0.001). There was no significant difference in the proportion of patients who experienced transaminitis, nephrotoxicity, mucositis, leukopenia, neutropenia, or pneumonitis.Conclusions: Hypoalbuminemia was associated with prolonged methotrexate clearance, and it remained significant after adjusting for the methotrexate dose and concomitant use of nephrotoxic agents. Patients with hypoalbuminemia were more likely to have bilirubin elevation, both markers of synthetic liver function. There were no differences in the toxicities of high dose methotrexate between groups owing to appropriate leucovorin rescue and good supportive care, demonstrating the safety of high dose methotrexate in patients with hypoalbuminemia, despite differences in the clearance time. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.