MSACL 2023 Abstract

Simultaneous Determination of Treosulfan and Fludarabine in Plasma by LC-MS/MS Shelby M. Hutcherson, Ryan C. Schofield, and Dean C. Carlow Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, United States Shelby Hutcherson, PhD (Presenter) Memorial Sloan Kettering Cancer Center >> POSTER (PDF)

Presenter Bio: Shelby Hutcherson is a second-year Clinical Chemistry Fellow at Memorial Sloan Kettering Cancer Center in New York, New York. She earned her PhD in Biochemistry and Cellular and Molecular Biology from the Johns Hopkins University School of Medicine after graduating with a B.S. in Biochemistry from the University of Virginia. She is passionate about clinical chemistry because it offers her the chance to help patients receive the best medical care possible by working at the interface of science and medicine. As a chemistry fellow, her responsibilities include interpreting serum and urine protein electrophoresis and immunofixation results, serving as the chemist on call to answer questions from clinicians and laboratory staff, and reviewing quality control and method validation data. She also participates in quality meetings and initiatives, teaches residents and members of MSK’s Lab Scholars program, and conducts research. She has validated an automated immunoassay to detect mesothelin, a biomarker used to monitor mesothelioma patients, on the Protein Simple Ella platform and is currently developing a Turbulent Flow LC-MS/MS method to quantitate treosulfan, a busulfan analog, in plasma. After completing her fellowship, Shelby hopes to continue working to improve patient care through service and research as a clinical laboratory director.

Introduction:
Prior to hematopoietic stem cell transplantation (HSCT), a conditioning regimen consisting of high doses of one or more myeloablative agents is given to most patients. This serves to both prevent rejection of the transplanted stem cells and reduce the tumor burden in patients with hematological malignancies. However, myeloablative agents can also cause serious and sometimes irreversible side effects. Treosulfan (Treo) is a structural analog of the alkylating agent busulfan which has been shown in clinical trials to have an improved safety profile compared to busulfan while still maintaining comparable myeloablative activity. Unlike busulfan, Treo is a prodrug which undergoes a pH- and temperature-dependent, nonenzymatic, sequential conversion into two epoxide species with alkylating activity. Because plasma concentrations of Treo exhibit significant interindividual variability, therapeutic drug monitoring (TDM) is necessary to ensure dosages are given that maximize efficacy while minimizing toxicity. Treo is currently being considered for approval by the FDA in combination with fludarabine (Flu) as a conditioning regimen prior to allogeneic HSCT and is also currently being evaluated in combination with Flu and/or other agents in several clinical trials in the United States. As the use of Treo increases in clinical trials and eventually in routine clinical practice, it will be important to be able to quantitate it rapidly and accurately in the clinical laboratory.

Objective:
The objective of this study was to develop an accurate liquid chromatography-tandem mass spectrometry (LC-MS/MS) assay capable of simultaneously quantifying Treo and Flu in plasma across a wide range of concentrations in order to facilitate TDM of both compounds.

Methods:
Deuterated Treo (Treo-d4) and Flu-C13N15 were used as the internal standards (IS) for Treo and Flu, respectively. To prevent the degradation of Treo into the epoxide species, 1 M citric acid was added to all plasma used for validation studies (50 μL per 1 mL of plasma). Drugs were extracted from 100 μL of acidified plasma via protein precipitation with methanol containing IS. After centrifugation at 13,000 rpm for 10 min., 10 μL of the supernatant was injected into the liquid chromatography system and subjected to chromatographic separation followed by electrospray ionization (ESI) tandem mass spectrometry. Treo and Flu measurements were performed on a Thermo Scientific TLX-2 HPLC system coupled to a TSQ Quantum Ultra mass spectrometer in positive ion mode, and compounds were detected using multiple reaction monitoring (MRM) mode. MRM transitions were as follows: Treo 296.0 > 183.1 and 296.0 > 87.1 m/z; Flu 286.1 > 154.0 and 286.1 > 134.0 m/z. Drug concentrations were calculated using a six-point calibration curve. Accuracy was evaluated by spiking Treo and Flu into acidified blank plasma at three concentrations spanning each compound’s analytical measurement range (AMR) and calculating the percent recovery after performing LC-MS/MS. Inter-day imprecision was assessed at three concentrations spanning the AMRs by running an LC-MS/MS batch daily for 20 days and calculating the coefficient of variation (CV). To test for the presence of matrix effects, five waste patient plasma samples were acidified and extracted with methanol. Five calibration curves were prepared by spiking Treo and Flu into the plasma extract and five calibration curves were prepared by spiking Treo and Flu into methanol. The mean signal obtained for each compound from the calibration curves prepared in plasma extract after LC-MS/MS was then compared to the mean signal obtained for each compound from the calibration curves prepared in methanol. The stability of Treo at three concentrations spanning the AMR was determined by spiking Treo into acidified or non-acidified blank plasma and performing drug extraction and LC-MS/MS at regular intervals over the course of 24 hours.

Results:
Treo was stable at 4°C in acidified plasma for up to 12 hours but degraded rapidly in non-acidified plasma. Calibration curves for both Treo and Flu were linear over each drug’s AMR with R2 values exceeding 0.99. Recovery for both compounds ranged from 95-100% at all three concentrations tested. Similarly, inter-day imprecision was <10% for Treo and Flu at each of the three concentrations tested. Imprecision for Treo ranged from 1.5-9.0%, and imprecision for Flu ranged from 3.9-8.3%. The differences observed when comparing Treo and Flu signal in calibration curve samples prepared in plasma extract to signal obtained from calibration curve samples prepared in methanol were negligible, indicating that no matrix effects are present.

Conclusion:
Our LC-MS/MS method is able to simultaneously quantify Treo and Flu across a wide range of concentrations. Although the validation is ongoing, the assay exhibits acceptable accuracy, imprecision, and linearity, and it is not impacted by matrix effects. The relative instability of Treo and the necessity of acidifying plasma samples directly after collection presents important logistical considerations that will need to be taken into account when implementing Treo measurements clinically. This assay will enable simultaneous TDM of Treo and Flu when used in combination for conditioning prior to HSCT in clinical trials and routine clinical practice.