49 Improved Low Concentration Precision and Interassay Correlation between ARK and TDx Methotrexate Assays Following a Laboratory Modification. (11th January 2018)

• Record Type: Journal Article
• Authors: Shajani-Yi, Zahra; Morrison, Michael; Curtis, Mark; Cervinski, Mark
• Is Part Of: American journal of clinical pathology. Volume 149(2018)Supplement 1
• Page Start: S189
• Page End: S189
• Publication Date: 2018-01-11
• Subjects: Diagnosis, Laboratory -- Periodicals; Pathology -- Periodicals; 616.07
• DOI: 10.1093/ajcp/aqx149.418
• Languages: English
• ISSNs: 0002-9173

Abstract

Abstract: Background: Methotrexate (MTX), a chemotherapeutic folate antagonist, is infused at high doses (>1 g/m 2 ) for treatment of acute lymphoblastic leukemia, lymphoma and osteosarcoma. MTX is primarily eliminated by the kidneys, and prolonged exposure to high concentrations leads to nephrotoxicity. Accurate measurement of plasma MTX concentration is crucial to coordinate the timing and dosage of leucovorin, and to determine when patients can be safely discharged. Currently, we measure MTX via the TDx Methotrexate II assay (Abbott Laboratories, Abbott Park, IL), which will unfortunately be discontinued in June 2017. Previously we validated the ARK Methotrexate assay (ARK Diagnostics, Sunnyvale, CA) on our Roche cobas 6000 analyzer and found that the ARK assay had poor precision at low concentrations. In an attempt to improve the precision of the assay we increased sample volume and concomitantly reduced reagent volume. Objective: Our study objective was to assess the precision of the ARK Methotrexate assay on low concentration samples (<0.3 mmol/L) on the Roche cobas 6000 analyzer by increasing sample volume and reducing reagent volume. Methods: We conducted method comparison and limit of quantitation (LOQ) studies of the ARK Methotrexate assay approved by the Food and Drug Administration, a laboratory modified ARK Methotrexate assay, and the TDx assay. For the method comparison, patient samples with MTX concentrations ranging from 0.02 to 0.77 mmol/L were analyzed on all three assays. The LOQ was determined by assaying five serum specimens spiked with MTX from pooled patient samples at final concentrations ranging from 0.03 to 0.3 mmol/L once daily for 17 days, and LOQ calculated by plotting the CV against the mean and fitting a curve to the data using EP Evaluator. Results: The method comparison yielded equations (95% confidence intervals) of unmodified ARK = 1.006(0.941 to 1.070)TDx + 0.018(–0.004 to 0.040), r = .9859, modified ARK = 0.895(0.829 to 0.96)TDx + 0.031(0.008 to 0.054), r = .9826. At low concentrations (<0.25 mmol/L), the modified ARK assay performed slightly better. The LOQ (20% CV) of the unmodified ARK assay was 0.06 mmol/L. The LOQ of the modified ARK assay could not be calculated as the CV was <20% at all concentrations tested. The LOQ for the TDx could not be calculated due to wide confidence intervals of the fitted CV. At a target concentration of 0.1 mmol/L the measured CVs were 12.4% (TDx), 6.0% (unmodified ARK) and 3.6% (modified ARK). At a target concentration of 0.03 mmol/L the measured CVs were 23.6% (TDx), 44.1% (ARK), and 14.1% (modified ARK). Conclusions: Our modification of the FDA-approved ARK assay on the Roche cobas 6000 analyzer demonstrated similar low concentration accuracy but improved precision. Despite this improvement, our results also support that the unmodified ARK assay is suitable for monitoring MTX at the medical decision points used at our institution.