The Challenge of 1,25-dihydroxyvitamin D: Reducing Complexity of Extraction While Simultaneously Increasing Sensitivity Kari Gurtner (Presenter) Mayo Clinic

Authors: Kari M. Gurtner, Stefan Grebe, Ravinder J. Singh
Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN

The challenge to measure 1,25-dihydroxyvitamin D (1,25D) at low concentrations, generates the necessity for excellent detection sensitivity. As assay volumes increased, it became obvious that the method was not scalable indefinitely, requiring redesign. We sought to simultaneously reduce pre-analytical complexity and improve low-end sensitivity in our assay. Of 1022 patients compared, 983 (96%) 1,25D2 and 911 (89%) 1,25D3 were in agreement with the current method. Our new method reduces the pre-analytical process to an antibody purification and derivatization approach, while at the same providing a significantly improved signal-to-noise ratio, rendering troubleshooting both simpler and less frequent for 1,25D analysis.

Problem

The low concentration of 1,25-dihydroxyvitamin D (1,25D) in human serum is a challenge for the laboratory in terms of assay maintenance. Regardless of instrumentation, high accuracy, low imprecision and excellent detection sensitivity must be achieved. This demands excellent recovery from pre-analytical processes, which are usually a complex combination of techniques, increasing room for error and often resulting in troubleshooting difficulties and delayed resolution of issues that arise. Additionally, the scalability of complex processes becomes its own issue as volumes increase within the clinical lab. When a high testing volume assay, such as 1,25D, fails in such a setting, test-reporting delays can quickly spiral out of control. In order to address these difficulties, we sought to simultaneously reduce pre-analytical complexity and increase low-end sensitivity of our 1,25D assay.

Method Information

The analysis 1,25D D2 and 1,25 D3 in patient serum was complex and time consuming, requiring three bench technologists and 5 hours of sample preparation followed by a 3-hour analytical run time for each set of 96 patients. The extraction started with addition of 500mcL of 0.2M HCL to 500mcL of patient serum to dissociate 1,25D from its binding protein. 100mcL of 8 ng/mL D2-d6, 2 ng/mL D3-d3 internal standard solution was added, followed by a short incubation. Patient mixtures were purified using solid phase extraction (SPE) and dried under nitrogen. 100mcL of monoclonal mouse anti-1,25D slurry (IDS) was then added to the dried patient extracts and incubated for 1 hour at room temperature. Once incubation was complete, slurry mixtures were transferred to a filter plate and washed with buffer and two washes of water. Finally, 1,25D was eluted with absolute ethanol and dried under nitrogen. Extracts were reconstituted and derivatized with 250mcL of 0.02g/mL 4-phenyl-1,2,4-triazole-3,5-dione (PTAD) solution at room temperature for 30 minutes. The solution was evaporated under nitrogen and reconstituted prior to measurement by liquid chromatography-tandem mass spectrometry (LC-MS/MS) with a Zorbax XDB-C8 column (Agilent) and a SCIEX Triple Quad 6500 mass spectrometer operating in multiple reaction monitoring mode.

Troubleshooting Steps

The new method reduces the pre-analytical process to simply antibody purification and derivatization, performed in a single 96-well format. It reduces the number of technologists to two and the preparative time to 4 hours. 500mcL of patient serum and 40mcL of 1ng/mL D2-d6 and D3-13C3 internal standard solution are added to 100mcL antibody bead slurry (Immundiagnostik) in a 96-well filter plate (Impact, Phenomenex). Once incubation is complete, liquid is evacuated through the filter while the beads remain above, followed by three washes of the beads with water. Finally, 1,25D is eluted with absolute ethanol and dried under nitrogen. Extracts are then treated identically to the previous method for derivatization and reconstitution. To improve chromatography and detection sensitivity, the LC-MS/MS method has been modified to use a Kinetex-C8 column (Phenomenex) with methylamine added to the LC mobile phases to further increase ionization efficiency of 1,25D.

Outcome

Previous methodologies, while sufficient for consistent 1,25D analysis, were complex and error prone. As volumes increased, it became increasingly obvious that the method was not scalable indefinitely, and required redesign. Applying a major redesign to an existing assay is expensive; in terms of both cost and time, but the necessity to stabilize assay sensitivity and reduce the rate of repeats was urgent in our case. We considered the potential contributing factors to troubleshooting this assay. Not only is the previous method complicated by several reagents and extractions, its complexity is further increased with the substitution of manifolds, tubes, and plates; involving various exchanges on and off an automated liquid handler. Performed in a 96-well format, the new process reduces variability between technologists, as well as variability between reagent and consumable lots. A method comparison was performed between the current 1,25D and the new methods. Of 1022 patients tested, 983 (96%) 1,25D2 and 911 (89%) 1,25D3 were in agreement with the current method (<20%). Within-assay precision was measured over five levels and had a mean CV of 4% (D2) and 4.7% (D3). Between-assay precision was also measured over five levels with a mean CV of 5.1% (D2) and 5% (D3). Signal to noise ratio increased for both analytes by an average of 250% for 1,25D2 and 110% for 1,25D3. The impact of the new method at very low concentrations was significant, as previously D2 peaks from 4-20pg/mL were hardly distinguishable. Increased sensitivity was achieved through modifications to the analytical portion of the assay; including new chromatographic methods and mobile phase modifiers. Additionally, moving to a new antibody source enabled us to completely redesign the pre-analytical portion of the assay, eliminating multiple potential sources of error. Our new assay greatly reduces complexity, and simultaneously increases scalability, rendering troubleshooting both simpler and less frequent for 1,25D analysis.