MSACL 2018 US Abstract

Topic: Troubleshooting

Evaluation of Surrogate Matrices for Serum 25-Hydroxyvitamin D Calibrators by LC-MS/MS Method

Bei-bei ZHAO (Presenter)
KingMed Diagnostics

Authorship: ZHAO Bei-bei (1,2); LI Wei (1,2); CHENG Ya-ting (1,2); SHE Xu-hui (1,2); DONG Heng (1)
(1) KingMed Diagnostics, Guangzhou, China; (2) KingMed College of Laboratory Medicine of GMU, Guangzhou, China

Short Abstract

It was problematic to obtain analyte-free serum for preparing 25-hydroxyvitamin D calibrators in our laboratory using LC-MS/MS method. We selected two surrogate matrices, 4% BSA and 30% ethanol, and determined their suitability by three steps: (1) comparing the internal standards peak areas in the 2 matrices; (2) mixing matrix effect verification in the presence of authentic serum samples; and (3) DEQAS proficiency testing accuracy validation. Only 4% BSA was a suitable surrogate matrix for preparing 25-hydroxyvitamin D calibrators and has been adapted to our routine LC-MS/MS method.

Long Abstract

Problem

It was problematic to obtain analyte-free serum for preparing 25-hydroxyvitamin D (25-OHD) calibrators in our laboratory using LC-MS/MS method. For routine use, a surrogate matrix for calibrator preparation is desirable in terms of both convenience and cost.

Method Information

1.Surrogate matrices : 4% bovine serum albumin (BSA) in PBS buffer and 30% ethanol

2.Sample preparation: (1) protein precipitation of authentic serum or surrogate matrices by acetonitrile solution containing internal standards (IS for d6-25-OHD3 and d3-25-OHD2); and (2 ) liquid-liquid extraction with hexane.

3.LC-MS/MS:

Shimadzu LC and SCIEX API4000

Mobile Phase-A: 0.1% Formic Acid in H2O

Mobile Phase-B: 0.1% Formic Acid in Methanol

7 min gradient LC program, 0.5 mL/min flow rate

Column: 100 x 2.1 mm, 2.7µm C18, with guard cartridge at 55 ℃

Injection volume 40 µL

Quantitative MRM acquisition

Troubleshooting Steps

Step 1 : Peak areas of the 2 internal standards were measured in six authentic serum samples, seven 25-OHD calibrators in both surrogate matrices. Kruskal-Wallis analysis was performed to compare the differences in the 2 IS peak areas for the 3 matrices. There was no significant difference in the peak areas of both IS between 4% BSA calibrators and authentic serum samples. But peak areas of calibrators in 30% ethanol were significantly lower (about 50%) than those in 4% BSA and the authentic serum samples, showing ion suppression in the 30% ethanol matrix.

Step 2: Six authentic serum samples were individually mixed with a surrogate matrix spiked with the 25-OHD2 and 25-OHD3 both at 30 ng/mL in a 1:1 ratio. The authentic serum samples, the surrogate matrix spiked with pure standards and the 1:1 mixed samples were extracted in triplicate and run by the LC-MS/MS method. The criterion for passing is the measured response ratio (analyte/IS) of each 1:1 mixture being within 20% of the mean response of the patient samples and the surrogate matrix. The 4% BSA passed this mixing test but the 30% ethanol failed.

Step 3: Five DEQAS proficiency testing samples were measured and the results were determined by calibration curves prepared in both surrogate matrices. The bias between measured value and DEQAS target value of each sample was determined for the 2 sets of calibration curve. The criterion for acceptable bias performance was set within ±ALE (i.e. ±25%). For the 4% BSA calibrators, all 5 samples had acceptable bias performance; while only 3 samples had acceptable bias performance for the 30% ethanol calibrators. Results measured by the 30% ethanol calibrators were 20% higher than those measured by the 4% BSA calibrators.

Outcome

Only 4% BSA was suitable for preparing 25-OHD calibrators and has been adapted to our routine LC-MS/MS method.


References & Acknowledgements:


Financial Disclosure

DescriptionY/NSource
Grantsno
Salaryno
Board Memberno
Stockno
Expensesno

IP Royalty: no

Planning to mention or discuss specific products or technology of the company(ies) listed above:

no