MSACL 2018 US Abstract

Topic: Troubleshooting

Is your internal standard pulling its weight? Top tips for clinical research method development

Stephen Balloch (Presenter)
Waters Corporation

Bio: I am a Senior Applications Scientist at Waters, Wilmslow, with several years' experience in developing and validation methods to CLSI requirements.

Authorship: Stephen A Balloch, Gareth W Hammond and Lisa J Calton
Waters Coproration, Wilmslow, UK

Short Abstract When it comes to measuring an analyte in a biological matrix for clinical research, every single sample is different. This is a result of biological variability, and can manifest itself as the dreaded ‘matrix effects’, rendering analytes susceptible to ionization suppression or enhancement when performing LC-MS/MS analysis. This matrix variability is challenging and unpredictable. As such, correct selection and use of a stable isotope-labeled internal standard is critical for a successful method. Is your internal standard pulling its weight?

Long Abstract

Problem

We have encountered a number of challenges during method development in the past relating to the choice and use of internal standard. Two examples of specific cases are presented:

1)“Why is this calibration line bending? I’m sure I prepared my calibrators correctly.” Poor linearity for analysis of 21-deoxycortisol in serum used in clinical research

2)"Why is there a response in the blank? Surely this will affect the LoQ of my method.” Interference from the internal standard affecting quantification of sirolimus in whole blood for clinical research

Method Information

1)21-Deoxycortisol in serum:

• Serum samples (100 µL) extracted using Waters Oasis PRiME HLB SPE

• Calibrator range 0.72–144 nmol/L, using 2H8-21-deoxycortisol as internal standard

• Waters ACQUITY UPLC® I-Class FTN and Xevo® TQ-S micro mass spectrometer

• Water/methanol/ammonium acetate mobile phase gradient

• Waters ACQUITY UPLC HSS T3 2.1 x 50 mm column, 1.8 µm with VanGuard T3 pre-column

2)Sirolimus in whole blood:

• Whole blood samples (50 µL) prepared using protein precipitation

• Calibrator range 1–30 ng/mL, using 13C2H3-sirolimus as internal standard

• Waters ACQUITY UPLC® and ACQUITY TQD mass spectrometer

• Water/methanol/ammonium acetate mobile phase gradient

• Waters MassTrak™ TDM C18 2.1 x 10 mm column

Troubleshooting Steps

1)21-Deoxycortisol in serum:

Calibrators were found to have a quadratic line of best fit. A number of possibilities were considered.

• Were calibrators prepared correctly? Yes

• Was the detector saturated? No

• Was ionization efficiency compromized at the higher concentrations? Possible, but no improvement using higher ESI capillary voltage

• Was the extraction recovery the same across the measuring range? Yes

• Were there solubility/stability issues with the analyte? No

• Were there interference/matrix effects? Yes, qualitative and quantitative matrix effect studies indicated ion suppression when the analyte eluted

• Did the internal standard pull its weight? No, it (2H8-21-deoxycortisol) did not co-elute exactly with the analyte and the determined matrix factor was not improved when using internal standard adjusted values

2) Sirolimus in whole blood:

Early method development studies used the internal standard, 13C2H3-sirolimus. It soon became evident that a response for sirolimus at its expected retention time was present in blank samples.

• Was the mass spectrometer optimized correctly? Yes, mass calibration, mass position and source parameters were optimized correctly and used in MS experiments

• Was there contamination (of whole blood, reagents, mobile phases and washes, mass spectrometer etc?). Possibly, all reagents and matrix analyzed with no addition of internal standard; no analyte response observed, therefore no contamination was present

• Was the internal standard interfering? Possibly, purity of the internal standard was investigated by assessing MS scans to look for unlabeled material, also prepare solvent solutions of labeled internal standard at appropriate concentrations and inject these monitoring the MRM(s) for unlabeled analyte; there ought to be no peaks

• Did the internal standard pull its weight? No, peaks in blank samples meant it could not be used in a clinical research method

Outcome

1)21-deoxycortisol in serum:

Use of the less deuterated 2H4-21-deoxycortisol narrowed the time between internal standard and analyte elution, which in turn provided improved linearity (typical r2 of 0.999) as a direct result of better compensation for matrix interference. It is critical to select an appropriate internal standard for your clinical research method; if your internal standard elutes at a significantly different retention time to your analyte, it may be subject to interferences and regions of ion enhancement or suppression not experienced by the analyte. Use of alternative stable labels e.g. 13C or 15N when available is advised (1).

2)Sirolimus in whole blood

Mass spectra obtained as part of troubleshooting confirmed the raw material contained a large amount of unlabeled sirolimus, highlighting an issue with the quality of the material and rendering it unsuitable for use in a clinical research method. Use of an alternative raw material 2H3-sirolimus in sample preparation yielded blanks free of sirolimus peaks and we were able to achieve the LoQ goals. Method developers should therefore always evaluate the purity of incoming raw materials as part of a successful method development.

References & Acknowledgements:

(1) CLSI. Liquid Chromatography-Mass Spectrometry Methods; Approved Guideline. CLSI document C62-A. Wayne, PA: Clinical and Laboratory Standards Institute; 2014.

Financial Disclosure

IP Royalty: no

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

yes