LC-MS/MS Bioanalysis

---

Introduction
The continuous increase in sensitivity of the new generation of LC-MS/MS instruments and the possibility of using a wider dynamic range has also drastically increased the carryover problems in bioanalytical methods. Carryover is caused by the residual analyte from injection of the previous sample.1,2,3,4,5,6,7 The evaluation of carryover is usually done by injecting a reconstitution solvent blank immediately after the upper limit of quantitation (ULOQ) of the standard curve. For sufficient accuracy at the lower limit of quantitation (LLOQ), any carryover peak must be less than 20% of the LLOQ response. Carryover should be thoroughly evaluated during method development. The source should be investigated and minimized by testing different rinsing solvents and autosampler parameters. If it were not possible to reduce the carryover by these means, the dynamic range of standard curve should be modified by dropping the ULOQ.8
Although a new “zero-carryover” generation of autosamplers is now available on the market, numerous LC-MS/MS users are still employing autosamplers where carryover can arise from a number of different locations.7, 9,10,11 It was first discovered that a specific aggressive solvent mixture must be associated with injection valve switching in order to efficiently remove the carryover.12 Based on this discovery, an original washing solvent mixture, washing solvent chronological order, and additional instrument plumbing (high-flow washing pump and bypass valve) were tested, and were found to eliminate autosampler carryover during the bioanalysis of two proprietary compounds A and B.13 It was assumed that the elevated carryover found during the method development of compound A and B was due to the presence of the “sticky” moieties reported in Figure 1.

Autosamplers

• Gilson Inc.’s (Middleton, WI) 215 liquid handler/injector


• CTC Analytics AG (Zwingen, Switzerland) and Leap Technologies Inc.’s (Carrboro, NC) PAL HTS or LC

HPLC pumps

• Additional quaternary pump of different brands — Waters Corp. (Milford, MA), Shimadzu Corp. (Kyoto, Japan) and Agilent Technologies Inc.’s (Palo Alto, CA) HP 1100 — were used to deliver the washing solvents


• -2 analytical Shimadzu LC-10AD

Column

• The chromatography was achieved on Phenomenex’s (Torrance,CA) Jupiter C18 5um 300Å 50x 2 mm column in isocratic or gradient conditions

Tandem mass spectrometers

• API 2000™ mass spectrometer from PerkinElmer Sciex (a joint venture between Wellesley, Mass.-based PerkinElmer Inc. and Concord, Ont.-based MDS Sciex) equipped with a turbo ion spray source, using Sample Control 1.4 software for the acquisition of data analysis and MacQuan 1.6 software for the integration of chromatographic peaks


• Waters’ Micromass Quattro Ultima mass spectrometer, equipped with      the orthogonal Z-spray interface, using MassLynx software

Washing mixtures
Four washing mixtures were used:

• Washing mixture 1: 30/30/40 methanol/aceton- trile/DMSO


• Washing mixture 2: 1% formic acid


• Washing mixture 3: deionized water


• Washing mixture 4: mobile phase

Washing sequence
The autosampler injection valve, after the injection, was washed with a flow of 5ml/min delivered by an additional quaternary pump delivering the above-mentioned washing mixtures in the following chronological order:

• Washing mixture 1 for one minute


• Washing mixture 2 for 0.5 minute


• Washing mixture 3 for neutralization for one minute


• Washing mixture 4 for equilibration from 0.5 minute

Step 1: Loop loading (Figure 2)

• Injection needle: Loading the sticky compound A or B into the autosampler loop
  
• Injection nalve: In “load” position


• 10-port valve: Bypass OFF


• Eluting pump: Equilibrating the analytical column


• Washing pump: OFF

Step 2: Injecting (Figure 3)

• Injection needle: Sequentially washed by washing mixture 1, 2, 3, and 4


• Injection valve: In “inject” position


• 10-port valve: Bypass OFF


• Eluting pump: Equilibrating the analytical column


• Washing pump: OFF

Step 3: Washing autosampler injector and HPLC lines (Figure 4)

• Injection needle: Sequentially washed by washing mixture 1, 2, 3, and 4


• Injection valve: Switching between “inject” and “load” positions, and sequentially washed by washing mixture 1, 2, 3, and 4 at 5mL/min


• 10-port valve: Bypass ON


• Eluting pump: Performing an isocratic or gradient chromatography according to the method used


• Washing pump: ON

Step 4: Washing 10-port valve and equilibrating the system (Figure 5)

• Injection Needle: Washed by mobile phase (W4) and ready for next      injection


• Injection valve: In “load” position, washed by mobile phase (W4), and ready for next injection


• 10-port valve: Bypass ON/OFF flushed by mobile phase coming from the eluting pump or washing pump to remove possible residual carryover


• Eluting pump: Equilibrating the column with the initial Mobile Phase


• Washing pump: ON, equilibrating the column with the initial mobile phase and at the same flow of the eluting pump

Results
The carryover was calculated considering the mean area of the chromatographic peaks at analytes retention time of the two mobile phases placed after the two ULOQ. This mean area must be less than or equal to 20% of the mean area of the chromatographic peaks at analyte retention time of the two LLOQ.8 Compound A had shown an initial carryover, during the analysis of a 96-sample test batch, between 15 and 48% of the LLOQ using a Gilson autosampler and between 20 and 76% using a CTC autosampler (Table 1). Compound B had shown an initial carryover, during the analysis of a 96-sample test batch, between 96 and 120% of the LLOQ using a CTC autosampler (Table 2). For both compound A and B the carryover has been reduced to three to 7% of LLOQ during the validation employing the procedure described (Table 1 and 2).

Conclusion
The use of an original washing solvent mixture, washing solvent chronological order, injection valve switching, and instrument additional plumbing (a high-flow washing pump and a bypass valve) were able to reduce carryover problems in two different brands of “non-zero-carryover” autosamplers during the bioanalysis of two proprietary very sticky compounds.

Acknowledgements
Pang, H., M. McIntosh, E. Wong, M. Kennedy, A. Shwajch and R. Grant of Eli Lilly Canada (Toronto, ON).
Colombo L., L. Carrano, and A. Amarù of Vicuron Pharmaceuticals Inc. (Gerenzano, Italy).

References
1.Dolan, J. W. “Autosampler carryover.” LCGC,
19.2 (2001). Retrieved September 4, 2005 <http://www.lcgcmag.com/lcgc/data/articlestandard/lcgc/482001/3016/article.pdf>.

2.Dunn-Meynell, K. W., S. Wainhaus and W.A. Korfmacher. “Optimizing an ultrafast generic high-performance liquid chromatography/tandem mass spectrometry method for faster discovery pharmacokinetic sample throughput.” Rapid Communication in Mass Spectrometry, 19.20 (2005): 2905-2910.

3.Forni, S., B.A. Roadcap, W. Zeng, A.Q. Wang, J.J. Zhao, and D. G. Musson. “Dual channel parallel on line turbulent flow extraction LC/MS/MS determination of geometrical isomers.” Proceedings of 2003 ASMS Conference. Retrieved September 4, 2005 <www.asms.org>.

4.Gao, S., M. Yuan, J. Chi, Z.P. Zhang, and T. Karnes. “An application of ion-pairing reagents to a LC/MS/MS bioanalytical assay as a mechanism of carryover reduction for amine possessing compounds.” Proceedings of 2005 ASMS Conference. Retrieved September 4, 2005 <www.asms.org>.

5.Saha, M. and R.W. Giese. “Primary contribution of the injector to carryover of a trace analyte in high-performance liquid chromatography.” Journal of Chromatography, 631.1-2 (1993): 161-163

6.Thurmond, M. and J.C. Hodgin. “A carryover-elimination method for a broad range of analytical samples.” The Application Notebook. Retrieved September 4, 2005 <http://www.lcgcmag.com/lcgc/article/articleDetail.jsp?id=159538>.

7.Vallano, P. T., S.B. Shugarts, E.J. Woolf, and B.K. Matuszewski. “Elimination of autosampler carryover in a bioanalytical HPLC-MS/MS method: a case study.” Journal of Pharmaceutical and Biomedical Analysis, 36.5 (2005): 1073-1078.

8.Garofolo, F. “Bioanalytical method validation.” Analytical method validation and instrument performance verification. Wiley InterScience, 2004.

9.Brus, T., B.D. Beato, and S.H. Maleki. “LC plumbing strategies for reducing carryover in LC/MS/MS methods.” Proceedings of 2003 ASMS Conference. Retrieved September 4, 2005 <www.asms.org>.

10.Gruver, M., K. MacFarland, P. Wang and D. Desai-Kriege. “Evaluation, Optimization and carryover reduction of on-line SPE/LC/MS/MS approaches for the quantitation of drug levels in rat plasma.” Proceedings of 2004 ASMS Conference. Retrieved September 4, 2005 <www.asms.org>.

11.Kuhlenbeck, D. L., T.H. Eichold, S.H. Hoke, T.R. Baker, R. Mensen and K.R. Wehmeyer. “On-line solid phase extraction using the Prospekt-2 coupled with a liquid chromatography/tandem mass spectrometer for the determination of dextromethorphan, dextrorphan and guaifenesin in human plasma.” European Journal of Mass Spectrometry, 11.2 (2005): 199-208.

12.Garofolo, F., H. Pang, M. McIntosh, E. Wong, M. Kennedy and A. Marland. “An effective procedure for carryover sites determination and elimination in TFC-HPLC-MS/MS.” Proceedings of 2002 AAPS Annual Meeting and Exposition. Retrieved September 4, 2005 <www.aaps.org> or <http://www.aapspharmaceutica.com/>.

13.Garofolo, F., M.T. Gilbert, L. Colombo, Carrano and A. Amarù. “A new, general, simple, low cost, and efficient strategy for reducing serious autosampler carryover in LC-MS/MS bioanalysis.” Proceedings of 2004 ASMS conference. Retrieved September 4, 2005 <www.asms.org>.

Fabio Garofolo, PhD is the head of Bioanalysis and PK department at Vicuron Pharmaceuticals, and can be reached at fabiogarofolo@hotmail.com.

Wei Garofolo works for LC-MS Pharmaceutical Development Inc. (Toronto, ON).