Use of Online SPE for Clinical Research in the Automated Analysis of Mycophenolic Acid by LC-MS

Significance of the Topic

Mycophenolic acid (MPA) is a widely used immunosuppressant in transplant medicine. Accurate and high-throughput quantification of MPA in biological matrices is crucial for clinical research, therapeutic drug monitoring, and pharmacokinetic studies. Conventional methods often require manual sample preparation, limiting throughput and reproducibility. Integrating online solid-phase extraction (SPE) with liquid chromatography–mass spectrometry (LC–MS) addresses these challenges by streamlining sample cleanup and analysis into an automated workflow.

Objectives and Study Overview

This study aimed to develop and validate an automated LC–MS method for measuring MPA in plasma using the ACQUITY UPLC Online SPE Manager. The method's performance was compared to a routine LC–MS procedure at an independent laboratory, focusing on sensitivity, specificity, throughput, and agreement between methods.

Methodology and Instrumentation

• Sample Preparation:
  • 50 µL plasma aliquots treated with 500 µL 30% methanol containing deuterated MPA internal standard and 0.2 M ZnSO₄
  • Protein precipitation by centrifugation and collection of supernatant
• Online SPE:
  • Waters MassTrak™ Online SPE Manager with XBridge® C18 cartridge
  • Cartridge conditioning (methanol, water), sample loading, wash with 25% methanol, and elution to UPLC column
• Chromatography:
  • ACQUITY UPLC HSS SB C18 (2.1 × 30 mm, 1.8 µm) at 50 °C
  • Gradient: 70:30 to 15:85 A/B in 1.6 min, then to 2% A, total cycle 3.0 min
• Mass Spectrometry:
  • Xevo® TQD, positive ESI, MRM transitions: 321.2→207.2, 321.2→159.2 for MPA; 324.2→210.2 for D₃-MPA
  • Capillary voltage 0.5 kV; desolvation 450 °C, 800 L/h; cone gas 25 L/h

Main Results and Discussion

• Retention time of MPA and D₃-MPA: 1.47 min with baseline resolution from glucuronide metabolites.
• Sensitive detection: signal-to-noise >400:1 at 0.97 µg/mL.
• Linearity over 0.01–50 µg/mL (r² ≥ 0.999); LOD 2 ng/mL; LLOQ 7 ng/mL.
• Inter-assay CVs: 5.8% (low), 8.4% (mid), 6.5% (high) across 25 replicates over 5 days.
• High throughput: analysis of 96 samples in ~5.5 h; injection-to-injection time 3.0 min.
• Method comparison via Deming regression versus external LC–MS: y = 0.99x – 0.01, bias ~1%; Bland-Altman showed no significant bias.

Benefits and Practical Applications

• Fully integrated, automated sample preparation and analysis increases throughput and reduces manual error.
• High sensitivity and specificity support low-level MPA quantification in clinical research.
• Robust performance with minimal interference, suitable for large-scale pharmacokinetic and drug monitoring studies.

Future Trends and Opportunities

• Extension of online SPE–LC–MS workflows to multiplexed drug panels for comprehensive immunosuppressant monitoring.
• Integration with high-resolution mass spectrometry to enhance specificity and metabolite profiling.
• Development of real-time data processing and feedback to support adaptive dosing in clinical trials.

Conclusion

The automated online SPE–LC–MS method provides a rapid, sensitive, and reproducible platform for MPA analysis in plasma. Performance metrics align closely with conventional assays while offering significant gains in throughput and automation, making it a valuable tool for clinical research and pharmacokinetic investigations.

References

• Eastwood M., Calton L., Chusney G. Use of Online SPE for Clinical Research in the Automated Analysis of Mycophenolic Acid by LC-MS. Waters Corporation; 2014.