Early-stage drug metabolite quantitation without radiolabels

Significance of the topic

The ability to quantify drug metabolites early in discovery without radiolabels or authentic metabolite standards accelerates candidate triage and reduces cost and complexity. Charged aerosol detection (CAD) offers a near-universal, structure‑independent response for nonvolatile and semivolatile analytes, enabling semi-quantitative or quantitative metabolite profiling where UV or MS alone are limited by chromophores or ionization variability.

Goals and overview of the study

This application note demonstrates an HPLC‑CAD workflow to quantify clozapine and its microsomal metabolites without radiolabels or individual metabolite standards. The study compares CAD and UV quantitation, uses MS as an orthogonal tool for metabolite identification, and describes method optimization to maximize CAD linearity and sensitivity using the Thermo Scientific Vanquish CAD P series.

Used instrumentation

• Thermo Scientific Vanquish Flex UHPLC system (System Base, Dual Gradient Pump F, Split Sampler F, Column Compartment H)
• Thermo Scientific Vanquish Diode Array Detector F (DAD)
• Thermo Scientific Vanquish Charged Aerosol Detector HP (CAD HP), CAD P series operation with four PV channels
• Thermo Scientific ISQ EM mass spectrometer (HESI, positive polarity) for confirmation
• Hypersil GOLD C18 column (150 × 2.1 mm, 1.9 μm)
• Chromeleon 7.3.2 chromatography data system

Methodology and sample preparation

• Microsomal incubations: human liver microsomes (50 donor pool) in freshly prepared 50 mM ammonium bicarbonate buffer (pH ≈ 7.4).
• Reaction composition per tube: 193 μL buffer, 2 μL clozapine stock (2 mg/mL in acetonitrile), 5 μL microsomes; reactions initiated with 10 μL NADPH (20 mM) and incubated at 37 °C with gentle agitation.
• Incubation times tested: 30, 60, 90, 120, and 180 min; samples vortexed, centrifuged (14,000 rpm, 10 min), and supernatant transferred to vials for UHPLC analysis.
• Chromatography: Hypersil GOLD C18, mobile phases 0.1% formic acid in water (A) and acetonitrile (B); gradient with inverse gradient compensation (IGC offset 538.50 μL) to maintain constant composition at the detector during the gradient.
• Detector configuration: DAD and CAD in series; a CAD diverter valve directs early salt-rich flow to waste and permits switching to MS for confirmation. CAD coupling mode was used to lower the charging-detection module temperature to ~5 °C above the evaporation tube temperature.

Key method settings and optimizations

• CAD evaporation temperature (EvapT): 25 °C (selected to preserve response for potential semivolatile metabolites).
• CAD power values (PV): four simultaneous channels collected at PV = 1.5, 1.65, 1.8, 1.95 (legacy PFV 1.0–1.3); PV = 1.5 chosen for final quantitation based on residual plots and quasi-linear response.
• Diverter timing: first 1.6 min directed to waste to protect CAD from matrix salts; valve switches permit directing flow to MS for identification.
• MS: ISQ EM operated with HESI positive ionization, alternating full scans m/z 90–400 and 400–1000; used for MS-based assignment only, not for quantitative work due to ionization variability.

Main results and discussion

• CAD response characteristics: linear, quasi-linear response across ~2 orders of magnitude in the selected calibration range. Linear fits for PV channels produced coefficients of determination r2 ≥ 0.998; residual analysis guided PV selection (PV 1.5 chosen).
• Limits of quantification: CAD LOQ set by lowest calibration standard at 0.67 ng/μL (3.4 ng on column). UV LOQ at 260 nm estimated at 0.09 ng/μL (0.45 ng on column) based on S/N ≥ 10 for the smallest standard used. CAD LOQ was reported conservatively as the lowest calibration standard.
• Method precision and recovery: clozapine peak area RSD 4.2% after 180 min incubation (n = 3). Total peak area recovery for target compounds was ~93% by CAD and ~92% by UV when comparing 0 and 180 min total areas.
• Metabolite detection and quantitation: three metabolite-related peaks observed. CAD quantified metabolite concentrations up to 1.52 ng/μL; several metabolites were at or below the LOQ for CAD/UV. MS provided tentative assignments: features consistent with N‑demethylclozapine (m/z ~313.1) and clozapine‑N‑oxide (m/z ~343.1); an additional feature at m/z ~352.1 remained unassigned and would require MSn/fragmentation for structure elucidation.
• Comparison of detectors: CAD delivers more uniform response across nonvolatile metabolites and allows quantitation using a single standard (here clozapine) because response is largely independent of chromophore or ionization differences. UV quantitation with a single calibrant can misestimate metabolites when chromophores differ. MS is essential for identification but is not appropriate as a single‑calibrant quantifier due to variable ionization efficiencies.

Practical benefits and applications

• Enables early-stage metabolite quantitation without radiolabels or authentic metabolite standards, reducing turnaround time and resource requirements.
• CAD's near-universal response allows semi-quantitative workflows where UV or MS would require multiple standards or complex calibration strategies.
• Integrated diverter valve and IGC permit automated protection of CAD from salt-rich matrices and facile switching to MS for metabolite confirmation within the same run.
• Method sensitivity (few ng on column) supports detection and quantitation of low-level metabolites produced in microsomal incubations.

Future trends and potential applications

• Wider adoption of CAD P series detectors in DMPK workflows for single‑standard metabolite screening, especially when metabolites lack chromophores.
• Combining CAD quantitation with higher-resolution MS/MS or MSn fragmentation to unequivocally identify unknown metabolites while preserving quantitation robustness.
• Further lowering LOQs by including additional low-level calibration points (e.g., 0.2–0.3 ng/μL) and refining CAD PV/evaporation settings for particular chemotypes.
• Integration into automated, high-throughput microsomal screening platforms for earlier decision-making in lead optimization.

Conclusions

The described UHPLC workflow using Vanquish CAD P series provides a practical and robust approach to quantify clozapine and its nonvolatile microsomal metabolites without radiolabels or individual metabolite standards. Optimal conditions identified were 180 min incubation, CAD EvapT 25 °C, and PV 1.5; these gave LOQ ~0.67 ng/μL (3.4 ng on column), good precision, and ~93% recovery. CAD offers a more uniform detector response than UV or MS for nonvolatile metabolites, while the in-line diverter valve and concurrent MS support rapid confirmation of metabolite identities.

References

1. Amatobi K., Lovejoy K. Method transfer and optimization of deoxycholic acid analysis using HPLC-CAD. Thermo Fisher Scientific, Technical Note TN003816; 2025.
2. Cai H., Crafts C., Ramanathan R., Humphreys W., Bailey B., Josephs J. A Novel Detection Technology: Charged Aerosol Detection (CAD) Coupled with HPLC, UV, and LTQ‑Orbitrap MS for Semi‑Quantitation of Metabolites in Drug Discovery Metabolism Studies. ASMS oral presentation; Salt Lake City, May 2010.