Application Note - Agilent

1 AuthorsMichael C. Zumwalt, PhDAgilent Technologies, , COUSAC hristine Moore, PhDImmunalysis, , CAUSAA bstract A fast, sensitive and reproducible technique for confirm-ing the presence of drugs of abuse (DOA) in oral fluids(OF) using the Agilent G6410AA Triple Quadrupole MassSpectrometer (QQQ) is presented. The sensitivity of theQQQ easily meets the cutoff levels required by the UnitedStates Substance Abuse and Mental Health ServicesAdministration (SAMHSA) for workplace testing. TheDOA analyzed in this work include THC, cocaine, amphet-amine, methamphetamine, and MDMA ("Ecstasy") in OFs,which have been prepared using solid phase extraction(SPE). The sample preparation is then followed byreverse-phase LC/MS/MS using a m, C18 columnfor high-chromatographic resolution with high-speedseparation. As a result, elution times for both analytes andinternal standards are less than minutes for THC, andless than minutes for the remaining drugs. The tech-nique is applied successfully to the quantification of quality 2004, the United States SAMHSA, proposed anew rule that would allow Federal agencies to useRapid Analysis of Drugs of Abuse by LC/Triple Quadrupole Mass SpectrometryApplication Notesweat, saliva, and hair in Federal drug testing pro-grams that now only test urine [1].

2 This initiativeeffectively confirmed the analysis of oral fluids asa viable test matrix for the determination of druglevels in humans in the workplace, which is logi-cally extended to other areas of testing includingpolice checkpoints for possible driving while underthe influence of drugs (DUID) the presence of DOA in OF using liquidchromatography/tandem mass spectrometry(LC/MS/MS) provides a faster analysis than gaschromatography/mass spectrometry (GC/MS)because the sample derivatization step, usuallyrequired for GC/MS analysis, is bypassed withoutsacrificing required levels of sensitivity. The use ofa C18 column with m particle size for liquidchromatography (LC) results in nicely resolved,symmetric peaks at high flow rates. The multiplereaction monitoring (MRM) capability of the QQQallows for the highly selective MS/MS analysis ofcoeluting analyte compounds and their corre-sponding internal standards, along with monitor-ing more abundant product ions for quantificationand less abundant product ions as qualifier ionsfor confirmation.

3 The MRM provides for highly spe-cific detection in a complex matrix such as this work five DOA are analyzed in two separateruns of less than minutes for THC (tetrahydro-cannabinol) and less than minutes for cocaine,amphetamine, methamphetamine, and MDMA (3,4-methylenedioxymethamphetamine). The sensi-tivity requirements set forth by SAMHSA for thesedrugs are easily met. The corresponding cutofflevels are shown in Table Toxicology2Ta b l e 1 .S A M H S A C u t o f f L e v e l s f o r D r u g s o f A b u s eNote that the objective of this work was to testQQQ instrument capability and not the quality ofthe extraction procedure. Therefore, it was decidedthat spiking blank OF extracts with both referenceand ISTDs after the extraction process would elim-inate the variability of sample recovery. However,QCs were spiked with both analytes and ISTD sbefore the extraction, and the unknown sampleswere only spiked with ISTDs before the extraction. Compounds AnalyzedThe target compounds and their molecular ionmasses are given in Figure , C21H30O2(M+H)+ = , C17H21O4(M+H)+ = , C9H13N(M+H)+ = , C10H15N(M+H)+ = , C11H15NO2(M+H)+ = level(ng/mL of OF)THC2 Cocaine8 Amphetamine50 Methamphetamine50 MDMA50 ExperimentalSample PreparationFor each sample, 1 mL of OF is collected using theFDA-approved QuantisalTMcollection device, whichis then dissolved in 3 mL of a proprietary buffersolution already contained in the sample collectiondevice.

4 One mL of this sample is used for furtheranalysis, which corresponds to 250 L of OF. Forthe quality control (QC) samples, reference solu-tions of each analyte are added to drug-free OF,along with the internal standard (ISTD) at low andmedium concentrations of each drug. To theunknown samples only internal standards areadded, and for the calibration standards the pre-scribed levels of analytes and ISTDs are addedafter the extraction method is the same as used foranalysis of these drugs by GC/MS, with any deriva-tization step omitted and the final residue dis-solved in the initial mobile phase rather than in atypical GC the OF/buffer aliquot 2 mL of M potassiumphosphate buffer is added and then vortexed. TheSPE (part number 691-0353T, SPEWare, San Pedro,CA), is conditioned with mL of methanol forTHC, and 3 mL of methanol for cocaine, etc., fol-lowed by 100 L of M acetic acid for THC, and 2 mL of M phosphate buffer for cocaine, SPE is performed by adding the sample to theSPE column followed by successive washes, whichinclude methanol and deionized water, followed by98:2 hexane:acetic acid for THC, 78:20:2CH2Cl2/IPA/NH4OH for cocaine, or 2% NH4OH inethyl acetate for amphetamine, methamphetamine,and evaporating the sample to dryness, it isreconstituted in the initial LC mobile phase ( formic acid/water).

5 For the calibration stan-dards, analytes, ISTDs, and mobile phase areadded to make 1-mL InstrumentationThe LC/MS/MS system used in this work consistsof an Agilent 1100-series vacuum degasser, binarypump, well-plate autosampler, thermostattedcolumn compartment, the Agilent G6410AA TripleQuadrupole Mass Spectrometer, and an electro-spray ionization source (ESI). System control anddata analysis is provided by the Agilent QQQ Con-trol (R&D version), Qualitative and QuantitativeData Analysis software programs. Detailed LC andMS conditions are shown objective of the method development was toobtain a fast and sensitive analysis for quantifyingand confirming the presence of drugs of abuse inoral fluids. For speed, while maintaining goodchromatographic resolution and peak symmetry,different solvents, flow rates, and column parame-ters were optimized. It was found that not onlywould a simple solvent system using water, acetonitrile, and formic acid, work very well, but avery fast 1-minute gradient on a m particleFigure compound structures, and their molecular ion ConditionsColumn: Agilent ZORBAX SB-C18, RRHT 50 mm, m (p/n 822700-902)Column temp:40 CMobile phase:A = Formic acid in waterB = Formic acid in acetonitrileFlow mL/min Gradient:5% B at 0 min95% B at 1 min95% B at 6 minPost run time = minInjection vol:80 L (THC); 20 L (for cocaine, etc)MS ConditionsMode: Positive ESI using the AgilentG1948A ionization sourceNebulizer:40 psigDrying gas flow:10 L/minDrying gas temp:350 C Vcap:4000 VQ1 amu (FWHM)Q2 amu (FWHM)Collision energy:23 V (THC); 5 V (all other analytes)MRM:4 transitions for THC.

6 16 transitionsfor cocaine, amphetamines,methamphetamines, and MDMAas shown in Table 2size C18 column would elute the compounds intimes very competitive with most techniques available in GC/MS as well as Method DetailsDetermination of the optimal MRM transitions forboth quantifier and qualifier ions was carried outby infusing the individual standards at concentra-tion levels around 1 ng/ L. The quantifier ion waschosen as the most abundant product ion and thequalifier ion was chosen as the second-most abundant product the time of this writing, the preliminary versionof software only allowed one collision energy andone time segment for the entire chromatographicrun. Therefore, a single fragmentation energy of 23 V was used for all transitions of for THC andISTD, and 5 V was used for all of the transitions ofthe cocaine, etc., compounds and their associatedISTDs, even though these settings were not optimalfor each transition. In addition, MRM transitionswere monitored continuously throughout the entirerun.

7 As a result, while the data shown here satis-fies the requirements of SAMHSA, even better sensitivity should be achievable with optimizationof collision energy and time programming of b l e 2 .D a t a A c q u i s i t i o n P a r a m e t e r s f o r M R M Tr a n s i t i o n sPseudo-QuantitationQualifierRT molecular ionproduct ionproduct ionCompound(min)(M+H)+(m/z)(m/z) and DiscussionThe chromatograms corresponding to one-half thecutoff value for THC, or 1 ng/mL, are shown inFigure 2. This level is easily seen and the on-column injection amount corresponds to 20 area reproducibility among three injections The root-mean-squared (RMS) signal-to-noise(S/N) is estimated conservatively as five times theRMS S/N. This corresponds to a S/N value of 32 limit of quantitation (LOQ) is about half thisvalue, which corresponds to ng/mL, and wasconfirmed by injecting smaller ion chromatograms for THC and D3-THC. Generation of chromatograms and integration ofpeaks is automated with opening of data file by the Agilent Qualitative Analysis software.

8 Peak elution times less than minutes. No smoothing level standard at1 ng/mL OF easily to 20 pg estimated conservatively at 5 x RMS(95% confidence levelIStd = 40 ng/mLTHC QuantifierTHC QualifierD3-THC QuantifierD3-THC area RSD at this area RSD at this levelLOQ calculated at 10 pg on-column,or ng/mL in OF5In Figure 3, and using the same reasoning for THC,the LOQs for cocaine (coc), MDMA, methampheta-mine (meth), and amphetamine (amp) are esti-mated to be , , , and ng/mL in OF, ion chromatograms for lowest level standard containing cocaine, D3-cocaine, MDMA, D5-MDMA,amphetamine, D5-amphetamine, methamphetamine, and D5-methamphetamine. Peak elution times less minutes. No smoothing QuantCoc. QualD3-Coc. QuantD3-Coc. QualMDMA QuantMDMA QualD5-MDMA QuantD5-MDMA QualMeth. QuantMeth. QualD5-Meth. QuantD5-Meth. QualAmp. QuantAmp. QualD5-Amp. Qual4 ng/mL, RSDLOQ calculated at 1 pg on-column,or ng/mL in OF4 ng/mL, RSDLOQ calculated at pg on-column,or ng/mL in OF50 ng/mL, RSDLOQ calculated at 3 pg on-column,or ng/mL in OF50 ng/mL, RSDLOQ calculated at pg on-column,or ng/mL in OF50 ng/mL6 Along with the quantifier ions for each of the com-pounds and associated ISTDs, the qualifier ionsare also shown in Figure 4.)

9 The requirement foreach qualifier ion is that its measured area fallswithin a range of specified ratios with respect tothe area of the quantifier ion. For example, withthe THC qualifier ion, as determined experimen-tally by the Agilent G6410AA instrument, the ratioof its measured area to that of the THC quantifierion should be 22%. Applying a window of accep-tance that is 20% gives an overall range of As long as the ratio of the areas fallswithin this range, the acceptance criteria for Figure confirmation of THC, the qualifier ion area must be 22% that of the quantifier ion area and within a window of 20%of that value, or from to overall. The two ways to display this for fast confirmation in the Quantitative Analy-sis software is normalized by area (left) and un-normalized (right), both of which show the overlap of the qualifier ion onthe quantifier ion. If the ion ratio is outside the window of acceptance, the integrated area of qualifier ion will be shadedblue, but transparently to still observe is met.

10 For all THC compounds, bothcalibration standards and QCs, this criteria wassatisfied. A similar criteria was established for the the remaining compounds, the qualifier ionarea ratio criteria were established as 4% forcocaine, 9% for MDMA, 95% for methamphetamine,and 26% for amphetamine. As was the case forTHC, criteria were established for the associatedISTDs as well. All calibration standards and QCsmet these criteria. THCD3-THCW indow of acceptance7 The calibration curves generated for all com-pounds are shown in Figure 5. The most conserva-tive fitting options are used to generate the line;that is, a linear fit with no weighting and no origintreatment. Each line is based on calibration levelsextending across nearly two orders of curves for each DOA using a linear line fit with no weighting and no origin > > > > > Forensic Use is subject to change without notice. Agilent Technologies, Inc. 2006 Printed in the USAMay 18, reproducibility for THC is shown in Table 3,and as expected, the %RSD values are lower forhigher concentrations.