Cross-Reactivity of Stimulants Found in Sports Drug Testing by Two Fluorescence Polarization Immunoassays

Transcription

1 Cross-Reactivity of Stimulants Found in Sports Drug Testing by Two Fluorescence Polarization Immunoassays Rafael de la Torre*, Roser Badia, Gerard Gonz~lez, Montse Garcia, Maria J. Pretel, Magi FartS, and Jordi Segura Department of Pharmacology and Toxicology, Institut Municipal d'lnvestigacid MOdica (IMIM), Universitat Autdnoma de Barcelona, Barcelona, Spain I Abstract We investigated the usefulness of immunological methods for presumptive detection of stimulants found in sports drug testing. The ingestion of substances that show no cross-reactivity in tests commercially available for the detection of amphetamines can produce positive results in the urine. Human metabolism contributes to the positive results of some urine samples when the parent compound does not cross-react with the antibodies of the assay. Urine samples from healthy volunteers given stimulants were tested by chromatographic methods and by two different fluorescence polarization immunoassays (FPIA) from Abbott Laboratories for the analysis of amphetamines. According to the results obtained, we classified stimulants into four groups: detectable stimulants that gave rise to amphetamine by human metabolism (group 1); detectable ephedrines and related compounds, appearing in the urine either as parent compounds or originated by metabolism (group 2); detectable stimulants that displayed actual cross-reactivity with amphetamine tests (group 3); and stimulants not detected by FPIA (group 4). Most of the true doping cases due to the ingestion of stimulants may be detected by FPIA. The specificity of the results may be increased by combining immunological assays with different antibodies. Introduction Some of the advantages of immunological analytical methods include rapid and simple performance, relatively low cost of reagents, high sensitivity when necessary, and inexpensive equipment. Immunological techniques that use antibodies with cross-reactivity toward a broad group of substances may be useful for drug screening. In some cases, chemically related substances not originally included in the design of the method may even be analyzed (1,2). Screening a group of substances is particularly useful when the expected prevalence of positive cases for each individual substance is low, such as in sports drug testing. Immunological methods, however, are not routinely used in the pre- Address correspondence and reprint requests to Rafael de la Torte, Deparlamenl de Farmacologia i Toxicologia, Instilut Municipal d'invesligaci6 ML~clica (IMIM), Doctor Aiguader 80, E4~8003 Barcelona. Spain. sumptive screening of drugs in human doping control (3-5) as compared with other related fields in clinical toxicology, such as drugs-of-abuse testing or horse dope testing. This may be a result of the extensive list of substances to be analyzed (6) as compared, for example, with those in the field of drugs-ofabuse testing (7). We have shown that enzyme immunoassay with polyclonal antibodies performed for the detection of amphetamine and methamphetamine could detect a number of related substances of interest in sports drug testing, and most of these substances are phenylalkylamine derivatives (8,9). Moreover, specificity of the test could be increased by using an additional chemical reaction. The pretreatment of samples with alkaline periodate eliminated the cross-reaction of cr phenyl ethylamines (i.e., ephedrine derivatives) (10,11). Chemical reactions can therefore be combined with immunological techniques to enhance specificity. We also found that some substances, such as amfepramone, for which no cross-reactivity has been stated by the manufacturers, gave rise to urinary metabolites with high cross-reactivity when administered to humans (8). It seems that there is a marked difference between results of cross-reactivity studies performed by manufacturers on urine specimens spiked with the parent compound and those obtained from healthy subjects given the same compound. The concept of biological cross-reactivity, which is based on the cross-reactivity of a particular compound and its metabolites, refers to the capability of the immunological test to detect the presence of this substance in a biological fluid. There are many factors that influence the final urinary excretion of a given substance, including the dose administered, interindividual variations in drug metabolism and body clearance, and sample collection time. However, when crossreactivity studies of substances found in sports drug testing are performed using the natural urinary metabolites, a better picture of the actual detection capability of a particular immunological test is obtained (12). We conducted this study to identify doping agents that can be detected by fluorescence polarization immunoassay (FPIA) in urine for the following purposes: (a) to assess the presence of Reproduction (photocopyingl of editorial content of this journal is prohibited without publisher's permission. 165

2 cross-reactivity after the ingestion of stimulants; (b) to identify urinary metabolites that can explain some unexpected results in terms of cross-reactivity as compared with the cross-reactivity of the parent compound through the use of chromatographic techniques; and (c) to evaluate the combined use of different immunological assays for enhancing specificity of the test. Experimental Samples Excretion studies involving drug administration and urine collection were performed with healthy male volunteers. The study protocol was approved by the Ethical Committee of Hospital del Mar (Barcelona, Spain) and authorized by the National Department of Health (reference number 88/135). Compounds classified as banned substances (stimulant class) were administered at the doses recommended by the International Olympic Committee (ICO) Medical Commission. Doping class A stimulants are shown in Table I (those used in this study are shown in bold face along with the dose, which was given orally). The collection periods performed were usually 0-8 and 8-24 h, and spot urine specimens were collected after 24 h. Volunteers were under medical supervision throughout the study. Procedure Urinary excretion studies were tested simultaneously by FPIA (Abbott Laboratories, Chicago, IL) and by chromatographic methods. Urine specimens were analyzed by FPIA tests using a TDx instrument, and the instructions of the manufacturer were followed (13). FPIA assays used a six-point calibration curve. The cutoff values established for positive cross-reactivity corresponded to the sensitivity limits reported by the manufacturer. Positive results could also be reported in quantitative terms, that is, a response equivalent to a given amount of the calibration substance. Two FPIA assays for the detection of amphetamines were used, "amphetamine class" and "amphetamine/methamphetamine Ir' tests. The amphetamine class test is calibrated against d,l-amphetamine, and the amphetamine/methamphetamine II test is calibrated against d- amphetamine. According to the manufacturer, both tests have the same limit of sensitivity at 100 ng/ml, although the amphetamine class test was designed for the detection of drugs structurally related to amphetamine (amphetamine congeners, substitutes, analogues, and derivatives), and the amphetamine/ methamphetamine II test is a more specific assay designed to detect amphetamine and methamphetamine. Two screening chromatographic procedures were used. The first consisted of the analysis of urine made alkaline and extracted with diethyl ether with a salting out effect (14). The final extracts were analyzed by capillary gas chromatography in conjunction with a specific nitrogen-phosphorous detector. This procedure is suitable for the analysis of highly volatile nitrogen compounds, such as those contained in most stimulants. In the second chromatographic method, samples were extracted with a solid-phase extraction procedure (15) after being submitted to an enzymatic hydrolysis. The final residue was selectively derivatized with N-methylsilyltrifluoroac- etamide and N-methyl bistrifluoroacetamide (MSTFA/MBTFA) and analyzed by gas chromatography-mass spectrometry (GC-MS) (15). This procedure is suitable for substances that are mainly excreted in urine as glucuronide conjugates and in which the derivatization step enhances the chromatographic properties of hydrolyzed compounds (i.e., some of the most polar metabolites of stimulants). In both procedures, the identification of all stimulants and their metabolites was ultimately confirmed by GC-MS. Results Stimulants were classified into four groups according to the metabolites detected by chromatographic methods and the immunological response obtained. Group 1 included stimulants that gave rise to amphetamine in their metabolism. Even if some of the parent compounds administered to healthy volunteers did not show cross-reactivity with antibodies (i.e., spiked urine specimens with unchanged clobenzorex, fenproporex, and mefenorex gave negative responses in both tests), the final result was positive because of the presence of amphetamine via biotransformation. Metabolites observed by chromatographic methods and their quantitation and results obtained with the amphetamine class and amphetamine/methamphetamine II Table I. Excretion Studies Evaluated (from the International Olympic Committee's List of Doping Classes. Class A: Stimulants; Revision, May 1992)* Amfepramone (25 mg) Amfetaminil (10 mg) Amineptine (100 mg) Amiphenazole (100 mg) Amphetamine (10 rag) Benzphetamine Caffeine Cathine (25 mg) CNorphentermine Clobenzorex (30 mg) Clorprenaline Cocaine Dimetamfetamine (30 mg) Ephedrine (50 rag) Etafedrine (50 mg) Etamivan (20 mg) Etilamfetamine Fencamfamin (20 rag) Fenetylline (25 mg) Fenfluramine t (40 mg) Fenproporex (10 mg) Furfenorex Mefenorex (40 rag) Mephentermine * (30 mg) Mesocarb (250 rag) Methamphetamine (10 mg) Methoxyphenamine (40 rag) Methylephedrine Melhylphenidate (20 mg) Morazone (70 mg) Nikethamide (250 rag) Pemoline Pentetrazole Phendimetrazine (30 rag) Phenmetrazine (25 mg) Phentermine (30 rag) Phenylpropanolamine (50 mg) Pholedrine ( mg) Pipradrol (2 mg) Prethcamide: f Cropropamide* and Crotetamide * (400 and 400 mg) Prolintane (20 rag) Propylhexedrine Pseudoephedrine (60 mg) Pyrovalerone Strychnine (2 rag) and related compounds * Bold face indicates urine specimens included in the present study. The oral dose of stimulants given to heahhy volunteers to generale the urine specimens is in parentheses. :1= Not explicitly included in the International Olympic Commitlee lists. ' Component of Microren. 166

3 Journal of Analytical Toxicology, Vol, 20, May/June 1996 tests are presented in Table II. Substances included in this group showed a positive response to both FPIA assays. Positive results in this table showed a response that was at least equivalent to 100 ng/ml amphetamine. A reference to the reported cross-reactivity for the parent compounds or their metabolites is also included when available. Group 2 included ephedrines or substances that display a chemical structure related to ephedrine (~-hydroxy-phenyl ethylamines), either as a parent compound or a metabolite. The substances of this group showed a positive response in the amphetamine class test and a negative response in the amphetamine/methamphetamine II test. FPIA results and the metabolites detected by chromatographic methods and their quantitation are detailed in Table III. Group 3 was a miscellaneous group that included substances detected by one or both FPIA tests. The following substances showed cross-reactivity in both amphetamine tests: fenfluramine, mephentermine, mesocarb, methoxyphenamine, and phentermine. Amineptine and pholedrine (p-hydroxymethylamphetamine) were only detected by the amphetamine/ methamphetamine II test. FPIA results and the metabolites found by chromatographic procedures are shown in Table IV. Group 4 included those stimulants not detected by any of the immunological tests used in this study. Antibodies of these tests are not capable of detecting the chemical structures of the parent compound or the metabolites generated. Amiphenazole, cropropamide and crotetamide, ethamivan, fencamfamin, methylphenidate, niketamide, pipradrol, prolintane, and strychnine were the compounds that showed no cross-reactivity in any of the FPIA tests used in the study. Table II. Cross-reactivities (Pure Compounds and Actual Urine Specimens) and Urine Metabolites Found by Chromatography for Substances Capable of Generating Amphetamine by Metabolism (Group 1) % Cross-reactivity of Generated urine t parent compound* FPIA results FPIA: FPIA urine (amphetamine equivalent) collection period amphetamine amphetamine collection (pg/ml) concentration Substance methamp.w II class period amphetamine amphetamine observed substances (pg/ml) administered assay assay (h) methamp. II class (GC-NPD) tl 1st 2nd Amphetaminil n+a+" n.a Amphetamine 0.3 1, , Amphetamine Amphetamine Clobenzorex n.d.** > 6.00 Amphetamine Clobenzorex p-hydroxyclobenzorex *+ - Dimethamphetamine n,a. n.a > Amphetamine Dimethamphetamine Methamphetamine Fenetylline n.a. n.a Amphetamine 1.7 Fenetylline 2.2 Fenproporex n.d Amphetamine Fenproporex 0.3 Mefenorex n.d > 8.00 > 6.00 Amphetamine > ,60 Mefenorex 2.8 Methamphetamine > Amphetamine Methamphetamine n,q, " Substances were tested by adding a known concentration of the test compound to a drug-free urine specimen. Cross-reactivities referred to Institut Municipal d'lnvestigaci6 Mc~lica (IMtM) and Abbott Laboratories studies, t Real urinary excretion studies. FPIA = Fluorescence polarization imrnunoassay. ~ Methamp. = Methamphetamine. GC-NPD = Gas chromatography-nitrogen-phosphorous detection. # n.a. = Not assayed. "n.d. = None detected. Response less than the sensistivity of the assay (100 n~ml). tt = Identified but not quantitated 167

4 Discussion Upon administration, substances corresponding to group 1 may be detected in urine using highly specific tests against amphetamine. The response obtained is mainly due to the detection of amphetamine generated by human metabolism. One of the interesting examples of substances in group 2 is amfepramone. As parent compound, it has a low cross-reactivity for the amphetamine class test, but urine specimens from individuals given this substance are positive for this test because most of its metabolites are ephedrine-related substances; amfepramone may even be absent from the urine because of its high hepatic clearance. In the miscellaneous group 3, positive responses obtained can be considered real cross-reactions, although antibodies were not originally directed to detect either the parent compounds or the metabolites generated. The present study confirms that human metabolism contributes to the positive results of some urine samples in cases in which the parent compound does not express cross-reactivity with the antibodies of the assay. This finding is of interest Table III. Cross-reactivities (Pure Compounds and Actual Urine Specimens) and Urine Metabolites Found by Chromatography for Substances Structurally Related to Ephedrine (Group 2) % Cross-reactivity of Generated urine ~ parent compound* FPIA results FPIA * FPIA urine (amphetamine equivalent) collection period amphetamine amphetamine collection (pg/ml) concentration Substance methamp,w II class period amphetamine amphetamine observed substances (pg/ml) administered assay assay (h) methamp. II class (GC-NPD)" 1st 2nd Amfepramone n.d? n.d 0-I 2 n.d n.d Amfepramone 0.4 ** Diethylnorephedrine Dinordiethylpropion N-Ethylnorephedrine Nordiethylpropion Norephedrine Cathine n.a. +t n.d Cathine n.d Ephedrine n.d n.d Ephedrine n.d Norephedrine Etafedrine n.a. n.a. 0-8 n.d Etafedrine n.d Ephedrine N-Ethylnorephedrine Norephedrine Morazone n.a. n.a Cathine 0.9 Phenmetrazine Phendimetrazine n.d n.d Phendimetrazine Phenmetrazine Phenmetrazine n.d Phenmetrazine Phenylpropanolamine n.d n.d n.d Phenylpropanolamine Pseudoephedrine n.d n.d Cathine n.d Pseudoephedrine * Substances were tested by adding a known concentration of the test compound to a drug-free urine specimen. Cross-reactivilies referred to Institut Municipal d'lnvestigaci6 M,~lica (IMIM) and Abbott Laboratories studies. t- Real urinary excretion studies. r FPIA = Fluorescence polarization immunoassay. ~ Methamp. = Methamphetamine. GC-NPD = Gas chromatography-nitrogen-phosphorous detection. # n.d. = None detected. Response less than the sensitivity of the assay (100 ng/ml). ** = Identified but not quantitated. 'i'tn.a. = Not assayed. 168

5 because of its application to immunological tests in sports drug testing and because it can be extended to use in the fields of forensic toxicology and drugs-of-abuse testing. In fact, a similar approach had long been applied when designing immunological tests for benzodiazepines in which oxazepam or nordiazepam were chosen as the metabolic intermediates of several benzodiazepines that were the target of these techniques. A few substances included in the International Olympic Committee's (IOC) stimulant class A were not available. Consequently, excretion studies in humans were not performed, and, therefore, the substances were not tested in the present protocol. However, from what is known about related substances, it can be anticipated that some of them will show a positive cross-reactivity with some of the tests evaluated in this study. Therefore, amphetamine derivatives such as benzphetamine, furfenorex, and ethylamphetamine can be included in group 1 because amphetamine is one of their urinary metabolites. Other nonrelated amphetamine compounds such as selegiline and famprofazone can also be included in group 1 because they give rise to methamphetamine when metabolized. Methylephedrine is an ephedrine derivative and can be classified in group 2. In the cases of chlorphentermine and propylhexedrine, the manufacturer has described a cross-reactivity with the parent compound in spiked urine specimens (13); the substances can thus be classified in group 3. Finally, the following additional substances, listed in stimulant class A of the IOC Medical Commission, were not included in the present protocol: caffeine, cocaine, clorprenaline, pemoline, pentetrazol, and pyrovalerone. Nevertheless, our previous experience with real positive cases in which these substances were ingested indicated that, as expected by their chemical structure, negative results would be obtained in both immunological tests. Specificity of the immunological results can be increased by combining results of techniques that use antibodies that are directed to structurally similar compounds with different crossreactivities. Examples are the substances included in group 2, which reacted to the amphetamine class assay but did not show any type of cross-reactivity in the amphetamine/methamphetamine II test. The contrasting case of pholedrine and Table IV. Cross-reactivities (Pure Compounds and Actual Urine Specimens) and Urine Metabolites Found by Chromatography for the Miscellaneous Group (Group 3) % Cross-reactivity of parent compound* FPIA results Generated urine t FPIA * FPIA urine (amphetamine equivalent) collection period amphetamine amphetamine collection (pg/ml) concentration Substance methamp.s II class peroid amphetamine amphetamine observed substances (pg/ml) administered assay assay (h) methamp. II class (GC-NPD) u 1st 2nd Amineptine n.a. # n.a. 0-8 n.d.** n.d. Amineptinemetabolitett n.d. Fenfluramine Desethylfenfluramine ** Fenfluramine 2.0 n,d, 1.3 Mephentermine Mephentermine 2.7 Phentermine 0.3 Mesocarb n.a. n.a p-hydroxymesocarb 3.5 Methoxyphenamine Diphenhydramine Methoxyphenamine nor-methoxyphenamine Phentermine Phentermine Pholedrine n.a. n.a n.d. Pholedrine * Substances were tested by adding a known concentration of the test compound to a drug-free urine specimen. Cross-reactivities referred to Institut Municipal d'investigaci6 Mc~lica (IMIM) and Abbott Laboratories studies. t Real urinary excretion studies. FPIA = Fluorescence polarization immunoassay. w Methamp. = Methamphetamine. II GC-NPD = Gas chromatography-nitrogen phosphorous detection. # n.a. = Not assayed. ** n.d. = None detected. Response less than the sensitivity of the assay (I00 ng/ml). ti'7-l(i0,11 -dihydro-sh-dibenzo (a,d)-cyclohepten-5-yllpentanoic acid. :l:l=identified but not quanlitated. 169

6 amineptine, which only showed cross-reactivity in this supposed "specific" test, is also of interest. Because IOC lists (6) are "open" to other substances showing pharmacological actions and/or chemical structures related to the specific substances listed (see Table [), biological crossreactivity may appear with new abused substances. In these circumstances, the use of immunological methods would be very helpful in early detection of new trends in drug abuse. Using immunological methods with low substance specificity to detect a broader range of substances can be very helpful if used judiciously. A subsequent confirmation of presumptive positive cases by chromatographic and spectrometric methods is necessary to determine the actual identity of the detected compounds. When substances that were reported in the last 4 years by IOC accredited laboratories as positive and belonging to the class A stimulants were reviewed, as many as 95% of cases detected by official chromatographic and spectrometric methods (16) would have been easily detected by combining both FPIA amphetamine tests. Acknowledgments We thank Marta Pulido, M.D., for editorial assistance and copy editing. We acknowledge the technical assistance of the following individuals for their collaboration during the 25th Barcelona Olympic Games: Joan Gum~, Meritxell Iglesias, Juan Maillo, Vinyet Margalef, Cristina Mate, Jesfis Salvador, Javier Sanagustfn, and Llufs Viadell. We also acknowledge the support given by Abbott Laboratories for the conclusion of this study. References 1. W. Ruangyuttikarn and D.E. Moody. Comparison of three commercial amphetamine immunoassays for detection of methamphetamine, methylenedioxyamphetamine, methylenedioxymethamphetamine, and methylenedioxyethylamphetamine. J. Anal. Toxicol. 12: (1988). 2. J. D'Nicuola, R. Jones, B. Levine, and M.L. Smith. Evaluation of six commercial amphetamine and methamphetamine immunoassays for cross-reactivity to phenylpropanolamine and ephedrine in urine. J. Anal. Toxicol. 16: (1992). 3. D.H. Catlin, R.C. Kammerer, C.K. Hatton, M.H. Sekera, and ].L. Merdnik. Analytical chemistry at the games of the XXlllrd Olympiad in Los Angeles Clin. Chem. 33: (1987). 4. J. Park, S. Park, D. Lho, H.R Choo, B. Chung, C. Yoon, H. Min, and M.J. Choi. Drug testing at the 10th Asian games and 24th Seoul Olympic games. J. Anal. Toxicol. 14:66-72 (1990). 5. J. Segura, J.A. Pascual, R. Ventura, J.I. Ustar~n, A. Cuevas, and R. Gonz~lez. International cooperation in analytical chemistry: experience of antidoping control at the Xl Pan American games. Clin. Chem. 39: (1993). 6. Medical Commission, International Olympic Committee. List of doping classes and methods. In International Olympic Charter against Doping in Sport. IOC, Laussane, Switzerland, 1990, updated June R.V. Blanke, C.N. Chiang, R.L. Hawks, J.E. Manno, and R.E. Willette. Testing for drugs of abuse. NIDA Research Monograph, Vol. 73. R.L. Hawks and C.N. Chiang, Eds., National Institute of Drug Abuse, Rockville, MD, J. Segura, R. de la Torre, M. Donike, J. Aubets, M. Mestres, and J. Cami. Value of enzyme immunoassay in the dope control of phenylalkylamines. Proceedings III World Conference Clinical Pharmacology. Stockholm, Sweden, 1986, p R. de la Torre, R. Badia, M. Garcia, G. Gonz~lez, M. Farr~, X. Lamas, and J. Segura. Screening of some dope agents by immunological methods. Proceedings 10th Cologne Workshop on dope analysis. June 7th to 12th 1992., Sport ed. M. Donike, H. Geyer, A. Gotzmann, U. Mareck-Engelke, and S. Rauth, Eds., Sport and Buch Straul~, KOIn, Germany, 1993, pp M.A. EISohly, D.F. Stanford, D. Sherman, H. Shah, D. Bernot, and C.E. Turner. A procedure for eliminating interferences from ephedrine and related compounds in the GC/MS analysis of amphetamine and methamphetamine. J. Anal ToxicoL 16: 109-]] (1992). 11. V.R. Spiehler, K. Maugh, and S. El Shami. Elimination of ephedrine and pseudoephedrine cross-reactivity in the Coat-A-Count methamphetamine radioimmunoassay. J. Anal. Toxicol. 17: (1993). 12. J. Segura and R. de la Torre. Methodology Evaluation for Urine Drug Testing. S.D. Ferrara and I. Sunshine, Eds., Syva Co., Palo Alto, CA, 1987, pp TDx/TDxFLx assays manual. Abbott Laboratories, Chicago, IL, D.S. Lho, H.S. Shin, B.-K. Kang, and J. Park. Systematic analysis of stimulants and narcotic analgesics by gas chromatography with nitrogen specific detection and mass spectrometry. J. Anat. Toxicol. 14:73-76 (1990). 15. A. Solans, M. Camicero, R. de la Torre, and ]. Segura. Comprehensive screening procedure for detection of stimulants, narcotics, adrenergic drugs and their metabolites in human urine. J. Anal. Toxicol. 19:104-I 4 (1995). 16. Medical Commission, International Olympic Committee. Requirements for accreditation and good laboratory practice. In International Olympic Charter against Doping in Sport. IOC, Laussane, Switzerland, 1990, updated June Manuscript received May 11, 1995; revision received August 3,