The Trazodone Metabolite meta-chlorophenylpiperazine Can Cause False-Positive Urine Amphetamine Immunoassay Results

Transcription

1 Technical Note The Trazodone Metabolite meta-chlorophenylpiperazine Can Cause False-Positive Urine Amphetamine Immunoassay Results Jason M. Baron*, David A. Griggs, Andrea L. Nixon, William H. Long, and James G. Flood Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts Abstract Amphetamines and methamphetamines are part of an important class of drugs included in most urine drugs of abuse screening panels, and a common assay to detect these drugs is the Amphetamines II immunoassay (Roche Diagnostics). To demonstrate that meta-chlorophenylpiperazine (m-cpp), a trazodone metabolite, cross-reacts in the Amphetamines II assay, we tested reference standards of m-cpp at various concentrations (200 to 20,000 μg/l). We also tested real patient urine samples containing m-cpp (detected and quantified by HPLC) with no detectable amphetamine, methamphetamine, or MDMA (demonstrated by GC MS). In both the m-cpp standards and the patient urine samples, we found a strong association between m-cpp concentration and Amphetamines II immunoreactivity (r = for the urine samples). Further, we found that patients taking trazodone can produce urine with sufficient m-cpp to result in false-positive Amphetamines II results. At our institution, falsepositive amphetamine results occur not infrequently in patients taking trazodone with at least 8 trazodone-associated false-positive results during a single 26-day period. Laboratories should remain cognizant of this interference when interpreting results of this assay. Introduction Testing for drugs of abuse plays an important role in the clinical management of several patient populations including those presenting to the emergency department in an altered mental state or following accidental or intentional trauma and those undergoing rehabilitation for drug abuse or addiction. In addition, testing for commonly diverted prescription substances * Author to whom correspondence should be addressed: Jason Baron, MD, Department of Pathology, Massachusetts General Hospital, 55 Fruit Street, GRJ-235, Boston, MA jmbaron@partners.org. is often used to ensure patients are taking, rather than selling or otherwise diverting their prescriptions. Outside the clinical setting, substance abuse test results are used in applications including civil and criminal legal proceedings and in employment qualification (1). Positive results in these settings have implications including loss of employment or even imprisonment, although many medicolegal applications include confirmatory testing algorithms to reduce the likelihood of falsepositive results (2). Given the stakes involved in both the clinical and forensic setting, quality control and assurance and knowledge of potential interferences or other sources of error in testing are of key importance. Amphetamine and methamphetamine are part of an important class of abused substances and assays to detect their use are typically included in urine drug abuse screening panels (1,2). Multiple methods are available to detect amphetamine (AMP) and methamphetamine (MAMP) use by identifying parent drugs or their metabolites in blood, urine, or other fluids. Methods include immunoassays, high-performance liquid chromatography (HPLC)-based techniques, and gas chromatography mass spectrometry (GC MS) (1,3). Immunoassays and GC MS are most commonly used for urine screening (1). Although immunoassays offer many benefits, one significant limitation of many of them is that they are subject to interference by cross-reactive compounds present in the sample tested (1). These cross-reactive compounds are usually medications that patients may be taking (or their metabolites) and may result in false-positive results. Among the commonly available urine amphetamine immunoassays is the Amphetamines II assay marketed by Roche Diagnostics and designed for use on the Roche or Hitachi automated chemistry analyzers. This assay detects AMP, MAMP, and methylenedioxymethamphetamine (MDMA, Ecstasy ). During a three-month initial experience with the Amphetamines II assay, we identified a significant number of possible or probable false-positive results that could not be explained by the known interferences in the assay. By observing that multiple patients with false-positive results were report- 364 Reproduction (photocopying) of editorial content of this journal is prohibited without publisher s permission.

2 edly taking trazodone and reading a report that the trazodone metabolite meta-chlorophenylpiperazine (m-cpp) cross-reacts in the EMIT II Ecstasy immunoassay (4), we hypothesized that trazodone or m-cpp may account for many of the Amphetamines II false-positive results. Trazodone is a popular antidepressant and sleep-aid medication. Trazodone is hepatically metabolized with metabolites excreted primarily in the urine. Very little unmetabolized trazodone passes into the urine (5). Elimination is biphasic with half-lives of 3 6 h and 5 9 h for the first and second phases, respectively (5). m-cpp is not only a metabolite of trazodone (5), but also is itself an abused stimulant and hallucinogen (6,7). In this report, we test the cross-reactivity of trazodone and m-cpp and confirm that m-cpp cross-reacts in the Roche assay, and can be present in high enough concentrations in the urine of patients taking trazodone to result in a false-positive urine amphetamine results. Methods Analytic standards The trazodone primary standard was a gift from Mead Johnson & Company. The m-cpp standard was from Sigma- Aldrich (product number , St. Louis, MO). A working methanolic stock of the m-cpp (1 g/l) was prepared and used to make diluted standard solutions via dilution with deionized water. Setting and patients All patient samples were submitted to the Massachusetts General Hospital (MGH) Core Laboratory for drug of abuse testing as ordered by the patient s treating clinician. The Core Laboratory processes samples from inpatients and outpatients at MGH, affiliated institutions and, on occasion, non-affiliated institutions. This work was performed as part of a clinical quality assurance program, and thus, specific institutional review board approval was not required. Amphetamines II immunoassay The Roche Amphetamines II immunoassay (Roche Diagnostics, Indianapolis, IN) is based on a kinetic interaction of microparticles in solution (KIMS) technique (8). The KIMS technique relies on a reaction solution containing soluble microparticles coated with antibodies directed against an antigen of interest (i.e., amphetamines in this case), soluble conjugates of the antigen of interest, and sample containing an unknown quantity of (non-conjugated) antigen (9). The soluble conjugates consist of multiple derivatives of the antigen covalently bound to a common carrier. When the test sample contains no antigen, the soluble antigen conjugates bind the antibody-coated microparticles, generating the maximum number of microparticle aggregates. Antigen in the test sample will also bind to the antibody beads, interfering in the bead s ability to bind the conjugates, thus reducing formation of bead aggregates. Aggregate formation (in this case, a function of urine amphetamines concentration) can be assayed as aggregates increase the scattering of light. Thus, by measuring changes in light transmission throughout the reaction, the concentration of amphetamine can be estimated. This assay is calibrated with 1000 μg/l of d-methamphetamine (Preciset DAT Plus I calibrator CAL4, Roche Diagnostics). Calibration at 300 and 500 μg/l is also available. We used qualitative reporting (positive or negative) in which the rate of absorbance of the positive calibrator is scaled to 0 milliabsorbance units per minute (mau/min). Positive samples then represent any absorbance rate greater than 0, and samples with absorbances less than or equal to zero are reported as negative. Samples containing some amphetamine, but at concentrations less than the positive calibrator, would be expected to result in an absorbance reading greater than a blank (0 μg/l) but less than zero mau/min and would thus be reported as negative. We used Amphetamines II reagent from lot # A (expiration 9/2011) or # (expiration 4/2011) for all studies reported here. Urine samples were tested using this assay as recommended by the manufacturer on our laboratory s COBAS c501 automated analyzers (Roche Diagnostics). Although we only report patient urine amphetamine results as positive or negative, for the purposes of this study, we defined a third category of partially positive. Partially positive samples had rates less than the positive calibrator, but nonetheless demonstrated significant immunoreactivity, with absorbance rates that were more than approximately 50% of the difference between the calibrator and blank. Partially positive samples are presumed to possibly contain amphetamines and cross-reactive substances at levels below the positive calibrator. HPLC testing HPLC testing of serum and urine samples was carried out, and AMP, MAMP, MDMA, and m-cpp were identified as described previously (3,10). The internal standard (protriptyline) elutes at approximate 9.5 min, whereas the m-cpp, AMP, MAMP, and MDMA peaks elute at 3.5, 5.2, 7.2, and 6.9 min, respectively. The detection limit in serum for AMP, MAMP, and MDMA are approximately 20, 35, and 75 μg/l, respectively. Serum trazodone concentrations were determined as reported in reference 10. Trazodone elutes at 8.8 min and the internal standard (lormetazepam) elutes at 6.8 min. Urine and serum m-cpp levels were quantified using a 1000 μg/l m-cpp calibrator by normalizing the peak height of m- CPP to the peak height of the internal standard in each sample and the calibrator. Independent urine amphetamine screen, GC MS testing The six patient urine samples tested for m-cpp in this study were sent to Mayo Medical Laboratories (Rochester, MN) for verification that amphetamines were not present by GC MS (and thus the Amphetamines II positive was a false positive). The Mayo test code used was and had a lower limit of detection of 50 μg/l for AMP, MAMP, and MDMA (11). Urine sample inclusion criteria We sought to identify whether the level of m-cpp interference demonstrated in the standards testing is clinically signif- 365

3 Table I. Data from Six Patients Whose Urine Exhibited Amphetamines II Immunoreactivity Urine Reported Amphetamines II Urine Urine Amphetamines Serum Serum Amphetamines II Immunoreactivity Creatinine m-cpp by GC MS m-cpp Trazodone Patient Result (mau/min) (g/l) (µg/l) (µg/l) (µg/l) (µg/l) A Negative Negative (< 50) B Positive Negative (< 50) C Positive Negative (< 50) D Positive Negative (< 50) E Negative Negative (< 50) F Negative Negative (< 50) icant (i.e., do real patient urine samples contain sufficient m- CPP to produce false-positive results?). To this end, we identified six patients who were tested in our laboratory for drugs of abuse during a two-week period that met the following criteria: 1. The patient had urine and serum collected with both tested for drugs of abuse by order of the treating clinician; the serum and urine samples were collected nearly simultaneously in four cases (patients A, B, C, and E in Table I), approximately 1 h apart in one case (patient D), and approximately one day apart in another (patient F). 2. The serum showed no evidence of amphetamines by HPLC (3). 3. Trazodone and m-cpp were detected in the serum by HPLC (3,10). Results Testing of analytic standards We tested a methanolic stock of m-cpp (1 g/l) diluted to five different concentrations using deionized water: 200, 2000, 5000, 10,000, and 20,000 μg/l. All dilutions and the deionized water blank used were tested in duplicate. The results of this testing are shown in Figure 1. They show that increasing m-cpp concentrations result in increasing Amphetamines II immunoreactivity. By fitting a logarithmic curve to the m-cpp verses immunoreactivity data, excluding the 0 μg/l point (using the Microsoft Excel 2003, scatter plot, Add Trendline feature, r > 0.99), we estimate that approximately 6700 μg/l of m-cpp would be sufficient to result in a positive Amphetamines II test result using a 1000 μg/l cutoff. In contrast, an extremely high concentration trazodone standard (200,000 μg/l) produced less than half the immunoreactivity of the 1000 μg/l methamphetamine calibrator, suggesting that the parent drug trazodone alone is unlikely to be responsible for clinically significant interference. Testing of patient urine specimens We tested the urine and serum samples from each of the six patients denoted A F for m-cpp. The results are shown in Table I. The third column from the left shows the urine Amphetamines II immunoreactivity. Subsequent columns list the Figure 1. Testing of m-cpp standards in the Amphetamines II assay. Shown are the concentrations of each m-cpp standard tested plotted against the immunoreactivity produced by the Amphetamines II assay. Figure 2. Patient urine Amphetamines II immunoreactivity. Shown are the concentrations of m-cpp in each urine sample measured by HPLC plotted against the immunoreactivity produced by the Amphetamines II assay (r = 0.990). 366

4 urine creatinine, the urine m-cpp concentration, the urine amphetamine results by GC MS, and the serum m-cpp and trazodone concentration. As discussed the 1000 μg/l positive calibrator is set to 0 mau/min and the negative (0 μg/l) blank controls used on the days that these urine samples were tested resulted in values between 307 and 228 mau/min. In addition, the dose-response curve plotting m-cpp against urine immunoreactivity is shown in Figure 2. m-cpp concentration is extremely well correlated with Amphetamines II immunoreactivity (r = 0.990, p < 0.001, calculated using Microsoft Excel 2007 in combination with standard formulas). Importantly, as expected based upon the serum results, none of the urine samples contained identifiable amphetamines (AMP, MAMP, and MDMA) by independent methods and all contained quantifiable levels of m-cpp. The urine of patients B, C, and D were reported as positive for Amphetamine because of their high Amphetamines II immunoreactivity. These three urines had the highest m-cpp concentrations of 7325, 4400, and 2500 μg/l, respectively. The urines reported as negative (A, E, and F) all had urine m-cpp levels less than 400 μg/l and immunoreactivity less than half that needed to be judged as positive using the 1000 μg/l cutoff. Discussion These data demonstrate that m-cpp cross-reacts in the urine Amphetamines II assay and can be present in the urine of patients taking trazodone at concentrations approaching or exceeding 6700 μg/l. This interference is not identified in the technical documentation (8) for the Amphetamines II assay. We believe the interference has not been previously reported. A previous case report demonstrates a false-positive result on an immunoassay for amphetamines produced by another manufacturer (Triage Drugs of Abuse Panel Plus Tricyclic Antidepressants, Biosite Diagnostics, San Diego, CA) that was attributed to trazodone (12). However, that case involved an overdose of trazodone. That is, although quantitative trazodone levels were not obtained, the patient had ingested 3000 mg of the drug (some of which was likely not absorbed secondary to intervention) (12). Of the three false-positives reported here, two had serum trazodone within the reference range we use ( μg/l) and the third was barely above the reference range (Patient B, 1793 μg/l). A 1996 editorial also notes spurious appearance of positive amphetamine results in the presence of trazodone on routine drug screening with polyclonal anti-bodies, although trazodone levels and the specific nature of the interference are not described (13). As noted previously, m-cpp was found to cross-react in the EMIT II Ecstasy urine immunoassay (4). Of note, our estimate that m-cpp at a concentration of approximately 6700 μg/l could result in a positive result, applies only when the Amphetamines II assay is calibrated at 1000 μg/l. Laboratories selecting the 300 or 500 μg/l calibration (also available for the assay) should expect that considerably lower concentrations of m-cpp may result in false-positive results. Based upon testing of our analytic standards, we estimate that nearly 6700 μg/l m-cpp would be required to result in a positive test result. Patients C and D had m-cpp levels of only 4400 and 2500 μg/l, respectively. Yet, they also were falsepositives. Likewise, the three other urine samples quantified were found to have somewhat lower levels of m-cpp than would be expected based upon their raw immunoreactivities. One likely explanation for this difference is that urine samples likely contain some modified forms of m-cpp (for example, glucuronide or sulfate conjugates) (5,6). Most m-cpp derivates will not be detectable using the HPLC method we used. The HPLC assay employs a hexane extraction at ph > 11. Aromatic ring-hydroxylated metabolites of m-cpp might be immunoreactive, but would likely go undetected in the HPLC assay we used. Similarly, glucuronidated m-cpp metabolites might contribute immunoreactivity in the Amphetamines II immunoassay, but would likely not extract or be detected in the HPLC assay. Primary standards and assays for these probable m-cpp derivatives were not available to us, so we could not test any further. Based on Patient D s m-cpp immunoreactivity, compared to our 6700 μg/l estimate needed for a false-positive, unmodified m-cpp may represent only 30 40% of the Amphetamines II immunoreactivity observed. We should anticipate urine levels of unmodified m-cpp much less than 6700 μg/l to frequently be associated with false-positive Amphetamines II results. In addition to m-cpp derivatives, other cross-reactive substances may be present in the urine and in part account for immunoreactivities higher than expected based upon our measured m-cpp concentration. For example, although trazodone appears to be minimally cross-reactive compared to m-cpp, any parent drug present in the urine would, nonetheless, likely contribute somewhat to the absorbance rate. Likewise, trazodone has metabolites in addition to m-cpp and its derivatives. Perhaps the most important of these is oxotriazolopyridinpropionic acid (OTPA). During N-dealkylation of trazodone (a key metabolic step), one portion of the parent molecule forms m-cpp while the other forms OTPA (5). Hydroxylation of trazodone produces a dihydrodiol metabolite which is also excreted in urine (5). It is unknown whether OPTA or other non-m-cpp metabolites of trazodone cross reacts in the Amphetamines II assay as primary standards were unavailable to us to test this possibility. Nonetheless, testing of the analytic standard proves that m-cpp is responsible for at least substantial immunoreactivity. Whether m-cpp alone or m-cpp in combination with other trazodone derivatives is responsible is unlikely to affect results interpretation in clinical practice. Although it is difficult to determine the precise frequency with which m-cpp interference results in false positive Amphetamines II results, we reviewed data at our institution to provide a rough estimate. During one 26 day period, more than 1200 urine Amphetamines II tests were performed. Of these, 48 were positive and another 44 were partially positive (92 total). We reviewed the electronic medical records for the patients involved in 89 of these 92 cases with significant immunoreactivity. Of the 89 cases reviewed, 27 could be explained by reported use of prescribed or illicit amphetamines or related and known cross-reactive compounds. Of the remaining 367

5 62 samples, at least 21 (8 positive; 13 partially positive; 34% of the 62 total) were from patients believed to be taking trazodone. Similar results were obtained on review of another month-long period. In addition, we postulate that many of these partially positive results likely would have also been resulted as positive, had we selected a less concentrated calibrator. Chart data were analyzed both before our laboratory investigation (using them to generate our hypothesis) and after it (to derive better estimates for the prevalence of this interference). We did not identify patterns within the medication histories of patients with false-positive results to make us especially suspicious of any single medication other than trazodone, and the only other medication we specifically investigated was bupropion. High concentration standards for two bupropion metabolites (threo-hydrobupropion and erythro-hydrobupropion each at 200,000 μg/l) were tested and demonstrated very minimal or no cross-reactivity. Taken together, our findings demonstrate a previously unreported and clinically significant interference in the urine Amphetamines II assay. It may be useful to investigate whether m-cpp also interferes in amphetamine immunoassays produced by other manufacturers. In addition, laboratories should take this information into account when evaluating positive urine amphetamine results from patients taking trazodone. Trazodone (Desyrel) was 6th on a list of the top 25 psychiatric drugs for 2009, with almost 19 million prescriptions (14). Laboratories serving facilities where trazodone is often prescribed may wish to select the 1000 μg/l calibration to reduce the rate of false positives caused by m-cpp. References 1. K.E. Moeller, K.C. Lee, and J.C. Kissack. Urine drug screening: Practical guide for clinicians. Mayo Clin. Proc. 83: (2008). 2. M.R. Pincus and N.Z. Abraham, Jr. Toxicology and therapeutic drug monitoring. In Henry s Clinical Diagnosis and Management by Laboratory Methods, R.A. McPherson, M.R. Pincus, and J.B. Henry, Eds. Saunders Elsevier, Philadelphia, PA, 2007, pp P.R. Puopolo, S.A. Volpicelli, D.M. Johnson, and J.G. Flood. Emergency toxicology testing (detection, confirmation, and quantification) of basic drugs in serum by liquid chromatography with photodiode array detection. Clin. Chem. 37: (1991). 4. B.K. Logan, A.G. Costantino, E.F. Rieders, and D. Sanders. Trazodone, meta-chlorophenylpiperazine (an hallucinogenic drug and trazodone metabolite), and the hallucinogen trifluoromethylphenylpiperazine cross-react with the EMIT II ecstasy immunoassay in urine. J. Anal. Toxicol. 34: (2010). 5. S. Rotzinger, M. Bourin, Y. Akimoto, R.T. Coutts, and G.B. Baker. Metabolism of some second - and fourth -generation antidepressants: iprindole, viloxazine, bupropion, mianserin, maprotiline, trazodone, nefazodone, and venlafaxine. Cell Mol. Neurobiol. 19: (1999). 6. H.H. Maurer, T. Kraemer, D. Springer, and R.F. Staack. Chemistry, pharmacology, toxicology, and hepatic metabolism of designer drugs of the amphetamine (ecstasy), piperazine, and pyrrolidinophenone types: a synopsis. Ther. Drug Monit. 26: (2004). 7. J. Kovaleva, E. Devuyst, P. De Paepe, and A. Verstraete. Acute chlorophenylpiperazine overdose: a case report and review of the literature. Ther. Drug Monit. 30: (2008). 8. Roche Diagnostics. Amphetamines II, (accessed December 2010, log in required). 9. N.T. Lu and B.G. Taylor. Drug screening and confirmation by GC MS: comparison of EMIT II and online KIMS against 10 drugs between US and England laboratories. Forensic Sci. Int. 157: (2006). 10. P.R. Puopolo, M.E. Pothier, S.A. Volpicelli, and J.G. Flood. Single procedure for detection, confirmation, and quantification of benzodiazepines in serum by liquid chromatography with photodiode-array detection. Clin. Chem. 37: (1991). 11. Rochester test catalog: 2011 online test catalog. Mayo medical laboratories, (January 2011). 12. R.J. Roberge, J.R. Luellen, and S. Reed. False-positive amphetamine screen following a trazodone overdose. J. Toxicol. Clin. Toxicol. 39: (2001). 13. R.J. Craig. Urine screening for drugs and trazodone. Br. J. Psychiatry 169: (1996). 14. J.M. Grohol. Top 25 psychiatric prescriptions for 2009, (accessed January 2011). Manuscript received March 1, 2011; revision received April 5,