Court filing
Article - Preston — USA v. Jittaphol (Dkt. 70-2, D. Mass.)
Filed April 6, 2022 in USA v. Jittaphol; one of 82 filings from this case.
Record facts
| Court | U.S. District Court for the District of Massachusetts |
|---|---|
| Filed | 2022-04-06 |
U.S. District Court for the District of Massachusetts · No. 1:21-cr-10270-MLW · Doc. 70-2 · 2022-04-06 · Docket on CourtListener
Full text
Journal of Analytical Toxicology, Vol. 23, September 1999 Monitoring Cocaine Use in Substance-Abuse-Treatment Patients by Sweat and Urine Testing Kenzie L. Preston 1,*, Marilyn A. Huestis 1, Conrad J. Wong 2, Annie Umbricht 1, Bruce A. Goldberger 3, and Edward J. Cone 1 I National Institute on Drug Abuse Intramural Research Program, Baltimore, Maryland; 2Department of Psychology, University of Vermont, Burlington, Vermont; and 3Department of Pathology, Immunology, and Laboratory Medicine, University of Florida College of Medicine, Gainesville, Florida l Abstract ] Sweat and urine specimens were collected from 44 methadone- maintenance patients to evaluate the use of sweat testing to monitor cocaine use. Paired sweat patches that were applied and removed weekly (on Tuesdays) were compared with 3-5 consecutive urine specimens collected Mondays, Wednesdays, and Fridays. All patches (N = 930) were extracted in 2.5 m/of solvent and analyzed by ELISA immunoassay (cutoff concentration 10 ng/mL); a subset of patches (N = 591) was also analyzed by gas chromatography-mass spectrometry (GC-MS) for cocaine, benzoylecgonine (BZE), and ecgonine methyl ester (EME) (cutoff concentration 5 ng/mL). Urine specimens were subjected to qualitative analysis by EMIT (cutoff 300 ng/mL) and subsets were analyzed by TDx (semiquantitative, LOD 30 ng/mL) and by GC-MS for cocaine (LOD 5 ng/mL). Results were evaluated to (1) determine the relative amounts of cocaine and its metabolites in sweat; (2) assess replicability in duplicate patches; (3) compare ELISA and GC-MS results for cocaine in sweat; and (4) compare the detection of cocaine use by sweat and urine testing. Cocaine was detected by GC-MS in 99% of El ISA-positive sweat patches; median concentrations of cocaine, BZE, and EME were 378, 78.7, and 74 ng/mt, respectively. Agreement in duplicate patches was approximately 90% by ELISA analysis. The sensitivity, specificity, and efficiency of sweat ELISA cocaine results as compared with sweat GC-MS results were 93.6%, 91.3%, and 93.2%, respectively. The sensitivity, specificity, and efficiency between ELISA sweat patch and EMIT urine results were 97.6%, 60.5%, and 77.7%, respectively. These results support the use of sweat patches for monitoring cocaine use, though further evaluation is needed. Introduction Cocaine use remains a significant public health problem in the United States with estimates of more than 3.5 million in- . Author to whom correspondence and requests for reprints should be addressed. dividuals using cocaine at least once in the past year and over 500,000 regular users (those who report use on 51 or more days in the past year) (1). Drug testing for cocaine use is em- ployed in many situations, including drug treatment, employ- ment screening, and the criminal justice system. The biological specimen most commonly used for drug testing is urine. Al- though urine testing is reliable, standardized, relatively inex- pensive, and widely available, it does have some disadvantages. For example, the window of detection in urine is dependent on the half-life of the drug and can be relatively short. Because of the widespread use of urine testing, tactics for subverting test results are well known and products for adulteration are mar- keted for sale (2), necessitating close monitoring of specimen collection, sometimes including direct observation. Sweat is an alternative matrix for drug testing that may pro- vide an additional tool for monitoring drug use. A number of drugs of abuse have been found to be excreted in sweat, in- cluding methadone (3,4), amphetamines (5-9), metham- phetamine (6,8,10-12), alcohol (10), morphine (4,10,13,14), heroin (9,15), cocaine (4,9,15-20), nicotine (13,21), and mari- juana (9,13). In addition, sweat on clothing has been used to detect drugs for forensic purposes (10,14,22). A promising method for routine sweat collections appears to be the sweat patch (15,16,23). The sweat patch device consists of an adhesive layer on a thin transparent film of surgical dressing and a rect- angular, absorbent, cellulose pad (14 cm2). The surgical dressing film allows oxygen, carbon dioxide, and water vapor to escape while the nonvolatile constituents in sweat are retained in the absorbent pad. Sweat patches are applied to nonhairy portions of body (e.g., abdomen) after cleansing of skin with an alcohol wipe to avoid contamination and improve adherence. The patches are tamper-resistant; they have identifying serial numbers and once removed cannot be reapplied to the skin. Because the patches are worn for up to one week, the window of detection can be longer than that provided by urine testing. Earlier research showed that cocaine appeared on the sweat Reproduction (photocopyin 8) of editorial content of this journal is prohibited without publisher's permission. 313 Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 1 of 10 patch within 1-2 h of a single dose and that the concentration peaked within 24 h in an apparently dose-dependent manner (15). Parent cocaine was the primary analyte excreted in sweat with smaller amounts of metabolites, ecgonine methyl ester (EME) and benzoylecgonine (BZE), also detected. Gas chro- matography-mass spectrometry (GC-MS) analysis of dupli- cate patches worn at the same time following controlled cocaine-dosing experiments provided similar cocaine concen- trations (15). Burns and Baselt (16) also found that the patch content yielded cocaine concentrations that accurately re- flected cocaine usage. However, the results from both studies (15,16) indicated high between-dose and between-subject vari- ability, leading to the conclusion that the patch technology was best suited to detect drug usage and should not be used to de- termine dose or specific time of use. The present study evaluated the use of sweat patches for monitoring cocaine use in a population of drug abuse treat- ment patients. Duplicate patches were applied to methadone maintenance patients participating in a clinical trial of a be- havioral treatment for drug abuse. Patches were worn for one week, removed, and analyzed for the presence of cocaine and its metabolites. Urine specimens were collected three times per week and analyzed by immunoassay (cocaine metabolite) and GC-MS (cocaine). Results were evaluated to (1) determine the relative amounts of cocaine and its metabolites in sweat, (2) as- sess replicability in duplicate patches, (3) compare two methods (ELISA and GC-MS) of assaying cocaine in sweat, and (4) compare the detection of cocaine use by sweat and urine testing. Methods Participants The participants were 44 patients (29 male, 15 female; 20 African American, 24 Caucasian) who were recruited from among participants in a clinical trial of a behavioral treatment for cocaine and heroin abuse (24). Individuals were eligible for the primary treatment study if they were between the ages of 18 and 65, if they qualified for methadone maintenance ac- cording to the Food and Drug Administration guidelines, and if they reported histories of intravenous opiate use. All pa- tients received standard methadone-maintenance drug-abuse treatment, which included daily methadone (50 to 80 mg/day orally) and weekly individual counseling, throughout the study. This study was approved by the local institutional review board for human research. All volunteers gave informed written con- sent prior to study participation and were paid $10 per week for wearing sweat patches. Details of the experimental treatment procedures have been described (24). Specimen collection Volunteers could participate for a maximum of 17 weeks. Participants visited the clinic seven days per week to receive methadone. Urine specimens were collected three days per week on Monday, Wednesday, and Friday under direct obser- vation by trained staff. An aliquot of each specimen was im- 314 Journal of Analytical Toxicology, Vol. 23, September 1999 mediately labeled and frozen at-30~ for later analyses; the re- mainder of the urine specimen was refrigerated for qualitative analysis within 24 h. Sweat patches were applied each Tuesday to the subject's lower abdomen and back after the skin was cleaned with an al- cohol wipe. The study number, the subject's identification code number, patch code number, and the date of application were recorded on a 3 x 5-in. index card. The following Tuesday, prior to removing the sweat patch, a technician examined the patch to determine its condition. Patch condition was rated ac- cording to the following categories: I, intact patch; C, curling (edges); E, exposed patch edge; M, missing patch. To remove patches, the technicians pulled the adhesive edge along the side of the patch. Once the adhesive along the side of the patch was away from the skin, the patch was pulled outward from both top corners equally on both the sides of the patch. Techni- cians were instructed to wear gloves and not to contaminate the absorbent pad by touching it. Technicians placed the patch adhesive-side down in the center of the index card, recorded the date the patch was removed and the condition of the patch, and placed the card in a resealable bag. The bag was sealed and stored in a refrigerator until it was transferred to a freezer at the end of the day. Urine specimen analyses All urine specimens were analyzed qualitatively for cocaine metabolite (EMIT d.a.u. TM, Behring Diagnostics, San Jose, CA) with a cutoff concentration of 300 ng/mL for cocaine metabo- lite. At the conclusion of the study, a subset of urine specimens were analyzed for BZE equivalents by fluorescence polarization immunoassay (FPIA) and for cocaine by GC-MS. Semiquanti- tative testing was conducted on freshly thawed aliquots. FPIA was performed with TDx | Cocaine Metabolite Assay reagents on a TDx instrument (Abbott Laboratories, Abbott Park, IL) ac- cording to manufacturer's recommended procedures. The cross-reactivity of this assay for BZE was 100% and < 1% for cocaine, EME, and ecgonine (25). The limit of detection of the assay was 30 ng/mL; the linear range for cocaine metabolite was 30 to 5000 ng/mL. Specimens that contained BZE equiv- alents greater than 5000 ng/mL were diluted with reagent buffer until results fell within the linear range of the assay. The 300-ng/mL cutoff concentration was also applied to the FPIA results. Urine cocaine concentrations were determined by a GC-MS procedure designed for rapid measurement of cocaine in urine specimens (26). The procedure consisted of a liquid-liquid extraction, followed by GC-MS analysis (Hewlett- Packard 5890 GC interfaced to a Hewlett-Packard model 5972 mass selective detector [MSD], Hewlett-Packard, Little Falls, DE) operated in the selected ion monitoring mode. Standard curves were prepared in blank urine with authentic drug stan- dards in eight concentrations. The limit of detection was ap- proximately 5 ng/mL, and the limit of quantitation was 25 ng/mL. Sweat patch analyses At the conclusion of the study, patches were sent to Pharm- chem Laboratories, Inc. (Menlo Park, CA) for analysis. The ab- sorbent pad was removed from the adhesive layer, placed in a Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 2 of 10 Journal of Analytical Toxicology, Vol. 23, September 1999 5-mL screw-cap plastic tube with 2.5 mL of a sweat extraction buffer (75% methanol/25% 0.2M sodium acetate [pH 5.0]), and shaken for 30 min to extract cocaine and its metabolites. The eluent was then analyzed according to package directions by competitive enzyme immunoassay (ELISA) with the STC Cocaine Metabolite Micro-Plate EIA (STC Diagnostics, Beth- lehem, PA). ELISA sweat results are reported as nanograms per milliliter of sweat extraction buffer according to the stan- dardized reporting format. The cross-reactivity of this assay was 100% for BZE, 102% for cocaine, 18% for EME, 20% for ecgonine, and 143% for cocaethylene (25). A subset of sweat specimens (591 patches) was confirmed by GC-MS analysis op- erated in the selected ion monitoring mode (Hewlett-Packard 5890 GC interfaced to a Hewlett-Packard MSD). Limits of de- tection were 3 ng/mL for cocaine and 2 ng/mL for BZE and EME. Limits of quantitation were 4 ng/mL for cocaine and 2 ng/mL for BZE and EME. Cutoff concentrations for positive specimens were 10 ng/mL for ELISA and 5 ng/mL for GC-MS. Data analysis Four main analyses were conducted: identification and quantitation of cocaine analytes in sweat by GC-MS; compar- ison of replicate patch results; comparison of detection of co- caine use in sweat by ELISA and GC-MS; and comparison of sweat test results to results from different combinations of urine collection periods. Median and maximum concentra- tions of cocaine analytes by GC-MS were determined. In those patches that were positive for cocaine and/or its metabolites, the percent positive for each combination of analytes was de- termined. Results of paired patches (front and back) were com- pared as ratios of analyte concentrations in front and back patches and by identifying those with discrepant results (i.e., one positive and one negative). For comparison of GC-MS and ELISA sweat results, a patch was considered positive by GC-MS if cocaine, BZE, or EME were detected at or above 5 ng/mL and positive by ELISA if re- suits were greater than or equal to 10 ng/mL. A true positive was defined as positive by both GC-MS and ELISA; a true neg- ative was defined as negative by both GC-MS and ELISA. A false positive was defined as a positive ELISA result and nega- tive GC-MS result; and a false negative was defined as negative ELISA result and positive GC-MS result. Sensitivity of the assay was calculated as true positives divided by true positives plus false negatives. Specificity was calculated as true negatives di- vided by true negatives plus false positives. Efficiency was cal- culated as true positives plus true negatives divided by the total number of analyzed patches. All ratios were multiplied by 100. The results of urine cocaine metabolite tests (EMIT) for four different monitoring intervals were compared to the results of sweat cocaine tests (ELISA) of patches worn for seven days. Sen- sitivity, specificity and efficiency were determined as described for four urine collection intervals: Wednesday, Friday, and Monday (WFM) urine collections after the Tuesday sweat patch application; Monday before and Wednesday, Friday, and Monday (MWFM) after the Tuesday patch application; Wednesday, Friday, Monday, and Wednesday (WFMW) after the Tuesday patch application; and Monday before and Wednesday, Friday, Monday, and Wednesday (MWFMW) after the Tuesday patch application (Figure 1). Comparisons were made only for those patch specimens for which all urine specimens collected during the comparison intervals were included. Actual cocaine use was assumed if any urine specimen collected in the interval tested positive for cocaine metabolite by EMIT; if all urine spec- imens were negative, drug abstinence was assumed. Sweat was determined to be positive if at least one of the two paired patches was positive for cocaine by ELISA (cutoff concentration 10 ng/mL). For the purpose of this comparison, a true positive was assigned if at least one patch and one urine specimen were positive; a false positive was assigned if at least one patch was positive, but all urine specimens were negative; a true negative was assigned if both patches and all urine specimens were neg- ative; and a false negative was assigned if both patches were neg- ative and at least one urine specimen was positive. Results Analyte analysis in sweat A total of 930 sweat patches was collected from 44 partici- pants and analyzed by ELISA; 611 (66%) specimens tested positive for cocaine and/or metabolites at concentrations > 10 ng/mL. A subset of 591 patches were analyzed by GC-MS for cocaine, BZE, and EME. Of the 591 sweat specimens, 499 (84.4%) tested positive (> 5 ng/mL) for one or more analytes and 92 were negative for all analytes. There were 362 specimens positive for all three analytes and 21 positive for both cocaine and BZE; 111 tested positive for cocaine only, and 5 tested pos- itive for BZE only. No specimens were positive for EME only, for EME and BZE only, or for cocaine and EME only. In the co- caine-positive sweat specimens (3/= 494), cocaine concentra- tions (plus or minus standard deviation) were as follows: mean 984.6 • 1005.8 ng/mL; median 378 ng/mL; range 5-26,490 ng/mL. In the BZE-positive sweat specimens (3/= 388), BZE concentrations were as follows: mean 133.7 • 108.3 ng/mL; Monitoring windows Sweat patch Sweat patch application Urine removal ~' collections ~1' M T W Th F S S M T W WFM MWFM ~- Ira, WFMW MWFMW Figure 1. Diagram of sweat patch and urine specimen collections. De- tection of cocaine in patches was compared with that detected in urine specimens collected over the four time intervals indicated by the hori- zontal lines. M, Monday; W, Wednesday; F, Friday. 315 Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 3 of 10 Journal of Analytical Toxicology, Vol. 23, September 1999 Subject A Subject B 100000 -]Cocaine - ELISA ~ ~ 1000-lCocalne - ELISA 10000"~ t ~ (Z "e .,ooo4 /I ~ 100 0 100000 ]Cocaine - GC-MS 1000 "1 Cocaine - GC--MS / t 10000 J ,oo 1ool b~./~ 10 0 , 0 100000 1BZE . GC-MS 100001 ~~176176 1 0 0 ~ . . . . . . 1000- BZE - GC-MS 100 0 ,-I E C 100000 EME- GC-MS 1000 EME- GC-MS 10000 1000 100 1 0 0 ~ 10 0 0 2 4 6 8 10 12 14 16 2 4 6 8 10 12 14 16 Weeks Weeks Figure 2. Quantitative results of sequential sweat patches collected from selected treatment patients over 17 weeks. Cocaine equivalent (ELISA), cocaine (GC-MS), BZE (GC-MS), and EME (GC-MS) concentrations were determined in replicate sweat patches applied to the front (closed symbols) and back (open symbols) abdomen for one week. Horizontal dashed lines indicate cutoff concentrations for positive specimens. Data points are not connected where spec- imen results were not available. 316 Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 4 of 10 Journal of Analytical Toxicology, Vol. 23, September 1999 median 78.7 ng/mL; range 5-2150 ng/mL. In the EME-positive sweat specimens (N = 362), EME concentrations were as follows: mean 116.0 • 90.3 ng/mL; median 74 ng/mL; range 5-774 ng/mL. Patterns of cocaine, BZE, and EME concentrations in sweat collected in patches applied to the abdomen and to the back across the course of the 17-week trial are illustrated for two representative patients in Figure 2. Cutoff concentrations for positive specimens are indicated by dashed lines (ELISA 10 Table I. Comparison of Paired Sweat Patches Applied to Abdomen and Lower Back with Discrepant ELISA Results* Abdomen Back ng/mL; GC-MS 5 ng/mL). Subject A had generally high levels of cocaine use during the study, indicating frequent, heavy use of cocaine. Cocaine, BZE, and EME were present in con- centrations well above the limit of quantitation in all but 1-3 specimens during this period. In contrast, Subject B had de- creasing cocaine use in the first half of the treatment and a low level of use in the last half of treatment. After the first three weeks of treatment, patch cocaine concentrations remained close to the cutoff values for both ELISA and GC-MS. Cocaine, BZE, and EME (GC-MS) concentrations varied over time in patterns similar to one another, with cocaine present in the highest quantity and EME present in the the lowest quantity. Patch EtlSA GC-MS ELISA GC-MS pair COC t COC BZE EME COC t COC BZE EME I 5 - - - 15 7 0 0 2 19 - - - 9 - - - 3 7 - - - 12 - - - 4 14 - - - 6 - - - 5 7 - - - 11 - - - 6 17 8 3 1 7 5 1 1 7 11 3 2 1 8 4 2 1 8 2 - - - 12 - - - 9 7 - - - 10 - - - 10 8 - - - 17 - - - 11 29 - - - 5 - - - 12 I0 - - - 5 - - - 13 14 - - - 9 - - - 14 7 - - - 13 - - - 15 13 - - - 5 - - - 16 9 6 1 0 27 12 3 0 17 10 5 0 0 4 - - - 18 10 4 1 0 5 3 2 0 19 10 5 0 0 6 5 0 0 20 15 5 1 0 9 8 2 0 21 8 - 13 - - - 22 5 0 0 0 12 8 0 0 23 4 - 16 - - - 24 6 - 15 - - - 25 8 - 21 - - - 26 5 0 0 0 14 14 4 I 27 6 7 0 0 26 79 5 I 28 10 0 0 0 6 0 0 0 29 10 - 7 - - - 30 8 - 16 0 18 0 31 6 0 0 0 10 0 0 0 32 7 0 0 0 18 33 4 0 33 23 - 5 - - - 34 17 40 1 0 8 8 0 0 35 7000 2340 2150 132 3 - - - 36 10 - 7 - - - 37 5 7 0 2 241 298 229 159 38 10 18 2 0 4 5 0 0 39 17 53 5 3 5 7 0 0 * All values are nanograms per milliliter. All specimens were tested by ELISA; a subset was confirmed by GC-MS; these results are shown where available. Specimens not analyzed by GC-MS are indicated by dashes (-), Eight additional discrepant specimen pairs with cocaine equivalent concentrations (ELISA) within 10% of the I 0-nglmL cutoff (8 to 12 nglmL) are not listed in the Table. t Cocaine equivalent. Comparison of duplicate sweat patches The replicability of patch testing was evalu- ated by comparing paired patches placed on the abdomen and lower back. The concentra- tions of cocaine and its metabolites tended to be similar in patches applied to the abdomen and lower back, as illustrated for two subjects in Figure 2. ELISA results were available for 455 pairs of patches. At the 10-ng/mL cutoff, both patches were positive in 278 pairs and both were negative in 130 pairs. Discrepant results (one positive and one negative) oc- curred in 47 pairs (10.2%). ELISA results in eight of these pairs were within 20% of the 10-ng/mL cutoff (8 to 12 ng/mL). The quanti- tative results for the remaining 39 pairs are listed in Table I. All but two specimen pairs were in similar concentration ranges. Two specimen pairs (numbered 35 and 37) showed substantial differences in concentration be- tween the front and back patches, suggesting the possibility of environmental contamina- tion. Among the 591 patches analyzed by GC-MS, there were 287 replicate pairs. The repro- ducibility of sweat patch results among these pairs was evaluated by comparing cocaine analyte concentrations. Discrepant results (one positive and one negative at the 5-ng/mL cutoff concentration) occurred in only six pairs (2%) of specimens (Table II). Sixfold or greater differences in cocaine concentration occurred in 12 pairs (4%) (Table II). Six of the 18 pairs listed in Table II were discrepant by ELISA analysis and are indicated by asterisks. Comparison of ELISA and GC-MS sweat results The presence of cocaine and/or its metabo- lites was determined by both ELISA and GC-MS in 591 sweat patch specimens. A total of 475 (79%) of the 591 specimens tested pos- itive by ELISA (_> 10 ng/mL cutoff); of these, 317 Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 5 of 10 Journal of Analytical Toxicology, Vol. 23, September 1999 Table II. Discrepant Sweat Patch Pairs and Pairs with Sixfold or Greater Differences in Cocaine Concentration Determined by GC-MS* Abdomen Back Patch Cocaine BZE EME Cocaine t Cocaine BZE EME Cocaine t Ratio pair GC-MS GC-MS GC-MS ELISA GC-MS GC-MS GC-MS ELISA Abdomen/Back Discrepant pairs 1 5 0 0 2 8 0 0 3 8 0 0 4 t 0 0 0 5' 0 0 0 6* 0 0 0 Sixfolddifferences 1 + 7 0 2 2 642 461 28 3t 7 0 0 4 128 7 11 5 10 11 3 6 352 42 30 7* 53 5 3 8 1747 110 51 9 175 0 0 10 197 9 10 11 780 43 48 12 26490 1661 264 8 0 0 0 5 9 0 0 0 3 5 0 0 0 7 5 8 0 0 12 5 14 4 1 14 7 33 4 0 18 * All values are nanograms per milliliter. f ELISA values discrepant. 5 298 229 159 241 0.024 916 22160 575 257 10000 0.029 6 79 5 1 26 0.092 58 1065 72 70 62 0.120 15 61 5 3 24 0.162 485 55 5 5 44 6.308 17 7 0 0 5 7.026 495 216 30 0 65 8.088 45 20 1 0 14 8.578 39 17 2 0 11 11.193 2619 22 2 I 28 35.470 7000 456 36 19 121 58.092 Sublect C Subject D 10,000 "1 Sweat 10,000 - I000- 100- 10- 1,000,000 - Sweat 100,000- . 10,000- 1000 - " 100" 10" 0 l I looo 100 - - 10- 0 | i l i i i l i i i i M i i i i i l J l i ! ! | ! i l i i i l | i i i 1 2 3 4 5 6 7 8 91011121314151617 2 3 4 5 6 7 8 91011121314151617 Urine Urine 1,000,000. r Cocaine, i00,000. D ~ ~ i ~ ~~ii qb_~;~.._ 9 ,oooo I000' ,~_ 10' --, , , ;7 i=i o . . . . . . . . . . . . . . . . 2 3 4 5 6 7 8 91011121314151617 2 3 4 5 6 7 8 9101 121314151617 Treatment weeks Treatment weeks Figure 3. Quantitative results of sequential sweat patch (top panels) and urine specimens (bottom panels) collected from selected treatment patients over 17 weeks. Cocaine equivalent concentrations (ELISA) were determined in sweat patches applied for one week on Tuesdays. BZE equivalent (open symbols; FPIA) and cocaine concentrations (closed symbols; GC-MS) were determined in urine specimens collected Mondays, Wednesdays, and Fridays. Horizontal dashed lines indicate cutoff concentrations for positive specimens (10 ng/mL for sweat patches; 300 ng/mL for immunoassay analysis of urine specimens; 5 ng/mL for GC-MS analysis of urine specimens). Data points are not connected where specimen results were not available. 318 Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 6 of 10 Journal of Analytical Toxicology, Vol. 23, September 1999 467 (98.3%) were positive by GC-MS (_> 5 ng/mL). In the entire sample set (N = 591), 79% were identified as true positives (ELISA positive and GC-MS positive), 1.4% as false positives (ELISA positive and GC-MS negative), 5.4% as false negative (ELISA negative and GC-MS positive), 14.2% as true negative (ELISA negative and GC-MS negative). Overall, sensitivity was 93.6%, specificity was 91.3%, and efficiency was 93.2%. The concentrations determined by ELISA (ascending order) and GC-MS for cocaine, BZE, and EME for the 32 specimens Table III. Concordance Between ELISA and GC-MS Analyses of Sweat Patches for Cocaine Monitoring: False Negatives and False Positives* ELISA GC-MS Specimen COC t COC BZE EME False negatives 1 3 35 1 0 2 3 5 0 1 3 4 6 0 1 4 4 12 3 3 5 4 5 0 0 6 4 5 0 0 7 5 5 0 1 8 5 8 0 0 9 5 8 0 0 10 5 7 0 0 11 5 7 0 2 12 6 7 0 0 13 6 9 0 0 14 6 5 0 0 15 7 4 5 0 16 7 12 1 0 17 7 5 1 1 18 7 20 0 0 19 7 9 1 1 20 7 6 1 1 21 8 5 0 0 22 8 11 3 0 23 8 11 1 0 24 8 0 5 0 25 8 8 0 0 26 9 6 1 0 27 9 6 0 0 28 9 8 0 0 29 9 12 0 0 30 9 13 1 0 31 9 4 8 0 32 9 8 2 0 False positives 33 10 4 1 0 34 10 0 0 0 35 10 0 0 0 36 10 0 0 0 37 11 3 2 1 38 11 0 0 0 39 13 4 0 0 40 25 4 0 0 * All values are nanograms per milliliter. ~" Cocaine equivalent concentrations. identified as false negatives and the eight specimens identified as false positives are listed in Table III. GC-MS-determined cocaine concentrations were greater than 10 ng/mL in only eight of the false-negative specimens. Three of the false-nega- tive specimens were positive for BZE only, and 29 were positive for cocaine only. There were nine replicate pairs (i.e., paired front and back applied the same week to the same subject) among the 40 specimens identified as false negative or false positive. In all cases, the paired specimens were either both false positive (one pair) or both false negative (eight pairs). In 7 of the 8 false positive sweat tests, ELISA results were lower than 15 ng/mL. Detection of cocaine use by sweat versus urine testing Figure 3 illustrates two representative patients' patterns of cocaine concentrations in sweat collected in patches applied for one-week periods to the abdomen and cocaine and BZE con- centrations in urine specimens collected three times weekly over the course of the 17-week trial. Dashed lines indicate the 10- and 300-ng/mL cutoff concentrations for the ELISA anal- yses in sweat and FPIA analyses in urine, respectively. Semi- quantitative cocaine concentrations by ELISA are included for comparison. Subject C used cocaine frequently in the first half of treatment and very infrequently in the last half of treat- ment. After the first eight weeks of treatment, patch cocaine concentrations remained close to the cutoff values, and all urine specimens were negative for cocaine use. Sweat con- tained cocaine above cutoff concentrations in at least three weeks during which no urine specimens tested positive for cocaine use. Subject D had generally high concentrations of cocaine metabolites and/or cocaine in both urine and sweat throughout the study, indicating frequent, heavy use of co- caine. Only two urine specimens tested below the 300-ng/mL cutoff concentration for BZE, and cocaine was above the 5-ng/mL limit of detection in many urine specimens. Cocaine was present in concentrations well above the 10-ng/mL ELISA cutoff concentration in all sweat specimens during this period. The relationship between the urine results and sweat re- sults was evaluated by comparing ELISA sweat results (10- ng/mL cutoff) from 355 patch pairs to EMIT results of urine Table IV. Mean Concordance of Cocaine Use Detection in Sweat and Urine in 44 Participants* WFM MWFM WFMW MWFMW Sensitivity 97.6% 96.5% 97.0% 96.0% Specificity 60.5% 62.1% 61.3% 62.9% Efficiency 77.7% 78.9.% 78.3% 79.4% True Positive t 45.4% 47.0% 46.2% 47.9% False Positive 21.1% 19.4% 20.3% 18.6% False Negative 1.1% 1.7% 1.4% 2.0% True Negative 32.4% 31.8% 32.1% 31.5% * Sweat specimens were analyzed by ELISA (10-ng/mL cutoff); urine specimens were analyzed by EMIT (300-ng/mL cutoff). True positive, at least one patch and one urine specimen were positive; false positive, at least one patch was positive and all urine specimens were negative; false negative, if both patches were negative and at least one urine specimen was positive; true negative, both patches and all urine specimens were negative. 319 Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 7 of 10 specimens collected over four different time periods: WFM, WFMW, MWFM, and MWFMW (see Figure 1 for timeline). Con- cordance between sweat and urine was acceptable and consis- tent across all four urine collection intervals (Table IV) with sensitivity ranging from 96.0 to 97.6%, specificity ranging from 60.5 to 62.9%, and efficiency ranging from 77.7 to 79.4% at the 10-ng/mL ELISA cutoff. Concordance was reduced when a 5-ng/mL ELISA cutoff was applied: 45.4% true positive; 40.3% false positive; 0.6% false negative; 13.2% true negative; 98.8% sensitivity; 24.7% specificity; and 58.9% efficiency for the WFM urine collection interval. At least one urine specimen tested positive for cocaine by EMIT at 300 ng/mL in 165 out of 355 sets of WFM urine spec- imens. This frequency of positives was lower than that found in the corresponding sweat patch specimens in which 236 of 355 patch pairs were found to have ELISA cocaine concentrations above 10 ng/mL. Comparison of ELISA sweat results to EMIT urine results gave a 45.4% true-positive rate, a 21.1% false-pos- itive rate, a 1.1% false-negative rate, and a 32.4% true-negative rate for the WFM urine comparison. In 75 cases, at least one of the ELISA sweat patches had a result greater than 10 ng/mL when the EMIT results were less than 300 ng/mL for all three urine specimens. Additional information was obtained to eval- uate these 75 cases. In 73 of the 75 cases, duplicate sweat patches were analyzed by ELISA. In only half of these 73 sweat patch pairs, the ELISA cocaine concentration was found to be greater than the 10-ng/mL cutoff concentration indicating low sweat cocaine concentrations in the patches. Urine FPIA test results were available for 33% (25 of 75) of the false positive cases; 40% (10 of 25) of these had at least one of three urine specimens with a concentration greater than 30 ng/mL, which is above the limit of detection but below the federally mandated cutoff of 300 ng/mL, indicating low level cocaine metabolite concentrations in the urine. GC-MS sweat analyses were avail- able for 56% (42 of 75) of the false-positive cases; 93% (39 of 42) were confirmed positive for cocaine, BZE, and/or EME above 5 ng/mL, substantiating the ELISA sweat results. Discussion Biological monitoring (drug testing) is needed in treatment to assess patient outcome. Urine testing is the current, stan- dard method for monitoring drug use, but it suffers from some disadvantages such as short detection time and difficulties in collection of unadulterated specimens. Thus, better methods for monitoring drug use are needed. This study is part of an on- going program to develop improved biological monitoring procedures. In the present study, we compared sweat testing (ELISA and GC-MS) to urine testing (EMIT, FPIA, and GC-MS) for monitoring cocaine use. High concentrations of cocaine and metabolites were de- tected in sweat by GC-MS collected from outpatient-treat- ment patients. The major analyte present was cocaine; similar results were found following acute administration of cocaine in controlled-clinical studies (15). Of the positive sweat speci- mens analyzed by GC-MS, only five specimens (1%) were iden- 320 Journal of Analytical Toxicology, Vol. 23, September 1999 tiffed as containing BZE without cocaine. BZE was frequently present in concentrations similar to or higher than those of EME. No specimens that contained EME in the absence of co- caine were identified. BZE was also found in higher concen- trations than EME in routine specimens from criminal justice and treatment settings (27). However, two earlier studies re- ported that EME was present in sweat in greater concentrations than BZE (15,23). The reason for this difference is not clear. All three studies used sweat patches manufactured by the same company. The participants in the study by Kintz (23) were drawn from a population of intravenous heroin abusers in a detoxification center; the cocaine detected in sweat patches was apparently from street use, as was true in our study. In the study reported by Cone et al. (15), only single doses of cocaine were administered. It is possible that the frequency of cocaine use in the current study was higher than either of these studies. With a higher frequency of use, BZE might selectively accumulate relative to EME because of its longer half-life. There may also have been greater degradation of cocaine to BZE in the present study than in the cocaine-administration study because degradation could have occurred while patients were wearing the sweat patches as well as during storage. Duplicate patches were applied to subjects in the current study to assess replicability of results from different regions of the body. The replicability of duplicate patches applied to the ab- domen and lower back was high in the present study, which is consistent with the finding of relatively low intrasubject vari- ability in quantitative assessment of duplicate patches in par- ticipants who received acute administration of cocaine in a laboratory setting (15). Although discrepant results (one posi- tive and one negative) occurred in about 10% of 455 pairs an- alyzed by ELISA, the discrepancies tended to occur for specimens with cocaine concentrations (ELISA) close to the 10- ng/mL cutoff concentration. Out of 47 discrepant pairs, eight pairs were within 20% of the 10-ng/mL cutoff (8-12 ng/mL) in both specimens, and only two pairs showed evidence that the discrepancy might have been due to contamination. Sixfold or greater differences in cocaine concentration occurred in only 4% of 287 pairs analyzed by GC-MS. The high replicability of the patch pairs suggests that the placement of patches does not significantly affect the outcome and therefore supports the utility of the procedure for monitoring drug use. GC-MS confirmation of ELISA sweat results was 93.2% with low percentages of false-positive (5.4%) and false-negative (1.4%) screening results. Therefore, the ELISA assay with a 10-ng/mL cutoff concentration appears to be a sensitive and specific screening assay for detecting cocaine and metabolites in sweat patches Because patches were applied on Tuesdays and urine speci- mens were collected on a Wednesday-Friday-Monday schedule, and given that cocaine and/or its metabolites reside in the body for several days, it was unclear which set of urine speci- mens should be compared with the patch data. Drugs taken after the Monday urine specimen and before patch removal on Tuesday could be detectable in the patch. Comparison of sweat patch results to four different intervals of urine monitoring re- suits showed high sensitivity, specificity, and efficiency for all four sets of results. Overall, sweat patches applied on Tuesday Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 8 of 10 Journal of Analytical Toxicology, Vol. 23, September 1999 and removed the following Tuesday adequately reflected co- caine use detected in the urine specimens collected on the Wednesday, Friday, and Monday while the patches were worn. Comparison of detection of cocaine use in sweat (ELISA, 10 ng/mL) to urine (EMIT, 300 ng/mL) resulted in an 80% con- cordance. More cocaine use was detected by sweat patch testing than by urine testing. The majority of sweat patches found positive for cocaine by ELISA in the absence of a cocaine-pos- itive urine were confirmed positive by GC-MS, suggesting that the occasions of drug use detected with sweat but not urine testing were true positives. Also supportive of some cocaine use was the finding that urine BZE concentrations (FPIA) were be- tween the 30-ng/mL limit of detection and 300-ng/mL cutoff concentration for a substantial proportion of the false-positive sweat patches. Increased detection of cocaine use with sweat testing may be due to higher sensitivity, external contamina- tion of the patch during application or removal, and/or differ- ences in analytical cutoffs. The correlation between sweat and urine cocaine results was higher with a 10-ng/mL ELISA cutoff concentration than with a 5-ng/mL cutoff. Few problems were encountered in using the patches in the outpatient population. In one patient, patches did not reliably adhere over a seven-day period and participation in the study was terminated. This individual was moderately obese and worked in a position that required repetitive motion that caused rubbing of the patch against his clothing. With a second subject, a skin reaction developed at the site of the patch. The rash appeared to be due to the use of the alcohol on the skin for cleansing; the rash resolved completely when the patch use was discontinued. Potentially more problematic was the possible external con- tamination of patches during application to the skin and re- moval. However, rates of discrepant results were quite low when handled properly according to manufacturer's direc- tions. False-positive sweat patches (compared to urine) were usually associated with low BZE urine concentrations, sug- gesting a threshold phenomenon rather than contamination. Only 12 out of 287 sweat patch pairs analyzed by GC-MS had greater than sixfold differences in cocaine concentration. In summary, sweat testing provides a useful alternative for monitoring drug usage. The patches were easy to apply and adhered appropriately to most participants. Cocaine in sweat detected by ELISA at a 10-ng/mL cutoff concentration was reliably confirmed by GC-MS at a 5-ng/mL cutoff concentra- tion. Finally, there was good correspondence in detection of co- caine use between thrice-weekly urine monitoring and weekty sweat patch monitoring. In fact, sweat patches appeared to detect more cocaine use than the frequent urine monitoring. Thus, sweat patch testing required fewer patient visits while having a higher detection rate than urine monitoring. Acknowledgment This study was supported by the Intramural Research Pro- gram of the National Institute on Drug Abuse. Sweat patches and analyses of patches were provided by Pharmchem Labora- tories, Inc. (Menlo, CA). We are grateful to Drs. Kenneth Sil- verman and Charles R. Schuster who helped design the clinical trial. References 1. Substance Abuse and Mental Health Services Administration (SAMHSA). National household survey on drug abuse. Population estimates 1995. U.S. Department of Health and Human Services, Public Health Service, Rockville, MD, 1996. 2. K.D. Eskridge and S.K. Guthrie. Clinical issues associated with urine testing of substances of abuse. Pharmacotherapy 17" 497-510 (1997). 3. G.L. Henderson and B.K. Wilson. Excretion of methadone and metabolites in human sweat. Res. Commun. Chem. Pathol. Phar- macol. 5:1-8 (1973). 4. V.S. Balabanova, E. Schneider, R. Wepler, B. Hermann, H.J. Boscher, and H. Scheitler. Die Bedeutung der Drogenbes- timmung in Pilocarpinschweiis fur den Nachweis eines zuruck- liegenden Drogenkonsums. Beitr. Gerichtl. Med. 50:111-115 (1992). 5. T.B. Vree, A.T. Muskens, and J.M. Van Rossum. Excretion of am- phetamines in human sweat. Arch. Int. Pharmacodyn. Ther. 199: 311-317 (1972). 6. J. Fay, R. Fogerson, D. Schoendorfer, R.S. Niedbala, and V. Spiehler. Detection of methamphetamine in sweat by EIA and GC-MS. J. Anal. Toxicol. 20:398-403 (1996). 7. T. Yamamoto, M. Terada, S. Yoshimura, 1-. Sato, H. Kitagawa, T. Yoshida, K. Aoki, and Y. Kuroiwa. Determination of sympath- omimetic amine by gas chromatography-mass spectrometry and the detection of the amine using this method from the gauze used for wiping the face and neck of habitual suspects. Eisei Kagaku 27:331-334 (1981 ). 8. K. Takahashi, M. Shimamine, M. Ono, Y. Kawasaki, K. Sekita, and T. Furuya. Microanalysis of amphetamines II detection of am- phetamines in methamphetamine administered monkey's body fluids. Eisei Shikenjo Hokoku 101 : 14-18 (1983). 9. P. Kintz, A. Tracqui, and P. Mangin. Sweat testing in opioid users with a sweat patch. J. Anal. Toxicol. 20:393-397 (1996). 10. I. Ishiyama, T. Nagai, E. Komuro, T. Momose, and N. Akimori. The significance of drug analysis of sweat in respect to rapid screening for drug abuse. Z. Rechtsmed. 82:251-256 (1979). 11. S. Suzuki, T. Inoue, H. Hori, and S. Inayama. Analysis of metham- phetamine in hair, nail, sweat, and saliva by mass fragmentog- raphy. J. Anal. Toxicol. 13:176-178 (1989). 12. K. Takahashi. Determination of methamphetamine and am- phetamine in biological fluids and hair by gas chromatography. Nippon Hoigaku Zasshi 38:319-336 (1984). 13. S. Balabanova and E. Schneider. Nachweis yon drogen im schweis. Beitr. Gerichtl. Med. 48:45-49 (1989). 14. A. Tracqui, R Kintz, B. Ludes, C. Jamey, and R Mangin. The de- tection of opiate drugs in nontraditional specimens (clothing): a report of ten cases. J. Forensic Sci. 40:263-265 (1995). 15. E.J. Cone, M.J. Hillsgrove, A.J. Jenkins, R.M. Keenan, and W.D. Darwin. Sweat testing for heroin, cocaine, and metabolites. J. Anal. Toxicol. 18:298-305 (1994). 16. M. Burns and R.C. Baselt. Monitoring drug use with a sweat patch: an experiment with cocaine. J. Anal. Toxicol. 19:41-48 (1995). 17. S. Balabanova, E. Schneider, G. Buhler, and H. Krause. Detection of cocaine in human sweat. Lab. Med. 13:479-481 (1989). 18. V. Spiehler, J. Fay, R. Fogerson, D. Schoendorfer, and R.S. Nied- bala. Enzyme immunoassay validation for qualitative detection of cocaine in sweat. Clin. Chem. 42:34-38 (1996). 19. D.A. Kidwell, M.A. Blanco, and F.P. Smith. Cocaine detection in 321 Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 9 of 10 Journal of Analytical Toxicology, Vol. 23, September 1999 a university population by hair analysis and skin swab testing. Forensic Sci. Int. 84:75-86 (1997). 20. G.L. Henderson, M.R. Harkey, C. Zhou, R.T. Jones, and P. Jacob, III. Incorporation of isotopically labeled cocaine into human hair: race as a factor. J. Anal. Toxicol. 22:156-165 (1998). 21. S.H. Gwent, J.F. Wilson, L.M. Tsanaclis, and J.F.C. Wicks. Time course of appearance of continine in human beard hair after a single dose of nicotine. Ther. Drug Monit. 17:195-198 (1995). 22. F.R Smith and R.H. Liu. Detection of cocaine metabolite in per- spiration stain, menstrual bloodstain, and hair. J. Forensic Sci. 31: 1269-1273 (1986). 23. P. Kintz. Drug testing in addicts: a comparison between urine, sweat, and hair. Ther. Drug Monit. 18:450-455 (1996). 24. K. Silverman, C.J. Wong, A. Umbricht-Schneiter, I.D. Montoya, C.R. Schuster, and K.L. Preston. Broad beneficial effects of cocaine abstinence reinforcement among methadone patients. J. Consult. Clin. Psych. 66:811-824 (1998). 25. E.J. Cone, S.L. Menchen, and J. Mitchell. Validity testing of the TDX cocaine metabolite assay with human specimens obtained after intravenous cocaine administration. Forensic Sci. Int. 37: 265-275 (1988). 26. D. Garside, B.A. Goldberger, K.L. Preston, and E.J. Cone. Rapid liquid-liquid extraction of cocaine from urine for gas chromato- graphic-mass spectrometric analysis. J. Chromatogr. 692:61-65 (1997). 27. N. Fortner, PharmChem Laboratories, Inc., Menlo Park, CA, per- sonal communication, 1999. Manuscript received December 8, 1998; revision received February 8, 1999. 322 Downloaded from https://academic.oup.com/jat/article/23/5/313/775779 by guest on 05 April 2022 Case 1:21-cr-10270-MLW Document 70-2 Filed 04/06/22 Page 10 of 10
File and source
- File
- gov.uscourts.mad.237964.70.2.pdf
- Size
- 920,959 bytes
- SHA-256
- 55ae5643153d4972c10c782a423ec58397fb04e1e77e93f034d59435bc0bb0ac
- Our copy
- gov.uscourts.mad.237964.70.2.pdf
- Original
- PACER (login required)