Cocaine profiling for strategic intelligence purposes, a cross-border project between France and Switzerland. Part I. Optimisation and harmonisation of the profiling method

Optimisation and harmonisation of analytical and statistical methodology have been carried out between two forensic laboratories (Lausanne, CH and Lyon, F) in order to provide drug intelligence for cross-border cocaine seizures. The aim was to improve the gas chromatographic analysis of cocaine samples for profiling. Some important validation parameters were tested to verify the developed method and demonstrate its profiling capacity: the selectivity of the method with retention time reproducibility, the choice of a derivatisation agent improving the chromatography (MSTFA, BSA, TMSI and BSTFA+TMCS 1%), the cutting agents influence (matrix effect), the influence of the sample storage conditions and the sample quantity to weigh for analyses. Eight main alkaloids, which represent the sample signature, have been selected: ecgonine methyl ester, ecgonine, tropacocaine, benzoylecgonine, norcocaine, cis- and trans-cinnamoylcocaine and 3,4,5-trimethoxycocaine. Their stability in the solvent used (CHCl(3)/pyridine) was demonstrated. In order to reach the final objective, which is the comparison of samples seized and analyzed in two different laboratories, the harmonisation of the profiling method between the two laboratories had to be ensured and is the subject of ongoing research.     

Further studies on spot tests and microcrystal tests for identification of cocaine

The presence of cocaine in illicit drug samples is still being determined in some laboratories using spot tests and microcrystal tests. Seventeen chemical species were tested using three different spot tests (Wagner, Marquis, and cobalt thiocyanate followed by stannous chloride reactions) and two microcrystal tests (gold chloride and platinic chloride) to determine whether the results could be differentiated from the results of these tests on cocaine. The data obtained indicated that nine of the 17 compounds gave results similar to those from cocaine using the three spot tests, but that the results from microcrystal testing allowed for differentiation of all nine compounds from cocaine.     

HAIRVEQ: an external quality control scheme for drugs of abuse analysis in hair

The Istituto Superiore di Sanità of Rome, Italy, in cooperation with Institut Municipal d'Investigaciò Mèdica of Barcelona, Spain, set up an external quality control program (HAIRVEQ) to evaluate reliability in hair testing for drug abuse by laboratories from the Italian National Health Service. Samples included in the program were real hair samples from drugs consumers. Prior to sending, hair samples were reduced to powdered form, mixed to ensure homogeneity and tested with GC/MS by four Reference Laboratories. Up to now, four different exercises have been concluded and 23 laboratories participated. Samples containing high and low concentrations of opiates, cocaine and metabolites, low concentrations of MDMA and two blank samples, were included in the intercomparison exercises performed in the first year of HAIRVEQ activities. Results show an insufficient performance of participating laboratories. About 82% of laboratories reported incorrect results on a qualitative basis (false positive and false negative results) for some of the submitted samples. More than one-half of laboratories reported quantitative results (60%). On the basis of the calculated z scores, only between 35 and 55% of results reported should be considered as satisfying. Guidelines have to be provided by Italian authorities for method validation as well as set of recommended cut-off concentrations to orientate laboratories in their quality objectives when developing analytical methodologies as tools to improve reliability and consequently performance of hair analysis.     

Headspace profiling of cocaine samples for intelligence purposes

A method for determination of residual solvents in illicit hydrochloride cocaine samples using static headspace-gas chromatography (HS-GC) associated with a storage computerized procedure is described for the profiling and comparison of seizures. The system involves a gas chromatographic separation of 18 occluded solvents followed by fully automatic data analysis and transfer to a PHP/MySQL database. First, a fractional factorial design was used to evaluate the main effects of some critical method parameters (salt choice, vial agitation intensity, oven temperature, pressurization and loop equilibration) on the results with a minimum of experiments. The method was then validated for tactical intelligence purposes (batch comparison) via several studies: selection of solvents and mathematical comparison tool, reproducibility and "cutting" influence studies. The decision threshold to determine the similarity of two samples was set and false positives and negatives evaluated. Finally, application of the method to distinguish geographical origins is discussed.     

The application of amino acid racemization in the acid soluble fraction of enamel to the estimation of the age of human teeth

The main objectives of the European project "Collaborative Harmonization of Methods for Profiling of Amphetamine Type Stimulants" (CHAMP) funded by the sixth framework programme of the European Commission, included the harmonization of MDMA profiling methods and the creation of a common database in a drug intelligence perspective. In the preliminary stages of this project, the participating laboratories analysed the physical characteristics, the chemical composition and the organic impurities of MDMA tablets, using the previously harmonized methods. The aim of the present work was to apply statistical treatments to the recorded data in order to evaluate their potential in the fight against drug trafficking. Comparable working procedures were applied on the different types of data. The first part of this article deals with organic impurities data, while the second part focuses on the potential of the physical characteristics. Organic impurities data were recorded by a harmonized Gas Chromatography/Mass Spectrometry (GC/MS) method previously developed. Statistical analysis provided a selection of pertinent variables among the 46 organic impurities identified in the chromatograms. Correlation coefficients were used to yield separation between populations of samples coming from the same synthesis batch and samples coming from different batches. It was shown that correlation measurements based on Pearson and cosine functions applied to the data pre-treated by normalisation to the sum of peak responses followed by the square root provided an excellent discrimination between the two populations. The statistical methods applied to organic impurities profiles proved to be excellent techniques to differentiate samples from different batches and to highlight operational links between samples.     

Drug intelligence based on organic impurities in illicit MA samples

One major objective of the European project "Collaborative Harmonisation of Methods for Profiling of Amphetamine Type Stimulants" (CHAMP), funded by the sixth framework programme of the European Commission, consisted of the harmonisation of a gas chromatography/mass spectrometry (GC/MS) method for the analysis of organic impurities found in illicit methamphetamine (MA) samples in a drug intelligence perspective. Statistical analysis provided a selection of pertinent variables among the 43 organic impurities identified in the chromatograms. As for the 3,4-MethyleneDioxyMethAmphetamine (MDMA) study, correlation coefficients were used as a discrimination tool between populations of linked samples (from the same seizure) and unlinked samples (from different seizures). It was also shown that correlation measurements based on Pearson and cosine functions applied to the data pre-treated by normalisation to the sum of peak responses followed by the square root provided excellent discrimination between the two populations. The organic impurities profiling method was proved to be relevant for the characterization of samples from different seizures and their synthesis route patterns.     

Proficiency test for the analysis of hair for drugs of abuse,organized by the Society of Hair Testing

Eighteen laboratories interested in the analysis of human hair for drugs of abuse participated in a proficiency test (PT) organized by the Society of Hair Testing (SoHT) in 2001.Samples sent to the participants included one drug-free hair sample and two samples from drug users, sent in the form of short segments previously checked for homogeneity by three reference laboratories. Participants were requested to analyze the samples following the standard procedure used routinely in their laboratories.The compounds present in the samples included opiates, cocaine and metabolite, cannabinoids and amphetamines. All the laboratories analyzed opiates, cocaine and benzoylecgonine (BE); only 10 analyzed amphetamines, and 9 cannabinoids. Various methods were used to extract drugs from the hair-enzyme treatment, acidic, basic and methanol extractions. All the laboratories employed GC-MS, with the exception of two which used GC-MS/MS and LC-MS/MS, respectively. Six laboratories performed initial screening tests by RIA, ELISA or EMIT.Results show that the laboratories performed well qualitatively, since they successfully identified all the analytes that they tested, with the exception of eight false results. However, the scatter of quantitative results was high.     

Solid-phase microextraction for the determination of cocaine and cocaethylene in human hair by gas chromatography-mass spectrometry

A method for the simultaneous determination of cocaine (COC) and cocaethylene (CE) in human hair was developed, using solid-phase microextraction (SPME) and gas chromatography-mass spectrometry (GC-MS) as analytical technique to identify and quantify the drugs. Selected ion monitoring (SIM) mode was used to obtain higher sensitivity. The deuterated-labeled analogues were used as internal standards. The detector response was linear for the drugs studied over the range 0.4-15 ng/mg, with correlation coefficients higher than 0.995. The coefficients of variation oscillated between 0.65% and 14.18% and the accuracy was in the range from 0.73% to 11.20%. The limits of quantitation and detection were found to be acceptable. Finally, this method was applied to 15 hair samples from cocaine users, obtaining positive results in all cases. The mean concentrations were 5.39 ng/mg (range: 0.43-8.98 ng/mg) for cocaine and 1.11 ng/mg (range: 0.42-2.23 ng/mg) for cocaethylene.     

The hair analysis proficiency testing program of the French Society of Analytical Toxicology

In an effort to improve laboratories performing hair analysis in forensic cases, the French Society of Analytical Toxicology (S.F.T.A.) has implemented a proficiency testing program since 1992. Actually about 10 laboratories are participating. Each survey is dedicated to one analyte or one pharmacological class: opiates (6-monoacetylmorphine, morphine and codeine), cocaine and benzoylecgonine, tetrahydrocannabinol, buprenorphine and norbuprenorphine, beta-blockers (metoprolol, atenolol), beta 2-agonists (salbutamol, clenbuterol). Animal hair was tested for clenbuterol. Prior to sending, hair samples were reduced to a powdered form, well mixed to ensure homogeneity, and then tested by GC/MS or HPLC/MS. Results confirm those obtained in a preliminary study on opiates and cocaine analysis in hair: a common analytical procedure has to be used by all the participants, including hydrolysis of hair. It is essential to work on authentic drug-positive hair samples and not on spiked samples. Participation at this program is free of charge and considered as an educational tool. Comparison of the results with those of other laboratories in Europe and USA shows that the analytical methods used during this program are in accordance with all the other procedures.     

Evaluation of the catalytic decomposition of H2O2 through use of organo-metallic complexes--a potential link to the luminol presumptive blood test

Hair testing for drugs of abuse is performed in Lombardy by eleven analytical laboratories accredited for forensic purposes, the most frequent purposes being driving license regranting and workplace drug testing. Individuals undergoing hair testing for these purposes can choose the laboratory in which the analyses have to be carried out. The aim of our study was to perform an interlaboratory exercise in order to verify the level of standardization of hair testing for drugs of abuse in these accredited laboratories; nine out of the eleven laboratories participated in this exercise. Sixteen hair strands coming from different subjects were longitudinally divided in 3-4 aliquots and distributed to participating laboratories, which were requested to apply their routine methods. All the participants analyzed opiates (morphine and 6-acetylmorphine) and cocainics (cocaine and benzoylecgonine) while only six analyzed methadone and amphetamines (amphetamine, methamphetamine, MDMA, MDA and MDEA) and five Δ(9)-tetrahydrocannabinol (THC). The majority of the participants (seven labs) performed acidic hydrolysis to extract the drugs from the hair and analysis by GC-MS, while two labs used LC-MS/MS. Eight laboratories performed initial screening tests by Enzyme Multiplied Immunoassay Technique (EMIT), Enzyme-linked Immunosorbent Assay (ELISA) or Cloned Enzyme Donor Immunoassay (CEDIA). Results demonstrated a good qualitative performance for all the participants, since no false positive results were reported by any of them. Quantitative data were quite scattered, but less in samples with low concentrations of analytes than in those with higher concentrations. Results from this first regional interlaboratory exercise show that, on the one hand, individuals undergoing hair testing would have obtained the same qualitative results in any of the nine laboratories. On the other hand, the scatter in quantitative results could cause some inequalities if any interpretation of the data is required.     

What constitutes a positive result in hair analysis: proposal for the establishment of cut-off values

Hair is still a seldom used specimen in most laboratories but its analysis has the potential of making a valuable contribution. Despite the many worthwhile reports, the scientific community at large still has reservations about the validity of hair analysis. Some of this is due to a lack of consensus among the active investigators on how to interpret the results from an analysis of hair. In USA, passive exposure seems to be a major problem, which can only be eliminated with difficulty. On the other hand, in Europe, scientists are performing standard decontamination procedures. It would be very helpful if a group of active researchers on hair analysis, representative of academic, government and private laboratories could define what are the areas of agreement and what are the issues that require further efforts to get a consensus. We propose the following guidelines: (1) a complete decontamination procedure, including the analysis of the wash solution; (2) two distinct analytical methods (immunoassay and GC/MS, or two different GC/MS methods); (3) the establishment of cut-off values (using 30-mg hair samples), 0.5 ng/mg of 6-MAM in the case of heroin abuse, and 1 ng/mg of cocaine in the case of cocaine abuse, which can be decreased to 0.5 ng/mg when use is supported by other evidence of drug intake.     

Results of the GEP-ISFG collaborative study on two Y-STRs tetraplexes: GEPY I (DYS461, GATA C4, DYS437 and DYS438) and GEPY II (DYS460, GATA A10, GATA H4 and DYS439)

A collaborative exercise was carried out by the Spanish and Portuguese ISFG Working Group (GEP-ISFG) in order to evaluate the performance of two Y-chromosome STR PCR tetraplexes, which include the loci DYS461, GATA C4, DYS437 and DYS438 (GEPY I), and DYS460, GATA A10, GATA H4 and DYS439 (GEPY II). The participating laboratories were asked to type three samples for the eight markers, using a specific amplification protocol. In addition, two control samples, with known haplotypes, were provided. The results obtained by the 13 different participating laboratories were identical, except for two laboratories that failed to type correctly the same two samples for GATA C4. By sequence analyses, two different GATA C4 allele structures were found. One control sample (allele 21) and two questioned samples (allele 22, correctly typed by all the laboratories, and allele 25) presented the following repeat structure: (TCTA)4(TGTA)2(TCTA)2(TGTA)2(TCTA)n, but different from the one found for allele 26 in one sample included in this exercise, as well as in the second control sample (allele 23), namely (TCTA)4(TGTA)2(TCTA)2(TGTA)2(TCTA)2(TGTA)2(TCTA)n. The collaborative exercise results proved that both Y-tetraplexes produce good amplification results, with the advantage of being efficiently typed using different separation and detection methodologies. However, since GATA C4 repeat presents a complex structure, with alleles differing in sequence structure, efficient denaturing conditions should be followed in order to avoid typing errors due to sizing problems.     

Elemental profiling using ICPMS of methylamphetamine hydrochloride prepared from proprietary medication using the Moscow and hypophosphorous synthesis

Illicit drugs manufactured from clandestine laboratories are often impure due to poor laboratory conditions, variations in synthesis and impure starting materials extracted either from common household products or pharmaceutical grade chemicals. Inductively coupled plasma mass spectrometry (ICPMS) can be utilised as a multi-element analytical tool to elicit the inorganic impurities which may be present in such samples, however the interpretation of the resultant data can be problematic and complex. This is particularly true when dealing with seized samples of unknown provenance. In this work, we have presented and interpreted inorganic profiles as a means to explore within and between batch variations in known provenance samples produced via two different popular synthetic routes. Samples were prepared from essential chemicals recovered from household materials and pharmaceutical medication available in the UK and extracted using different solvents. The presence or absence of elements in the final synthesised products could be linked to the synthesis route, salting out method and potentially the solvent used in the precursor extraction process.     

Hair analysis for opiates,cocaine and metabolites. Evaluation of a method by interlaboratory comparison

The aim of this study was to evaluate the performance of a technique for the simultaneous testing of opiates, cocaine and metabolites in hair by interlaboratory comparison. Sixteen forensic and clinical laboratories with different degrees of experience in hair analysis participated voluntarily in the study (no selection criteria were applied). The suggested analytical procedure, the one routinely used in our laboratory, consisted of incubation in HCl 0.1N (45 degrees C, overnight), solid phase extraction (with Bond Elut Certify) cartridges), derivatisation (trimethylsilyl (TMS) derivatives) and GC-MS analysis. Three different mixtures of finely cut (1 mm or less) hair were prepared using drug-users' and drug-free hair: one 'negative' sample (<0.1 ng/mg for morphine, 6-acetylmorphine (6AM), cocaine and benzoylecgonine (BE)), one 'low concentration' sample (between 0.5 and 2 ng/mg) and one 'high concentration' sample (>3 ng/mg). Accuracy and precision (CV% lower than 5.1, 9.9, 5.2, 3.8, 7.3 and 8.3% for morphine, 6AM, codeine, cocaine, BE, and methylecgonine (ME), respectively; range 0.5-5 ng/mg) of the method and homogeneity of the mixtures were evaluated in our laboratory by intraday (CV% lower than 12% for all analytes) and interday analyses (CV% lower than 17% for all analytes except 6AM, 25%). Participants in the study were grouped into: (1) laboratories (n = 6) obtaining the best qualitative and quantitative values, corresponding to those with long experience in hair analysis; (2) laboratories (n = 5) with no reported false positive and/or false negatives; (3) laboratories (n = 5) with one or more reported false positives/false negatives. The results obtained by the labs of the first group were used as reference values. The scatter of data was similar to those obtained in other published studies.     

High-performance liquid chromatography with column switching for the determination of cocaine and benzoylecgonine concentrations in vitreous humor

A sample concentration technique was adapted for the determination of cocaine and benzoylecgonine (BE) concentrations in vitreous humor. Vitreous humor (0.5 mL) was diluted 1:1 with water and applied through a filter onto a 3-cm preconcentration column. Following a simple wash step, the analytes were flushed directly onto a reversed-phase analytical high-performance liquid chromatography (HPLC) system. Absolute recoveries were high (above 90%) and the chromatograms were free from interference. Analysis for the drug and its breakdown product was performed using ultraviolet (UV) visible photodiode array detection, which allowed confirmation of peak identity. Recognizable UV spectra could be measured with as little as 20 ng on column. Comparison of the drug levels in 27 blood and vitreous humor samples showed that, while there was only a low correlation between the blood and vitreous concentrations (R = 0.70), vitreous cocaine and BE determinations were good indicators of antemortem cocaine use. In almost all cases, the vitreous BE concentrations were higher than the cocaine concentrations. The technique was easy to perform and the vitreous samples were especially compatible with this low-labor analytical procedure.     

Impurities,adulterants and cutting agents in cocaine as potential candidates for retrospective mining of GC-MS data

Cocaine is one of the most widely used illicit drugs worldwide. Cocaine powders seized by the Police may contain numerous other substances besides the drug itself. These can be impurities originating from the coca plant or the production process, or be purposely added to the drug formulation as adulterants and cutting agents. In forensic laboratories, identification of cocaine is routinely done through GC-MS analysis, but other components are often ignored even if the method allows for their detection. Yet, they can provide valuable insight into the history of a seizure and its potential connection to other samples. To explore this idea, an extensive review of common impurities and adulterants encountered in cocaine is presented. Based on their incidence, concentration in the end product and compatibility with GC-MS methods, their overall usefulness as candidates for the statistical investigation of existing forensic data is evaluated. The impurities cis- and trans-cinnamoylcocaine, tropacocaine, norcocaine and N-benzoylnormethylecgonine as well as the adulterants lidocaine, procaine, tetracaine, benzocaine, caffeine, acetylsalicylic acid, phenacetin, ibuprofen, levamisole, hydroxyzine and diltiazem are promising candidates to provide additional forensic intelligence. Future research on optimized routine GC-MS methods, signal reproducibility, comparison, statistics and databases is suggested to facilitate this concept. Ultimately, such an approach may significantly advance the amount of information that is extracted from routine casework data, elucidate developments in the cocaine markets in the past and facilitate Police work in the future. Preliminary assessment of existing data from the forensic laboratory of the Amsterdam Police has been included to show that the detection of the identified target impurities is feasible, and that small adjustments to the analysis method could significantly increase the detectability of these analytes in prospective drug screenings. Forensic intelligence based on retrospective data mining of cocaine containing casework samples may thus be realized with minimal additional laboratory efforts by using already available instrumentation, samples and data.     

Results from the NIST 2004 DNA Quantitation Study

For optimal DNA short tandem repeat (STR) typing results, the DNA concentration ([DNA]) of the sample must be accurately determined prior to the polymerase chain reaction (PCR) amplification step in the typing process. In early 2004, the National Institute of Standards and Technology (NIST) conducted an interlaboratory study to help assess the accuracy of DNA quantitation in forensic DNA laboratories. This study was designed with four primary purposes: (1) to examine concentration effects and to probe performance at the lower DNA concentration levels that are frequently seen in forensic casework; (2) to examine consistency with various methodologies across multiple laboratories; (3) to examine single versus multiple source samples; and (4) to study DNA stability over time and through shipping in two types of storage tubes. Eight DNA samples of [DNA] from 0.05 ng/microL to 1.5 ng/microL were distributed. A total of 287 independent data sets were returned from 80 participants. Results were reported for 19 different DNA quantitation methodologies. Approximately 65% of the data were obtained using traditional slot blot hybridization methods; 21% were obtained using newly available quantitative real-time PCR (Q-PCR) techniques. Information from this interlaboratory study is guiding development of a future NIST Standard Reference Material for Human DNA Quantitation, SRM 2372.     

Evaluation of immunoassay methods for the screening of cocaine metabolites in urine

Immunoassay kits for urine cocaine (and metabolite) screening, obtained from two commercial sources, were examined for correlation of their results, expressed in terms of equivalent benzoylecgonine concentration, with the gas chromatography/mass spectrometry (GC/MS) concentration of benzoylecgonine. The correlation coefficients obtained, based on 62 (out of a total sample population of 3295) highly relevant samples, were 0.467 and 0.766 for Abuscreen (ARIA) and TDx (TDX), respectively. The preliminary screen cutoff values, which correspond to 150 ng/mL benzoylecgonine (as determined by GC/MS), were calculated based on the resulting regression equations and found to be 380 and 190 ng/mL for ARIA and TDX, respectively. With these cutoff values, ARIA generates 5 false negatives and 16 unconfirmed presumptive positives, while TDX results in 3 false negatives and 6 unconfirmed presumptive positives.     

The intersection of racial profiling research and the law

There has been a significant increase in the litigation of selective enforcement cases based on racial profiling claims. This trend has resulted in two legal issues that are problematic for racial profiling research. First, selective enforcement claims that rely on statistical evidence must successfully measure “similarly situated persons” who were eligible for police stops to provide a comparison against those actually stopped by police. Second, the research must demonstrate “how much” statistical evidence of racial/ethnic disparities exists. Although these legal components are necessary for successful selective enforcement claims, the methodologies and statistical analyses currently used in racial profiling research cannot adequately address these issues. It is argued that the over-reliance on social science research, in general, and statistical techniques, specifically, to provide evidence of discrimination in selective enforcement cases places policing research and legal decision making at a crossroads.     

A quick and automated method for profiling heroin samples for tactical intelligence purposes

A computerized procedure is presented for the profiling and comparison of illicit heroin seizures. The system involves the derivation and gas chromatographic separation of five major constituents followed by a fully automatic data analysis and transfer to a PHP/MySQL database. Comparisons via the square cosine function between a single chromatographic profile and the whole database (several hundred of samples) are performed in a few seconds. Advantages of this profiling method compared to the classical minor constituents one are discussed.