Simultaneous Determination and Stability Analysis of Ten New Psychoactive Substances including Synthetic Cathinones, Phenethylamines, and Ketamine Substitutes in Urine Using Liquid Chromatography-Tandem Mass Spectrometry

Knowing the stability of drugs is important to ensure accurate and reliable results of drug concentrations. This study evaluated the stability of ten new psychoactive substances (NPSs) in urine and methanol/water at different storage temperatures. Quantitative analyses were performed using liquid chromatography-tandem mass spectrometry. Three replicates of each storage condition were analyzed at day 0 and after 7, 14-, 30-, 60-, and 90 days with storage at +25°C, +4°C, and −20°C. For each analyte, the percent difference at each time interval from day 0 was calculated for each storage condition. Para-methoxyamphetamine (PMA), para-methoxymethamphetamine (PMMA), deschloroketamine (DCK), and 2-fluorodeschloroketamine (2-FDCK) were stable in urine, even when stored for 90-day periods at various temperatures. For synthetic cathinones, the concentrations declined over time at room temperature (+25°C) in urine but were relatively stable in methanol solvent with 0.1% formic acid. The significant degradation was found at +25°C, and the most excellent stability was shown by samples stored at −20°C. Phenethylamines (PMA and PMMA) and ketamine substitutes (DCK and 2-FDCK) were relatively more stable than synthetic cathinones (mephedrone, butylone, pentylone, ephylone, 4-MEAPP, and eutylone).


Introduction
New psychoactive substances (NPSs) are a complex and diverse group of substances often known as either designer or synthetic drugs or "legal highs." NPSs are associated with several health and social harms on an individual and societal level [1,2]. Te number of NPSs is constantly growing, and trends and patterns of use change over time. Abuse of NPSs has become a critical threat to health and security in recent decades [3]. Some NPS, such as synthetic cathinones, paramethoxymethamphetamine, and ketamine, have posed a risk to public health and a challenge to drug policy [4]. Chewing khat leaves is common in the Arabian Peninsula and Eastern Africa. Te primary psychoactive compositions in khat leaves cause sympathomimetic activity, mild euphoria, and excitation [5]. Synthetic cathinones are designer analogs of the natural active principle of khat. In recent years, synthetic cathinones have become the most seized class of NPS reported to the United Nations Ofce on Drugs and Crime (UNODC) Early Warning Advisory System [6]. Para-methoxymethamphetamine (PMMA) and paramethoxyamphetamine (PMA) are the para-methoxylated analogs of methamphetamine and amphetamine and have been found in tablets and capsules of the 3,4-methylenedioxymethamphetamine (MDMA) sold as "ecstasy." A number of deaths have been attributed to tablets that contained PMMA and PMA [7][8][9]. 2fuorodeschloroketamine (2-FDCK) and deschloroketamine (DCK) as substitutes for ketamine have emerged among drug abusers in recent years and are considered to have similar abuse potential as ketamine [10]. Due to its recent appearance, little research has been done on the compound.
Te urine specimen is commonly used for NPS analysis. Te time from sample collection to analysis might be several days due to transportation time. Sometimes, repeat testing due to unexpected results or additional drug analysis makes the time between collection and analysis increase by more than several days. Because of this time interval, to ensure that results are accurate, it is necessary to demonstrate that specimens are stable during this time frame [11,12]. Moreover, the preparation of quality control and calibration samples in the process of illicit drug analysis is timeconsuming. Te working solutions of the analyte reference materials were often prepared using methanol and water and then stored for several days to months at refrigerator or freezer temperature. If preanalytical changes in concentration occur between specimen collection and analysis or between the initial and retests due to drug instability or degradation, the results of the re-test may vary from the original test results [13,14]. Terefore, understanding the stability of analyzed drugs in testing samples is critically important.
Most published studies on NPS stability focused on single-category [14][15][16][17][18]. Few studies addressed the stability among diferent NPS categories. Te stability studies among diferent solvents were also scarce. Methods to detect NPSs in biological matrices are challenging due to their low concentrations. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) has become a tool of choice for quantitative bioanalytical assays due to its inherent selectivity and sensitivity and has proven to be fast and accurate [19,20]. In this study, we used LC-MS/MS to develop a quantitative method for simultaneous determination of multiple NPSs and evaluate the stability of these NPSs, classifed as synthetic cathinones, phenethylamines, and ketamine, in 50% methanol/water with or without 0.1% formic acid, and urine, at diferent storage temperatures.

Sample Preparation.
Te study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Boards of Kaohsiung Medical University Hospital (KMUHIRB-E(I)-20221077). After informed consent was obtained, the midstream urine was obtained from fve healthy volunteers simultaneously and pooled together for validation and stability experiments. Terefore, the urine matrices in this study were the same in all test samples. Pooled blank urine was confrmed to be negative for any medications using LC-MS/MS testing and stored at −20°C. Te calibration standards and quality control samples were prepared using pooled blank urine. Te working solution of ten analytes (10 mg/L) was prepared in 50% methanol/water with 0.1% formic acid from reference materials of each analyte. Spiking solutions were prepared from the working solutions as mixtures of the ten NPSs. Te internal standards (ISs), including mephedrone-D3, butylone-D3, MDA-D5, and ketamine-D4, were prepared in 50% methanol/water with 0.1% formic acid for each fnal concentration with 25 ng/mL.
Protein precipitation was performed before LC-MS/MS analysis. In a microcentrifuge tube, 50 μL of urine was mixed with 50 μL methanol. After centrifugation at 14,000 × g for 10 min to separate layers, 20 μL of the supernatant was transferred to a new microcentrifuge tube and mixed with 160 μL of 25 ng/mL internal standard and 20 μL of 50% methanol/water and 0.1% formic acid solution. Ten, 20 μL of the sample was injected into the LC-MS/MS system. Te analytes were then quantifed using a Xevo ® TQ-XS tandem mass spectrometer (Waters Corporation, Milford, MA US). Te mass spectrometer was operated in multiple reaction monitoring modes (MRMs). Te MRM transitions and conditions for the analytes and ISs, as well as the retention times, target ions, and qualifer ions used for identifcation and quantifcation, are shown in Table 2. International Journal of Analytical Chemistry

Method Validation.
Te validation procedures were based on guidelines for bioanalytical method validation [21,22] and the SWGTOX validation guidelines for urine [23]. All method validations were carried out by LC-MS/MS, and the data were analyzed by Microsoft Excel software (version 15.0.5363.1000).

Specifcity and Selectivity. Specifcity and selectivity
were assessed by spiking each drug and individual IS to test for interference. Exogenous interferences were evaluated by spiking blank matrices with twelve analytes, including dehydronorketamine, ethylone, dibutylone, funitrazepam, nimetazepam, nitrazepam, phenylpropanolamine, ephedrine, pseudoephedrine, phentermine, cocaine, and ecgonine methyl ester, to ascertain if exogenous analytes may interfere with the method analytes.

Sensitivity and Linearity.
Te lower limit of quantifcation (LLOQ) was determined as the lowest concentration with chromatographic signal-to-noise (S/N) value of 10 : 1 and met the accuracy limit of ±20%. Linearity was assessed by analyzing ten separate calibration curves by spiking blank urine with concentrations at 3.1, 6.2, 12.5, 25, 50, 62.5, 75, and 100 μg/L for all compounds. Calibration curves produced per batch were generated by plotting the peak area ratio (PAR) versus the spiked analyte concentration. A blank matrix containing only IS was analyzed with each batch but not included in the calibration curves. Te correlation coefcient (R 2 ) was calculated and with an acceptability criteria of >0.99.

Precision and Accuracy.
Accuracy and precision were calculated by running calibration standards alongside fve replicates of 25, 50, and 75 ng/mL. Both intra-and interday precision were assessed across fve batches with an acceptability criterion of ≤10%. A method accuracy limit of ±15% was used [24].
According to the guideline on bioanalytical method validation [22], the matrix efects were estimated with ten urine samples from ten individuals. For each analyte and the IS, the matrix factor (MF) was calculated for each sample by calculating the ratio of the peak area in the presence of the matrix (measured by analyzing a matrix blank spiked after extraction with an analyte) to the peak area in the absence of the matrix (pure solution of the analyte). Te IS-normalized MF was also calculated by dividing the analyte's MF by the IS's MF. Te CV of the ISnormalized MF from the ten diferent urine samples should not be greater than 15%.

Sample Stability.
Analyte stability was analyzed in diferent matrices and temperatures, including solutions of 50% methanol, 50% methanol/water with 0.1% formic acid, and urine, and storage at 25°C (room temperature), 4°C (refrigerator), and −20°C (freezer). Stability was assessed by monitoring the PARs of each analyte to the IS and the individual peak areas themselves. Tree replicates at each concentration were analyzed at day 0 and after 7, 14, 30, 60, and 90 days. For each analyte, the percent diference at each time interval from day 0 was calculated and averaged for triplicate samples at each storage condition. Tere were no exclusion criteria; the percent diference calculations included every specimen concentration result. A concentration change of ±20% was deemed unstable.  Table 2. Te total ion chromatogram (TIC) of the analytes and internal standards are shown in Figure 1. No endogenous or exogenous interferences were observed from the pooled drug-free matrices analyzed. None of the analytes or ISs interfered with the other analytes' peak areas or retention times within this method.    Table 3. 4-MEAPP, eutylone, and 2-FDCK were linear over a concentration range of 3.1-100 μg/L with a correlation coefcient (R 2 ) > 0.99. Te calibration curves for other analytes were linear over a concentration range of 6.3-100 μg/L (R 2 > 0.99). Te linearity curves of all analytes are shown in Figure S1.

Precision and Accuracy.
According to quantitative mass spectrometry for pharmacokinetic studies, the bioanalytical method of choice must be specifc, precise, and reproducible to the intended analyte in a given matrix . Te current best industry practice of validating an LC-MS/MS method and applying it for sample analysis suggested that accuracy is determined by replicate analysis of samples (QCs) containing known amounts of the analyte. Te mean measured value should be within 15% (bias) of the nominal value for all QC concentration levels except for the LLOQ, where the bias (%) should be within 20% [24,25].

Carryover and Matrix Efects.
No carryover was found in the blank samples following duplicate injection of the highest calibrator. Te concentrations of all blank samples were 0 μg/L. Matrix efect data of all analytes are shown in Table 4. Te matrix efects of these analytes ranged from 84%-109%.

Stability of the Analytes in Diferent Matrices and
Temperatures. Te stability of each analyte in diferent matrices and temperatures are shown in Figures 2 and 3.     Table S1.

Te Stability Study of Mephedrome.
Te stabilities of the synthetic cathinones in diferent matrices and storage temperatures are shown in Figure 2. For mephedrone, when the samples were stored at freezer temperature (−20°C), the percent diferences were within 20% of the original measurement for all solvents, even by day 90. When the samples were stored in 50% methanol/water with 0.1% formic acid, the percent diferences were within 20% of the original measurement for all storage temperatures for 90 days (Figure 2(a)). Mephedrone in urine stored at 25°C showed a 68.65 ± 1.21% loss of the initial concentration by day 7 and a 96.43 ± 0.06% loss of the initial concentration by day 14. Mephedrone in urine stored at 4°C showed a 20.9 ± 3.66% loss of the initial concentration by day 14 and a 42.66 ± 3.10% loss of the initial concentration by day 30 (Table S1).

Te Stability Study of Butylone, Pentylone, and
Eutylone. For butylone, pentylone, and eutylone, when the samples were stored at −20°C, the percent diferences were within 20% of the original measurement for all solvents, even by day 90. When stored in 50% methanol/water (with or without 0.1% formic acid), the percent diferences were within 20% of the original measurements at 4°C and −20°C (Figures 2(b)-2(d)). Butylone in urine stored at 25°C showed a 48.42 ± 1.90% loss of the initial concentration by day 14 and an 88.15 ± 0.73% loss of the initial concentration by day 30. Butylone in urine stored at 4°C showed a 30.01 ± 1.64% loss of the initial concentration by day 90. Butylone in 50% methanol/water without 0.1% formic acid stored at 25°C showed a 24.43 ± 1.61% loss, a 40.74 ± 1.03% loss, and a 64.50 ± 0.31% loss of the initial concentration by day 30, day 60, and day 90, respectively. Pentylone in urine stored at 25°C showed a 27.22 ± 2.19% loss of their initial concentration by day 7 and a 60.16 ± 1.44% loss of the initial concentration by day 14. Pentylone in urine stored at 4°C showed a 35.54 ± 1.24% loss of the initial concentration by day 90. Pentylone in 50% methanol/water without 0.1% formic acid stored at 25°C showed a 27.84 ± 1.19% loss, a 61.03 ± 1.70% loss, and a 79.37 ± 0.69% loss of the initial concentration by day 30, day 60, and day 90, respectively.
Eutylone in urine stored at 25°C showed a 42.74 ± 3.10% loss of the initial concentration by day 14 and a 79.26 ± 1.09% loss of the initial concentration by day 30. Eutylone in urine stored at 4°C showed a 23.92 ± 0.64% loss of their initial concentration by day 90. Eutylone in 50% methanol/water without 0.1% formic acid stored at 25°C showed a 25.78 ± 2.29% loss, and a 50.21 ± 0.37% loss of the initial concentration by day 60, and day 90, respectively (Table S1).

Te Stability Study of Ephylone and 4-MEAPP.
For ephylone and 4-MEAPP, when the samples were stored at −20°C, the percent diferences were within 20% of the original measurement for all solvents, even by day 90. When the samples were stored in 50% methanol/water with 0.1% formic acid, the percent diferences were within 20% of the original measurement for all storage temperatures for 90 days (Figures 2(e) and 2(f )).
Ephylone in urine stored at 25°C showed a 24.32 ± 0.90% loss of the initial concentration by day 7 and a 53.00 ± 1.85% loss of the initial concentration by day 14. Ephylone in urine stored at 4°C showed a 36.11 ± 1.47% loss of the initial concentration by day 90. Ephylone in 50% methanol/water stored at 25°C showed a 25.17 ± 2.29% loss of the initial concentration by day 30 and a 71.02 ± 0.41% loss of the initial concentration by day 90. 4-MEAPP in urine stored at 25°C showed a 37.81 ± 1.52% loss of the initial concentration by day 7 and a 74.29 ± 1.69% loss of the initial concentration by day 14. 4-MEAPP in urine stored at 4°C showed a 29.59 ± 1.28% loss of their initial concentration by day 60 and a 53.98 ± 1.09% loss of the initial concentration by day 90. 4-MEAPP in 50% methanol/ water stored at 25°C showed a 21.05 ± 3.44% loss of the initial concentration by day 14 and a 90.3 ± 0.19% loss of the initial concentration by day 90 (Table S1).

Te Stability
Study of PMA and PMMA. PMA and PMMA samples were stable for more than 90 days at 4°C and −20°C in all solvents evaluated in this study (Figures 3(a) and  3(b)). In PMA samples, the percent diferences were all within 20% of the original measurements in all solvents, even at room temperature by day 90. In PMMA samples, a more than 20% diference was found only in urine at room temperature by day 60.

Te Stability Study of DCK and 2-FDCK. DCK and 2-
FDCK samples were stable for more than 90 days at 4°C and −20°C in all solvents evaluated in this study (Figures 3(c) and

3(d)
). In the DCK and 2-FDCK samples, more than 20% diferences were found only in 50% methanol/water with 0.1% formic acid at room temperature by day 90 and day 60, respectively. Te stability study showed that phenethylamines (PMA and PMMA) and ketamine substitutes (DCK and 2-FDCK) were relatively stable than synthetic cathinones (mephedrone, butylone, pentylone, ephylone, 4-MEAPP, and eutylone). PMA, PMMA, DCK, and 2-FDCK were stable in urine, even when stored for 90-day periods at various temperatures ( Figure 4). Similar results were reported previously about the long-term stability of amphetaminetype stimulants (ATS) [26,27]. Tus, delayed testing, repeat testing, and add-on testing for PMA, PMMA, DCK, and 2-FDCK in urine specimens can yield reliable results for up to 90 days following the urine collection date.
Te stability results for synthetic cathinones showed that the detectable concentrations would decline over time ( Figure 2). In urine samples, all the synthetic cathinones tested in this study declined within 7-14 days at room temperature. Most synthetic cathinones, except mephedrone, are stable in urine for at least 30 days when stored at refrigerator temperature. Mephedrone concentrations in urine signifcantly declined within 14 days, even at refrigerator temperatures (Figure 4). All tested synthetic cathinones were stable at freezer temperature in both urine and 50% methanol/water, no matter with or without 0.1% formic acid. Synthetic cathinones dissolved in 50% methanol with 0.1% formic acid were relatively more stable than those stored without formic acid in the present study. Previous studies reported similar results that synthetic cathinones are relatively stable under acidic conditions [14].   Many commercially available NPS reference substances are dissolved in methanol. Considering the polarity and solubility distribution of NPSs, we chose 50% methanol as one of the solvents for drug stability evaluation. Some previous studies that investigated the stability of synthetic cathinones used preservatives or bufers containing metal ions for diferent pH solutions [14,28]. However, metal ions may interfere with mass spectrometry or damage the instrument. In this study, we used formic acid to maintain the pH of solvents.
Te chemical behavior of synthetic cathinone is mainly determined by the ketone and amine groups in its structure. Synthetic cathinones have a "beta-keto" structure and two tautomers of "keto" and "enol" form in solution. Many factors, such as solvent properties, matrix, pH value, aromaticity, conjugation, hydrogen bonds, and substitutions of synthetic cathinones, would afect the equilibrium position between the keto and enol forms [29]. A previous study considered that solvent efects are due to the strong tendency of the enol form to hydrogen bond intramolecularly, while the keto form may hydrogen bond with protic solvents, providing stabilization [30]. Acidic conditions favor the formation of an enol form. It has been reported that acid makes the carbonyl group more electrophilic, increasing the acidity of alpha-protons and facilitating enol formation [29]. Previous studies demonstrated that mephedrone could form dihydromephedrone through carbonyl reduction of its ketone group [31,32]. We speculate that synthetic cathinones might partially be transformed into the enol form due to low pH in 50% methanol with 0.1% formic acid. Te reaction of the ketone functional group will not be able to proceed under acidic conditions, thus leading to an increase in stability ( Figure S2).
Using unstable reference material might lead to unreliable results that could signifcantly impact the justice system and the individuals whose samples are tested in casework. Storing the samples with 0.1% formic acid would improve the stability of the analytes, especially in synthetic cathinones. Te most signifcant degradation was found at room temperature (25°C), and the most excellent stability was stored in the freezer (−20°C). Although unopened reference material may be stable for long, this work highlights the importance of regularly updating reference material once opened.

Conclusion
In this study, we simultaneously determined ten NPSs, including synthetic cathinones, phenethylamines, and ketamine substitutes, and evaluated their stability in various solvents and storage temperatures. Mephedrone is the least stable analyte tested in this study. Phenethylamine and ketamine substitutes are more stable than synthetic cathinones, synthetic cathinones are more stable under acidic conditions, and all analytes are stable within 90 days at freezer (−20°C) temperature. Knowing the stability of drugs is essential to ensure accurate and reliable results of drug concentrations. Tese fndings highlight the importance of the storage environment for reference materials and biological samples in forensic laboratories when performing NPS analysis.

Data Availability
Te data presented in this study are available on request from the corresponding author. Data may be available upon request to interested researchers. Please send data requests to Yi-Ching Lin, MD, PhD, Department of Laboratory Medicine, Kaohsiung Medical University Hospital, Kaohsiung Medical University.

Ethical Approval
Te study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of Kaohsiung Medical University Hospital (KMUHIRB-E(I)-20221077).

Consent
Informed consent was obtained from all subjects involved in the study.

Conflicts of Interest
Te authors declare that they have no conficts of interest.