Original Article
Evaluation of the stability and cross-reactivity of zopiclone and eszopiclone using immunoassay kits
2026 Volume 51 Issue 3 Pages 173-181
Details
2026 Volume 51 Issue 3 Pages 173-181
Zopiclone (ZOP) and eszopiclone (EZOP) are hypnotics belonging to the cyclopyrrolone class, which are metabolized to zopiclone-N-oxide (ZOPNO) and N-desmethylzopiclone (NDZOP), respectively, and finally to 2-amino-5-chloropyridine (2A5C) after long-term storage, freezing, and conservation. This study aimed to examine the stability of ZOP and EZOP using a commercially available immunoassay urine screening kit for ZOP. ZOP and EZOP, as well as their metabolites and 2A5C, were analyzed in urine using liquid chromatography coupled with tandem mass spectrometry (LC–MS/MS). Spiked urine samples were used to confirm cross-reactivity for the immunoassay test and to compare the quantitative results of the clinical urine samples. The results indicate that urine samples stored for an extended period or altered postmortem may lead to false-negative results for ZOP, providing important information for interpreting results in cases involving ZOP.
Zopiclone (ZOP) and its S-isomer, eszopiclone (EZOP), belong to the cyclopyrrolone class of hypnotic drugs (Fig. 1). EZOP is a nonbenzodiazepine sedative-hypnotic that was approved in Europe in 1986 (Rösner et al., 2018) and by the US Food and Drug Administration on December 15, 2004. It is used for treating insomnia in patients aged 18 years and older (Najib, 2006). In Japan, ZOP is regulated as a psychotropic drug, but EZOP is not.
Chemical structures of zopiclone, eszopiclone, 2-amino-5-chloropyridine, N-desmethylzopiclone and zopiclone-N-oxide.
ZOP is extensively metabolized in the human liver into two major metabolites, N-desmethylzopiclone (NDZOP) and zopiclone-N-oxide (ZOPNO). (Becquemont et al., 1999) (Fig. 1). ZOPNO has low pharmacological activity, whereas NDZOP is pharmacologically inactive. ZOP, EZOP, and their metabolites can be detected in urine using a ZOP50 Rapid Test Cassette (urine) (ZOP kit). However, the cross-reactivity of these compounds is currently unknown.
Complete decomposition of ZOP and its metabolites results in the formation of 2-amino-5-chloropyridine (2A5C) (Fig. 1), following long-term storage, freezing, and conservation. Galloway et al. detected ZOP and 2A5C in the urine of ZOP users using gas chromatography (Galloway et al., 1999). Furthermore, ZOP exhibits instability even after extraction from biological fluids (Kratzsch et al., 2004); therefore, the results of the ZOP kit may vary depending on the condition of the preserved samples.
In this study, we investigated the changes in the concentrations of ZOP, EZOP, and their metabolites in urine during storage. We also examined their cross-reactivity when detected using the ZOP kit. Furthermore, Mannaert et al. reported that 2A5C in urine only originates from ZOP, NDZOP, or ZOPNO (Mannaert et al., 1997). However, edoxaban and betrixaban (Fig. 2) are oral anticoagulants that contain 2A5C in their structure, so 2A5C may be detected in the urine of people taking these drugs. Therefore, we confirmed whether 2A5C could be detected under various preservation conditions when ZOP, EZOP, their metabolites, edoxaban and betrixaban were present in urine.
Chemical structures of edoxaban and betrixaban.
ZOP, EZOP, NDZOP, N-desmethyleszopiclone (NDEZOP), and ZOPNO were purchased from Toronto Research Chemicals (Toronto, ON, Canada) and diluted in methanol to obtain the appropriate concentrations. ZOP-d8 (ZOP-d8), NDZOP-d8 (NDZOP-d8), and ZOP-d8-N-oxide (ZOPNO-d8), also purchased from Toronto Research Chemicals, were used as internal standards (ISs). 2A5C was purchased from Kanto Chemical Co., Inc. (Tokyo, Japan). Acetic acid, acetonitrile, betrixaban, chloroform, edoxaban, and methanol for LC–MS were purchased from Sigma-Aldrich (St. Louis, MO, USA). The ZOP50 Rapid Test Cassette for urine samples was purchased from AllTest Biotech Co., Ltd. (Hangzhou, China). Captiva EMR-Lipid was used for lipid removal from urine samples (Agilent Technologies Inc., Folsom, CA, USA). Blank urine, collected from healthy laboratory personnel, was used after confirming by LC-MS/MS that no drugs were detected. This study has been approved by the Research Ethics Committee of Tokai University School of Medicine (approval number: 24R105). Before urine sample collection, written informed consent was obtained from the patients.
Confirmation of cross-reactivity and selection of clinical urine samplesBlank urine samples were collected anonymously from healthy personnel and used were collected anonymously from healthy personnel and used only after LC–MS/MS analysis confirmed the absence of drugs. Patient urine samples were also anonymized before use. Blank urine samples spiked with standards for ZOP, EZOP, and their metabolites, as well as clinical urine samples, were used to assess the cross-reactivity of the ZOP kit.
The presence of ZOP or EZOP in clinical samples, indicating positive results, was confirmed using chiral LC–MS/MS analysis. Among the ZOP kit-negative urine samples, those in which ZOP was detected using the Poroshell 120 EC-C18 (3.0 × 100 mm, 2.7 μm, Agilent Technologies) column were validated using the chiral Kromasil 3-AmyCoat RP column (150 mm x 3.0 mm) (Akzo Nobel Pulp and Performance Chemicals AB, Bohus, Sweden) to confirm the presence of ZOP and EZOP. Subsequently, the clinical samples were divided into aliquots and stored under various conditions.
Storage conditionsThe stability of clinical urine samples was tested under the following storage conditions: at 4°C, room temperature, and −30°C for up to 3 months. Their stability was also analyzed after three freeze (−30°C) and thaw (room temperature) cycles.
The stability of the stored samples was initially assessed using the ZOP kit. Furthermore, the decomposition of each compound into 2A5C was confirmed using LC–MS/MS analysis. The concentrations of each compound were compared with those of the standard sample, which was stored at −30°C and analyzed simultaneously.
Detection of 2A5C from compounds other than ZOP/EZOPThe urine samples of patients treated with betrixaban and edoxaban, which share a 2A5C motif in their chemical structure, might contain 2A5C. Therefore, these compounds were added to the urine samples at concentrations of 100 and 1000 µg/mL, and the presence or absence of 2A5C was confirmed 1 month later.
Sample preparationTo prepare the urine samples, 20 μL of urine was mixed with 5 μL of ISs (1 μg/mL) and 75 μL of acetonitrile, and then vortexed for 15 sec. The urine sample was loaded into Captiva EMR-Lipid. The filtered sample was supplemented with 20 μL of 0.1% acetic acid before LC–MS/MS analysis. For compounds that exceeded the calibration curve, urine was diluted to bring them within the calibration curve range. Then they were reanalyzed.
LinearityCalibration standards for urine samples were prepared in the ranges of 2–200 ng/mL and 50–5000 ng/mL, depending on the sensitivity of the analytes. Regression lines were calculated using a weighted (1/concentration) linear regression model. For quantifying the stored samples—both before and after storage—a new calibration curve was prepared prior to each analysis.
LC–MS/MS conditionsThe samples were analyzed using an Agilent 1290 Infinity HPLC system interfaced with an Agilent 6410 Triple Quad LC–MS/MS (Agilent Technologies), coupled to an electrospray ionization (ESI) source in positive mode.
The presence of ZOP and EZOP was confirmed using a Kromasil 3-AmyCoat RP column at 40°C. The mobile phase consisted of 0.1% acetic acid (A) and acetonitrile (B), in a 90:10 ratio, and was run isocratically. The flow rate was 0.3 mL/min, and the injection volume was 5 μL.
ZOP, EZOP, and metabolites were analyzed using a Poroshell 120 EC-C18 column at 40°C. The mobile phase consisted of 0.1% acetic acid (A) and acetonitrile (B) at an 80:20 ratio. Gradient elution was performed by increasing B to 100% within 10 min, then returning to 20%; post‐time 4 min. The ESI configuration was as follows: gas temperature, 300°C; gas flow rate, 10 L/min. Multiple Reaction Monitoring (MRM) analysis was performed in positive mode.
Statistical analysisChanges in the concentrations of ZOP, EZOP, and their metabolites in urine samples stored under various conditions were measured twice for each condition. The average values are expressed as a percentage of the initial concentrations measured before storage.
Table 1 shows the mass transitions for each compound. Two transitions for each substance were chosen for identification, and the most intense one was used for quantification. For all compounds investigated, peak-area ratios of quantifier to qualifier were found to be reproducible. For the deuterated ISs, a single MRM transition was used. Fig. 3 shows the ZOP and EZOP chromatograms analyzed on the Kromasil 3-AmyCoat RP column. Fig. 4 shows the chromatograms of ZOP and other compounds analyzed using on the Porosell 120 EC-C18 column.
;%0A%09%09%09newWindow.document.open();%0A%09%09%09newWindow.document.write('<img src=%22./Graphics/51_173-t1.jpg%22>');%0A%09%09%09newWindow.document.close();%0A%09%09)
Chromatograms of zopiclone (top) and eszopiclone (bottom) standards analyzed by multiple reaction monitoring using the Kromasil 3-AmyCoat RP column.
Chromatogram of multiple reaction monitoring analysis of zopiclone, eszopiclone, zopiclone-N-oxide, N-desmethylzopiclone, N-desmethyleszopiclone, zopiclone-d8, zopiclone-d8-N-oxide, N-desmethylzopiclone-d8, and 2-amino-5-chloropyridine standards on the Poroshell 120 EC-C18 column.
Table 2 summarizes the cutoff concentrations for each standard ZOP compound (NDZOP, ZOPNO, EZOP, and NDEZOP) using the ZOP kit. These results show that the ZOP kit exhibits cross-reactivity, even with metabolites. According to the instructions for the ZOP kit, the cutoff concentration for ZOP and/or ZOPNO in urine is 50 ng/mL, but in our tests, the concentration was slightly higher. The ZOP kit demonstrated no cross-reactivity to 2A5C.
;%0A%09%09%09newWindow.document.open();%0A%09%09%09newWindow.document.write('<img src=%22./Graphics/51_173-t2.jpg%22>');%0A%09%09%09newWindow.document.close();%0A%09%09)
ZOP showed cross-reactivity at lower concentrations than EZOP. However, ZOPNO and NDZOP cross-reacted at half the concentration of ZOP. While ZOPNO and NDZOP showed similar cross-reactivity, it was lower than that stated in the instruction manual of the ZOP kit. Moreover, NDEZOP shows ten times lower cross-reactivity than NDZOP. Moreover, ZOP exhibits a greater affinity for the antibody than EZOP, resulting in a lower estimated concentration. Because the cutoff concentration of the ZOP kit is 50 ng/mL and ZOPNO and NDEZOP cross-react at half the concentration of ZOP, ZOPNO and NDZOP may yield positive results in the ZOP test even if the ZOP concentration is below the cutoff.
Ten clinical urine samples in which ZOP or EZOP were previously detected were used for testing sample stability. The pH values of the clinical urine samples were between 6.6 to 7.3. Two representative samples from ZOP and EZOP are presented in Tables 3 and 4, illustrating the changes in concentration of each compound as well as the cross-reactivity to the ZOP kit after storage under various conditions. Supplementary data showed similar results. The cumulative concentrations of 2A5C were also detected in these samples.
;%0A%09%09%09newWindow.document.open();%0A%09%09%09newWindow.document.write('<img src=%22./Graphics/51_173-t3.jpg%22>');%0A%09%09%09newWindow.document.close();%0A%09%09)
;%0A%09%09%09newWindow.document.open();%0A%09%09%09newWindow.document.write('<img src=%22./Graphics/51_173-t4.jpg%22>');%0A%09%09%09newWindow.document.close();%0A%09%09)
LC–MS/MS results show that the concentrations of ZOP, EZOP, NDZOP, NDEZOP, and ZOPNO tended to decrease. Urine spiked with edoxaban and betrixaban did not show any cross-reactivity with the ZOP kit at any concentration. Furthermore, 2A5C was not detected in urine stored at −30°C, but was detected in urine samples containing added edoxaban that were stored under other conditions (Table 5). For urine samples with added betrixaban, 2A5C was not detected under any storage conditions.
;%0A%09%09%09newWindow.document.open();%0A%09%09%09newWindow.document.write('<img src=%22./Graphics/51_173-t5.jpg%22>');%0A%09%09%09newWindow.document.close();%0A%09%09)
Up to 75% of the total oral dose of racemic ZOP is excreted in the urine, primarily as metabolites, with less than 7% excreted renally as the parent compound (Brielmaier, 2006). While the systemic elimination of EZOP is poorly studied, it is likely to be comparable to that of racemic ZOP. Drug screening kits use monoclonal antibodies, which have precise structural recognition properties. The monoclonal antibodies in the ZOP kit exhibit low cross-reactivity with S-isomers relative to racemic forms. Therefore, they also have poor cross-reactivity with S-isomer metabolites. Therefore, the cross-reactivity of EZOP metabolites may be similar to that of EZOP rather than ZOP, resulting in less cross-reactivity with antibodies than ZOP metabolites. Indeed, the results of the ZOP kits revealed that the concentration at which NDZOP exhibits cross-reactivity is 10 times lower than that for NDEZOP. Although we did not use a standard for EZOPNO, we speculate that the concentration at which ZOPNO shows cross-reactivity is likely to be lower than that of EZOPNO.
Mannaert et al. reported that when urine containing approximately 300 ng/mL of ZOP at pH 6 was stored at room temperature, the ZOP concentration decreased rapidly, and the pH rose after about 10 days. However, when the same urine sample was stored at 4°C, no change in concentration or pH was observed (Mannaert et al., 1997). In the former case, ZOP was most likely decomposed into 2A5C, resulting in its decline. In our study, urine samples stored at 4°C for 4 weeks contained lower levels of 2A5C than those stored at room temperature. Despite 2A5C accumulation, ZOP or EZOP was also detected. Furthermore, high-concentration clinical urine samples stored for one month at 4°C showed positive results with the ZOP kit. However, no traces of ZOP, EZOP, or its metabolites were detected in either prepared or clinical samples after storage at room temperature for over 3 months (data not shown), yielding negative results with the ZOP kit. However, the results varied depending on the initial concentration of each compound in urine. In particular, we obtained clinical urine samples from patients suffering from drug overdose. Therefore, the concentration of ZOP, EZOP, and its metabolites can be expected to be lower when samples are obtained from patients receiving a normal amount of medication.
Moreover, we confirmed that after three months, ZOP, EZOP, and metabolites remained stable in all samples stored at −30°C (Tables 3 and 4). However, they appeared to be unstable at higher temperatures. If the patient is in a febrile state after taking ZOP or EZOP, or if the patient is temporarily exposed to a hyperthermic environment after death, urinary 2A5C may be elevated. Nilsson et al. reported that decomposition of ZOP, NDZOP, and ZOPNO in urine results in the formation of 2A5C. However, 2A5C is produced in large amounts at pH >8.2 instead of pH <6.5 (Nilsson et al., 2014).
This decomposition pattern was also observed in blood. Liu et al. investigated the stability of EZOP in rat plasma. They reported that, when a 90% phosphate buffer solution (pH 3.3) adjusted with 0.1 M phosphate buffer (pH 7.4) and 10% acetic acid was added to rat plasma, the decomposition of EZOP did not occur, and 2A5C was not detected (Liu et al., 2020). Nilsson et al. (Nilsson et al., 2015) reported that the decomposition of ZOP to 2A5C in blood is greatly affected by the urine pH, storage conditions, and duration. ZOP did not decompose to 2A5C in blood and urine under acidic conditions. Similar studies found a 45% decrease in ZOP concentrations in serum samples stored for 31 days at -4°C (Segawa et al., 2023). Although they did not analyze serum 2A5C concentrations, the depletion of ZOP, EZOP, and metabolites is most likely due to the formation of 2A5C.
Mannaert et al. reported that 2A5C is a specific degradation product of ZOP, NDZOP, and ZOPNO, since the concentrations of ZOP, NDZOP, and ZOPNO reflect their original concentrations in urine only in the absence of 2A5C (Mannaert et al., 1997).
Urine spiked with edoxaban or betrixaban did not show cross-reactivity with the ZOP kit at any concentration. In urine samples containing added edoxaban, 2A5C was detected when stored under conditions other than −30°C, whereas those spiked with betrixaban did not contain 2A5C, regardless of storage conditions.
Although urine samples obtained from individuals taking edoxaban were not evaluated in this study, these samples are likely to contain 2A5C. Therefore, detection of 2A5C in urine alone would not prove the presence of ZOP or EZOP.
Furthermore, we reported that the anti-histamine fexofenadine is partly excreted as azacyclonol (Saito) in the urine samples of patients taking this drug. The use of azacyclonol, a positional isomer of pipradrol, is internationally restricted. Therefore, we suspected that azacyclonol and pipradrol might be misidentified in the LC-MS analysis of urine samples from patients taking fexofenadine. When analyzing urine, it is crucial to distinguish between metabolites derived from other compounds and identify the parent compounds to prevent misidentification. It is necessary to verify the analytical conditions for separating and analyzing each isomer using its respective standard sample. Although it is a standard practice in forensic toxicology to validate the positive results obtained using immunoassay kits, it is essential to exercise caution, as there may be no cross-reactivity in urine samples containing EZOP. Moreover, it is essential to consider these limitations when 2A5C is detected in urine specimens, as there is a risk of quantitative underestimation, which can lead to false negativity.
The ZOP kit is an effective method for screening ZOP and EZOP in urine. The primary objective of this study was to evaluate the stability of ZOP, EZOP, and their metabolites in urine samples tested using the ZOP kit. Therefore, if the patient has been prescribed ZOP or EZOP, steps should be taken to avoid potential false results (especially in cases with a significant post-mortem elapsed time). The presence of 2A5C in urine samples that tested negative should be further investigated using LC-MS/MS for re-validation. This is crucial for minimizing the occurrence of false negative cases and ensuring the accuracy of the analysis. We also evaluated the cross-reactivity profiles for ZOP and EZOP, which are not included in the manufacturer’s instructions for the ZOP kit. Our results showed limited cross-reactivity toward ZOP, NDZOP, ZOPNO, EZOP, and NDEZOP.
In summary, our findings show that this screening test is not always sufficient for accurately identifying these drugs. In forensic cases, especially cases involving the use of ZOP, a confirmatory test for 2A5C by instrumental analysis is necessary even in cases where the immunoassay is negative.
This work was supported by JSPS KAKENHI Grant Number JP23K09776.
FundingWe have not received any research funding other than from JSPS KAKENHI.
Conflict of interestThe authors declare no conflicts of interest regarding this manuscript.
Data availabilityThe data in this study are included in the article/supplementary materials. Contact the corresponding author(s) directly to request the underlying data.
Author contributionsAll authors contributed to the study conception and design. Experimental and instrumental analysis were performed by Takeshi Saito. The first draft of the manuscript was written by Takeshi Saito, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Ethical approval and consent to participateThis study has been approved by the Research Ethics Committee of Tokai University School of Medicine (approval number: 24R105). Before urine sample collection, written informed consent was obtained from the patients.
Patient consent for publicationWe obtained consent from all patients.
