A proposed limit test for p-chloroaniline impurity in paracetamol pharmaceutical formulations

Abstract

P-chloroaniline is a polar organochlorine compound and an important member of aromatic amines, widely used in various industries, including pesticides, dyes, and pharmaceuticals. Exposure to high levels of P-chloroaniline can cause severe damage to the liver and kidneys and negatively affect the central nervous system. Recently, concerns have been raised about P-chloroaniline as a contaminant in paracetamol pharmaceutical formulations. Interestingly, this impurity is neither controlled in the API nor in the finished paracetamol products. Consequently, a rapid, sensitive, and selective liquid chromatography mass spectrometry (LC/MS) method was developed and applied as a limit test for the determination of P-chloroaniline in 11 paracetamol pharmaceutical formulations. The LC separation was carried out with a C18 column at 35 ℃ with a mobile phase of 0.1% Methanol: 0.1% formic acid (50:50 v/v). Additionally, P-chloroaniline ions were monitored at 127.9/93 and 127.9/111.0 as qualifier and quantifier ions, respectively, using an ESI ion source. The method was validated as a limit test according to ICH Q2 (R1) guidelines and showed high sensitivity and specificity for the determination of P-chloroaniline in paracetamol formulations. Implementing the method for commercial sample analysis revealed its suitability as a new strategy to ensure product compliance with the quality standards.

Introduction

p-chloroaniline (PCA) (Fig. 1) is a polar organic compound and is an important member of aromatic amines that are widely used in several industries, including pesticides, dyes, and pharmaceuticals^1,2. Para-chloroaniline (4-chloroaniline) is an organic molecule used as a crucial intermediary in the manufacture of urea herbicides and insecticides, as well as medicinal and cosmetic products such as chlorhexidine and triclocarban. Triclocarban has been used as an antibacterial agent in soaps, lotions, and deodorants, whereas PCA has been used as an intermediary in the production of triclocarban. This may result in the presence of residues of PCA in finished pharmaceutical and cosmetic products^3,4. For instance, PCA has been identified as an impurity in the antimalarial agent Proguanil Hydrochloride⁵. Typical occurrence of PCA is in mouthwash preparations containing chlorhexidine formulations, where chlorhexidine acts as an antimicrobial agent⁶. The presence of PCA in chlorhexidine formulations could be attributed to many reasons. For instance, it may be due to either its utilization as a starting material in the synthesis of chlorhexidine or the degradation of chlorhexidine to PCA². Furthermore, it could be encountered because of microbial biotransformation of chlorhexidine digluconate³.

Fig. 1.

Chemical Structure of p-chloroaniline (PCA).

According to the United States Pharmacopeia–National Formulary (USP-NF), the content of PCA in chlorhexidine gluconate oral rinses must not exceed 3.0 µg/mL, while in bulk 20% chlorhexidine gluconate solution the limit is set at 500 µg/mL (≈ 500 ppm); similarly, for topical solutions derived from this bulk material, the specification requires that the PCA response corresponds to not more than 500 ppm in the source solution^6,7. In a survey of seven marketed chlorhexidine products analyzed by HPLC, PCA was below the method detection limit in all cases, with a reported LOD of 0.0035 mg/L (3.5 µg/L) and LOQ of 0.0105 mg/L (10.5 µg/L), indicating that commercial preparations typically contain PCA at levels well below pharmacopeial thresholds⁸. Furthermore, an industry technology dossier noted that bulk chlorhexidine digluconate solutions were sometimes manufactured to a tighter in-house specification of 20 ppm PCA, substantially below the compendial limit, to minimize downstream impurity levels in gels and rinses⁹. Regulatory assessments, such as those by the European Medicines Agency, also recognize PCA (listed as Impurity P in the European Pharmacopoeia) as a critical impurity that increases with storage time, reinforcing the importance of strict quality control throughout the product lifecycle¹⁰.

The International Agency for Research on Cancer (IARC) has classified PCA as a group 2B carcinogen, indicating that it is possibly carcinogenic to humans, based on sufficient evidence of carcinogenicity in experimental animals¹¹. To manage its potential risk in pharmaceutical products, the International Council for Harmonisation (ICH) has established a permissible daily exposure (PDE) of 34 µg/day¹². For context, this threshold is several orders of magnitude lower than the typical therapeutic dose of commonly administered drugs, such as a single 500–1000 mg paracetamol tablet, illustrating the extremely low acceptable exposure level for PCA relative to standard pharmaceutical doses. This safety threshold varies depending on the frequency and route of administration of the drug product^6,7. Furthermore, PCA has been detected in industrial effluents¹³, representing a potential ecological hazard, particularly in heavily industrialized water bodies such as the Rhine River in the European Union¹⁴. In addition, the United States Environmental Protection Agency (US EPA) has established a reference dose for chronic oral exposure of 3 mg/kg/day, which provides an additional benchmark for risk assessment and regulatory decision-making¹⁵. Collectively, these considerations highlight the critical importance of rigorous monitoring and control of PCA in both pharmaceutical manufacturing and environmental contexts.

In mid-2020, concerns were raised about PCA contamination in paracetamol, a pharmaceutical ingredient manufactured by certain companies¹⁶. The United States Pharmacopoeia (USP) described a paracetamol monograph, while PCA is not listed as a potential impurity in the drug product¹⁷.

The presence of PCA in paracetamol products may be related to the use of p-aminophenol, the primary starting material in paracetamol synthesis. Typically, p-aminophenol is acetylated with acetic anhydride to produce paracetamol¹⁸. During the production of p-aminophenol, trace contamination with chlorinated analogues, such as PCA, may occur, which can subsequently undergo acetylation to form p-chloroacetanilide (Compound J) (Fig. 2). Compound J is listed among the paracetamol-related impurities in the USP, and the presence of PCA and compound J is often correlated¹⁹. Deacetylation of Compound J can regenerate PCA²⁰. Another paracetamol-related USP impurity, N-phenylacetamide (Compound D, Fig. 3), can also serve as a precursor for PCA formation. Additionally, some studies describe alternative synthetic routes of paracetamol in which PCA is used as a starting material. The method involves heating PCA with ethanol, followed by treatment with formic acid and acetic acid, which produces paracetamol after hydrolysis and dehydration, yielding light brown crystals²¹. Consequently, PCA may be present as an impurity in paracetamol formulations.

Fig. 2.

Acetylation of p-aminophenol and p-chloroaniline.

Fig. 3.

Paracetamol-related substances listed in the USP, PCA was not listed in USP as paracetamol related compounds, however, the acetylated derivative of PCA was listed as compound J in USP.

According to the International Council for Harmonisation (ICH), impurities in pharmaceutical products can be evaluated using one of two principal approaches described in ICH Q2(R1). The first approach is the limit test, which focuses on determining whether an impurity is present above a specified threshold. This approach requires validation of critical method parameters, including specificity, detection limit, and accuracy, to ensure reliable identification of impurities at or above the limit. The second approach is the quantitative impurity test, which involves precise measurement of impurity levels within the product. Both approaches are considered suitable for assessing whether the level of impurities in a pharmaceutical product remains within safe and acceptable limits²².

Searching the literature reveals that several analytical techniques have been applied for the determination of PCA. For instance, a rapid and simple method was developed and validated to determine PCA by using high-performance liquid chromatography (HPLC) and isocratic reversed-phase (RP). This is useful for the quantification of PCA in various pharmaceutical formulations. The study did not report LOD, and the LOQ of 0.25 ppm suggests inadequate sensitivity for detecting PCA at trace levels (i.e., ppb)²³. Thus, to achieve higher sensitivity, an HPLC system or GC system coupled with mass spectrometry, utilizing ionization techniques such as electrospray ionization (ESI) or atmospheric pressure chemical ionization (APCI) is demanded. Different reported studies employed LC-MS as an analytical instrument, utilizing atmospheric pressure chemical ionization (APCI) and electrospray ionization (ESI) for the detection of PCA. With these instruments, LOD was achieved at 27.5 ppb, while the LOQ was reported to be 50 ppb^24,25.

Most of the impurities that are mentioned in different pharmacopeia require very sensitive methods to detect, determine, and quantify these impurities, even though some impurities were not mentioned in certain products in these pharmacopeias. Perhaps some methods were not sensitive enough to determine these impurities owing to the very low concentrations. Only highly sensitive methods can detect such types of impurities. Moreover, to establish a practical and appropriate Limit Test method for the determination of a specific compound, the method needs to be capable of detecting at ppm or even in ppb, depending on the detection limit of the analytical technique employed.

The purpose of the current work is to propose a Limit Test using LC-MS, following ICH requirements for the potential evaluation of paracetamol contamination with PCA. Such an approach could be utilized for screening potential contaminated drug products and to explore unexpected contamination, as well as offering fast results and specific detection capabilities. In addition, this work complies with international standards and regulatory requirements. The LC/MS technique has been applied in this study. This methodology provides high sensitivity, selectivity, and mass accuracy, offering a suitable approach for trace-level detection of PCA.

Materials and methods

Chemicals and reagents

P-chloroaniline (CAS: 106-47-8) was purchased from the United States Pharmacopeia (USP, USA), and methanol (≥ 99.8%, CAS 67-56-1; Fisher Scientific, UK) was obtained from Fisher Scientific (UK). Formic acid (98%) was supplied by Merck (Germany), and water was obtained from VWR International (France).

Instrumentation

The analysis was performed with electrospray ionization (ESI) in positive ion mode. The following conditions were applied: ion spray voltage of + 5 kV, curtain gas pressure of 20 psi, collision energy of 4.5 eV, source temperature of 550 °C, nebulizer gas pressure of 40 psi, and desolvation gas pressure of 45 psi. The liquid chromatography pump was acquired from Agilent Technologies (Agilent Double Binary Pump), and the mass spectrometry from AB Sciex Qtrap 6500. The separation column was C18 -Thermo Fisher Scientific with Particle Size: 3 μm and dimension (Length 50 mm x Diameter 4.6 mm).

Standard solution preparation

A stock solution of PCA was prepared in a concentration of 1000 ppm (1 mg/mL) by dissolving 10 mg of PCA in 10 mL of methanol, and the working solutions were prepared by dilution of the stock solution using the mobile phase, which consisted of (1:1) methanol and water containing 0.1% formic Acid to prepare the calibration solution.

Sample preparation

Sample preparation is a crucial aspect that requires precise adjustment. It is essential to establish a threshold for each product, taking into account daily PCA exposure, detection limits, and the daily product dose. The sample was prepared by measuring an amount equivalent to the established PCA limit, expressed in milligrams for solid samples and microliters for liquid samples. This quantity was then dissolved in an appropriate volume of the mobile phase, which is 10 mL in this study. The amount to be diluted was determined based on the daily dose of each product, as detailed in Table 1. The sample was diluted in accordance with the daily dose of the drug product to ensure that the impurity limit was accurately represented in the test. The calculation of the sample quantity for dilution was performed using the following equation.:

Table 1.

Various dosage forms of Paracetamol and the established threshold of PCA; the calculations were based on PDE 34 µg/day.

	Dosage form	Unit dose	Paracetamol mg/Unit dose	Co-drug class	Product daily dose (Number of units/Day)	PCA Threshold µg/unit dose	Established Threshold ppm	Sample taken
1	Syrup	5 mL	125.00	Antihistamine Decongestant	3.00	11.33	2.266	22.07 µl
2	Syrup	5 mL	120.00	None	6.00	5.67	1.133	44.13 µl
3	IV Solution	100 mL	1000.00	None	4.00	8.50	0.085	588.24 µl
6	Capsule	Content: 558.71 mg	300.00	Muscle Relaxant	6.00	5.67	10.148	4.93 mg
8	Tablet	578.42 mg	300.00	Antihistamine Decongestant	3.00	11.33	19.594	2.55 mg
9	Tablet	721.34 mg	500.00	Stimulant	8.00	4.25	5.892	8.49 mg
10	Sachet	6146.66 mg	600.00	Decongestant	6.00	5.67	0.922	54.23 mg

The limit of quantification (LOQ) was determined to be 0.005 ppm. The established thresholds are detailed in Table 1, expressed in parts per million (ppm) as weight/weight (w/w) for solid dosage forms and weight/volume (w/v) for liquid dosage forms. The peak area of PCA will serve as the acceptance criterion. The sample solution should either contain no PCA or exhibit a PCA peak below quantifiable levels.

Liquid chromatography-mass spectrometry parameters

Preliminary LC-MS parameters were adapted from the method reported by Gonçalves²³. The reported method employed a C18 column for the determination of PCA and chlorhexidine, utilizing gradient elution. The mobile phase consisted of 0.1% formic acid in water as mobile phase A, and 100% methanol as mobile phase B. The gradient elution commenced with 30% mobile phase B, linearly increasing to 95% over a period of four minutes. However, as chlorhexidine is beyond the scope of our method, we utilized isocratic elution with a mobile phase: methanol with 0.1% formic acid and water with 0.1% formic acid (50:50, v/v). The octadecyl silane (C18) column was held at a 35 °C column chamber. The elution was achieved at a flow rate of 0.5 mL/min with a total run time of 5 min. the injection volume was 5 µL. Mass spectrometry was set to monitor 127.9/93 and 127.9/111.0 ions (M + H) using an ESI ion source at 550° C.

Validation parameters

Validation was performed according to ICH Q2(R1) Guidelines²². Specificity and detection limits are obligatory requirements for a Limit Test for each sample matrix. Other parameters, including linearity, accuracy, and repeatability, were optionally added to evaluate the consistency of the system and the method. In addition, the Limit of Quantitation (LOQ) was estimated to evaluate the recovery of the matrix spike, which may indicate the possibility of ion suppression. Furthermore, chlorohexidine tablets, a product known to contain PCA, were used as an independent positive control sample to validate the established threshold.

Results

Method validation

LC-MS was calibrated using standard solutions ranging from 5 to 250 parts per billion (ppb) of PCA solution, and the response was linear (R² > 0.99). To evaluate the specificity of the developed method, each sample matrix was injected with both spiked and unspiked samples, as well as a blank solvent solution. No visible interferences were noticed for all samples (Fig. 4). One ppb was the method limit of detection, while the signal-to-noise ratio was S/N > 1:3. The LOQ was estimated based on the signal-to-noise ratio and the calibration curve standard deviation. To demonstrate the accuracy of the current method, PCA was injected at five different QC concentrations: 10, 30, 50, 100, and 250 ppb (ng/mL). Each QC level was injected three times to determine the relative standard deviation (RSD%) and to evaluate the capability of the Limit Test method accuracy, as shown in Table 2. RSD% was found to be 2.67%, which complies with the acceptance criteria of < 5 at a concentration of 10 ppb. RSD% was found to be 3.58%, which complies with the acceptance criteria of < 5 at a concentration of 30 ppb. RSD% was found to be 2.04%, which complies with the acceptance criteria of < 5 at a concentration of 50 ppb. RDS% was found to be 1.76%, which complies with the acceptance criteria of < 5 at a concentration of 100 ppb and at a concentration of 250 ppb. RSD% was found to be 1.78%, which complies with the acceptance criteria of < 5.

Fig. 4.

Calibration curve for six concentrations of PCA standard solutions.

Table 2.

Results of the QC of PCA contaminant injected three times at five various concentrations – 10,30,50,100, and 250 Ppb (ng/mL) – for the evaluation of the accuracy and method performance.

Level	Target conc. (ppb (ng/mL)	Calculated conc. (ppb (ng/mL)	Average Bias %	% Average	RSD %
1	10.000	10.323	5.406667	10.54067	2.67
1	10.000	10.440
1	10.000	10.859
2	30.000	28.545	−0.86889	29.73933	3.58
2	30.000	30.589
2	30.000	30.084
3	50.000	54.136	9.824	54.912	2.04
3	50.000	56.194
s3	50.000	54.406
4	100.000	100.845	0.595667	100.5957	1.76
4	100.000	98.715
s4	100.000	102.227
5	250.000	255.825	0.9928	252.482	1.78
5	250.000	254.249		10.54067
5	250.000	247.372

The optimized method was validated according to ICH requirements for Limit Test development. Accordingly, the determination of the limit of detection and specificity was required for each sample matrix. A blank sample was utilized for this purpose. Positive control was included in the validation process and had been previously tested by a method to confirm the presence of PCA within the PDE. According to the USP, the limit of PCA in chlorohexidine depends on the purpose of use; however, the daily exposure to PCA should not exceed 34 µg/day.

To determine the accuracy of recovery at the LOQ, a quality control negative sample was evaluated by spiking 5 ppb of PCA standards into the matrix. The accuracy of recovery at the LOQ concentration was within the acceptance criteria of 90–110%, with RSD 1.88% < 5% (Table 3).

Table 3.

Results of the injections for the evaluation of the accuracy of the recovery of the QC negative sample spiked at the LOQ concentration.

Accuracy of LOQ
QC negative sample	Spike (ppb)(ng/mL)	Range	Calculated conc. ppb (ng/mL)	Recovery
1	5	100%	5.367	107.34%
2	5		5.221	104.42%
3	5		5.413	108.26%
Average	106.98
Standard deviation	0.0188
RSD%	1.88

According to the sample preparation described earlier, duplicate injections of each sample matrix, both spiked and unspiked, were conducted to evaluate the specificity and the limit of detection. The method was specific for all sample matrices, and LOD was above 1 ppb, yielding a signal-to-noise ratio of S/N > 1:3. PCA was detected in a retention time of 0.64 min. The developed test was recruited for the analysis of 41 samples representing a variety of dosage forms and sample matrices. All samples were spiked very close to the quantitation limits to evaluate the possibility of ion suppression. It is clear from the great number of data collected that if the method was developed for quantitative purposes, the LOQ would not be less than 5 ppb with S/N > 1:10.

Fig. 5.

Extracted chromatograms and multiple reactions monitoring chromatograms of PCA obtained from blank, negative, positive unspiked, and spiked samples to evaluate specificity and detection limits.

Method application

Samples from the potentially contaminated batches were collected from the local market for further analytical investigation. A total of 41 batches representing 8 finished products from 5 different manufacturers were collected from the market. Since this contamination case was unpredictable, developing a quantitative method that covered these samples was challenging. This was particularly difficult because samples came in five different dosage forms, and some products came in combination with other co-drugs like antihistamines. Effervescent products and syrups acted as an additional challenge due to the presence of colorants and flavoring agents. Furthermore, the level of contamination also could not be predicted.

Discussion

The current Limit Test has been developed and validated as per ICH Guidelines, offering a simple and fast application for screening PCA in real samples. This test is also user-friendly and enables easy identification and determination of PCA in 5 min run time. Furthermore, the reported approach can be applied to evaluate the degree of contamination by any unpredicted contaminants while a rapid and accurate response is demanded.

The sensitivity of the current Limit Test surpasses previous quantitative methods for PCA analysis. The limit of quantification (LOQ) in the proposed study is 10 times lower than that achieved by Gosetti et al.²⁴, where it was 50 ppb (ng/mL). Another study utilized an LC-DAD instrument, achieving a limit of detection (LOD) of 66.5 ppb (ng/mL) and a LOQ of 222 ppb (ng/mL)²⁵, However, the proposed method offers much better sensitivity.

In the current research, PCA impurity analysis achieved a limit of detection above 1 ppb (ng/mL) and a limit of quantification at 5 ppb (ng/mL), one of the lowest concentrations observed compared to previous studies. The method exhibited a calibration range between 5 and 250 ppb, with a linear response (R² > 0.99). As per ICH Q2 (R1) Guidelines, the specificity and limit of detection were sufficient for validating the Limit Test for each sample matrix.

Overall, this study presents a reliable and promising approach for the fast screening of impurities, offering rapid results with a high level of confidence and applicability across a wide range of sample matrices. Subsequently, this study was extended to analyze 41 sample batches, encompassing various dosage forms collected from diverse sources. However, all 41 sample matrices yielded negative results for PCA contamination, which demonstrated that no significant contamination was observed in the market.

The study sought to expand its scope by increasing the sample size and including different dosage forms, some of which were combined with other co-drugs like antihistamines. Effervescent products and syrups posed additional challenges due to the presence of colorants and flavoring agents, making the analysis tedious and time-consuming. Despite these challenges, the method proved reliable and promising for detecting PCA impurities, even at concentrations lower than 4 ppb. The study’s meticulous design and execution are evident in the results presented in Fig. 5, which includes blank samples, negative samples, spiked negative samples with 5 ppb, and unspiked positive samples.

Conclusion

The Limit Test based on USP and ICH using LC-MS was developed for the detection of potential contaminant PCA in 11 different pharmaceutical paracetamol products. The purpose of this work was to investigate the possibility of paracetamol contamination by PCA. It was challenging to develop a quantitative method for each drug product because the validation requirement was long and tedious. Conversely, the Limit Test had only 2 validation parameters and could be evaluated easily in cases where a variety of samples were requested for testing and investigation. The purpose of the Limit Test is to establish a limit for a certain contaminant for each drug product based on certain factors. In this research, a limit was established for each drug product based on permissible daily exposure. Consequently, the limit was validated for each drug product. Notably, the research approach could be applicable in assessing other unpredicted contamination scenarios where fast and reliable results are imperative, and no sufficient validated method exists. In addition to the advantages offered by this method compared to previously reported methods—such as its rapid analysis with a five-minute run time at a 0.5 mL/min flow rate, straightforward sample treatment through dilution with the mobile phase, and its ease of application—it also results in the use of fewer chemicals, a reduced need for analytical tools, and decreased effort required by laboratories and analysts. Future studies could explore areas for enhancement, such as incorporating an internal standard and examining other possible impurities in the developed method.

Author contributions

Yahya M. Alshehri conceptualized the study design and sourced instruments and samples. Thamer S. Alghamdi & Sultan K. Alshammari performed and validated the analytical experiments. Ibrahim A. Al Othaim and Nehad A. Abdallah conducted data analysis and wrote a draft manuscript. Fahad S. Aldawsari and Naif Aljuhani reviewed & edited the manuscript. All authors have read and approved the final version of the manuscript.

Data availability

The data generated or analyzed during this study are included in this published article, any further data request are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Disclaimer

The views expressed in this paper are those of the author(s) and do not necessarily reflect those of the SFDA or its stakeholders. Guaranteeing the accuracy and the validity of the data is the sole responsibility of the research team.