*Corresponding Author:
Annapoorna Vadlamani
Department of Chemistry
Koneru Lakshmaiah Education Foundation
Green Fields, Vaddeswaram-522 502, India
E-mail: [email protected]
Date of Accepted 14 April 2020
Date of Revised 17 February 2020
Date of Received 08 December 2019
Indian J Pharm Sci 2020;82(3):429-434  

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as the author is credited and the new creations are licensed under the identical terms


A novel, simple, sensitive, specific and rapid reverse-phase liquid chromatography-electrospray ionization-mass spectrometry method was established and validated for the ultra-trace analysis of efavirenz-related compound C, 8-hydroxy efavirenz and its impurity 1 which are potential genotoxic impurities associated with efavirenz. These were separated on high performance liquid chromatography and detected with tandem mass spectrometry in the positive ionization mode with MRM transitions, m/z 314.15>244.10, 212.10>57.10, 305.70>69.05 and 364.15>291.25. A gradient method was developed to separate the impurities using 0.01 M ammonium acetate buffer and methanol at 1 ml/min flow-rate. Samples were analyzed using a Hypersil C-18 column (5 µm, 250×4.6 mm). The calibration curves have shown good linearity in the concentration range of 2.5-78 ppb and a correlation coefficient of 0.99. Mean intra and inter day precision was 2.6-3.2 ppm and 2.5-2.6 ppm, respectively. For the 3 genotoxic impurities the limit of detection and limit of quantitation were 0.04 ppm and 0.125 ppm, respectively. The liquid chromatography-mass spectrometry/mass spectrometry method developed was specific, sensitive, precise and accurate. The developed method could be used for quantification and monitoring of genotoxic impurities of efavirenz.


Efavirenz-related genotoxic impurities, Liquid chromatography-mass spectrometry/mass spectrometry method, Trace-level, Validation

Efavirenz (EFV, (S)-6-chloro-(cyclopropylethynyl)- 1,4-dihydro-4(trifluoromethyl)-2H-3,1-benzoxazin- 2-one) is an effective antihuman immunodeficiency virus type 1 agent that acts as a non-nucleoside reverse transcriptase inhibitor (NNRTI). Normally, NNRTI combination is the best suited for treating naive HIV patients. Some well-known NNRTI combinations are, combinations of EFV with antiretroviral (ARV) drugs like tenofovir, lamivudine, zidovudine and emtricitabine[1]. Furthermore, the mechanism of action of ARV drugs is by diffusion and circulation into the genital tract[2-4].

Pharmacopeia of Brazil, USA, British and Japan, indicated restrictions for active pharmaceutical ingredients (APIs) and formulations for allowable levels of impurities. Moreover, the Food and Drug Administration (FDA) and International Council for Harmonisation (ICH) mentioned strategies for the identification and quantification of impurities along with residual solvent in any in novel dosage forms[5-9]. Moreover, some impurities in trace levels could affect the efficacy and safety of API, as well as be carcinogenic[10,11]. Hence, monitoring and control of trace impurities in any API turn into a very tough assignment. Therefore, the process of minimizing such carcinogenic substances became important in pharmaceutical toxicology[12].

Based on the above documented facts, researchers mainly focused on minimizing the production of impurities in any API manufacturing process[13]. However total elimination of impurities in any process is difficult. So, the method development for accurate identification of impurities is the only option for pharmaceutical industries. Furthermore, USFDA and agency of European medicines proposed toxicological threshold limits to be 1.0-1.5 μg/d[14,15] for genotoxic impurities.

To date, several analytical methods are defined for the quantification of EFV in biological fluids, highperformance liquid chromatography-ultraviolet/ visible (HPLC-UV/Vis) detection[16-20], HPLC[21], liquid chromatography-mass spectrometry (LC-MS)[22], LC-MS/MS[23,24] and ultra-performance liquid chromatography-mass spectrometry/mass spectrometry (UPLC-MS/MS) methods[25,26]. In the present work, we aimed to determine genotoxic impurities in EFV, EFV-related compound C (CPP), 8-hydroxy EFV (CPE), and EFV impurity 1 (EFI, fig. 1) using LC-MS/MS method for quantification of these impurities in EFV formulations.


Figure 1: Chemical structures of analytes
Chemical structures of A. efavirenz (EFV), B. efavirenz-related compound C (CPP), C. 8-hydroxy efavirenz (CPE), D. Efavirenz impurity 1 (EFI)

Materials and Methods

HPLC grades of ammonium acetate and methanol were purchased from Merck (Mumbai, India). CPE, EFV, CPP and EFI were obtained from Perfomics Analytical Labs (Hyderabad, India). A Shimadzu LC-MS/MS- 8050 system associated with the Nexera X2 HPLC and Lab Solutions software v.5.6 was used. Separations were accomplished on a 5 μm particle size of Hypersil C18 column (4.6×250 mm) purchased from Thermo Fisher Scientific.

Method development:

Generally, for any analysis, sample preparation plays an important role; it affects the sensitivity, as well as better recovery of impurities. So, preferable combinations of acetonitrile, water, ammonium acetate, and methanol were used as diluents for chromatographic efficiency. In the present work, 0.01 M ammonium acetate in methanol was chosen as a diluent with column oven at 40º due to good response and recovery for impurities CPE, CPP, and EFI. Also, both isocratic and gradient modes of elution were performed.

Nevertheless, from the observations, it was noticed that all the impurities were effectively separated by the gradient method. Similarly, a column of dimensions, namely Zorbax C8, Hypersil C18 column Phenomenox, Kromasil C8 and C18, were also investigated for resolutions. Finally, Hypersil C18 column was selected due to its better response, peak shape, linearity, and reproducibility even at a lower concentration. Moreover, the positive mode of ESI resulted in improved signal intensities and lesser noise background for impurities, when compared to negative mode.

Method optimization:

Mobile phase A used was prepared by dissolving 0.77 g of ammonium acetate in 1000 ml Milli-Q water by sonication followed by filtration (0.22 μm). Pure HPLC grade methanol was used as a mobile phase B. An LC-MS/MS system, coupled with an 8050 triple quadrupole detector, was used. Separation was achieved on a 5 μm Hypersil C18 column (250×4.6 mm) with injection volume 10 μl, 1 ml/min flow rate, sample cooler temperature at 15° and column oven temperature 40°. Table 1 summarized conditions of MRM, Valco valve, and source gas parameters for mobile phases A and B under gradient mode against the blank solution (diluent). In order to safeguard the ESI sourced from high EFV concentrations, the valve ailment is kept under enabling transferral of eluent to unwanted.

Gradient program
Time (minute) Mobile Phase A Mobile Phase B
0.01 45 55
8.00 45 55
10.00 20 80
13.50 Total Flow 0.8 ml
15.00 45 55
17.00 45 55
17.00 Total Flow 0.8 ml
20.00 Total Flow 1 ml
21.50 20 80
32.00 45 55
37.00 Controller Stop
Multiple reactions monitoring conditions
Impurity MRM Q1 Prebias CE Q3 Prebias Dwell Time (milliseconds)
CPP IMP 212.10>57.10 -26.0 -22.0 -24.0 100
CPE IMP 305.70>69.05 20.0 22.0 24.0 100
Efavirenz Impurity 1 364.15>291.25 22.0 22.0 30.0 100
Efavirenz 314.15>244.10 20.0 19.0 26.0 100
Valco Valve Condition for sample method
Time (min) Command Value
17.00 FCV2= 1
24.00 FCV2= 0

Table 1: Gradient Programme

Standard preparation:

Separately, 2.6 mg of CPE, CPP and EFI were weighed accurately and dissolved completely in 100 ml diluent via sonication. One milliliter of the above impurity/ intermediate standard stock solution was further diluted to 100 ml with diluent. One hundred milligrams of accurately weighed EFV was diluted to 5 ml. To evaluate the system suitability parameters, 10 μl of the above-prepared solution was separately injected namely blank, standard, and sample preparations and their peak area responses were monitored. As per the pharmacopeias, the average peak area response of % relative standard deviation (RSD) of CPE, CPP, and EFI impurities should not be more than 15.0.

Method validation:

Method was validated according to the USFDA and ICH guidelines. The appropriateness and efficacy of the chromatographic scheme were obtained from the system suitability test and it is proficient in the investigation without any bias. To guarantee the capacity of the chromatographic systems, these must placate pre-defined acceptance conditions to implement the examination of various samples. In the contemporary experiment, CPE, CPP, EFI impurity solutions were injected into the LC-MS/MS system for determining system suitability parameters such as peak area and its RSD and retention time, which were detailed after data incorporation using software (Table 2; fig. 2).

System suitability parameters CPP CPE EFI
%RSD of Peak areas obtained from six replicate injections of the standard solution 0.8 1.2 1.6

Table 2: Results for System Suitability


Figure 2: System suitability

The CPE, CPP and EFI impurities were also checked for specificity by injecting them against the blank solution. The outcomes showed that the chosen method is unbiased concerning the presence of further components and interestingly, no nosiness was recorded at the RTs of CPE, CPP, and EFI impurities (Table 3 and fig. 3).

Name Retention time (min) Interference found at the retention time of CPP, CPE, and EFI
CPP 12.940 No
CPE 14.952 No
EFI 27.831 No
Blank NA No

Table 3: Blank Interference Results for CPP, CPE, and EFI


Figure 3: Specificity

Results and Discussion

At limit of quantification (LOQ), 4 levels of the precision method, namely system precision, intermediate precision (ruggedness), method precision (repeatability) and precision were evaluated. System precision suggested inconsistency in the dimensions of the analytical system, while repeatability (method precision) indicated the reproducibility of the method. Standard solution was prepared with CPE, CPP, and EFI impurities and injected (n=6) into the LC-MS/MS system from which peak area and RSD were derived, whereas form method precision, the % RSD data was obtained (Table 4). The results of % RSD of CPE, CPP and EFI impurities were set up to be within the acceptance limits (Table 5).

Preparation CPP (ppm) CPE (ppm) EFI (ppm)
1 2.5 2.8 3.0
2 2.6 2.9 3.1
3 2.7 3.0 3.2
4 2.7 2.9 3.2
5 2.8 3.1 3.4
6 2.5 2.7 3.1
Average 2.6 2.9 3.2
% RSD 4.6 4.9 4.3

Table 4: Method Precision Results for CPP, CPE, and EFI

Preparation CPP (ppm) CPE (ppm) EFI (ppm)
1 2.5 2.6 2.5
2 2.5 2.6 2.5
3 2.6 2.6 2.5
4 2.5 2.6 2.5
5 2.5 2.6 2.5
6 2.6 2.7 2.6
Average 2.5 2.6 2.5
%RSD 2.0 1.6 1.6
%RSD for twelve determinations (Precision & Intermediate Precision)
Preparation CPP (ppm) CPE (ppm) EFI (ppm)
1 2.5 2.8 3.0
2 2.6 2.9 3.1
3 2.7 3.0 3.2
4 2.7 2.9 3.2
5 2.8 3.1 3.4
6 2.5 2.7 3.1
7 2.5 2.6 2.5
8 2.5 2.6 2.5
9 2.6 2.6 2.5
10 2.5 2.6 2.5
11 2.5 2.6 2.5
12 2.6 2.7 2.6
Average 2.6 2.8 2.8
% RSD 4.0 6.5 12.4

Table 5: Intermediate Precision Results for CPP, CPE, and EFI

Furthermore, to create an in lines connection amongst the concentration/quantity of analyte existing in the sample taken and to the detector response, a set of standards of calibration were equipped. Moreover, linearity was calculated by formulating 5 to 150 % of standard concentrations.

The peak area of each sample was noted and plotted against respective concentrations. The Eqn. y=mx+b, defined the linear relation between impurity concentration (x) and respective peak area (y). From this analysis, a correlation coefficient (must be above 0.99) and slope-intercept values were derived (Table 6). Hence, the coefficient for CPE, CPP and EFI impurity peak was found to be 0.99 each, which indicated good linearity (Table 6). Similarly, a signal-to-noise ratio for impurities of limit of detection (LOD) and LOQ were predictable at 3:1 and 10:1, respectively (Table 7). Also, Table 8 and 9 shows the % RSD values of impurities (n=6), and the RSD of peak areas, respectively. The S/N ratio for performing precision of LOQ and LOQ solutions (n=6) was found to be 10:1.

Linearity results
Conc. in ppb Area Conc. in ppb Area Conc. in ppb Area
5 % 2.508 259187 2.531 117538 2.569 305901
10 % 5.224 298024 5.272 137684 5.352 388002
25 % 12.538 1214979 12.653 619137 12.845 1482624
50 % 25.075 2661829 25.306 1341802 25.690 3110678
75 % 37.613 3873591 37.958 1991286 38.534 4712314
100 % 50.046 4817911 50.506 2451722 51.272 5857968
125 % 62.688 6522284 63.264 3349419 64.224 8271495
150 % 75.748 8502919 76.444 4296518 77.604 11062322
Slope 107848 55129 129684
Intercept -165458.4 -87086.3 -234051.5
Correlation 0.9974 0.9994 0.9976

Table 6: Linearity of CPP, CPE, and EFI

Impurity name Limit of detection Limit of quantification
in ppm in ppm
CPP 0.04 0.125
CPE 0.04 0.125
EFI 0.04 0.125

Table 7: LOD and LOQ Establishment Results

LOQ Preparation CPP (ppb) CPE (ppb) EFI (ppb)
Preparation-1 1597 1268 1396
Preparation-2 1601 1301 1403
Preparation-3 1583 1342 1405
Preparation-4 1576 1296 1413
Preparation-5 1605 1304 1426
Preparation-6 1597 1344 1419
Average 1593.2 1309.2 1410.3
%RSD 0.7 2.2 0.8

Table 8: LOQ Precision Results

LOQ preparation CPP (ppm) CPE (ppm) EFI (ppm)
Preparation-1 110.6 116.2 115.3
Preparation-2 114.1 117.1 112.6
Preparation-3 115.7 117.5 111.5
Preparation-4 112.9 115.1 111.6
Preparation-5 116.7 121.6 115.8
Preparation-6 113.7 116.4 115.0

Table 9: LOQ Accuracy Results

A novel LC-MS/MS method was established and validated for the ultra-trace-level identification of three PGI, namely CPE, CPP, and EFI in EFV. For this analysis, a switching valve was used not only to guard the ESI source but also to afford favorable analysis conditions. Furthermore, the MRM-mode method afforded good sensitivity and the LOD and LOQ values were measured to be minimal for all the 3 impurities. Based on the data of analysis, it could be justified that this method is validated completely with acceptable parameters like accuracy, linearity, and precision. Moreover, this method was also found to be simple and sensitive. Hence, this analytical method could be used as a tool for quality control, as well as checking of PGI in EFV.


The authors would like to thank the following for providing necessary facilities to carry out this research work. Koneru Lakshmaiah Education Foundation, Green Fields, Vaddeswaram-522 502, Guntur, Andhra Pradesh, India, and Perfomics Analytical Labs LLP, Nacharam, Hyderabad, Telangana 500076, India.

Conflict of Interest

The authors report no conflict of interest in this work.