The acquired immune deficiency syndrome (AIDS) is a collection of symptoms and infections resulting from the specific damage to the immune system caused by the human immunodeficiency virus (HIV) in humans (1).
The main goal of the anti-HIV therapy is to prevent disease progression and to reduce morbidity and mortality. Indeed, several classes of anti-HIV drugs, including protease inhibitors (PIs), nucleoside reverse transcriptase inhibitors (NRTIs), non-nucleoside reverse transcriptase inhibitors (NNRTIs), CCR5 antagonists, and integrase inhibitors, have been developed (2).
To ameliorate patient management, the determination of anti-HIV drug plasma levels on a routine basis (i.e., therapeutic drug monitoring; TDM) could represent a useful clinical tool, enabling the study of the relationship between drug plasma levels, drug toxicity, virological response failure, and treatment fine-tuning (3–5). However, although TDM has been incorporated into anti-HIV treatment guidelines (3, 4, 6–8), prospective randomized clinical trials assessing the usefulness of this strategy have shown contradictory results (3, 9). Moreover, the determination of anti-HIV drug concentration in other biological samples, such as peripheral blood mononuclear cells, allows drug monitoring at their pharmacological action sites (10).
Here, the setting up and validation of a new high-performance liquid chromatography-mass spectrometry (HPLC-MS/MS) method, using an hybrid quadrupole time-of-flight mass analyzer (QqTOF), to achieve the reliable, sensitive and selective quantification of lamivudine, lopinavir, ritonavir, and zidovudine in plasma from HIV-infected patients is reported. The great advantage of this method is the possibility to achieve a very high specificity toward the selected anti-HIV drugs, despite the simple and rapid sample preparation, and to extend easily the analysis to co-administrated drugs.
Lamivudine (from Iaf Biochem. Int./Glaxo Wellcome, London, UK), lopinavir (from Abbott, Abbott Park, IL), ritonavir (from Abbott), and zidovudine (from Glaxo Wellcome) were obtained through the NIH AIDS Research Reagent Program, Division of AIDS, NIAID, National Institute of Health (Bethesda, MD). All drugs were of analytical grade. Renin was purchased from PE-Sciex (Concord, ON, Canada). Acetonitrile, methanol, and formic acid (from Carlo Erba, Milano, Italy) were of HPLC grade. Ultrapure water (18 MΩ, TOC <100 ppb) was produced on-site by means of a Millipore MilliQ apparatus (Bedford, MA). All solvents (i.e., acetonitrile, methanol, and formic acid) were degassed by sparging with helium prior to use in the chromatographic apparatus.
All solvents were of analytical grade and were used without further purification.
The chromatographic apparatus for HPLC-MS/MS was constituted by a Series 200 micro LC Pump equipped with a Series 200 Autosampler and a Series 200 Vacuum Degasser from Perkin Elmer (Norwalk, CT). Analytes were chromatographed on a Vydac column (250 × 1 mm i.d.) packed with 3.0-μm C18 particles from Lab Service Analytica (Bologna, Italy).
Mobile Phase Solutions
The mobile phase was composed of acetonitrile as phase A and ultrapure water as phase B, both solvents contained 0.2% (v/v) formic acid. Both the solutions were degassed by sparging with helium. The mobile phase was delivered at 70 μL/min. Gradient elution was performed by linearly increasing the percentage of acetonitrile from 5 to 100% in 16 min.
The PE-Sciex QSTAR Pulsar Hybrid Tandem-MS system (equipped with a TurboIon-Spray source operating in positive ion mode) consisted of a quadrupole mass analyzer (Q1) followed by a quadrupolar collision cell (Q2) and a reflectron time-of-flight (TOF) unit, as the second mass analyzer (QqTOF). The ion spray voltage was 5,500 V. High-purity nitrogen was used as curtain and collision gas, air was used as nebulizer and heater gas. Heater gas was set at 7.0 L/min, and TurboIon-Spray probe temperature was maintained at 300 °C. The settings for the nebulizer and curtain gases were 1.2 and 1.0 L/min, respectively; the gas pressure in the collision cell was 8 mTorr. Mass axis calibration of the mass-resolving quadrupole Q1 was performed by infusion of a polypropylene glycol solution (from Perkin Elmer) at 10 μL/min. Unit mass resolution was established and maintained in the mass resolving quadrupole by keeping a full width at half maximum of ∼0.7 Da. Mass calibration for the TOF analyzer and resolution for TOF-MS and TOF-MS/MS acquisition mode were performed by infusion of the renin solution. Mass resolution calculated at 829.5348 amu was 8,333 full width at half maximum. All the analyses were performed in TOF-MS/MS acquisition mode and in a new working mode that combines TOF-MS and TOF-MS/MS acquisition in a single chromatographic run. The analyses in TOF-MS/MS acquisition mode were performed by four separate product ion scan experiments with the Q2 pulsing function turned on. Values of the declustering potential, the optimum collision energy, the m/z ratio of parent ions selected in the Q1 quadrupole and the Q2 pulsing setting are reported in Table 1. The mass spectrometry data handling system used was the AnalystQS software from PE-Sciex.
Table 1. Experimental conditions for the QqTOF product ion scan determination of anti-HIV drugs
Q2 pulsing setting
Q1 quadrupole (m/z)
Ion release delay (μsec)
Ion release width (μsec)
Declustering potential (Volts)
Collision energy (Volts)
Stock, Working, and Plasma Solutions
Stock solutions of lamivudine, lopinavir, ritonavir, and zidovudine (1.0 mg/mL) were prepared by dissolving 5.0 mg of each anti-HIV drug in 5.0 mL of methanol. The anti-HIV drug concentration ranged between 0.47 and 20 ng/mL for lamivudine, 0.28 and 20 ng/mL for lopinavir, 0.30 and 20 ng/mL for ritonavir, and 0.66 and 20 ng/mL for zidovudine. All solutions were stored at 4.0°C and were stable for at least 6 months.
According to the protocol previously approved by the Ethics Committee of the National Institute for Infectious Diseases I.R.C.C.S. “Lazzaro Spallanzani” (Roma, Italy) and with the written informed consent of the patients, blood samples were taken from HIV-infected patients who were instructed not to take their morning pills prior to the consultation; therefore, the previous drug administration occurred 10 to 12 h before.
The samples (6.0 mL) were collected in monovettes Li heparinate and centrifuged at 3,000 rpm for 20 min at room temperature and the plasma was transferred into Falcon tubes. The plasma was stored at –20°C up to time of analysis.
A solid-phase extraction (SPE) was used for the clean-up of the plasma samples using Waters Oasis HLB 1cc extraction cartridges (Milano, Italy). The chemical composition of the Oasis HLB 1cc extraction cartridges was the lipophilic divinylbenzene and the hydrophilic N-vinylpyrrolidone. The SPE cartridges were conditioned with 1.0 mL methanol followed by 1.0 mL of ultrapure water. A 600-μL plasma sample was diluted with 100 μL methanol in an Eppendorf microvial. The resulting solution was vortex mixed and centrifuged at 13,000 rpm for 6 min. The supernatant was diluted with 1.0 mL ultrapure water and load in the cartridge. Then, the cartridges were washed with 1.0 mL of 5% (v/v) methanol in ultrapure water through the cartridge. Analytes were eluted from the cartridge with 2.0 mL methanol. The eluate was evaporated in a water bath at 36 °C under a stream of nitrogen. The extracted sample was reconstituted with 100 μL methanol and transferred to an injection vial. Finally, 5.0 μL of the sample were injected into HPLC-MS/MS system.
The calibration curves were obtained over the 0.47–20 ng/mL range for lamivudine, 0.28–20 ng/mL range for lopinavir, 0.30–20 ng/mL range for ritonavir, and 0.66–20 ng/mL range for zidovudine. For each amount injected, measurements were made in triplicate.
The absolute recovery of lamivudine, lopinavir, ritonavir, and zidovudine were determined at 5.0 and 10 ng/mL. Six blank samples were fortified with the appropriate amount of mixed standard solution. These samples plus six unfortified blank samples were extracted as previously described. The unfortified blank samples were fortified after being processed (simulated samples) and used as the reference to avoid the matrix effect, causing analyte ion suppression. Therefore, the recovery values were calculated by comparing the areas of the fortified samples versus the simulated samples.
RESULTS AND DISCUSSION
The HPLC-MS/MS method here reported allows the simultaneous measurement of the lamivudine, lopinavir, ritonavir, and zidovudine concentration present in blood samples of HIV-infected patients. The optimized gradient elution profile provides a good separation of anti-HIV drugs, the retention time for lamivudine, zidovudine, lopinavir, and ritonavir was 2.01, 4.40, 14.47, and 14.99 min, respectively. All peaks were symmetrical and well resolved. Figure 1 shows the chromatogram of a healthy donor plasma sample spiked with 100 μL of lamivudine, lopinavir, ritonavir, and zidovudine (5.0 ng/mL). For NRTIs (i.e., lamivudine and zidovudine) and PIs (i.e., lopinavir and ritonavir), the precursor ions [M+H]+ resulted from the addition of a proton to form the positively charged molecular ion. As expected, the full scan mass spectral analyses of lamivudine, lopinavir, ritonavir, and zidovudine shows protonated molecular ions of 230.2, 629.4, 721.9, and 268.1 m/z, respectively. The optimized instrumental parameters are reported in Table 1. A preliminary fragmentation study (Fig. 2) was performed to find the instrumental conditions affording the best sensitivity and thus enabling the unequivocal identification of the very small amounts of analytes present in plasma samples (Table 2).
Table 2. Analyte fingerprints
Parent peaks (m/z)
Major fragment ions (m/z)
This procedure allowed to analyze plasma samples of HIV-infected patients. Figure 3 shows the chromatographic profiles of the plasma sample from HIV-infected patients taking lamivudine (150 mg), lopinavir (133.3 mg), ritonavir (33 mg), and zidovudine (300 mg) twice a day. The plasma concentration of lamivudine, lopinavir, and ritonavir was 390, 940, and 620 ng/mL, respectively. Zidovudine was not detected in plasma of patients since its half-life is about 1 h (11, 12). Values of plasma concentrations of anti-HIV drugs determined by the HPLC-MS/MS method here set up agree well with those reported in the literature (13).
Specificity and Selectivity
Blank samples showed no interfering endogenous substances eluting at the retention time of anti-HIV drugs. Moreover, the selectivity was determined by injecting onto the HPLC column all currently prescribed anti-HIV drugs and/or drugs employed in the treatment/prophylaxis of opportunistic infections, with no evidence of false positivities. These results indicate that the HPLC-MS/MS instrument allows to provide the unambiguous identification of anti-HIV drugs by TOF-MS/MS analysis, full-scan mass spectra throughout each TOF chromatogram, and accurate mass measurements.
The standard curves for lamivudine, lopinavir, ritonavir, and zidovudine are satisfactorily described by unweighted least-squares linear regression. The response was linear between the selected ranges; Table 3 shows the calibration curve parameters.
Table 3. Anti-HIV drug calibration curve parameters
Response range (ng/mL)
y = 7.771x + 29.10
y = 192.0x + 282.3
y = 13.09x + 6.117
y = 8.544x + 25.58
Limits of Detection and Quantification
The limit of detection (LOD) of anti-HIV drugs in plasma is defined as the concentration that yields a signal-to-noise ratio of 3:1. For as the concentration to be accepted as the lowest limit of quantification (LOQ) the percent deviation from the nominal concentration (measure of accuracy) and the relative standard deviation (measure of precision) has to be less than 20%. The LOQ values were 0.47, 0.28, 0.30, and 0.66 ng/mL for lamivudine, lopinavir, ritonavir, and zidovudine, respectively. These values are comparable with the best ones presented in literature (4).
Human plasma proteins were precipitated by adding methanol to the sample and removed by centrifugation. Then, samples were cleaned-up by SPE, a reliable way of eliminating interfering species. The recovery of the anti-HIV drugs lamivudine, lopinavir, ritonavir, and zidovudine ranged between 91 and 107%, according to literature (14) (Table 4).
Table 4. Recovery of anti-HIV drugs after extraction from human plasma
The standard deviation was calculated taking into account six experiments for each concentration.
99.3 ± 5.9
91 ± 2.0
98.1 ± 2.9
107.2 ± 5.0
96.3 ± 4.5
98.0 ± 1.1
102.0 ± 6.2
99.1 ± 3.2
Precision and Accuracy
Intraday and interday precision and accuracy were studied at three different concentrations, ranging between 0.28 and 20 ng/mL. The precision was calculated as the relative standard deviation within a single run (intraday) and between different assays (interday). The accuracy was calculated as the percentage of deviation between the nominal and the found concentration; the results are shown in Table 5. For all the anti-HIV drugs precision was <15% and accuracy ranged between 94 and 110% (14).
Table 5. Intraday and interday anti-HIV drug determination
The standard deviation was calculated taking into account six experiments for each concentration.
0.50 ± 0.02
0.48 ± 0.04
2.43 ± 0.12
2.6 ± 0.12
19.7 ± 1.4
18.9 ± 1.73
0.3 ± 0.009
0.29 ± 0.04
2.48 ± 0.31
2.53 ± 0.36
20.2 ± 0.62
20.1 ± 2.9
0.33 ± 0.04
0.33 ± 0.03
2.54 ± 0.25
2.48 ± 0.31
20.8 ± 1.4
19.9 ± 0.91
0.63 ± 0.08
0.65 ± 0.08
2.6 ± 0.33
2.53 ± 0.3
19.8 ± 1.20
22 ± 2.9
The combination of two NRTIs and one ritonavir-boosted PI/ is currently recommended for first-line therapy of HIV infection. Ritonavir is used at low doses in combination with most PIs to enhance or “boost” the plasma level and to prolong the half-life of the PI. This strategy generally decreases the dosing frequency and the number of pills required, and improves the activity of some PIs. Generally, highly active antiretroviral treatment (HAART) should be selected with consideration of both patient and medication factors. The patient characteristics, adherence history, HIV resistance pattern, and self-defined goals of HAART should be considered in selecting a regimen to which the patient will best adhere. The patient comorbid conditions and potentially interacting medications should be evaluated for possible contraindications or synergism. The antiretroviral history and all resistance profiles should be reviewed carefully so that a regimen that will be likely to achieve durable viral suppression can be chosen (2).
Although the DHHS Guidelines suggest that zidovudine/lamivudine+lopinavir/ritonavir should be considered as an “alternative” and not a “preferred” regimen for initial therapy of HIV-infection, it is still widely used for the treatment of HIV-positive population, few years ago representing a “standard of care.” However, zidovudine/lamivudine+lopinavir/ritonavir isrecommended as a “preferred” regimen in certain patient populations, such as HIV-infected pregnant women and is also commonly used as a postexposure prophylaxis (2). In particular, the HIV-infected patients enrolled in this study took lamivudine (150 mg), lopinavir (133.3 mg), ritonavir (33 mg), and zidovudine (300 mg) twice a day.
SPE with Oasis HLB followed by HPLC-MS/MS, using an hybrid quadrupole/TOF apparatus, is an appropriate method for the determination of anti-HIV drug levels in plasma. This method was validated with regard to specificity, selectivity, linearity, LOD, recovery, precision, and accuracy. The quantification limits obtained for the TOF experiment were poorer, in some cases, than those obtained using a triple-quadrupole mass spectrometer operating under multiple-reaction-monitoring (15). However, the availability of full-scan mass spectra throughout each TOF chromatogram and accurate mass measurements provides qualitative information that could be used to secure identification and quantification of lamivudine, lopinavir, ritonavir, and zidovudine in plasma from HIV-infected patients.
Although a number of analytical methods have been developed for anti-HIV drugs determination in human plasma (4, 10, 16–24), the HPLC-MS/MS method here reported, using a QqTOF mass analyzer, displays great advantages including the possibility to obtain very high specificity towards the selected analytes (e.g., the anti-HIV drugs considered), despite a simple and rapid sample preparation. This target was achieved by the rational exploitation of the chromatographic and mass-spectrometric tools. Moreover, this method is easily extendible to the analysis of co-administrated drugs and to new anti-HIV drugs, the development of anti-HIV drugs being an active field.
The authors thank Dr. Alessandra di Masi for helpful discussion. This work was partially supported by a grant from the Ministry for Health of Italy (National Institute for Infectious Diseases I.R.C.C.S. “Lazzaro Spallanzani”, Roma, Italy, Ricerca corrente 2009 to P.A.)