Pharmacokinetics and Bioequivalence Study of Hydroxychloroquine Sulfate Tablets in Chinese Healthy Volunteers by LC–MS/MS

Introduction Hydroxychloroquine (HCQ), 4-aminoquinoline, is an antimalarial drug and has become a basic therapy for rheumatic disease treatment. It can stabilize the condition of SLE patients and reduce the chances of patient relapse through its immunosuppressive function and antiinflammatory effects. This drug was absorbed completely and rapidly by oral administration, but has a prolonged half-life for elimination. The objective of this study was to evaluate the pharmacokinetic parameters and relative bioequivalence of a new generic (test) formulation with the branded (reference) formulation of HCQ in healthy Chinese male volunteers. This study was designed to acquire regulatory approval for the test formulation. Methods This study was conducted with a randomized, single-dose, two-period, and crossover design. The male subjects were randomly assigned to two groups at a 1:1 ratio to receive 0.2 g hydroxychloroquine sulfate tablets (0.1 g/piece) of the two formulations after a 3-month washout period then administered the alternate formulation. Study drugs were administered after overnight fasting (over 10 h). Plasma concentrations of hydroxychloroquine were measured by a validated LC-MS/MS method. The following pharmacokinetic properties were determined by a noncompartmental pharmacokinetic method: C max, T max, AUC0–t, AUC0–∝, and t 1/2. The bioequivalence between the test and reference products was assessed based on the following parameters: C max, AUC0–60d, and AUC0–∝ using the ANOVA method. If the 90% CI for AUC0–t was within 80–125% and for C max was within 70–143% of the statistical interval proposed by the SFDA, the two formulations were assumed bioequivalent. Concerning the main pharmacokinetic charateristics of hydroxychloroquine, a long half-life drug, the pharmacokinetic parameters of 0–72 h were determined according to the FDA. Furthermore, a comparison was made between the parameters at 0–60 days and 0–72 h to evaluate whether a truncated AUC method can be applied to estimate the relative bioavailability of HCQ. Tolerability was assessed by monitoring vital signs and laboratory tests and by questioning subjects about adverse events. Results The 90% CI of C max for HCQ is 103.8–142.3%; the AUC0–60 is 100–114.2% and AUC0–∝ 100–115.5%. Both met the criteria according to the SFDA’s guidelines for bioequivalence. The relative bioavailability was 109.5% (according to AUC0–60d) and 110.7% (according to AUC0–∝). No serious or unexpected adverse events were observed. Conclusions In this study, the pharmacokinetic studies and results were conducted so that the test and reference formulations of HCQ met the Chinese criteria for assuming bioequivalence. Both formulations were well tolerated in the population studies.

antiinflammatory effects. This drug was absorbed completely and rapidly by oral administration, but has a prolonged half-life for elimination. The objective of this study was to evaluate the pharmacokinetic parameters and relative bioequivalence of a new generic (test) formulation with the branded (reference) formulation of HCQ in healthy Chinese male volunteers. This study was designed to acquire regulatory approval for the test formulation.
Methods: This study was conducted with a randomized, single-dose, two-period, and crossover design. The male subjects were randomly assigned to two groups at a 1:1 ratio to receive 0.2 g hydroxychloroquine sulfate tablets (0.1 g/piece) of the two formulations after a 3-month washout period then administered the alternate formulation. Study drugs were administered after overnight fasting (over 10 h). Plasma concentrations of hydroxychloroquine were measured by a validated LC-MS/MS method. The following pharmacokinetic properties were determined by a noncompartmental pharmacokinetic method: C max , T max , AUC 0-t , AUC 0-µ , and t 1/2 .
The bioequivalence between the test and reference products was assessed based on the following parameters: C max , AUC0-60d, and AUC 0-µ using the ANOVA method. If the 90% CI for AUC 0-t was within 80-125% and for C max was within 70-143% of the statistical interval proposed by the SFDA, the two formulations were assumed bioequivalent.  [3,4]. Now HCQ is considered the secondline treatment for SLE, but it is quite effective therapeutically. One study showed that HCQ can reduce the risk of clinical SLE flares and severe SLE exacerbations [5]. Insulin resistance occurs more frequently in RA and SLE and is a common risk factor for cardiovascular disease (CVD) and diabetes mellitus (DM) [6][7][8]. A study showed that HCQ has a beneficial effect on insulin sensitization by using HCQ in nondiabetic obese subjects for 6 weeks [9]. HCQ also appears to protect against the occurrence of thrombotic events [10]. In addition, the main mechanism of HCQ applied in rheumatic diseases is the inhibition of stimulation of Toll-like receptors (TLRs). TLRs are cellular receptors for microbial products that induce inflammatory responses by activating the innate immune system [11]. Two types of side effects may be encountered with HCQ treatment: one is gastrointestinal intolerance, which usually disappears with dose reduction; the other is rare but potentially severe and involves various combinations of retinal, neuromuscular, cardiac, and hematological impairments. Compared with CQ treatment, retinopathy's incidence in HCQ treatment is very small [12]. HCQ is rapidly and almost completely absorbed after oral administration (the absorption rate through the gastrointestinal tract is 70-80%). However, it has a prolonged half-life (between 40 and 50 days) and low blood clearance (96 ml/min) because of its PK properties [13]. Approximately 50% of the HCQ in plasma is bound to plasma proteins. In the liver, HCQ is metabolized to three active metabolites: desethyl-chloroquine (DCQ), desethyl-hydroxychloroquine (DHCQ), and bis-desethyl-hydroxychloroquine (BDCQ) [14].
Although the pharmacokinetic properties of HCQ have been well identified in previous studies, few studies have been conducted on the PK characteristics of HCQ among healthy Chinese individuals. Although one similar study reported on the bioequivalence of HCQ, it had a parallel study design. However, according to the findings of Tett et al. [16], the bioavailability of HCQ is consistent in each individual at different times but variable between subjects. The FDA also advises that a crossover study design can make the variables determined by physiological factors (such as the clearance, volume of distribution, and absorption) have less interoccasion variability than that arising from formulation performance. Therefore, differences between two products because of formulation factors can be determined [17]. Thus, this study was conducted using a crossover design.
Compared with the preceding study design, this one could eliminate individual differences.
Before a generic product can be marketed in China, the State Food and Drug Administration (SFDA) requires a bioequivalence experiment.
The aim of this study was to compare the relative bioavailability of a new generic formulation (test) of HCQ and the branded formulation (reference) in a Chinese population to meet the SFDA's requirement for marketing the generic formulation in China.

Study Design and Procedures
Healthy Chinese males aged 18-40 years with BMIs between 19 and 24 kg/m 2 were enrolled in the study. Subjects were considered healthy on the basis of medical history, full physical examination, clinical laboratory tests (especially for renal and hepatic function), vital signs (oral body temperature, heart rate, respiratory rate, and sitting blood press), and 12-lead ECGs. Subjects were excluded if they had any impairment of a major organ; had used or abused an illegal drug or alcohol; had psoriasis, active bleeding, colds, or clinically significant abnormalities or hydroxychloroquine-like eye lesions after inspection of the fundus; had a history of mental or neurological disease or glucose-6phosphate dehydrogenase (G-6-PD) defects; had an allergy or sensitivity to 4-aminoquinoline compounds; or had participated in a clinical trial within 2 weeks or donated blood within 2 months prior to the study. The subjects had been informed about the details, including the risks and benefits of this study, and they were free to withdraw at any time.
The study was conducted according to a randomized, open-label, single-dose, twoperiod, and crossover design. Subjects were assigned randomly to two groups to receive a single dose of 0.2 g hydroxychloroquine sulfate tablets (0.1 g/piece) with 250 ml water in the test formulation (Jiangsu Shenhua Pharmaceutical Co., Ltd., Jiangsu, China; lot no. 20090721) or the reference formulation (Shanghai Zhongxi Pharmaceutical Co., Ltd., Shanghai, China; lot no. 081203). Subjects were required to take tablets after overnight fasting (over 10 h) and without breakfast, and they were not allowed to drink alcohol, coffee, and juice, but were allowed to drink water 2 h after administration. They were allowed to have a standard meal 4 h after administration. There was a 3-month drug-free washout period followed by administration of the initial formulation after the alternate formulation had been administered. At the end of the test, subjects were scheduled to have reexaminations of routine blood tests and alanine aminotransferase, aspartate aminotransferase, and creatinine levels.

Chromatographic Conditions
The samples were injected onto a 50 9 2.1-mm 2.6-lm C18 column (Kinetex, Phenomenex) at room temperature. The mobile phase comprised 0.6% formic acid aqueous solution (phase A) and methyl alcohol (80:20, v/v) (phase B) at a flow rate of 0.5 ml/min. The column temperature was at room temperature.

Mass Spectrometric Conditions
The analytes were quantified by mass spectral detection using a mass spectrometer (API 5000, Eventually, we calculated the accuracy and RSD. The results suggested that the lower limit of quantitation could meet the acceptance criteria (Fig. 1).

Specificity
Six randomly selected control blank human plasma samples were processed by a similar extraction procedure and analyzed to determine whether the hydroxychloroquine and chloroquine peaks were well shaped and no  Fig. 2 suggest that the conditions provided high specificity and sensitivity (Table 1) and can accurately determine the concentration of plasma hydroxychloroquine.

Recovery and Matrix Effect
The recovery was evaluated by the response of the analyte recycled from the biological sample matrix divided by the response of the pure standard. The matrix effect experiments were performed by evaluating the ratio between the spiked mobile phase solutions and un-extracted samples spiked on plasma residues (Table 2).

Accuracy and Precision
Both the accuracy and precision evaluations were performed by repeated analysis of hydroxychloroquine in human plasma. The run consisted of a calibration curve and six  Table 3 indicate that the assay method is reproducible for replicate analysis of hydroxychloroquine in human plasma.

Stability
The stability of hydroxychloroquine in plasma was evaluated in the following studies: a stability study at room temperature, a stability study in an auto-sampler, and a freeze-thaw study. LLOQ and QC samples (0.500, 80.0 ng ml -1 ) of hydroxychloroquine were assayed among three batches.

Tolerability
Throughout the study, subjects were monitored by two doctors, two pharmacists, and two nurses. Tolerability was assessed based on vital signs (blood pressure, heart rate, breathing rate), clinical laboratory tests, and 12-lead ECGs.
Physical examinations were performed at baseline and after completion of the study; subjects were interviewed about symptoms of possible adverse events (AEs). Once any undesirable symptoms occurred in subjects, the information would be recorded on the CRF, and the subjects would countinue to be monitored until their physical condition returned to normal.

Study Population
Twenty-one healthy Chinese male volunteers participated in this study; 20 volunteers eventually completed the study. The patient

LC/MS/MS Method Validation
The linear range of hydroxychloroquine was 0.20-100 ng/ml (r = 0.9911). Moreover, the accuracy of the range from 0.20 to 100 ng/ml was between 87.60 and 104.13%, the interanalysis RSD was \6.0%, and the intra-analysis

0-60d
The mean ± SD main pharmacokinetic parameters C max (index of the rate of absorption), AUC0-60d, and AUC 0-µ of the test and reference formulations (n = 20) are shown in Table 4. The mean (SD) concentration-time curves of HCQ after administration of the two formulations are shown in the Fig. 3. The 90% CIs of the ratios (test:reference) for the log-transformed C max and AUC0-60d were 103.8-142.3% and 100.0-114.2%, respectively. The relative bioavailability was 109.5% (according to AUC 0-60d ) and 110.7% (according to AUC 0-µ ).

0-72h
The mean ± SD main pharmacokinetic parameters C max (index of the rate of absorption), AUC0-72h, and AUC 0-µ of the test and reference formulations (n = 20) are shown in Table 5. The mean (SD) concentration-time curves of HCQ after administration of the two formulations are shown in the Fig. 4. The 90% CIs of the ratios (test:reference) for the log-transformed C max and AUC0-72h were 103.8-142.3% and 104.8-117.2%, respectively. The relative bioavailability was 111.8% (according to AUC 0-72h ) and 105.9% (according to AUC 0-µ ).
Both the mean values and standard deviations of the main pharmacokinetic parameters such as C max , T max , AUC 0-60d , and AUC 0-µ were found to be close between the test and reference preparations. In addition, the calculated 90% confidence interval for mean C max , AUC 0-60d , and AUC 0-µ of the two drugs lay within the SFDA's accepted range of 80-125%. Therefore, it could be concluded that the two hydroxychloroquine preparations analyzed were bioequivalent in terms of the rate and extent of absorption.

Tolerability
No serious or unexpected adverse events were observed.

DISCUSSION
The aim of this study was to compare the bioavailability of the test formulation with that of the reference formulation, intending to acquire regulatory approval for the test formulation of HCQ. In this study, the AUC 0-t , AUC 0-µ , and C max of HCQ were defined as the main parameters in order to assess the bioequivalence between both preparations.
The criteria according to the SFDA's guidelines for bioequivalence are the 90% CIs of the test/ reference geometric means ratio in the range of 80-125% for the AUC and 70-143% for C max [15]. The ANOVA results of this study showed that the formulation, period, and sequence had no statistically significant effect on the AUC 0-t , AUC 0-µ , and C max of HCQ. Chinese regulatory authorities do not require the testing of food effects in relative bioavailability studies. Therefore, we only conducted this study under the fasting condition.
In the bioequivalence study of HCQ sulfate tablets (0-60 days), the ANOVA analysis results suggested that the main pharmacokinetic parameters are in accordance with the pharmacokinetics characteristic of a long halflife drug. As previously mentioned, the intra-

CONCLUSION
Based on the pharmacokinetics and the results of this study, it was concluded that the test and reference formulations of HCQ met the Chinese criteria for assuming bioequivalence. Both formulations were well tolerated in the population studies. Moreover, the results of applying a truncated AUC method in this study showed that this method is acceptable for estimating the relative bioavailability of HCQ.

ACKNOWLEDGMENTS
This study was conducted to meet the regulatory guidelines for bioequivalence before allowing a drug on the Chinese market and was approved by the SFDA (approval document no. 2009L00031). Funding for this study was