Introduction
Hepatitis C virus (HCV) infection is one of the main causes of chronic liver disease, with the number of chronically infected persons worldwide estimated to be about 185 million.1 HCV carries a significant health burden, contributing to 27% of liver cirrhosis and 25% of hepatocellular carcinoma globally.2 The World Health Organization (WHO) has set the ambitious goal of eliminating HCV as a major global public health threat by 2030.3 Timely and effective antiviral therapy for patients with HCV, including those with chronic hepatitis C (CHC), can reduce liver damage and relieve liver fibrosis, thus controlling development of the disease.4
Sofosbuvir, a representative of the direct-acting antivirals, is a uridine nucleotide analogue inhibitor of HCV NS5B polymerase that potently inhibits RNA replicons of genotypes 1–6 and demonstrates a high genetic barrier to resistance. It is phosphorylated within the host hepatocyte to the active triphosphate form, and by competing with the natural nucleotides, causes premature RNA chain termination in the viral genome.5 The brand-named sofosbuvir (Sovaldi®) was approved by the U.S. Food and Drug Administration (FDA) for the treatment of HCV infection in 2013 and then was marketed in China in 2017. It has been incorporated as a first-line agent in the latest guidelines for HCV treatment.6 Because of its excellent pharmacokinetic (PK) profile, sofosbuvir can be administered orally in a single daily dose. In vitro, it exerts potent antiviral effects against HCV.5,7 Previous clinical studies have shown that compared with the traditional pegylated interferon with ribavirin program, sofosbuvir, either in combination with pegylated interferon/ribavirin or in interferon-free combinations, such as sofosbuvir plus vipatavir, shortens the treatment time and improves the sustained virologic response and safety profile of patients.5,8–11
In China, where 10 million individuals are infected with HCV, sofosbuvir is one of the most commonly used drugs for HCV.12 However, the cost of the brand-named sofosbuvir (Sovaldi®) is high (CNY 19,660 or approximately USD 2,800, for one course), and cannot be reimbursed by the medical insurance system in China,5 which restricts the application of this highly efficacious drug in the country. Generic consistency evaluation (GCE) is a key program in China substituting expensive brand-named original drugs with cheap generic drugs. With GCE approval by the China Food and Drug Administration (CFDA), generic drugs can replace brand-named original drugs in clinical practice. Thus, the GCE program is an effective approach for assessing the quality of generic drugs, thus ensuring the safety and efficacy of drugs and saving medical costs.13 Therefore, the development of a cheap generic sofosbuvir that has an equivalent PK profile, with similar clinical efficacy and safety, would reduce the economic burden, increase the application of the drug for patients with HCV infection, and consequently help achieve the goal set by the WHO. Nanjing Chia Tai Tianqing Pharmaceutical Co., Ltd. developed a generic sofosbuvir product, which was approved by the CFDA in 2016 for phase II clinical trials in treating CHC. Moreover, studies on the PKs and bioequivalence of the product in healthy subjects are also required by the CFDA.
Therefore, the aim of this study was to evaluate the bioequivalence and safety of this generic sofosbuvir, with reference to the brand-named sofosbuvir (Sovaldi®), in healthy Chinese volunteers.
Subjects and methods
Subjects
Healthy male and female volunteers aged 18 to 45 years and with a body mass index (BMI) ranging from 19 to 26 (minimal body weight of 45 kg for females and 50 kg for males), were eligible to participate in the study. Subjects were excluded if they were found to have significant cardiac, hepatic, renal, pulmonary, neurological, gastrointestinal, or hematologic disorder or any other relevant clinically significant medical condition, as evaluated by a physical examination and clinical laboratory tests. In addition, subjects with current or previous tobacco or alcohol addictive consumption, drug abuse, or who were taking any prescription, over-the-counter, or investigational medications within 14 days of screening were excluded. Female subjects who were pregnant or lactating were also excluded. Heterosexually active subjects who participated in the study agreed to use protocol-specified contraceptive methods 2 weeks before, during, and 6 months after the study.
The clinical protocol was approved by the Institutional Review Board on Ethics of the Third Hospital of Changsha, following the Clinical Trial Authorization of the CFDA (Approval No. 2016L03881). It was also registered at ChinaDrugTrials.org.cn (identifier: CTR20170117). Written informed consent was provided by all subjects, and the clinical study was conducted in accordance with the ethical principles for human studies of the Declaration of Helsinki, the International Conference on Harmonization Guidelines for Good Clinical Practice, and the CFDA Guidelines and regulations for Good Clinical Principles.
Study design
This study was performed with a single-dose, randomized, open-label, two-sequence, four-period, crossover trial design, to evaluate the bioequivalence and safety of the generic sofosbuvir under both fasting and fed conditions. Eligible subjects were screened and enrolled for the fasting and fed conditions. Under each condition, subjects were randomized to receive initial treatment according to either the test-reference-test-reference (TRTR) or reference-test-reference-test (RTRT) sequence, and then the treatment was switched to the other sequence after a 7-day washout period (Fig. 1). Under the fasting condition, a single dose of either the generic or test sofosbuvir (400 mg tablet, Lot No. 180601; manufactured and supplied by Nanjing Chia Tai Tianqing Pharmaceutical Co., Ltd., Nanjing, China) or the brand-named or reference sofosbuvir (Sovaldi®, 400 mg tablet, Lot No. 17SB001D1; manufactured by Gilead Sciences Cork, Ireland and supplied by Nanjing Chia Tai Tianqing Pharmaceutical Co., Ltd.) was administered to each subject with 240 mL water after at least 10-h overnight fasting. Under the fed condition, subjects consumed an FDA-standardized high-fat, high-calorie meal containing 800–1,000 kcal (approximately 150 kcal protein, 250 kcal carbohydrates, and 500–600 kcal fat) within 30 min after at least 10-h overnight fasting. This was followed by a single dose of either test or reference product administered to each subject with 240 mL water (Fig. 1).
Blood collection and sample preparation
Venous blood samples for sofosbuvir plasma concentrations were collected by inserting a catheter into the forearm vein prior to administration. Under the fasting condition, blood samples were collected within 1 h before dosing, and then 0.167, 0.33, 0.5, 0.75, 1, 1.25, 1.5, 1.75, 2, 2.5, 3, 3.5, 4, 6, 8, 12, 24, 48, and 72 h after dosing. Under the fed condition, blood samples were collected within 1 h before dosing and high-fat, high-calorie breakfast, and then 0.25, 0.5, 0.75, 1, 1.25, 1.5, 1.75, 2, 2.5, 3, 3.5, 4, 5, 6, 8, 12, 24, 48, and 72 h after dosing. At each timepoint, a volume of 3 mL blood was collected into heparinized tubes and centrifuged at 1,500 g and 4 °C for 10 min. The collected plasma was maintained at the study site in an ultra-deep freezer (below −60 °C) for PK analysis.
Determination of plasma sofosbuvir concentrations and PK parameters
The plasma concentrations of sofosbuvir were determined by a validated liquid chromatography-tandem mass spectrometry (LC-MS/MS) method. Sample separation was performed on the ultimate XB-C18 column (2.1 mm × 100 mm, 3 µm; Welch Materials Inc., West Haven, CT, USA) and an analytical column connected with a C18 guard column (Guard Cartridge System, USA) at a temperature of 40 °C. The gradient system consisted of solvent A (0.2% formic acid containing 10.0 mM ammonium acetate) and solvent B (methanol) at a flow rate of 0.3 mL/min. For the analysis of sofosbuvir, sofosbuvir-13C-d3 was used as the internal standard. The gradient program was as follows: 28% solvent B (0–0.4 min), 70% solvent B (0.5–3.2 min), 100% solvent B (3.3–4.3 min), and 28% solvent B (4.4–6.0 min). Positive-mode electrospray ionization was selected for mass spectral analysis, and m/z 530.2→243.1 and m/z 534.2→247.1 for sofosbuvir and sofosbuvir-13C-d3, respectively, were chosen as the mass transitions to detect ions in the multiple reaction monitoring mode. The standard calibration curves for sofosbuvir in human plasma was established within the concentration range of 3.0–2,000.0 ng/mL. The intra-day and inter-day precision and accuracy values of the samples were acceptable, and the relative standard deviations were all <15.0% for the low quantity, middle quantity, and high quantity controls. The intra-day and inter-day precision, relative standard deviations, and accuracy values of the linearity and the lower limit of quantifications were all <20.0%. The recoveries were ranged from 93.2% to 100.6%. The utilized method as indicated by the validation was suitable for large amounts of biomedical samples and provided a reference for several clinical applications and PK studies of sofosbuvir. Noncompartmental PK analysis was performed using the validated Phoenix WinNonlin version 7.0 software to derive PK parameters from plasma sofosbuvir concentrations versus time; the primary endpoints were Cmax, AUC0–t and AUC0–∞, and the secondary endpoints were Tmax, T1/2z, and λz.
Safety assessment
All subjects randomized to receive any of the study drugs in the study were included in the safety analysis. Safety was evaluated by continuously monitoring vital signs, electrocardiography, hematology, biochemistry, urinalysis, and clinical manifestations at baseline, during the dosing and sampling at the designated time points following drug administration, and the day after study completion. Adverse events (AEs) were defined as abnormalities that were considered clinically significant by the investigators after randomization and classified by their intensity into mild, moderate, and severe. Subjects with AEs were further followed up until recovery or improvement.
Statistical analysis and determination of bioequivalence
The PK parameters (Cmax, AUC0–t and AUC0–∞) were derived using unadjusted plasma concentration-time curve, and logarithmically converted. Then, analysis of variance was conducted using a fixed effects model for a four-way crossover design to evaluate the ratio and 90% confidence interval (CI) of the test/reference. According to the bioequivalence standards of the FDA, if SWR (within-subject variability of reference product, within-subject variability (CV%) =100∗eSWR2−1) was <0.294 for either AUC0–t, AUC0–∞ or Cmax of the reference product, then the average bioequivalence (ABE) method was used to determine the bioequivalence between the test and reference products, which was defined if the 90% CI of Cmax, AUC0–t and AUC0–∞ fell within 80–125%. If SWR was ≥ 0.294 for either AUC0–t, AUC0–∞ or Cmax of the reference product, then the reference-scaling average bioequivalence (RSABE) method was used for that measure, and the bioequivalence was defined if the 95% upper confidence limit was ≤0 and the predicted geometric mean ratio for the test/reference ratio was within the range of 0.80–1.25.14
Results
Subjects
Overall, 168 subjects were screened, 91 for the fasting condition and 77 for the fed condition. Sixty-five and fifty-one subjects were excluded from the fasting and fed conditions, respectively. Thus, a total of 52 eligible subjects were enrolled in the study, 26 and 26 for the fasting and fed conditions, respectively. Then these subjects for each condition were equally randomized to follow the TRTR (n = 13) or RTRT (n = 13) sequence. All subjects completed the study according to the study protocol (Fig. 1). There were no differences between the subjects for the two conditions in demographic characteristics, height, weight, and BMI (Table 1)
Table 1Demographic characeristics and body mass index of subjects recruited to the study for the fasting and fed conditions
| Demographics | Fasting (n = 26) | Fed (n = 26) |
|---|
| Male/Female (N) | 18/8 | 18/8 |
| Age (years) | 27.50 ± 7.61 (18–44) | 26.40 ± 7.35 (18–45) |
| Height (cm) | 164.42 ± 6.71 (152–178) | 163.96 ± 6.64 (153–176) |
| Weight (kg) | 58.81 ± 6.87 (46.5–74.4) | 59.93 ± 6.12 (49.5–72.5) |
| BMI (kg/m2) | 21.70 ± 1.65 (19.2–25.3) | 22.29 ± 1.82 (19.7–25.8) |
PK parameters
The PK parameters of the generic and brand-named sofosbuvir products are summarized in Table 2; there were no significant differences in t1/2, Tmax, Cmax, AUC0–t, and AUC0–∞ between the two groups. The mean plasma concentration-time curves and semi-logarithmic curves of the products after a single dose are shown in Figures 2 and 3, and there were no significant differences in the mean plasma concentration between the two groups at the various time points.
Table 2Pharmacokinetic parameters of generic (or test) and reference products of sofosbuvir under fasting and fed conditions
| Parameter | Fasting
| Fed
|
|---|
| Test | Reference | Test | Reference |
|---|
| Tmax (h) | 0.42 (0.333,2.25) | 0.42 (0.333,1.375) | 1.63 (0.5,5) | 1.56 (0.875,5) |
| Cmax (ng/mL) | 980.77 ± 547.32 | 1,026.97 ± 495.51 | 722.75 ± 308.15 | 807.85 ± 361.85 |
| AUC0–t (h·ng/mL) | 847.55 ± 323.00 | 815.61 ± 304.85 | 1,070.67 ± 330.90 | 1,078.71 ± 303.30 |
| AUC0–∞ (h·ng/mL) | 858.35 ± 319.92 | 821.30 ± 305.39 | 1,085.47 ± 334.79 | 1,083.84 ± 309.95 |
| λz (1/h) | 1.42 ± 0.32 | 1.45 ± 0.25 | 1.50 ± 0.20 | 1.54 ± 0.24 |
| T1/2Z(×10−1h) | 5.35 ± 2.05 | 5.26 ± 1.60 | 4.79 ± 0.68 | 4.65 ± 0.79 |
Bioequivalence of the generic product
Under the fasting condition, SWR values for Cmax, AUC0–t, and AUC0–∞ were 0.540, 0.326, and 0.255, respectively, with Cmax and AUC0–t exceeding the regulatory limit of 0.294. Therefore, the RSABE method was applied for Cmax and AUC0–t. It was shown that the 95% upper confidence limit was <0 for both the Cmax and AUC0–t, and the test/reference geometric least square (GLS) mean values for the point estimates of Cmax and AUC0–t were 0.89 and 1.04, respectively, which were within the range of 0.80–1.25. As the AUC0–∞ was below the regulatory limit of 0.294, the ABE method was applied for the analysis of AUC0–∞. It was shown that the ratio of test/reference GLS mean value was 105.29% (90% CI: 97.31–113.93%) for AUC0–∞, which was within the accepted bioequivalence limit (Tables 3 and 4). In the fed condition, SWR values for Cmax, AUC0–t, and AUC0–∞were 0.435, 0.166, and 0.168, respectively; the Cmax exceeded the regulatory limit of 0.294. So the RSABE method was applied for Cmax, the 95% upper confidence limit and the test/reference GLS mean values for the point estimates of Cmax were <0 and 0.91, which were within the range of 0.80–1.25. The AUC0–t and AUC0–∞ were <0.294; the ABE method was conducted analyzing the AUC0–t and AUC0–∞. The ratios of the test/reference GLS means were 99.11% (90% CI, 94.33–104.13) and 100.13% (90% CI, 95.08–105.44%) for AUC0–t and AUC0–∞, respectively, which were within the accepted bioequivalence limits (Tables 3 and 4).
Table 3Bioequivalence analysis of sofosbuvir by logarithmic transformation after a single 400 mg dose under fasting and fed conditions, as determined by the RSABE method
| Parameters | Fasting
| Fed
|
|---|
| N | SWR | Point estimate (90% CI) | Criteria limit of 95% CI | N | SWR | Point estimate (90% CI) | Criteria limit of 95% CI |
|---|
| Cmax(ng/mL) | 26 | 0.54 | 0.89 (0.76–1.05) | −0.119 | 26 | 0.44 | 0.91 (0.82–1.02) | −0.083 |
| AUC0–t(h*ng/mL) | 26 | 0.33 | 1.04 (0.97–1.11) | −0.054 | 26 | 0.17 | 0.99 (0.94–1.04) | −0.014 |
| AUC0–∞(h*ng/mL) | 25a | 0.26 | 1.05 (0.99–1.12) | −0.029 | 24b | 0.17 | 1.00 (0.95–1.05) | −0.015 |
Table 4Bioequivalence analysis of sofosbuvir by logarithmic transformation after a single 400 mg dose under fasting and fed conditions, as determined by the ABE method
| Parameters | Fasting
| Fed
|
|---|
| N | GMR | 90% CI | Power% | N | GMR | 90% CI | Power% |
|---|
| AUC0–t (h*ng/mL) | – | – | – | – | 26 | 99.11 | 94.33–104.13 | >99.99 |
| AUC0–∞ (h*ng/mL) | 25a | 105.29 | 97.31–113.93 | 97.41 | 24b | 100.13 | 95.08–105.44 | >99.99 |
Safety
Throughout the study, a total of 10 AEs were observed (Table 5). Of the 26 subjects under the fasting condition, 1 subject had a mild elevated triglyceride while taking the reference sofosbuvir, which seemed unrelated to the study drug. Of the 26 subjects under the fed condition, 6 (23.1%) experienced nine AEs: three AEs (two mild elevated bilirubin and one diarrhea) were likely related to the study drug, and the other AEs were considered unrelated to the study drug. Throughout the course of this study, all reported AEs were mild in intensity, and no subject withdrew from the study.
Table 5Incidence of adverse events in the subjects while taking the test or reference formulation of sofosbuvir under fasting and fed conditions
| Adverse event | Fasting (N = 26)
| Fed (N = 26)
| Total (N = 52) |
|---|
| Test | Reference | Test | Reference |
|---|
| Elevated triglyceride | 0 | 1 | 0 | 1 | 2 |
| Elevated bilirubin | 0 | 0 | 1a | 1a | 2 |
| Rash | 0 | 0 | 1# | 0 | 1 |
| Decreased globulin | 0 | 0 | 1# | 0 | 1 |
| Elevated total bile acid | 0 | 0 | 1 | 0 | 1 |
| Stomachache | 0 | 0 | 1* | 0 | 1 |
| Diarrhea | 0 | 0 | 0 | 1*a | 1 |
| Elevated serum cholesterol | 0 | 0 | 1* | 0 | 1 |
| Total | 0 | 1 | 6 | 3 | 10 |
Discussion
In this study, all PK parameters including Cmax, AUC0–t, and AUC0–∞ were similar between the test and reference products in healthy Chinese subjects under both fasting and fed conditions. Despite differences in Tmax, there were no significant differences in Cmax, AUC, and t1/2 between the fasting and fed conditions. Overall, the mean plasma concentration-time curves of the generic sofosbuvir and Sovaldi® were similar, regardless of whether the drug was administered under fasting or fed conditions. Throughout the study, AEs were observed in 13.5% of the subjects (3.8% and 23.1% under fasting and fed conditions, respectively). All reported AEs were mild in intensity, and no subject withdrew from the study.
A drug is regarded as a highly variable drug (HVD) if the CV% in the PK measures of Cmax and/or AUC is greater than 30%.15 Since 2006, the FDA has recommended the RSABE method to estimate the bioequivalent of HVDs, which requires that subjects receive twice the reference product to account for with-subject variability. Therefore, a partial replicated (three-way crossover: TRR, RTR or RRT) or a fully replicated (four-way crossover: TRTR or RTRT) study should be performed, and the number of subjects enrolled must be at least 24. The equivalence threshold criteria for HVDs set in The Guidance for Bioequivalence Studies of Highly Variable Drugs (draft),16 published by the CFDA in 2018, are consistent with that by the FDA. As sofosbuvir is a highly variable drug, the present bioequivalence study was performed with a single-dose, randomized, open-label, two-sequence, four-period, crossover trial design, under fasting and fed conditions. According to the results of the present study, the main PK parameters, including Tmax, Cmax, AUC0–t and AUC0–∞, were similar between the test and reference products under both conditions. Therefore, the two products of sofosbuvir meet the equivalence criteria under both fasting and fed conditions. It should be stated that although the elimination half-life of sofosbuvir is less than 1 h, and blood samples 8 h post-dosing would be sufficient for the bioequivalence study as shown in Figure 2, collection of the blood samples was extended for up to 72 h post-dosing for determination of PKs of the generic product and its metabolite.17
When administered in the fasting condition, the test and reference sofosbuvir products were rapidly absorbed, both with a Tmax of 0.42 h. In addition, Cmax, AUC0–t and AUC0–∞ were similar between the test and reference formulations. Both formulations were rapidly eliminated, with half-lives of 0.535 and 0.526 h, respectively. Previously, it was reported that this generic sofosbuvir had similar Tmax, Cmax and AUC0–∞ to that of the brand-named drug in Chinese subjects,18 but these parameters were slightly larger than those of the brand-named drug in Japanese and Caucasian populations.19,20 It is likely that the differences in population and time of blood collection accounted for the different results. When administered in the fed condition, the Cmax, AUC, and Tmax were also similar between the test and reference sofosbuvir products in this study, concordant with the PK parameters reported in the FDA reports for the brand-named sofosbuvir.21 It was observed that, under the fed condition, the absorption of sofosbuvir was delayed with the Tmax of 1.6 h, and the sofosbuvir exposure (AUC) was increased by 1.29-fold, compared to the fasting condition. These findings may be explained by that fact that high-fat foods retard gastric emptying and inhibit epithelial efflux transporters, leading to the blockade of sofosbuvir export and suggesting that the administration of sofosbuvir with food may increase treatment efficacy.22
In this study, sofosbuvir was well tolerated in healthy Chinese volunteers for both formulations. AEs were observed in 3.8% and 23.1% under fasting and fed conditions, respectively; the reasons for the higher rate observed in the fed condition may be related to the subjects' physiological status. According to the physician who determined if an AE was related to the study drug, only three AEs observed under the fed condition were considered to be related to the study drug: one (mild elevated bilirubin) occurred on the generic product, and two (mild elevated bilirubin and diarrhea) on the reference product. After appropriate management, all AEs were recovered. There were no significant differences in the severity and frequency of AEs between the two formulations. Subjects with AEs were further followed up until recovery or improvement. These data suggest that the test sofosbuvir is safe and can offer a new alternative to the brand-named sofosbuvir for millions of patients with HCVs and thus support its clinical development.
Future directions
The future studies are mainly about the re-evaluation in large clinical samples, including effectiveness, safety, and pharmacoeconomic evaluation.
Conclusions
The generic sofosbuvir 400-mg tablet is bioequivalent to the reference sofosbuvir (Sovaldi®) 400-mg tablet under both fasting and fed conditions. In addition, the generic sofosbuvir is as safe and well tolerated as the reference product in healthy Chinese volunteers. These findings suggest that these two products can be used interchangeably in clinical practice.
Abbreviations
- HCV:
hepatitis C virus
- CHC:
chronic hepatitis C
- WHO:
World Health Organization
- FDA:
the US Food and Drug Administration
- PK:
pharmacokinetic
- GCE:
generic consistency evaluation
- CFDA:
the China Food and Drug Administration
- BMI:
body mass index
- AEs:
adverse events
- GLS:
geometric least square
- CI:
confidence interval
- ABE:
the average bioequivalence method
- RSABE:
the reference-scaling average bioequivalence method
- HVD:
highly variable drug
Declarations
Acknowledgement
We would like to thank all volunteers for their participation in this clinical trial.
Funding
This study was supported by the National Major Scientific and Technological Special Project for “Development and Industrialization of Sofosbuvir” (Project: 2018ZX09302050) and the Changsha Science and Technology Plan Project for “Changsha Anti-Infective Drugs Engineering Technology Research Center” (Project: kq1801120).
Conflict of interest
The study drug, sofosbuvir tablets, was supplied by Nanjing Chia Tai Tianqing Pharmaceutical Co., Ltd. Yujie Liu is an employee of Nanjing Chia Tai Tianqing Pharmaceutical Co., Ltd. All other authors have no conflict of interests related to this publication.
Authors’ contributions
Xin Li, the principal investigator of this paper, directed this study. All authors contributed to the data analysis, drafting and revising the article, gave final approval of the version to be published, and agree to be accountable for all aspects of the work.