Abstract
Aim
The aim of this study was to compare the pharmacokinetics (PK) of DRL_BZ with that of EU‐approved (reference medicinal product; RMP) and US‐licensed (reference product; RP) bevacizumab (Avastin®) in healthy male subjects.
Methods
In this double‐blind, parallel‐group, Phase 1 study (BZ‐01‐001), men aged 20–45 years were randomized 1:1:1 to receive a single intravenous infusion of 1 mg kg−1 of bevacizumab as DRL_BZ, RMP or RP. A total of 149 subjects were randomized (DRL_BZ, 50; RMP, 50; RP, 49). Primary endpoints included maximum observed serum concentration (C max), area under the concentration–time curve from time zero (pre‐dose) extrapolated to infinity (AUC(0–∞)), and area under the concentration–time curve from time zero (pre‐dose) to last quantifiable concentration (AUC(0–t)). Secondary objectives were to compare the safety and immunogenicity of DRL_BZ with those of the reference products.
Results
Primary PK parameters were comparable across groups, and 90% confidence intervals for the geometric mean ratios of the primary PK endpoints were within the pre‐specified equivalence margins (80–125%) for all pairwise comparisons (DRL_BZ vs. RMP, DRL_BZ vs. RP and RMP vs. RP). No deaths or serious adverse events were reported. Similar numbers of subjects reported similar numbers of treatment‐emergent adverse events in the three treatment groups. One subject who received DRL_BZ had anti‐drug antibodies at the Day 85 visit; however, no anti‐drug antibodies were detected in this subject at the 12‐month follow‐up visit.
Conclusions
PK, safety and immunogenicity of DRL_BZ were comparable to EU‐approved and US‐licensed bevacizumab in healthy male subjects.
Keywords: oncology, Phase I, bioequivalence, pharmacokinetics
What is Already Known about this Subject
Bevacizumab (Avastin®) is a recombinant humanized monoclonal immunoglobulin G1 antibody that binds to vascular endothelial growth factor and inhibits angiogenesis.
It is approved in Europe and the United States for the treatment of various solid tumours.
For the development of bevacizumab biosimilars, establishing pharmacokinetic equivalence is required.
What this Study Adds
In this first‐in‐human study, pharmacokinetic equivalence of single 1 mg kg−1 intravenous doses of the candidate bevacizumab biosimilar DRL_BZ with the EU‐ and the US‐sourced innovator reference products was established.
The safety of the candidate bevacizumab biosimilar DRL_BZ was comparable to that of the EU‐ and the US‐sourced reference products in healthy human males.
Introduction
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=6771 (Avastin®; Roche Pharma AG, Grenzach‐Wyhlen, Germany; Genentech Inc., South San Francisco, CA, US) is a recombinant humanized immunoglobulin G1 (IgG1) monoclonal antibody 1. It was first approved in the United States (US) and European Union (EU) in 2004 and 2005, respectively, for use in combination with standard chemotherapy for metastatic colon cancer. Since then, it has been approved for use in various other solid tumours as single or combination therapy – metastatic breast cancer; non‐squamous non‐small cell lung cancer; advanced and/or metastatic renal cell cancer; epithelial ovarian; fallopian tube; primary peritoneal cancer; persistent, recurrent or metastatic cervical cancer; and glioblastoma. The recommendations for use 1, 2 vary across countries/regions; for example, metastatic breast cancer is an approved indication in the EU but not in the US, whereas glioblastoma is an approved indication in the US but not in the EU 1, 2.
Bevacizumab acts by binding to and blocking the activity of http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=5085, thereby impeding angiogenesis and inhibiting tumour growth 1. VEGF expression is upregulated and http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=324 are highly expressed in many tumours in humans 3, 4. VEGF plays major roles in a variety of physiological and pathological processes, particularly as a key regulator of angiogenesis 5, 6, leading to the growth of a tumour to a clinically detectable size 3. VEGF levels have been shown to correlate with clinical prognosis 7, 8, 9, 10. Hence, bevacizumab is an important component for oncologists across the globe in the treatment armamentarium of malignant tumours. However, access to bevacizumab is limited by cost in some regions, creating treatment access barriers. The development and introduction of a biosimilar provides an opportunity to increase patient access to biologic therapies and has the potential to generate savings and efficiencies for the healthcare system 11. To this end, Dr. Reddy's Laboratories Ltd., Hyderabad, India, is developing a potential biosimilar of Avastin® (bevacizumab concentrate for solution for infusion [DRL_BZ]). Extensive physicochemical and analytical comparability and preclinical evaluation have already shown DRL_BZ to be comparable to innovator bevacizumab (Avastin®) – US and EU (data on file, Dr. Reddy's Lab. Ltd., Telangana, India). In this first‐in‐human study, the pharmacokinetics (PK), safety, and immunogenicity of a single intravenous 1 mg kg−1 dose of DRL_BZ was compared with that of the reference medicinal product (RMP, EU‐approved Avastin®) and the reference product (RP, US‐licensed Avastin®) in healthy male subjects.
Regulatory agencies require extensive stepwise comparative evaluations of the proposed biosimilars with their reference products. These evaluations start by a comprehensive physicochemical, analytical comparability and pre‐clinical comparability evaluation 12, 13. Comparative PK studies designed to demonstrate a similar PK profile of the biosimilar and the reference product are considered an essential part of the development programme, and the evaluation should include pharmacodynamic (PD) similarity assessments if applicable 14, 15. After PK similarity has been proven, and unless there are surrogate markers for efficacy, the European Medicines Agency (EMA) guidance states that it is necessary to demonstrate comparable clinical efficacy of the biosimilar and the reference product in adequately powered, randomized, parallel‐group comparative clinical trials, preferably double‐blind, in a representative patient population of the approved product indications that is sensitive for detecting potential differences between the biosimilar and the reference product 14. The EMA guidance requests the collection of safety data prior to authorization, depending on the type and severity of safety issues known for the reference product. Care should be taken to compare the type, severity and frequency of the adverse reactions between the biosimilar and the reference product, particularly those described in the Summary of Product Characteristics of the reference product 14. The US Food and Drug Administration (FDA) guidance adds that specific safety or effectiveness concerns regarding the reference product and its class (including history of manufacturing‐ or source‐related adverse events [AEs]) may warrant more comparative clinical data 13.
Both EMA and FDA request the collection of immunogenicity data, depending on the experience gained with the reference product and the product class 12, the extent of analytical similarity between the proposed product and the reference product, and the incidence and clinical consequences of immune responses for the reference product 13. The FDA recommends a comparative parallel design (i.e., a head‐to‐head study) in treatment‐naïve patients to assess potential differences in the risk of immunogenicity 13. The EMA requires that anti‐drug antibody (ADA) responses are assessed in a comparative manner with those of the reference product. The duration of follow‐up should be justified based on the time course and characteristics of unwanted immune responses described for the reference product; for chronic dosing, at least 1 year is recommended 13, 14.
This study was designed primarily to provide the required PK similarity evaluation within the development programme of DRL_BZ. Several other single‐dose comparative PK studies in normal healthy subjects with various other candidate biosimilars have also been published, demonstrating the PK similarity of these products with the reference products with no remarkable safety findings 16, 17, 18, 19, except for a duodenal ulcer perforation 47 days after dosing in a subject with no apparent contributing factors other than a positive test for Helicobacter pylori in one of the studies 19.
A summary of the basis for the approval of a bevacizumab biosimilar (ABP215) in the US has been recently published 20. This published summary remarked that this bevacizumab biosimilar was approved on the basis of an extensive comparative analytical characterization, data obtained in a pharmacokinetic similarity study in healthy subjects, and a comparative clinical study in patients with non‐small cell lung cancer.
Methods
Study design
This randomized, double‐blind, single‐dose, three‐arm, parallel‐group, Phase 1 study was conducted at two clinical centres in New Zealand between November 2015 and June 2016. The study included an up to 21‐day screening period during which, after appropriate informed consent was obtained, all subjects were screened to ensure compliance with inclusion/exclusion criteria. Screening included collection of medical history; physical examination, including vital signs, 12‐lead electrocardiography, and clinical laboratory examinations, including thyroid function tests using standard techniques. If needed, washout of forbidden medications was completed before study drug administration, followed by a 15‐day in‐house stay at the clinical centre, and an 8‐visit/84‐day follow‐up period, with up to 12 months follow‐up for ADA‐positive subjects (Figure S1). Residential stay was required from the evening of the day before study drug administration (Day −1) until Day 15 to ensure subject safety and adherence to the protocol as well as to facilitate the study evaluations. Subjects were discharged on Day 15 (14 days after dosing), after completion of all examinations up to this time point. Block randomization was employed to ensure that the three treatment arms were balanced.
The study was reviewed and approved (Ethics ref: 15/STH/180) by an Independent Ethics Committee (Health and Disability Ethics Committees, Ministry of Health, Wellington, New Zealand) and the New Zealand regulatory authority, Medsafe, prior to initiation. The study was conducted in accordance with the ICH Harmonized Tripartite Guidelines for Good Clinical Practice, the ethical principles laid down in the Declaration of Helsinki, and applicable local regulations. The trial was prospectively registered on http://www.anzctr.org.au (ANZCTR Trial ID: ACTRN12615001204538). All subjects provided written informed consent.
Subjects
Key inclusion criteria were men in general good health, aged 20–45 years, body mass index (BMI) 18.0–28.5 kg m−2, and body weight 50–100 kg, with screening results (vital signs, physical examination, clinical laboratory tests, 12‐lead electrocardiogram [ECG], and thyroid function tests) within the normal or clinically acceptable range. Subjects had to be willing to use appropriate contraceptive measures for 7 months after administration of the study drug. Only male subjects were included, as is usual for bevacizumab biosimilar studies in healthy subjects 16, 17, 18, 19. This is due to the known potential risk of long‐lasting ovarian failure and impaired fertility, as well as the risks related to bevacizumab exposure during pregnancy, when bevacizumab is administered to females 1. Key exclusion criteria were prior exposure to bevacizumab or to any VEGF‐targeted treatment at any time, prior exposure to an investigational monoclonal antibody within 1 year before enrolment, live‐virus vaccination within 3 months prior to screening or intention to receive it within 3 months after administration of study drug, history of immunodeficiency or autoimmune disorders, ongoing or frequent/recurring clinically significant infection (defined as more than three events per year requiring treatment), hypertension, history or presence of kidney disease, clinically relevant non‐healed wounds or fractures, serum alanine aminotransferase and/or aspartate aminotransferase >1.5 times the upper limit of normal at screening or admission to the study centre, and history or evidence of any clinically significant disease. Subjects were required to abstain from alcoholic beverages from 2 days prior to study drug administration until Day 15 post‐dose.
Randomization and treatments
Subjects were randomized in a 1:1:1 ratio to receive a single dose of 1 mg kg−1 of DRL_BZ, RMP or RP. Master randomization was generated by using block randomization to ensure that assignment to each of the treatment groups was balanced. Study drug was administered as a single intravenous administration over 90 min using an electronic infusion pump with a maximum 3‐min interruption during infusion allowed. Sentinel dosing was used for the first three cohorts (3, 6 and 9 subjects were dosed in total, across the three groups in a sequential manner) as an additional safety precaution (Figure S1).
Study subjects were sampled for PK over a period of 85 days from study drug administration. This sampling period was considered sufficient, based on a published population PK model for bevacizumab 21, which found that a two‐compartment model with an initial component mean half‐life of 1.4 days and a second component half‐life of 19.9 days appropriately fit the profiles 21. Hence, the duration of the PK sampling schedule in our study covers more than four times the slowest component or terminal half‐life (t 1/2), and the parallel safety evaluations up to the same time ensure that subjects were followed for a sufficient time, so that little or no drug was remaining in their bodies. The later sample for ADA in most subjects was obtained at the same time to ensure a sufficient washout for proper sensitivity of the assay, while positive subjects were required to be followed up with further samples at quarterly intervals until two negative samples had been obtained or for 1 year, whichever was earlier.
Study endpoints
The primary endpoints were maximum observed serum concentration over the entire blood‐sampling interval (C max), area under the concentration–time curve from time zero (pre‐dose) extrapolated to infinity (AUC(0–∞)), and area under the concentration–time curve from time zero (pre‐dose) to last quantifiable (above lower limit of quantitation [LLOQ]) concentration (AUC(0–t)). Secondary endpoints included time to maximum concentration (t max), volume of distribution at steady state (V ss), t 1/2, and systemic clearance (CL). Safety and immunogenicity parameters were assessed as secondary endpoints – safety was assessed in terms of AEs as well as laboratory investigations, and immunogenicity was assessed using incidence of ADAs, ADA titres and incidence of neutralizing antibodies (NAbs).
Pharmacokinetic assessments
PK sampling time points were pre‐dose (i.e., prior to infusion); mid infusion (i.e., 45 min after start of infusion); at end of infusion; and then 1 h, 2 h, 5 h, 8 h, 12 h, 24 h, 48 h, 96 h, 168 h, 336 h, 22 days, 29 days, 43 days, 57 days, 71 days and 85 days after end of infusion.
A validated bioanalytical method using an enzyme‐linked immunosorbent assay (ELISA) was used for the quantitation of bevacizumab in serum. Briefly, the tested sample (standard, quality control [QC], or study sample) was added to plates coated with human anti‐bevacizumab antibody. Anti‐bevacizumab antibody conjugated with horseradish peroxidase was then added to detect the bound bevacizumab using a colorimetric read‐out. The colorimetric intensity was determined at wavelengths of 450 nm for detection of bound antibody and of 540 nm for non‐specific absorbance to minimize variability. The LLOQ was established at 156 ng ml−1, with a quantification range extending from the LLOQ to 2500 ng ml−1. The mean expected C max values of bevacizumab were about 24 600 ng ml−1 (one‐fifth of the value after administration of 5 mg kg−1 reported in patients by Wu et al. 22) on the basis of the known linearity of bevacizumab PK in the 1–5 mg kg−1 dose range. Hence, the assay LLOQ was expected to adequately cover the study conduct following the requirement for bioanalytical methods intended to be used in bioequivalence studies (and through cross‐referencing biosimilarity studies) of the EMA guideline on Bioanalytical Method Validation 23 of an LLOQ not higher than 5% of the C max. These expectations have been validated a posteriori by the study results (minimum observed C max value of 13 100 ng ml−1), 5% of this value being 655 ng ml−1 as compared to the 156 ng ml−1 LLOQ. Inter‐assay precision and accuracy were established from QCs at five levels (LLOQ, low, medium, high and upper limit of quantitation) in six different validation runs for DRL_BZ, RMP and RP each. Supplementary Table S1 summarizes the results. The performance of the bioanalytical method during this study is presented in Supplementary Table S2. Inter‐assay precision and accuracy confirmed that the performance observed was within acceptable limits. Incurred sample re‐analysis demonstrated reproducible quantitation of the drug in study samples.
For PK parameter calculations, pre‐dose samples that were below the lower limit of quantitation (BLQ) or missing were assigned a numerical value of zero. If any pre‐dose concentration was greater than 5% of that subject's C max, the PK parameters for the given subject were calculated and reported but excluded from statistical summaries and analyses.
For PK parameter calculations, the BLQ concentration values were considered as zero if observed preceding any quantifiable time point, as BLQ/2 if observed between quantifiable values, and as missing if observed after the last quantifiable time point.
PK parameters were calculated by non‐compartmental methods using actual elapsed time from start of study drug administration with Phoenix® WinNonlin® 6.4 (Certara, Princeton, NJ, US). For the calculation of AUC(0–t), the linear‐up/log‐down trapezoidal rule was used.
Safety assessments
The safety and tolerability of all treatments were assessed by means of vital signs, AEs, physical examination, ECG and clinical laboratory safety data. AEs were assessed for severity and relationship to the study drug and graded by the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.03.
Immunogenicity assessment
Blood samples for immunogenicity analyses were obtained within 1 h before the start of study drug infusion and post‐dose on Days 15, 29, 57 and 85 (end‐of‐study visit). The incidence of ADAs was measured. A two‐step testing strategy was used. First, all samples were tested using a screening assay. The samples that tested putative positive at the screening assay were further tested with a confirmatory assay and only considered positive if confirmed by the confirmatory assay. Confirmed ADA‐positive samples were further tested for ADA titre and the presence of NAbs. All subjects who were confirmed positive for anti‐bevacizumab antibodies were required to visit the study centre at 6 months (ADA follow‐up 1), 9 months (ADA follow‐up 2) and 12 months (ADA follow‐up 3) after study drug administration to provide a sample to confirm their ADA status, or until two consecutive samples were negative for anti‐bevacizumab antibodies, whichever came first.
The detection and assay of anti‐bevacizumab antibodies in human serum were performed using ELISA based on the affinity capture elution (ACE) principle. Following the ACE principle, the immune complexes were acid dissociated to release the ADAs bound to them, which were then captured by plates pre‐coated with DRL_BZ. The bound ADAs were further eluted using low pH buffers and transferred and coated onto the fresh plate and detected using digoxigenylated DRL_BZ, followed by anti‐digoxigenin POD (conjugate) and its substrate 3,3′,5,5′‐tetramethylbenzidine for detection. The data were acquired using a SpectraMax Plus 384 Microplate Reader (Molecular Devices, LLC, Sunnyvale, CA, US). Data capture and analysis were performed with SoftMax Pro Software, version 5.2 (Molecular Devices, LLC, Sunnyvale, CA, US).
Assay sensitivity was 3.48 ng ml−1 (without drug) with a drug tolerance of 80.0 μg ml−1 at the positive control low (PCL) of 62.5 ng ml−1. The overall inter‐run assay precision for the QC samples was 20.1%. An assay to detect the neutralizing capacity of ADA was also performed.
The NAb assay was established as a qualitative ligand‐binding assay with ACE procedures incorporated to extract the anti‐bevacizumab antibodies in controls and samples, followed by an ELISA to detect the NAbs. Briefly, the controls and samples were first treated with acid (dissociation step) and then neutralized with basic buffer followed by mixing with bevacizumab‐coupled Dyna beads (bevacizumab beads) and incubated. The ADAs bound to bevacizumab beads were eluted with acid and incubated with DRL_BZ conjugated with digoxigenin (DIG‐bevacizumab), then added to an ELISA plate pre‐coated with VEGF and incubated. With anti‐bevacizumab NAbs in samples and positive controls, the DIG‐DRL_BZ binds to the ADAs, which in turn is sequestered away from binding to the VEGF coated on the plate with reduced response and vice versa. The activity of anti‐bevacizumab NAbs in samples and controls is inversely proportional to the responses.
Statistical analyses
Information on the variability of PK parameters of bevacizumab in healthy subjects was not available at the time this study was designed. Published values for bevacizumab PK‐parameter variability were patient‐based estimates. In addition, a published population PK meta‐analysis of bevacizumab in patients with solid tumours reported that patients with low serum albumin and/or high serum alkaline phosphatase had higher clearance than the other patients 21, suggesting that bevacizumab clearance is dependent on the severity of the disease, a variability factor absent in normal healthy subjects.
Due to the above‐mentioned lack of information, a blinded sample size re‐estimation (BSSR) was performed after the study had enrolled approximately 150 subjects to ensure 120 evaluable subjects, a number considered to ensure a sufficiently accurate estimate of the intersubject variability expressed as geometric coefficient of variation (CV) of the primary PK endpoints.
The estimate of the number of subjects included in the study (n = 150), until the BSSR was performed, was based on the recommendations provided in the publication by Teare et al. 24 who, using a simulation approach, showed that there is little gain in the precision of the variability (pooled standard deviation) estimates for a normally distributed outcome by increasing the study size beyond 35 measured subjects per arm. Given that PK parameters usually follow a log‐normal distribution 25 (converting in a normal distribution after the logarithmic transformation used for analysis), this calculation would be expected to apply for a PK study. The 50 subjects per treatment arm sample size of the part of the study prior to BSSR thus aimed to ensure that at least 35 measurable subjects per study arm were available.
In the BSSR procedure, defined a priori in the study protocol and statistical analysis plan, the pooled geometric CV% for each primary PK parameter was first calculated. The sample size was then re‐estimated by using the maximum CV% of the primary PK parameters in a blinded manner. The BSSR aimed to provide at least 90% power for each of the pairwise comparisons and, as a result, approximately 80% overall power (three pairwise comparisons showing equivalence) for each of the primary PK parameters under the assumption of a DRL_BZ/Avastin® ratio equal to 0.95. If the BSSR results showed that the study had achieved the target statistical power for all primary PK parameters, recruitment had to be stopped. If the study had not yet achieved the target statistical power, the required number of additional subjects (up to a maximum of 300 total subjects to ensure a minimum of 240 evaluable subjects) were to be recruited.
The PK population included all randomized subjects who received the study‐drug infusion, had at least one of the primary PK parameters reliably calculated, and completed the study without major protocol deviations or factors that could significantly affect PK assessment. Exclusions from the PK population were based upon blinded review of the data. Subjects who were confirmed positive for anti‐bevacizumab antibodies were not excluded from the PK population. The safety population consisted of all subjects who were dosed and was analysed according to actual treatment received. The immunogenicity population comprised all subjects included in the safety population.
The primary PK parameters C max, AUC(0–∞), and AUC(0–t) were compared among treatment groups using an analysis of variance model with treatment as a fixed term. PK equivalence was demonstrated if the 90% confidence intervals (CIs) of the geometric mean ratios (GMRs) of C max, AUC(0–∞), and AUC(0–t) (DRL_BZ vs. RMP, DRL_BZ vs. RP and RMP vs. RP) were between 80% and 125%. PK equivalence was tested for the three pairwise comparisons.
Statistical analysis was performed using SAS version 9.4 (SAS Institute, Inc., Cary, NC, US). Figures were generated using SAS and Sigma Plot 12.5 (Systat Software, Inc., San Jose, CA, US).
Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY 26, and are permanently archived in the Concise Guide to PHARMACOLOGY 2017/18 27.
Results
Subject disposition
A total of 149 subjects were enrolled, randomized and received study drug (DRL_BZ, 50; RMP, 50; RP, 49). Subject disposition is shown in Figure 1. One subject, who was later identified as a case of cross‐participation, erroneously received two doses of the study drug (RP), one each at the two study centres, and was excluded from the PK analysis but not from the safety and immunogenicity assessments. The BSSR showed that the sample size was sufficient following the specifications detailed under the Statistical Analysis section, and for this reason recruitment was stopped. The final PK analysis, safety and immunogenicity populations comprised 147, 149 and 149 subjects, respectively.
Figure 1.
Subject disposition. DRL_BZ, Dr. Reddy's Laboratories bevacizumab; RMP, reference medicinal product (EU‐approved Avastin®); RP, reference product (US‐licensed Avastin®); PK, pharmacokinetics
Subject demographic characteristics
Demographic and baseline characteristics of subjects were well balanced among groups (Table 1). The mean (standard deviation) age and BMI of the subjects were 26 (6) years and 23.62 (2.18) kg m−2, respectively (Table 1).
Table 1.
Demographics and baseline characteristics (safety population)
| Variable/Category | DRL_BZ (n = 50) | RP (n = 49) | RMP (n = 50) | Total (N = 149) |
|---|---|---|---|---|
| Sex, n (%) | ||||
| Male | 50 (100.0) | 49 (100.0) | 50 (100.0) | 149 (100.0) |
| Race, n (%) | ||||
| White | 30 (60.0) | 34 (69.4) | 36 (72.0) | 100 (67.1) |
| Asian | 12 (24.0) | 10 (20.4) | 6 (12.0) | 28 (18.8) |
| Black or African American | 1 (2.0) | 0 (0.0) | 1 (2.0) | 2 (1.3) |
| Native Hawaiian or Other Pacific Islander | 0 (0.0) | 1 (2.0) | 1 (2.0) | 2 (1.3) |
| Unknown | 0 (0.0) | 1 (2.0) | 1 (2.0) | 2 (1.3) |
| Other | 7 (14.0) | 3 (6.1) | 5 (10.0) | 15 (10.1) |
| Ethnicity, n (%) | ||||
| Not Hispanic or Latino | 50 (100.0) | 48 (98.0) | 48 (96.0) | 146 (98.0) |
| Hispanic or Latino | 0 (0) | 1 (2.0) | 2 (4.0) | 3 (2.0) |
| Age (years), mean (SD) | 27 (7) | 26 (6) | 26 (5) | 26 (6) |
| BMI (kg m −2 ), mean (SD) | 23.60 (2.095) | 23.21 (2.348) | 24.04 (2.057) | 23.62 (2.181) |
BMI, body mass index; DRL_BZ, Dr. Reddy's Laboratories bevacizumab; RMP, reference medicinal product (EU‐approved Avastin®); RP, reference product (US‐licensed Avastin®); SD, standard deviation
Age was derived as the lower closest integer value of (informed consent date − birth date +1)/365.25
Pharmacokinetics
The overall in‐study performance of the bioanalytical method was acceptable, as summarized in Supplementary Table S2. In addition, 96.8% of the re‐analysed samples met the incurred sample re‐analysis acceptance criteria.
The time vs. serum drug–concentration profile curves for all tested products matched closely over the entire profiling interval (Figure 2). Concentrations remained quantifiable in all evaluated subjects until at least Day 57 after dosing. Several subjects (treated with RP and two treated with RMP) were excluded from analysis under blinded conditions due to missing key samples (the sample obtained at the end of infusion in one case, thus precluding any appropriate calculations; multiple samples at scheduled times starting from 48 h to 1344 h due to study withdrawal by the subject in the others, precluding a proper calculation of AUC(0–t), AUC(0–∞) and t 1/2) for an appropriate and comparable calculation of PK parameters and a further subject treated with RP was considered non‐evaluable due to duplicate study participation. In all subjects included in the final PK analysis, AUC(0–t) accounted for at least 80% of the total AUC (AUC(0–∞)). PK parameters were comparable across tested products (Table 2).
Figure 2.
Mean (±SD) serum concentration–time profiles for all treatments on (A) linear and (B) semi‐logarithmic scales (pharmacokinetic population). Note: Times are relative to start of infusion. DRL_BZ, Dr. Reddy's Laboratories bevacizumab; RMP, reference medicinal product (EU‐approved Avastin®); RP, reference product (US‐licensed Avastin®); SD, standard deviation
Table 2.
Pharmacokinetic parameters for all treatments (pharmacokinetic population)
| Treatment | AUC(0–∞) | AUC(0–t) | C max | t max | V ss | t 1/2 | CL |
|---|---|---|---|---|---|---|---|
| Statistic | (h*μg ml −1 ) | (h*μg ml −1 ) | (μg ml −1 ) | (h) | (l) | (h) | (ml day−1) |
| DRL_BZ (n = 50) | |||||||
| n | 50 | 50 | 50 | 50 | 50 | 50 | 50 |
| Mean | 6420.5 | 6275.6 | 21.29 | NC | 5.858 | 333.18 | 283.359 |
| SD | 1054.36 | 1027.74 | 3.7044 | NC | 0.9804 | 53.244 | 61.0444 |
| CV% | 16.4 | 16.4 | 17.4 | NC | 16.7 | 16.0 | 21.5 |
| Minimum | 3317 | 3234 | 15.3 | 1.50 | 3.81 | 243.4 | 179.03 |
| Median | 6474 | 6292 | 21.2 | 2.50 | 5.74 | 321.5 | 282.07 |
| Maximum | 8662 | 8528 | 30.8 | 13.50 | 8.24 | 501.0 | 544.08 |
| Geo Mean | 6328.4 | 6185.9 | 20.99 | NC | 5.777 | 329.29 | 277.639 |
| Geo CV% | 17.8 | 17.8 | 17.0 | NC | 17.1 | 15.4 | 20.2 |
| RP (n = 47) | |||||||
| n | 43 | 43 | 47 | 47 | 43 | 43 | 43 |
| Mean | 6540.5 | 6397.1 | 21.34 | NC | 5.609 | 329.22 | 269.485 |
| SD | 942.09 | 908.89 | 3.446 | NC | 0.9246 | 54.359 | 45.0248 |
| CV% | 14.4 | 14.2 | 16.1 | NC | 16.5 | 16.5 | 16.7 |
| Minimum | 4780 | 4644 | 15.1 | 1.50 | 3.93 | 231.4 | 211.52 |
| Median | 6632 | 6479 | 20.8 | 2.50 | 5.65 | 323 | 263.45 |
| Maximum | 8582 | 8400 | 31.8 | 13.50 | 8.27 | 454.3 | 415.6 |
| Geo Mean | 6473.5 | 6333.0 | 21.08 | NC | 5.534 | 324.95 | 266.074 |
| Geo CV% | 14.7 | 14.5 | 15.8 | NC | 16.8 | 16.4 | 16.0 |
| RMP (n = 50) | |||||||
| n | 48 | 48 | 50 | 50 | 48 | 48 | 48 |
| Mean | 6786.7 | 6610.0 | 21.79 | NC | 6.0 | 350.23 | 277.405 |
| SD | 1102.3 | 1023.42 | 4.7618 | NC | 0.979 | 59.505 | 52.2431 |
| CV% | 16.2 | 15.5 | 21.9 | NC | 16.3 | 17.0 | 18.8 |
| Minimum | 4831 | 4630 | 13.1 | 1.50 | 4.07 | 258.1 | 173.15 |
| Median | 6710 | 6479 | 20.9 | 2.50 | 5.95 | 341.4 | 275.8 |
| Maximum | 9951 | 9410 | 42.0 | 13.50 | 8.01 | 529.4 | 437.57 |
| Geo Mean | 6701.0 | 6533.1 | 21.33 | NC | 5.92 | 345.61 | 272.566 |
| Geo CV% | 16.2 | 15.6 | 20.5 | NC | 16.8 | 16.4 | 19.3 |
Concentration data from bioanalytical laboratory are in units of ‘ng ml−1’ but applicable PK parameters are presented in units of ‘h*μg ml−1’ and ‘μg ml−1’ for ease of interpretation
AUC(0–∞), area under the concentration–time curve from time zero (pre‐dose) extrapolated to infinity; AUC(0–t), area under the concentration–time curve from time zero (pre‐dose) to last quantifiable concentration; C max, maximum observed serum concentration over the entire blood‐sampling interval; CL, systemic clearance; CV, coefficient of variation; DRL_BZ, Dr. Reddy's Laboratories bevacizumab; Geo, geometric; NC, not calculated; PK, pharmacokinetics; RMP, reference medicinal product (EU‐approved Avastin®); RP, reference product (US‐licensed Avastin®); SD, standard deviation; t 1/2, terminal half‐life; t max, time to maximum concentration; V ss, volume of distribution at steady state
Note: Only n, minimum, median and maximum are presented for t max
The observed GMRs (90% CI) were: for the DRL_BZ vs. RMP comparison 98.40% (92.48–104.70) for C max, 94.69% (89.56–100.10) for AUC(0–t) and 94.44% (89.24–99.24) for AUC(0–∞); for the DRL_BZ vs. RP comparison 99.57% (94.25–105.20) for C max, 97.68% (92.42–103.23) for AUC(0–t) and 97.76% (92.47–103.35) for AUC(0‐∞). For the reference product comparisons (RMP/RP), the observed GMRs (90% CI) were 101.19% (95.19–107.57) for C max, 103.16% (97.92–108.68%) for AUC(0–t) and 103.52% (98.13–109.20%) for AUC(0–∞). Hence, the 90% CIs for the GMRs of the PK primary endpoints (AUC(0–∞), AUC(0–t), and C max) were within the pre‐specified acceptance margins (80–125%) for PK equivalence for all three comparisons: DRL_BZ vs. RMP, DRL_BZ vs. RP and RMP vs. RP (Table 3).
Table 3.
Statistical comparison of key pharmacokinetic parameters (pharmacokinetic population)
| Pairwise comparison | |||||||
|---|---|---|---|---|---|---|---|
| Parameter (units) | Treatment | n | Geometric least‐squares means | 95% CI | Pair | Ratio (%) | 90% CI |
| C max (μg ml −1 ) | DRL_BZ | 50 | 20.99 | 20.01, 22.02 | DRL_BZ/RMP | 98.40 | 92.48, 104.70 |
| RP | 47 | 21.08 | 20.13, 22.08 | DRL_BZ/RP | 99.57 | 94.25, 105.20 | |
| RMP | 50 | 21.33 | 20.14, 22.60 | RMP/RP | 101.19 | 95.19, 107.57 | |
| AUC (0–t) (h*μg ml −1 ) | DRL_BZ | 50 | 6186 | 5883, 6504 | DRL_BZ/RMP | 94.69 | 89.56, 100.10 |
| RP | 43 | 6333 | 6058, 6621 | DRL_BZ/RP | 97.68 | 92.42, 103.23 | |
| RMP | 48 | 6533 | 6246, 6834 | RMP/RP | 103.16 | 97.92, 108.68 | |
| AUC (0–∞) (h*μg ml −1 ) | DRL_BZ | 50 | 6328 | 6019, 6654 | DRL_BZ/RMP | 94.44 | 89.24, 99.94 |
| RP | 43 | 6473 | 6189, 6771 | DRL_BZ/RP | 97.76 | 92.47, 103.35 | |
| RMP | 48 | 6701 | 6395, 7021 | RMP/RP | 103.52 | 98.13, 109.20 | |
Results based on analysis of variance model with treatment as a fixed effect on the log‐transformed values of C max, AUC(0–t) and AUC(0–∞).
AUC(0–∞), area under the concentration–time curve from time zero (pre‐dose) extrapolated to infinity; AUC(0–t), area under the concentration–time curve from time zero (pre‐dose) to last quantifiable concentration; C max, maximum observed serum concentration over the entire blood sampling interval; CI, confidence interval; DRL_BZ, Dr. Reddy's Laboratories bevacizumab; RMP, reference medicinal product (EU‐approved Avastin®); RP, reference product (US‐licensed Avastin®)
Safety
No deaths or serious AEs were reported during the study, and no subjects had infusion discontinuation because of AEs. A similar number of subjects were reported to have a similar number of treatment‐emergent AEs (TEAEs) in the three treatment groups: DRL_BZ, 89 TEAEs were reported in 37 subjects (74.0%); RMP, 105 TEAEs were reported in 35 subjects (70.0%); and RP, 89 TEAEs were reported in 38 subjects (77.6%). The number of subjects who received RMP or RP and had TEAEs considered by the investigator as related to the study drug was comparable (11 subjects [22.0%] with 20 TEAEs and nine subjects [18.4%] with 13 TEAEs, respectively). The number of subjects with treatment‐related TEAEs was numerically fewer with DRL_BZ (five subjects [10.0%] with seven TEAEs).
The most frequently reported TEAE considered related to the study drug was headache (DRL_BZ, 10 TEAEs for nine subjects [18.0%]; RMP, 24 TEAEs for 20 subjects [40.0%]; and RP, 14 TEAEs for 12 subjects [24.5%]) (Table 4). Most of the TEAEs were grade 1 in severity (81 subjects [54.4%] with 248 TEAEs) and resolved by the end of the study. Only one grade 3 severity TEAE (RMP treatment group; abdominal pain on Day 76 that lasted for 8 days) was reported; however, it was deemed unrelated to the study drug.
Table 4.
Summary of all drug‐related TEAEs (occurring in any group >0%) by System Organ Class and Preferred Term for each treatment (safety population)
| Number (%) of subjects/Number of events | ||||
|---|---|---|---|---|
| Treatment group | ||||
| DRL_BZ (n = 50) | RP (n = 49) | RMP (n = 50) | Total (N = 149) | |
| Subjects with TEAEs | 5 (10.0) /7 | 9 (18.4) /13 | 11 (22.0) /20 | 25 (16.8) /40 |
| Cardiac disorders | 1 (2.0) /1 | 0 (0.0) /0 | 0 (0.0) /0 | 1 (0.7) /1 |
| Palpitations | 1 (2.0) /1 | 0 (0.0) /0 | 0 (0.0) /0 | 1 (0.7) /1 |
| Gastrointestinal disorders | 1 (2.0) /1 | 2 (4.1) /2 | 3 (6.0) /4 | 6 (4.0) /7 |
| Abdominal pain | 0 (0.0) /0 | 2 (4.1) /2 | 0 (0.0) /0 | 2 (1.3) /2 |
| Vomiting | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| Mouth ulceration | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| Nausea | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| Aphthous ulcer | 1 (2.0) /1 | 0 (0.0) /0 | 0 (0.0) /0 | 1 (0.7) /1 |
| Dysphagia | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| General disorders and administration site conditions | 1 (2.0) /1 | 2 (4.1) /2 | 0 (0.0) /0 | 3 (2.0) /3 |
| Catheter site inflammation | 1 (2.0) /1 | 0 (0.0) /0 | 0 (0.0) /0 | 1 (0.7) /1 |
| Fatigue | 0 (0.0) /0 | 1 (2.0) /1 | 0 (0.0) /0 | 1 (0.7) /1 |
| Influenza‐like illness | 0 (0.0) /0 | 1 (2.0) /1 | 0 (0.0) /0 | 1 (0.7) /1 |
| Infections and infestations | 0 (0.0) /0 | 0 (0.0) /0 | 2 (4.0) /2 | 2 (1.3) /2 |
| Oral herpes | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| Pharyngitis | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| Musculoskeletal and connective tissue disorders | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| Myalgia | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| Nervous system disorders | 2 (4.0) /3 | 3 (6.1) /4 | 5 (10.0) /7 | 10 (6.7) /14 |
| Headache | 1 (2.0) /1 | 2 (4.1) /2 | 5 (10.0) /6 | 8 (5.4) /9 |
| Dizziness | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| Sensory disturbance | 1 (2.0) /1 | 1 (2.0) /1 | 0 (0.0) /0 | 2 (1.3) /3 |
| Muscle contractions involuntary | 1 (2.0) /1 | 0 (0.0) /0 | 0 (0.0) /0 | 1 (0.7) /1 |
| Respiratory, thoracic and mediastinal disorders | 1 (2.0) /1 | 4 (8.2) /5 | 3 (6.0) /4 | 8 (5.4) /10 |
| Oropharyngeal pain | 0 (0.0) /0 | 1 (2.0) /1 | 1 (2.0) /1 | 2 (1.3) /2 |
| Epistaxis | 0 (0.0) /0 | 2 (4.1) /3 | 1 (2.0) /2 | 3 (2.0) /5 |
| Nasal congestion | 0 (0.0) /0 | 1 (2.0) /1 | 1 (2.0) /1 | 2 (1.3) /2 |
| Cough | 1 (2.0) /1 | 0 (0.0) /0 | 0 (0.0) /0 | 1 (0.7) /1 |
| Skin and subcutaneous tissue disorders | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /2 | 1 (0.7) /2 |
| Pruritus | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
| Rash | 0 (0.0) /0 | 0 (0.0) /0 | 1 (2.0) /1 | 1 (0.7) /1 |
System Organ Class and Preferred Term were from the MedDRA dictionary, Version 19.0. The numbers of subjects within each column cannot be added because a subject may have had more than one adverse event. A subject experiencing multiple occurrences of an adverse event was counted, at most, once per System Organ Class, Preferred Term and relationship
If a subject had more than one identical event that is both related and unrelated, then only the related event in each System Organ Class and Preferred Term was counted
Related TEAEs are defined as those assessed as related to study drug, or those for which the relationship is unknown or missing.
DRL_BZ, Dr. Reddy's Laboratories bevacizumab; MedDRA, Medical Dictionary for Regulatory Activities; RMP, reference medicinal product (EU‐approved Avastin®); RP, reference product (US‐licensed Avastin®); TEAE, treatment‐emergent adverse event
No clinically significant findings with respect to vital signs, ECG or laboratory parameters were reported in any of the treatment groups.
Immunogenicity
One subject who received DRL_BZ had ADA at the end‐of‐study visit (Day 85) with a titre of 1. The NAb result at the same time point was negative. At the 6‐month follow‐up, the ADA titre increased to 2 and the antibodies were proven to be neutralizing. At the 9‐month follow‐up, the titre decreased to non‐reportable levels and the antibodies were proven to be non‐neutralizing. At the final follow‐up at 12 months post‐dose, the subject was ADA‐negative. For no other subjects were ADAs detected in any sample.
As shown in Table 5, the PK parameters determined in the subject with positive ADA did not show any extreme values, and as shown in Figure 3, the bevacizumab time–concentration profile in this subject did not appear to be different from that of the other study subjects.
Table 5.
Selected pharmacokinetic‐parameters comparison between the subject showing a positive result for ADA with subjects in the full treatment arm (DRL_BZ)
| Parameter | Full‐treatment‐arm values | Value in the ADA‐positive subject | |
|---|---|---|---|
| C max (μg ml −1 ) | Minimum | 15.3 | 24.3 |
| Median | 21.2 | ||
| Maximum | 30.8 | ||
| AUC (0–∞) (h*μg ml −1 ) | Minimum | 3317 | 6702 |
| Median | 6474 | ||
| Maximum | 8662 | ||
| t 1/2 (h) | Minimum | 243.4 | 321.5 |
| Median | 321.5 | ||
| Maximum | 501.0 | ||
ADA, anti‐drug antibody; AUC(0–∞), area under the concentration–time curve from time zero (pre‐dose) extrapolated to infinity; C max, maximum observed serum concentration over the entire blood‐sampling interval; DRL_BZ, Dr. Reddy's Laboratories bevacizumab; t 1/2, terminal half‐life
Figure 3.
Spaghetti plots of serum bevacizumab concentration–time profiles in subjects with and without ADA (only shown for DRL_BZ; no subjects with positive ADA in other arms) on a semi‐logarithmic scale. Black lines: Subjects not positive for ADA. Red line: Subjects positive for ADA. ADA, anti‐drug antibody; DRL_BZ, Dr. Reddy's Laboratories bevacizumab
Discussion
This first‐in‐human study primarily aimed to comparatively evaluate the PK profile of DRL_BZ with those of the reference products, EU‐approved (Avastin®) and US‐licensed bevacizumab (Avastin®) as requested by the regulatory guidelines (reviewed earlier) in the setting of the development of a biosimilar product 12, 13, 14, 15, 20. The secondary objectives were to compare the safety and immunogenicity of DRL_BZ with those of the reference products.
In this study, PK equivalence was demonstrated for all comparisons (DRL_BZ vs. RMP, DRL_BZ vs. RP and RMP vs. RP), as all 90% CIs for the GMR of AUC(0–∞), AUC(0–t), and C max between products were within the predefined 80–125% bioequivalence limits.
The LLOQ of the assay (156 ng ml−1 or 0.156 μg ml−1) was lower than 5% of all measured C max values during the study (minimum value 13.1 μg ml−1 [Table 2]), thus fulfilling the requirement for bioanalytical methods used in bioequivalence studies (and through cross‐referencing biosimilarity studies) of the EMA guideline on Bioanalytical Method Validation 23 of an LLOQ not higher than 5% of the C max.
Establishing PK equivalence is necessary to establish similarity of a proposed biosimilar to the reference product and should be accomplished in a study evaluating a population, route of administration and dose that are adequately sensitive for detection of PK differences 13. A comparable PK profile (such as demonstration of comparable exposure) helps support the conclusion that there are no clinically meaningful differences between a proposed biosimilar and its reference product 13.
The use of normal healthy subjects allows a long enough sampling to properly elucidate the terminal phase of the profile, as in this study. Patients on treatment with bevacizumab need to be dosed at 2‐ to 3‐week intervals 1, 2, which, given the observed half‐life values (ranging from 231.4–529.4 h; i.e., >1 week to >3 weeks; Table 2), are too short for a proper elucidation, whereas the present study had a 12‐week sampling schedule post‐administration, which has been shown above to be sufficient for full PK profiling. In addition, patient populations have many potential confounding factors such as underlying disease burden, concomitant illnesses/co‐morbidities and concomitant medications 15. It is also known that factors related to disease severity are significant covariates for key bevacizumab PK parameters in patients 21. All these factors confirm the choice of a healthy population for this comparative PK study.
A parallel‐group design (as used in this study) is considered appropriate for products that have a long half‐life (values in this study ranging from 231.4–529.4 h; Table 2) or for products where repeated exposures can lead to an increased immune response that can affect the PK and/or PD similarity assessments 15.
The dose used to demonstrate bioequivalence should be in the range where linear PK behaviour is known for the reference drug if the reference drug displays linear PK behaviour in its therapeutically used dose range 28, 29. PK of bevacizumab is known to be linear in the 1.0–10 mg kg−1 dose range 2. While higher doses have been used in studies of other bevacizumab biosimilar candidates (including 3.0 mg kg−1 18, 19 and 5.0 mg kg−1 17), the lowest dose of 1 mg kg−1 in this linear dose range was considered for this study, in the safety interest of the subjects, while still allowing for appropriate PK assessments.
The results of this study show that the 1 mg kg−1 dose was appropriate for PK similarity evaluation. Measured AUC (AUC(0–t)) accounted for more than 80% of total AUC (AUC(0–∞)) in all subjects analysed for AUC (exclusions were due to missing samples or double participation). Both measured and total AUC were evaluated in 100% (50/50), 96.0% (48/50) and 91.5% (43/47) of subjects (Table 3) in the DRL_BZ, RMP and RP groups, respectively. Thus, the measured AUC accounted for more than 80% of total AUC in more than 80% of the subjects 29. Furthermore, the log‐linear regression of the concentration decline over time used to calculate the terminal phase rate had a square coefficient of correlation of at least 0.90 (data on file, Dr. Reddy's Lab. Ltd., Telangana, India) in all subjects not excluded in the Blinded Data Review Meeting due to missing samples, supporting that the terminal phases were well elucidated and that the 1 mg kg−1 dose is sufficient to fulfil the objectives of the study. Hence, the PK similarity conclusion from this study has been well established.
DRL_BZ, RMP and RP administered as a single 1 mg kg−1 intravenous dose were comparably well tolerated in the study population. No new or unexpected safety concerns were raised.
One subject was ADA‐positive at the end of the study (in the DRL_BZ group). The samples obtained at the 6‐month and 9‐month follow‐up were also positive for ADA, with the sample obtained at the 6‐month follow‐up showing NAbs. The subject became ADA‐negative at the 12‐month follow‐up. The PK results in this subject were not noticeably different from those in the other subjects. Because ADAs were observed in a single subject and they did not appear to have induced extreme PK results, they are not expected to have influenced the study results. Thus, the immunogenicity profile was comparable among DRL_BZ, RMP and RP, even if definite conclusions cannot be made due to the low incidence of ADAs given the sample size of the study, optimized for PK parameter comparisons. The low rate of immunogenicity observed in this study is in line with published data and the label/Summary of Product Characteristics 1, 2, 17, 18, even if one of the references describing other studies with candidate bevacizumab biosimilars 19 reported all subjects being ADA‐positive by Day 14, and seven and six subjects out of a total of 57 per arm remained positive by Day 78. The authors of that study commented that the high incidence of ADA on Day 14 in their study might be due to assay interference by bevacizumab‐induced VEGF homodimers 19.
In conclusion, PK, safety and immunogenicity of DRL_BZ were comparable to innovator EU‐approved and US‐licensed bevacizumab in healthy male subjects. The results support continued development of DRL_BZ as a candidate biosimilar to bevacizumab.
Competing Interests
C.W. and C.S. have received pharmaceutical company funding. S.S.B., L.L.L. and S.K. are employees of and own stock options in Dr. Reddy's Lab. Ltd.
This study was funded by Dr. Reddy's Laboratories Ltd. Medical writing support was provided by Steven Tresker of Cactus Communications, funded by DRL. The authors retained full control of the manuscript content. We thank Dr. Priyadarshini Roy (Dr. Reddy's Lab. Ltd., Biologics) for review support.
Contributors
C.W. was involved in the conception, design or planning of the study, acquisition of data, interpretation of the results, and drafting of the manuscript, and also was the Principal Investigator. C.S. was involved in the acquisition of data, interpretation of the results, and drafting of the manuscript. S.S.B., L.L.L. and S.K. were involved in the conception, design or planning of the study, interpretation of the results, and drafting of the manuscript. All authors critically revised the manuscript for important intellectual content and approved the final version of the article.
Supporting information
Table S1 ELISA inter‐assay precision and accuracy: validation
Table S2 ELISA inter‐assay precision and accuracy: performance during study sample analysis
Figure S1 Study schematic. DRL_BZ, Dr. Reddy's Laboratories bevacizumab; RMP, reference medicinal product (EU‐approved Avastin®); RP, reference product (US‐licensed Avastin®); PK, pharmacokinetics
Wynne, C. , Schwabe, C. , Batra, S. S. , Lopez‐Lazaro, L. , and Kankanwadi, S. (2018) A comparative pharmacokinetic study of DRL_BZ, a candidate biosimilar of bevacizumab, with Avastin® (EU and US) in healthy male subjects. Br J Clin Pharmacol, 84: 2352–2364. 10.1111/bcp.13691.
References
- 1. Avastin® (bevacizumab) prescribing information. Available at https://www.gene.com/download/pdf/avastin_prescribing.pdf. Genentech, Inc. 2016 (last accessed 14 November 2017).
- 2. Avastin® (bevacizumab) summary of product characteristics. Available at http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000582/WC500029271.pdf. Roche Registration Limited. 2015 (last accessed 14 November 2017).
- 3. Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1995; 1: 27–30. [DOI] [PubMed] [Google Scholar]
- 4. Sharma PS, Sharma R, Tyagi T. VEGF/VEGFR pathway inhibitors as anti‐angiogenic agents: present and future. Curr Cancer Drug Targets 2011; 11: 624–653. [DOI] [PubMed] [Google Scholar]
- 5. Ferrara N, Gerber H‐P. The role of vascular endothelial growth factor in angiogenesis. Acta Haematol 2001; 106: 148–156. [DOI] [PubMed] [Google Scholar]
- 6. Dvorak HF, Brown LF, Detmar M, Dvorak AM. Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 1995; 146: 1029–1039. [PMC free article] [PubMed] [Google Scholar]
- 7. Karayiannakis AJ, Syrigos KN, Polychronidis A, Zbar A, Kouraklis G, Simopoulos C, et al Circulating VEGF levels in the serum of gastric cancer patients: correlation with pathological variables, patient survival, and tumor surgery. Ann Surg 2002; 236: 37–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Chen J, Li T, Wu Y, He L, Zhang L, Shi T, et al Prognostic significance of vascular endothelial growth factor expression in gastric carcinoma: a meta‐analysis. J Cancer Res Clin Oncol 2011; 137: 1799–1812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Fleitas T, Martínez‐Sales V, Vila V, Reganon E, Mesado D, Martin M, et al VEGF and TSP1 levels correlate with prognosis in advanced non‐small cell lung cancer. Clin Trans Oncol 2013; 15: 897–902. [DOI] [PubMed] [Google Scholar]
- 10. Macedo F, Ladeira K, Longatto‐Filho A, Martins SF. Gastric cancer and angiogenesis: is VEGF a useful biomarker to assess progression and remission? J Gastric Cancer 2017; 17: 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Monk BJ, Lammers PE, Cartwright T, Jacobs I. Barriers to the access of bevacizumab in patients with solid tumors and the potential impact of biosimilars: a physician survey. Pharmaceuticals (Basel) 2017; 10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Committee for Medicinal Products for Human Use . European Medicines Agency (2014). Guideline on similar biological medicinal products. Doc. Ref.: CHMP/437/04 Rev. 1. Available at http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2014/10/WC500176768.pdf (last accessed 18 July 2018).
- 13. US Food and Drug Administration (2015) Scientific considerations in demonstrating biosimilarity to a reference product. Available at http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm291128.pdf (accessed 14 November 2017).
- 14. Committee for Medicinal Products for Human Use . European Medicines Agency (2014) Guideline on similar biological medicinal products containing biotechnology‐derived proteins as active substance: non‐clinical and clinical issues. Doc. Ref.: EMEA/CHMP/BMWP/42832/2005 Rev. 1. Available at http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2015/01/WC500180219.pdf (last accessed 18 July 2018).
- 15. US Food and Drug Administration . Guidance for Industry. Clinical pharmacology data to support a demonstration of biosimilarity to a reference product. December 2016, Biosimilars. Available at https://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM397017.pdf (last accessed 14 November 2017).
- 16. Hettema W, Wynne C, Lang B, Altendorfer M, Czeloth N, Lohmann R, et al A randomized, single‐blind, Phase I trial (INVICTAN‐1) assessing the bioequivalence and safety of BI 695502, a bevacizumab biosimilar candidate, in healthy subjects. Expert Opin Investig Drugs 2017; 26: 889–896. [DOI] [PubMed] [Google Scholar]
- 17. Knight B, Rassam D, Liao S, Ewesuedo R. A phase I pharmacokinetics study comparing PF‐06439535 (a potential biosimilar) with bevacizumab in healthy male volunteers. Cancer Chemother Pharmacol 2016; 77: 839–846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Markus R, Chow V, Pan Z, Hanes V. A phase I, randomized, single‐dose study evaluating the pharmacokinetic equivalence of biosimilar ABP 215 and bevacizumab in healthy adult men. Cancer Chemother Pharmacol 2017; 80: 755–763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Tajima N, Martinez A, Kobayashi F, He L, Dewland P. A phase 1 study comparing the proposed biosimilar BS‐503a with bevacizumab in healthy male volunteers. Pharmacol Res Perspect 2017; 5: e00286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Casak SJ, Lemery SJ, Chung J, Fuchs C, Schrieber SJ, Chow ECY, et al FDA's approval of the first biosimilar to bevacizumab. Clin Cancer Res 2018. 10.1158/1078-0432.CCR-18-0566. [DOI] [PubMed] [Google Scholar]
- 21. Lu JF, Bruno R, Eppler S, Novotny W, Lum B, Gaudreault J. Clinical pharmacokinetics of bevacizumab in patients with solid tumors. Cancer Chemother Pharmacol 2008; 62: 779–786. [DOI] [PubMed] [Google Scholar]
- 22. Wu JY, Wu XN, Ding L, Zhao YB, Ai B, Li Y, et al Phase I safety and pharmacokinetic study of bevacizumab in Chinese patients with advanced cancer. Chin Med J (Engl) 2010; 123: 901–906. [PubMed] [Google Scholar]
- 23. Committee for Medicinal Products for Human Use . European Medicines Agency (2011). Guideline on bioanalytical method validation. Doc. Ref.: EMEA/CHMP/EWP/192217/2009 Rev. 1 Corr. 2**. Available at http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2011/08/WC500109686.pdf (last accessed 18 July 2018).
- 24. Teare MD, Dimairo M, Shephard N, Hayman A, Whitehead A, Walters SJ. Sample size requirements to estimate key design parameters from external pilot randomised controlled trials: a simulation study. Trials 2014; 15: 264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Lacey LF, Keene ON, Pritchard JF, Bye A. Common noncompartmental pharmacokinetic variables: are they normally or log‐normally distributed? J Biopharm Stat 1997; 7: 171–178. [DOI] [PubMed] [Google Scholar]
- 26. Harding SD, Sharman JL, Faccenda E, Southan C, Pawson AJ, Ireland S, et al The IUPHAR/BPS guide to PHARMACOLOGY in 2018: updates and expansion to encompass the new guide to IMMUNOPHARMACOLOGY. Nucl Acids Res 2018; 46: D1091–D1106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Alexander SPH, Fabbro D, Kelly E, Marrion NV, Peters JA, Faccenda E, et al The Concise Guide to PHARMACOLOGY 2017/18: Catalytic receptors. Br J Pharmacol 2017; 174 (Suppl 1): S225–S271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Ebbers HC, van Meer PJ, Moors EH, Mantel‐Teeuwisse AK, Leufkens HG, Schellekens H. Measures of biosimilarity in monoclonal antibodies in oncology: the case of bevacizumab. Drug Discov Today 2013; 18: 872–879. [DOI] [PubMed] [Google Scholar]
- 29. Committee for Medicinal Products for Human Use . European Medicines Agency (2010). Guideline on the Investigation of Bioequivalence. Doc. Ref.: CPMP/EWP/QWP/1401/98 Rev. 1/Corr **. Available at http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2010/01/WC500070039.pdf (last accessed 18 July 2018).
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1 ELISA inter‐assay precision and accuracy: validation
Table S2 ELISA inter‐assay precision and accuracy: performance during study sample analysis
Figure S1 Study schematic. DRL_BZ, Dr. Reddy's Laboratories bevacizumab; RMP, reference medicinal product (EU‐approved Avastin®); RP, reference product (US‐licensed Avastin®); PK, pharmacokinetics