TROIKA‐1: A double‐blind, randomized, parallel group, study aimed to demonstrate the equivalent pharmacokinetic profile of HD201, a potential biosimilar candidate to trastuzumab, versus EU‐Herceptin® and US‐Herceptin® in healthy male subjects

Abstract Prestige Biopharma Ltd (Singapore) has developed HD201, a proposed biosimilar to reference product trastuzumab. As a part of the stepwise approach to ensure comparability between the biosimilar candidate and the reference medicinal product, a phase I study in healthy subjects was conducted to demonstrate the pharmacokinetic (PK) equivalence (NCT03776240). The primary objective of the study was to demonstrate (PK) equivalence of HD201, EU‐Herceptin®, and US‐Herceptin® given at 6 mg/kg as a 90‐min i.v. infusion to healthy male subjects. A pairwise comparisons based on the primary endpoint AUC0–inf and secondary PK endpoints, AUC0–last and C max were undertaken. PK equivalence was to be concluded if the 90% confidence interval (CI) for the ratio of geometric means for each criterion were within the equivalence margin of 80% to 125%. Secondary objectives included assessment of other PK parameters, safety, tolerability, and immunogenicity in the three arms. A total of 105 healthy male subjects (35/treatment) were randomized in this study. The 90% CI for the ratios of AUC0–inf, C max and AUC0–last, were within 80%–125% for the comparisons of HD201 to EU‐Herceptin® or US‐Herceptin® and EU‐Herceptin® to US‐Herceptin®. The frequency of subjects with TEAEs of special interest was slightly lower in the HD201 group (20.0%) compared to the other treatment groups (EU‐Herceptin®: 34.3%; US‐Herceptin®: 31.4%). Only 1 subject (EU‐Herceptin® group) developed anti‐drug antibodies prior to dosing. Overall, HD201 demonstrates PK similarity to both EU‐Herceptin® and US‐Herceptin®. The three study drugs also demonstrated similar safety profiles.


| INTRODUC TI ON
Trastuzumab containing regimens are the backbone for the treatment of human epidermal growth factor Receptor 2 (HER2)-positive breast cancer, providing significant clinical benefit for metastatic breast cancer and increasing the proportion of cured patients in the adjuvant setting for early breast cancer. 1,2 As the patents on the licensed trastuzumab expired, there has been an increasing interest in developing biosimilar as the approval of biosimilar can facilitate patient access to high-quality biologic medicine at a lower cost.
Prestige Biopharma Ltd. (Singapore) has developed HD201, a potential biosimilar to reference product trastuzumab. The European Medicines Agency (EMA) and the US Food and Drug Administration (FDA) have developed specific guidelines for a biologic drug to be approved as a biosimilar. 3,4 Among the stepwise approach to ensure comparability between the biosimilar candidate and the reference medical product, a phase I study in healthy subjects is recommended to demonstrate the pharmacokinetic (PK) equivalence. EAGLE-I-12, a first phase I study demonstrated PK equivalence between HD201 and European Union sourced Herceptin ® (EU-Herceptin ® ) (EudraCT number 2012-000805-56). 5 Since, to ensure the production of the biologics at an industrial scale, some changes have been introduced in the production process of HD201. These changes, induced the need to repeat the overall comparability exercise. A new phase I study (TROIKA-1) in healthy subjects was designed with at the request of the US regulatory agency (US FDA) the addition of an US-sourced Herceptin ® . In this study (NCT03776240), pairwise comparisons were conducted between HD201 and both US-sourced and European-sourced Herceptin.

| Study design
This was a single center, phase I, double-blind, unstratified, randomized, single-dose, three-arm parallel group study in healthy adult male subjects. The primary objective of the study was to demonstrate PK equivalence of HD201 (Prestige Biopharma. Ltd, Singapore), EU-Herceptin ® , (Roche Pharma AG), and US-Herceptin ® (Genentech).
Secondary objectives included assessment of other PK parameters, safety, tolerability, and immunogenicity in the three arms.
For inclusion into the study, subjects had to be between the ages of 18 and 55 years. All subjects had to have normal screening results for vital signs, physical examination, and hematologic, renal, and hepatic functions. A normal 12-lead ECG and a left ventricular ejection fraction (LVEF) >60% according to echocardiogram (ECHO) were required. Subjects who had a history of cardiac disease, cancer, or any clinically significant disease were excluded.
This double-blinded randomized study was conducted in compliance with Good Clinical Practice and the Declaration of Helsinki.
The study protocol and its amendments were approved by the Independent Ethics Committee of Australia, and all of the participants provided written informed consent prior to initiation of any study-related procedures.
Subjects were confined from at least 10 h before dosing until after the 48-h post-dose blood draw. The duration of the study for each subject was approximately 14 weeks: 4 weeks for screening, and 8 weeks for single-dose PK and safety assessments. The PKC population included the cases with PK samples collection even if not enough samples were available to estimate all PK parameters. PKP population included cases with all PK parameters estimated.

| Pharmacokinetic assessments
An immunoassay method was validated for the determination of HD201 or Herceptin ® in human serum using an ELISA method. This ELISA was designed to quantify HD201 or Herceptin ® in human serum. Plates were coated with anti-trastuzumab antibody and then blocked to minimize any non-specific binding. Diluted human serum samples were then added and the plate was incubated. Subsequently, plates were washed and any Trastuzumab present in the samples were bound by the immobilized antibody and unbound substances were washed away. Analyte was detected by subsequent addition of mouse anti-human IgG (Fc) CH2 domain antibody labeled with horseradish peroxidase (HRP). A colorimetric signal was generated by further addition of Tetramethylbenzidine (TMB) substrate and stop solution. Plates were read in a microplate reader at 450 nm with a reference wavelength of 620 nm. The signal produced was proportional to the amount of analyte present and it was interpolated from the calibration curve using a four-parameter logistic curve-fitting program. Experiments were carried out to determine bioanalytical parallelism in the ligand-binding assay. Minimum required dilution, dilutional integrity, selectivity, and specificity in normal male serum and matrix effects (healthy and diseased populations) were considered in the PK validation assay.
The method was successfully validated over the calibration range 1.00-100 μg/ml (LLOQ: 2.00 μg/ml, ULOQ: 70.00 μg/ml) and precision and accuracy for all validation parameters passed the acceptance criteria.
A total of 13 blood samples at predefined timepoints were drawn from each subject for PK analyses. Blood samples were collected prior to drug administration, 1.5 and 3 h, (from the start of infusion), 8,24,48,96,168,336, 504, 672, 1008, and 1272 h. Blood samples for PK analysis or immunogenicity were centrifuged at 1300 to 2000 g for 10 min at 20℃. Two aliquots of atleast 0.5-ml serum were transferred to appropriate tubes and stored at approximately -80℃ until the sample analysis for PK. and immunogenicity. PK and ADA analysis was performed by the Agilex Biolabs (28 Dalgleish Street).
NAb analysis was not performed for this study as the positive ADA was from the pre-dose sample thus it is concluded as unrelated to the treatment. These bioanalyses were performed in compliance with the GCP and also in accordance with the current regulations as per the industry standards: Guidelines on Bioanalytical Method Validation, Good Laboratory Practices (GLP) and Guideline for GCP ICH E6.

| Pharmacokinetic parameters
The primary endpoint for this study was AUC 0-inf (the area under the

| Statistical analysis
Pharmacokinetic analysis was performed using Phoenix ® WinNonlin ® , which is validated for bioequivalence/bioavailability studies. Inferential statistical analyses were performed using SAS ® according to EMA and FDA guidelines.
Analysis of variance (ANOVA) was applied to natural logtransformed data for AUC 0-inf, AUC 0-last , and C max . Treatment was incorporated in the model as a fixed factor with three levels. The normal distribution of values was assessed and based on least-squares means from the ANOVA, the geometric mean and 95% confidence interval (CI) for each treatment, and the ratio of geometric means and 90% CI for the ratio of geometric mean were calculated. These were presented after back-transformation to the original scale. As foreseen in the protocol and the Statistical Analysis Plan, no correction for multiplicity was applied in the analysis.
Pairwise comparisons were performed between HD201 and US-Herceptin ® groups, HD201 and EU-Herceptin ® groups and US-Herceptin ® and EU-Herceptin ® groups. Bioequivalence (similarity) was achieved if 90% CI of the ratio of geometric means of log-transformed values, based on least-squares means from the ANOVA, were included within the interval 80.00% to 125.00%.
Based on literature and sponsor data, inter-subject coefficient of variation was estimated to be 17% and 19% for AUC and C max , respectively. With those expected coefficients of variation, assuming a ratio of AUC and C max between 0.925 and 1.08 and a power of at least 85%, 29 evaluable subjects per group, 87 in total, were required to show PK similarity. Accounting for possible dropouts, 35 subjects were to be included per group, 105 in total. Because equivalence can be claimed if only all pairwise comparisons are contained between the prespecified interval, no testing for multiplicity is required. The study has been designed to be analyzed in two independent submission dossiers in EU and in US, respectively. Therefore, the comparison of HD201 versus the respective reference products has been performed at an uncorrected level of significance and without calculating the power based on the two main comparisons. In addition, safety, tolerability, and immunogenicity data will be reported using descriptive statistics (arithmetic means, SD, CV%, min., max., and median).

| Safety evaluations
The safety population consisted of all randomized subjects who had at least one dose of study drug. All adverse events (AEs) reported during the study were coded according to the Medical Dictionary for Regulatory Activities (version 21.1). Severity was graded as mild, moderate, and severe as defined in the protocol.

| Subject characteristics and disposition
A total of 105 healthy male subjects (35 in each arm) were randomized in this study and their baseline demographic characteristics were similar among the three treatment groups (Table 1). Figure 1 reports the distribution of the randomized subjects according to their completion of PK. All randomized subjects were included in safety and immunogenicity analysis. Serum trastuzumab concentrations summaries were based on the PKC population (HD201-34, EU-Herceptin ® -35, and US-Herceptin ® -32). Two subjects were prematurely withdraw due to infusion issues resulting in partial administration, PK samples were not collected due to venous access issues in two subjects which were also excluded from the PKC population. Pharmacokinetic parameter summary statistics and assessment of PK equivalence were performed on the PKP population (HD201-32, EU-Herceptin ® -34, and US-Herceptin ® -31). Three cases with missing PK samples were excluded from the PKP population. The incomplete collection was due to a lack of compliance without any reasons but in one case related to a fracture of the left thumb requiring surgery.

| PK results
After administration of HD201, EU-Herceptin ® , or US-Herceptin ® , the percentage of the AUC 0-inf due to extrapolation (residual area) was 4.6%. 5.5%, and 4.5% of AUC 0-inf . This indicates that the applied sampling schedule ensured the majority of AUC was captured and the range of times across which K el was estimated was greater than twice the resultant t 1/2 . Nevertheless, there is one subject in the EU-Herceptin group with 23% of residual area. Since less than 20% of the subjects had residual area <20%, no sensitivity analysis was performed in which this subject was excluded. All summarized PK parameters were therefore considered to be reliably estimated.
The inter-subject variability based on AUC 0-inf and AUC 0-last , was characterized by a geometric CV ranging from 15.3% to 17.6%.  Pairwise comparisons for PK endpoints (AUC 0-inf , AUC 0-last , and C max) in all groups provided confidence intervals included between the prespecified equivalence margin (Table 3). Overall, systemic exposures, based on AUC 0-inf , AUC 0-last , and C max after administration of HD201, EU-Herceptin ® , or US-Herceptin ® were similar.

| Safety
Frequencies of subjects with TEAEs, treatment-related TEAEs, and mild or moderate TEAEs were similar across the three treatment groups (

| Immunogenicity
One subject in the EU-Herceptin ® group tested positive for antidrug antibody (ADA) at baseline (prior to dosing). This subject did not test positive at any time after receiving the study drug. No test for neutralizing antibodies was performed for this subject as ADA was detected prior to study drug administration.

| DISCUSS ION
The purpose of this study was to compare the pharmacokinet-  EAGLE-I-12 Phase I study was conducted in 2012 with more than sufficient coverage of theoretical drug clearance to make sure that the PK comparison can be obtained between reference herceptin and HD201. Over the last decade, many scientific evidence have been compiled to prove that PK similarities can be fully assessed with a lesser duration of profiling. A shortened PK sampling day was considered in TROIKA-1 study as other trastuzumab biosimilar phase I studies. Consequently, the considered C max in this study did not represent a stringent determination of this parameter but the concentration provided in the sample collected at 1.5 h. This rounded value appeared acceptable because the main objective was to perform a pairwise comparisons of PK and not to characterize the PK profile of trastuzumab.
Overall exposure to trastuzumab, assessed by AUC 0-inf , AUC 0-last , and peak systemic exposure to trastuzumab assessed by C max , was shown to be comparable after administration of HD201 and EU-Herceptin ® , and after administration of HD201 and US-Herceptin ® . For both comparisons, the 90% CIs for the ratio of the geometric mean of AUC 0-inf , AUC 0-last , and C max , were all contained within the prespecified margin of 80.00% to 125.00%. In addition, the US-Herceptin ® and EU-Herceptin ® were equivalent based on a similar comparison process. All pairwise comparison was included in the prespecified margins then the PK equivalence can be stated between HD201, US-Herceptin ® , and EU-Herceptin ® . In addition, the other secondary PK parameters including residual area, t max , t ½ el , K el , Cl, and V d were similar between the three treatment groups.
The secondary objectives of this study were to assess the safety, tolerability, and immunogenicity of HD201 and the EU and US refer-   All author provided a final approval of the version to be published; AND agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

DATA Q UA L I T Y A S S U R A N C E
The clinical site has established Quality Control (QC) and Quality Assurance (QA) systems with written SOPs to ensure that the study was conducted and data were generated, recorded, and reported in compliance with the protocol, GCP, and applicable regulatory requirements. A rigorous QC program was applied to ensure the accuracy of all data and reports.
In accordance with the principles of GCP and GLP, the study could be inspected by the QA unit of the clinical sites, regulatory authorities, the Sponsor and Syneos Health. The Sponsor was entitled to access information.

RO LE O F TH E FU N D I N G S O U RCE
The funding source validated the study as designed by the trial's steering committee as well as subsequent amendments. The

DATA AVA I L A B I L I T Y S TAT E M E N T
Data collected for the study, including individual participant data and a data dictionary defining each field in the set, will be made available to others; all available data will be de-identified participant data. The study protocol, statistical analysis plan, informed consent form, and ethics committee approval are available. To access the data, a request should be submitted at Prestige Biopharma with a scientific proposal including objectives. The data will be shared after approval and by the steering committee of the trial. Approval by the ethics committee might be required according to the type of proposal and objectives.