This article does not contain any studies with human participants or animals performed by any of the authors. All SB8 samples were prepared and experiments conducted aseptically in a biological safety cabinet.
Stability of Opened/Pierced Vials
Three batches of 100-mg vials and three batches of 400-mg vials were used. Each 400-mg vial contains 16 ml, and 15 ml was withdrawn aseptically from each vial using a syringe with a 21G needle. Each 100-mg vial contains 4 ml, and 3 ml was withdrawn aseptically from each vial using a syringe with a 21G needle. The vials were incubated at 5 ± 3 °C under light protection (shielded chamber) for 72 h. Test intervals were 0, 24 and 72 h.
Stability of Unopened vials
One batch of SB8 100-mg vials manufactured on January 20, 2017, and stored upright and light-protected for 36 months at 5 ± 3 °C was subsequently stored upright and light-protected for 1 month at 30 ± 2 °C with 65 ± 5% relative humidity beginning January 29, 2020. At each time point (day 0, week 2 and month 1), vials were pulled from the shielded chamber for testing.
Stability of Diluted SB8
Three batches of 100 mg unopened vials were stored at 30 ± 2 °C with 65 ± 5% relative humidity for 24 h in the shielded chamber before aseptically preparing the dilutions (1.4 mg/ml and 16.5 mg/ml). Infusion bags (n = 21) for each concentration were prepared (7 bags per batch). Bag 1 was used for day 0 sampling, bag 2 for day 7 sampling, bag 3 for day 14 sampling, bag 4 for day 21 sampling, bag 5 for day 35 sampling, bag 6 for day 37 sampling and bag 7 for both days 47 and 48 sampling. The bags were stored at 5 ± 3 °C in the absence of light (shielded chamber) from days 0–45 with the exception of bag 6, which was stored at 30 ± 2 °C with 65 ± 5% relative humidity on days 36 and 37. For bag 7, on day 45, storage was changed from 5 ± 3 °C in the absence of light to 30 ± 2 °C for 3 days.
The average overfill volume of 100 ml 0.9% NaCl was determined and removed from each polyolefin bag (B. Braun). For the 1.4 mg/ml concentration, 5.6 ml of 0.9% NaCl was removed, and the same volume of SB8 was added from two unopened vials (each 100 mg vial has 4 ml). For the 16.5 mg/ml concentration, 66 ml of 0.9% NaCl was removed and the same volume of SB8 was added from 17 unopened 100-mg vials. After SB8 addition, the bag was slowly inverted ten times. Sampling occurred on days 0, 7, 14, 21, 35, 37, 47 and 48. Only the purity and impurity tests were conducted on days 7, 21 and 47.
For the 1.4 mg/ml bags, a part of the sample was concentrated to > 10 mg/ml using Amicon Ultra centrifugal filters (30 K) (Merck Millipore) because the concentration was too dilute for the capillary electrophoresis-sodium dodecyl sulfate (CE-SDS), imaged capillary isoelectric focusing (icIEF), VEGF binding assay and VEGF neutralization assay.
Biological and Physicochemical Analyses
For both the unopened and opened vials, color was tested according to Ph. Eur. 2.2.2, clarity was tested according to Ph. Eur. 2.2.1, and visible particles were tested according to Ph. Eur. 2.9.20. For the diluted samples only, subvisible particulate assessment was by MFI, and the particle standard was analyzed with samples to confirm the system suitability. Dynamic light scattering (DLS) was used to determine submicronic aggregation only for the diluted samples, and the latex standard was analyzed with samples to confirm the system suitability.
The following tests were done for all three experiments: pH according to Ph. Eur. 2.2.3, protein concentration, VEGF neutralization assay, VEGF binding assay, and subvisible particulate ≥ 10 µm and ≥ 25 µm in size quantification via light obscuration with a HIAC 9703 + and HRLD 400 instrument (Beckman Coulter, Inc.) per Ph. Eur. 2.9.19; particulates ≥ 2 µm and ≥ 5 µm were also counted for the diluted samples. Purity and impurity tests consisted of SE-HPLC, icIEF and CE-SDS in nonreducing conditions with capillary gel electrophoresis. A summary of the tests conducted is provided in Table 1. Details of conductance of the purity and impurity tests as well as the biological function tests are provided below.
Forced degradation studies were conducted to assess the stability indicating nature of physicochemical and biological assays used. Specifically, SB8 was subjected to heat stress (40 ± 2 °C with 75 ± 5% relative humidity and stored for up to 3 months), oxidation (treatment with 0.01% final concentration of hydrogen peroxide and incubation at 5 ± 3 °C for 24 h), acid stress test (1 N hydrochloric acid treatment to pH 4.0 and incubation at 25 ± 2 °C for 3 days) and base stress test (1 N sodium hydroxide treatment to pH 10.0 and incubation at 25 ± 2 °C for 3 days).
Size Exclusion High-Performance Liquid Chromatography (SE-HPLC)
Bevacizumab has irreversible and reversible aggregates, and the formation of reversible aggregate depends on pH, concentration and temperature. Nondiluted SB8 samples were used to determine the total aggregate levels, which is the sum of irreversible and reversible aggregates. In contrast, diluted sample at 2 mg/ml was used for monomer level determination, so that reversible aggregates would dissociate. The dilution buffer used was 50 mM sodium acetate, pH 5.0. As dilution was not needed for the low concentration (1.4 mg/ml), the percentage total aggregate and percentage monomer were determined by one analysis.
For SE-HPLC, samples were injected onto a TSK gel G3000SWXL column (5 µm/7.8 mm × 300 mm; TOSOHAAS) with a flow rate of 0.5 ml/min and a mobile phase that consisted of 200 mM potassium phosphate, 250 mM potassium chloride, pH 6.23. Detection occurred at 280 nm.
Imaged Capillary Isoelectric Focusing (icIEF)
An imaged capillary isoelectric focusing system (iCE3, Protein Simple) was used for determination of the distribution of charged species. Samples were mixed with the carrier ampholytes, injected into a transparent column (capillary), retained in a microscope-slide cartridge and separated by the system using free-solution isoelectric focusing. Specifically, the samples were introduced onto the fluorocarbon-coated capillary (100 μm/50 mm, Protein Simple), a pH gradient was created inside the capillary by small ampholyte molecules, and charged proteins electrophoretically moved to the point where the pH of the capillary was equal to the isoelectric point of the protein. The movement was recorded by a charge-coupled device camera.
Capillary Electrophoresis-Sodium Dodecyl Sulfate (CE-SDS)
A high-performance capillary electrophoresis system (PA 800 Plus Pharmaceutical Analysis System; Beckman Coulter, Inc.) was used for CE-SDS analyses. Alkylated, non-reduced SB8 in the presence of sodium dodecyl sulfate with the addition of a 10-kDa internal standard was used. Each sample was electrokinetically introduced onto a capillary (50 µm/30.2 cm; Beckman Coulter, Inc.). UV radiation (wavelength, 220 nm) was passed through the capillary window aperture (100 × 200 µm; Beckman Coulter, Inc.) and was detected by the PA800 Plus.
VEGF Binding Assay
An ELISA plate had VEGF-A 165 adsorbed, nonspecific binding to the plate surface blocked by bovine serum albumin-containing buffer, serially diluted SB8 added to bind to VEGF-A 165, an HRP-conjugated anti-hIgG (Fc specific) antibody added to bind to bound SB8 and an appropriate chromogenic substrate added to produce a colored product by peroxidase. The test plate was then read on a spectrophotometer at 450 nm. The optical density, which was directly proportional to the SB8 concentration, was yielded, and the relative VEGF-A 165 binding activity of SB8 was determined by analyzing a four-parameter curve plotted from data in the parallel line analysis software.
VEGF Neutralization Assay
A VEGFR2-dependent reporter gene system was used to determine the neutralization potency of VEGF-A 165 by SB8. Specifically, the HEK293-NFAT-VEGFR2-Luc cell line contained the nuclear factor of activated T-cell binding sequence in the promoter region of the luciferase reporter gene. Bound VEGF to VEGFR2 led to VEGFR2-dependent signal pathway activation and the expression of the luciferase reporter gene. Thus, the neutralization potency of VEGF-A 165 by SB8 was calculated based on dose-dependent changes in luminescence levels that corresponded to the amount of luciferase in the cell line.