Analytical similarity acceptance criteria and calculation of the quality ranges for each attribute for BAT1806/BIIB800 to the RP are reported elsewhere [2]. Briefly, the quality range of 70–130% was based on the criticality of product quality attributes, method capability, and lot-to-lot variability of the RP. A tier system was implemented to classify each quality attribute. Tier 1 included critical quality attributes related to the primary mechanism of action; Tier 2 comprised quantitative assays measuring high-impact critical quality attributes; and Tier 3 covered quantitative assays measuring non-critical quality attributes and semi- or non-quantitative assays. Equivalence tests varied between tiers [2].

2.1 Materials

SAR studies were conducted for glycation, glycosylation, charge variants, hydrophobicity, oxidation, and deamidation using one lot each of BAT1806/BIIB800 and TCZ (one lot of EU-sourced only; SAR studies for glycation used one lot each of CN-, EU-, and US-sourced TCZ). Details regarding the BAT1806/BIIB800 and TCZ formulations are reported elsewhere [2]. The BAT1806/BIIB800 drug product was manufactured by Bio-Thera Solutions Ltd (Guangzhou, China) and the TCZ lots were purchased from the CN, EU, and US markets, and stored according to the manufacturers’ instructions.

2.2 Glycation2.2.1 Intact and Reduced Deglycosylated Mass Analysis by Liquid Chromatography-Mass Spectrometry

Intact mass liquid chromatography-mass spectrometry (LC–MS) can be used for qualitative and relative quantitative analysis of glycated and non-glycated antibodies and can distinguish mono- and poly-glycated antibodies. Samples of BAT1806/BIIB800 and TCZ were subjected to an intact mass LC–MS analysis to determine glycation content. Additionally, stress-glycated samples were analyzed by intact mass LC–MS to allow further assessment of the potential biological impacts of glycation modification of BAT1806/BIIB800 and TCZ. Control samples were incubated without glucose solution, and stress-glycated samples were incubated with 0, 50, and 200 mM glucose solutions, prepared with 1 M of glucose and 250 mM of ammonium bicarbonate solution (total volume of 100 μL; final concentration of 10 mg/mL). Samples were incubated at 37 °C for 24 h, then purified in ultrapure water using 3-kD ultrafiltration tubes (Millipore, Burlington, MA, USA) and stored at −20 °C.

2.2.2 Reduced and Non-reduced Peptide Mapping by Liquid Chromatography–Tandem Mass Spectrometry

Reduced and non-reduced peptide mapping were performed as described elsewhere [2]. A site-specific post-translational modification (PTM) analysis was used to identify glycation modification sites and pyroglutamic acid at the N-terminus. Pre-treated samples were separated by liquid chromatography–tandem mass spectrometry (LC–MS/MS) and primary and secondary mass spectral signals of peptides were analyzed using Biopharmalynx software. Post-translational modification content of target peptides was calculated from the intensity of the primary mass spectral signal. Content of pyroglutamic acid and the N-terminus was calculated using the response value of the N-terminal peptides with pyroglutamic acid modifications/(the response value of the N-terminal peptide with pyroglutamic acid modification plus the response value of the N-terminal peptide without pyroglutamic acid modification).

2.3 Glycosylation2.3.1 Sample Processing

Non-deglycosylated samples were prepared by adding 100 mM of Tris (pH 7.95) solution to 1 mg of BAT1806/BIIB800 or TCZ to give a final concentration of 10 mg/mL. Deglycosylated samples were initially prepared in the same manner (test samples diluted with 100 mM of Tris), then the PNGase F enzyme (New England Biolabs, Inc., Ipswich, MA, USA) was added (the mass:volume ratio of the PNGase F enzyme was 50 µg:1 µL) and incubated at 37 °C for 4 h.

2.3.2 Magnetic Bead Sample Purification

To allow measurement of the biological activity of deglycosylated BAT1806/BIIB800 and TCZ, samples underwent magnetic bead purification to remove PNGase F enzyme immediately after deglycosylation. Magnetic bead sample purification was performed according to the manufacturer’s instructions for the magnetic bead kit (BeaverBeads™ Magrose Protein A; BEAVER, Suzhou, Jiangsu, China), and stored at 2–8 °C. The degree of deglycosylation (capillary electrophoresis sodium dodecyl sulfate under reducing conditions), sample purity, and the effect of enzyme removal (size exclusion chromatography-high performance liquid chromatography [HPLC]) were measured, as detailed elsewhere [2]. Briefly, for capillary electrophoresis sodium dodecyl sulfate, samples were treated with sample buffer and 2-mercaptoethanol and incubated at 70 °C for 10 min. Samples were then separated at −16.5 kV for 30 min in a capillary electrophoresis instrument (Agilent CE7100; Agilent Technologies, Santa Clara, CA, USA) coupled with a bare fused-silica capillary with a length of 24.5 cm (SCIEX, Framingham, MA, USA). For size exclusion chromatography-HPLC, diluted samples were injected directly onto an analytical column (TSK gel G30000 SWXL [7.8 × 300 mm, 5 μm], Tosoh) with an HPLC (1260II; Agilent Technologies) flow rate of 0.5 mL/min.

2.3.3 Sample Ultrafiltration

Samples were further purified using ultrafiltration tubes and centrifugation. Briefly, 450 μL of ultrapure water was added to ultrafiltration tubes and centrifuged twice at 12,000 rpm for 6 min. Following this, 450 μL of the purified sample was added to the tube and centrifuged at 12,000 rpm for 6 min, then ultrapure water (450 μL) was added to each tube and centrifuged three times at 12,000 rpm for 6 min. Samples were concentrated to 100 μL, and the concentration was determined by a high-throughput microprotein concentration analyzer (Lunatic; Unchained Labs, LLC, Pleasanton, CA, USA).

2.4 Charge Variants2.4.1 Charge Variant Identification Using Ion Exclusion Chromatography-High Performance Liquid Chromatography

Ion exclusion chromatography-HPLC (IEC-HPLC) is used to evaluate the charge variants of BAT1806/BIIB800 and TCZ samples with and without carboxypeptidase B (CpB) treatment, prepared as detailed elsewhere [2]. Samples were treated with CpB to remove C-terminal lysine residues and reduce lysine-induced charge heterogeneity between BAT1806/BIIB800 and TCZ. CpB-digested BAT1806/BIIB800 and TCZ samples were loaded onto a semipreparative strong cation exchange column (Proteomix SCX, 10 μm, 21.2 mm × 250 mm) on an Ultimate 3000 HPLC (Thermo Fisher Scientific, Waltham, MA, USA) and eluted at a flow rate of 10 mL/min to collect charge variant fractions. Samples were analyzed on an analytical column (Thermo MAbPac SCX-10, 10 μm, 4 × 250 mm). Sample charge variant fractions were concentrated by low-temperature ultrafiltration using a 30-kDa ultrafiltration centrifuge tube and exchanged into BAT1806/BIIB800 or TCZ formulation buffer (pH 6.2; final concentration ≥ 2 mg/mL). Samples were stored at −60 °C.

2.5 Hydrophobicity2.5.1 Hydrophobic Interaction Chromatography-High Performance Liquid Chromatography

Hydrophobic interaction chromatography-HPLC was used to monitor hydrophobic variants, per the methods detailed elsewhere [2]. The collection of hydrophobic variant fractions of BAT1806/BIIB800 and TCZ was performed on an Ultimate 3000 HPLC (Thermo Fisher Scientific, Waltham, MA, USA) using a phenyl sepharose high performance (GE HealthCare Technologies Inc., Chicago, IL, USA) packed preparative column. The antibody (150 mg) was loaded onto the column and eluted at a flow rate of 2.5 mL/min. Fractions were collected, concentrated by low-temperature ultrafiltration using a 30-kDa ultrafiltration centrifuge tube, and exchanged into BAT1806/BIIB800 formulation buffer (without polysorbate 80, pH 6.2; final concentration 2 mg/mL or higher). Samples were stored at −60 °C.

2.6 Oxidation and Deamidation2.6.1 Forced Oxidation Sample Preparation

H2O2 was added to BAT1806/BIIB800 and TCZ samples to give final sample concentrations of 0.0005%, 0.001%, 0.002%, 0.005%, and 0.01% H2O2, respectively. Samples were incubated at 25 °C in the dark for 0, 1, and 2 days, respectively, and analyzed using LC–MS/MS.

2.6.2 Forced Deamidation Sample Preparation

Ammonium bicarbonate (250 mM) was added to BAT1806/BIIB800 and TCZ samples to give a final concentration of 50 mM of ammonium bicarbonate. Samples were incubated at 37 °C for 0, 2, 4, 6, and 8 days, respectively, and analyzed using LC–MS/MS.

2.6.3 Characterization of Oxidation and Deamidation by Liquid Chromatography–Tandem Mass Spectrometry

Liquid chromatography–tandem mass spectrometry was used to analyze site-specific PTMs for BAT1806/BIIB800 samples. Samples (5 µL) were denatured with 15 µL of 6 M of guanidine hydrochloride (100 mM Tris, pH 7.95), reduced with 0.5 µL of 0.5 M of dithiothreitol, and alkalized with 1 µL of 0.5 M of iodoacetamide. Following this, C-terminal lysine and arginine peptide bonds were hydrolyzed with a mass ratio of protein:trypsin of 1 μg:25 μg at 37 °C, pH 7.95. The reaction was then terminated with 5 µL of 5% formic acid. Separation and detection of fractions were performed using HPLC coupled with MS (Waters ACQUITY UPLC H-Class tandem with Xevo G2-S QToF). The chromatographic column (ACQUITY UPLC® Peptide BEH C18 Column [300Å, 1.7 µm, 2.1 mm×150 mm, 1/pkg]) was set at 60 °C, mobile phases A and B were 0.1% formic acid-water and 0.1% formic acid-acetonitrile, at a flow rate of 0.2 mL/min. The elution time was 2–122 min, 0.2–40% of phase B was used for effective elution, and the total gradient was 130 min. Data acquisition was performed using MassLynx software. Data were analyzed using BiopharmaLynx software.

2.7 Binding Affinity and Functional Assays2.7.1 Binding Affinity to Fragment Crystallizable Receptors and a Complement Component 1q Binding Kinetic Analysis by Bio-layer Interferometry and Surface Plasmon Resonance

Binding affinity was tested to evaluate whether deglycosylation of BAT1806/BIIB800 and TCZ impacts binding affinity, and if test samples behave similarly. The affinity of BAT1806/BIIB800 and TCZ with FcγRIIIa (158F/V), FcγRIIa (131R/H), FcγRIIb, FcγRIa, neonatal FcR (FcRn), and complement component 1q (C1q) were determined by bio-layer interferometry using an Octet QKe platform (Sartorius, Göttingen, Germany) or surface plasmon resonance (SPR) using a Biacore™ T200 (GE HealthCare Technologies Inc., Chicago, IL, USA). Bio-layer interferometry and SPR methodological details are described in full elsewhere [2].

2.7.2 Antibody-Dependent Cell-Mediated Cytotoxicity and Complement-Dependent Cytotoxicity Analyses

A cell-based reporter gene assay was used to evaluate the ADCC effects of BAT1806/BIIB800 and TCZ. In the assay, TF-1 cells (human leukemic cells expressing membrane IL-6R) were used as target cells, and engineered Jurkat/NFAT-Luc+FcRIIIa-158V cells were used as effector cells.

A separate cell-based assay was used to evaluate the CDC effects of BAT1806/BIIB800 and TCZ, with TF-1 cells used as target cells, and human serum complement used to analyze CDC function. Ofatumumab (anti-CD20 mAb; lot number DM8H; Novartis Pharma GmbH, Nürnberg, Germany) was used as a positive control and CD20-expressing lymphoma (Raji) cells were used as positive cells. Pertuzumab (anti-human epidermal growth factor receptor 2 mAb; lot number H0212H01; Roche, Munich, Germany) was used as a negative control.

2.7.3 Binding to the Soluble Interleukin-6 Receptor by an Enzyme-Linked Immunosorbent Assay

Binding to the soluble IL-6R by an enzyme-linked immunosorbent assay (ELISA) was performed as described elsewhere [2]. Results were reported as a percentage of the relative binding potency against the reference standard.

2.7.4 Binding Kinetics to the Interleukin-6 Receptor by Surface Plasmon Resonance

The IL-6R binding kinetics of BAT1806/BIIB800 and TCZ were determined by SPR methods. A carboxy-methylated sensor chip with pre-immobilized Protein A was used to capture the Fc region of each sample, then bound with serial concentration IL-6R. Results were reported as a percentage of the relative binding potency against the reference standard.

2.7.5 Inhibition of Interleukin-6-Mediated Proliferation in TF-1 Cells

The functional activity of BAT1806/BIIB800 and TCZ to inhibit IL-6-mediated cell proliferation was evaluated by a TF-1 cell-based assay. TF-1 cells were incubated with serial concentration of reference standard or test samples in the presence of IL-6. After incubation, a CellTiter-Glo® Luminescent Cell Viability Assay (Promega Corporation, Madison, WI, USA) solution was added, and luminescence measured with a microplate reader. Results were reported as a percentage of the relative binding potency against reference standard.

2.7.6 Interleukin-6 Blockage Activity by a Secreted Embryonic Alkaline Phosphatase Reporter Gene Assay

The functional activity of BAT1806/BIIB800 and TCZ to block IL-6-mediated signaling was evaluated with a secreted embryonic alkaline phosphatase (SEAP) reporter gene cell-based assay. HEK-BlueTM IL-6 cells (InvivoGen, Toulouse, France) expressing membrane IL-6R, signal transducer and activator of transcription 3 (STAT3) cDNA, and interferon-β response element-driven SEAP reporter genes were incubated with varying concentrations of reference standard or test samples in the presence of IL-6-mediated activation of STAT3. Downstream SEAP was measured by the addition of QUANTI-Blue™ (InvivoGen, Toulouse, France) solution and quantified with a SpectraMax M5e microplate reader (Molecular Devices, San Jose, CA, USA). Blockage activity was calculated against a dose–response curve of an internal reference standard by SoftMax Pro Software (Molecular Devices, San Jose, CA, USA) and reported as percent relative potency.