Recombinant Factor VIII Product and Sample Preparation
This study used commercially available turoctocog alfa dry powder batches representing the lowest, middle, and highest strengths of the drug product (250, 1500, and 3000 IU).
Study Design
The stability of turoctocog alfa dry powder batches under varying storage conditions was investigated over the full 30-month shelf-life period [15, 16], with sampling throughout the study period (see below for further details) (Fig. 1). All drug products were initially stored at 5 °C for 24 months to enable testing toward the end of the product’s shelf-life. Turoctocog alfa test samples were then cycled ten times between 5 °C with ambient humidity (AH) and 40 °C with 75% relative humidity (RH) over a period of 2 months. One cycle was equal to 96 h at 5 °C and 72 ± 2 h at 40 °C. After 10 temperature cycles, test samples were stored at 40 °C/75% RH for 3 months and then at 5 °C/AH for 1 month (the 1-month period at 5 °C allowed the full 30-month shelf-life to be reached). Reference samples were stored at 5 °C, whereas test samples underwent temperature cycling and were then stored at 40 °C/75% RH for 3 months and at 5 °C/AH for 1 month, alongside the test samples. For the test samples, purity, high molecular weight proteins (HMWPs), and oxidized forms were assessed after each temperature cycle, once a month during storage at 40 °C (3 months), and at study end (after 1 month at 5 °C). Potency and water content were assessed after temperature cycles 1, 3, 5, 7, 9, and 10, once a month during storage at 40 °C, and at study end (after 1 month at 5 °C). Reference samples were assessed at day 31 (equivalent to cycle 5) and day 66 (equivalent to cycle 10), once a month during storage at 40 °C (3 months), and at study end (after 1 month at 5 °C).
Assay Methods
The study was performed according to current International Conference for Harmonisation bracketing design guidelines [17]. The parameters assessed include those from the drug product specification that are susceptible to change during storage and/or those likely to influence product quality. The results from all assays were compared with turoctocog alfa 250-, 1500-, and 3000-IU reference samples and evaluated against predetermined specification limits, calculated using appropriate statistical methods [17]. The specific limits for each assay are noted within the appropriate following sections.
Potency
Potency was investigated using the chromogenic kit Coamatic® FVIII (Chromogenix, Instrumentation Laboratory, Bedford, MA, USA) on the ACL® Elite Pro analyzer (Instrumentation Laboratory) in accordance with European Pharmacopoeia Assay of human coagulation FVIII using a product-specific standard as calibrator. Turoctocog alfa product samples and the product-specific calibrator were reconstituted in 4.3 ml 0.9% sodium chloride before undergoing three dilution steps: (1) predilution to approximately 11 IU/ml using the Coamatic® FVIII kit buffer solution; (2) dilution to approximately 1 IU/ml using FVIII-deficient plasma; (3) dilution to approximately 0.005 IU/ml using the Coamatic® FVIII kit buffer solution. Blank samples were prepared by dilution of 20-µl FVIII-deficient plasma with 4000-µl buffer solution.
Turoctocog alfa product samples, calibrator, and blank samples were analyzed in triplicate on the ACL® Elite Pro analyzer. Absorbance readings for product and reference samples and calibrator were used to calculate the potency (FVIII:C) using a slope-ratio analysis. The acceptance criteria for the potency of the three turoctocog alfa samples were defined as follows: 200–313 IU/vial for the 250-IU dose; 1200–1875 IU/vial for the 1500-IU dose; 2400–3750 IU/vial for the 3000-IU dose.
Purity
The purity of reconstituted turoctocog alfa product samples was assessed using reverse-phase high-performance liquid chromatography (RP-HPLC). Analysis was performed on an HPLC system equipped with processing software and a 4.0 × 250-mm, C4 5-μM, 300 Å column (Novo Nordisk Pharmatech A/S, Køge, Denmark). The column temperature was set at 40 °C, with a detection wavelength of 215 nm. A gradient of 35–100% eluent B (0.09% trifluoracetic acid [TFA] in 80% acetonitrile in purified water) and 65–0% eluent A (0.1% TFA in purified water) was applied over a duration of 40 min at a flow rate of 1 ml/min. The composition of 100% eluent B was then maintained for 5 min before the composition was changed back to the initial conditions over 1 min and the column was then equilibrated for 14 min, resulting in a total run time of 60 min. The purity of the turoctocog alfa product was calculated as the sum of area percentages of the following components on the resulting chromatograms: turoctocog alfa light chain; turoctocog alfa single chain; and three heavy chain (HC) components (nontruncated form, one with the C-terminal at amino acid 740 [HC_740], and one with the C-terminal at amino acid 720 [HC_720]). The acceptance criterion was ≥ 89.4% purity.
Oxidized Forms
Oxidized forms within the reconstituted turoctocog alfa product were assessed using RP-HPLC. The RP-HPLC system, column, HPLC parameters (column temperature, detection wavelength, mobile phase eluents), and elution gradients used to assess oxidized forms were the same as those used to assess product purity. Oxidized forms were calculated as the percentage area on the resulting chromatograms. The acceptance criterion was ≤ 6.8% oxidized forms.
High Molecular Weight Proteins
Turoctocog alfa samples were analyzed by size exclusion-HPLC (SE-HPLC) to determine the presence of protein aggregates. Prior to SE-HPLC analysis, product samples were reconstituted in either 1.0 (250-IU samples) or 4.3 ml (1500- and 3000-IU samples) 0.9% sodium chloride solution. SE-HPLC measurements were performed using an HPLC system equipped with a BioSep SEC S3000 7.8 × 300-mm, 5-μm, 290 Å column (Phenomenex, Torrance, CA, USA) or a Shodex PROTEIN KW-803, 8 × 300-mm column (Shodex) or equivalent. Elution was employed at a flow rate of 0.4 ml/min, using a column temperature of 30 °C and excitation and emission detection wavelengths of 285 and 335 nm, respectively. The eluent buffer consisted of 10 mM TRIS, 10 mM CaCl2, 300 mM sodium chloride, and 5% 2-propanol pH 7.0; the injection volume of turoctocog alfa was 100 µl, independent of protein concentration. The running times were ≥ 70 and ≥ 80 min for the turoctocog alfa product reconstituted in 4.3- and 1.0-ml 0.9% sodium chloride, respectively. HMWP content was determined by calculating the area percentage of the HMWP peak on the resulting chromatogram. The acceptance criterion was ≤ 3.9% HMWP.
Water Content
The water content of turoctocog alfa dry product samples prior to reconstitution was evaluated by near-infrared (NIR) spectroscopy using an FT-NIR Spectrometer (MPA, Bruker, Billerica, MA, USA) equipped with an integrating sphere (or equivalent). Turoctocog alfa samples were scanned in the frequency ranges of 7502.1–6098.1 and 5450.1–4597.7 nm at a resolution of 8 cm−1, and an average of 32 scans was recorded for each spectrum. The spectrometer was equipped with OPUS software (Bruker) (or equivalent), and data were analyzed using a partial least squares fit method. Turoctocog alfa samples were analyzed without any pretreatment; however, NIR spectra data were pretreated using vector normalization and a first derivative to enhance spectral information and correct interferences from the analyzed material that might otherwise induce baseline drift and changes in maximum absorbance [18]. Karl-Fischer coulometry was used as the reference method for determining the calibration function and to analyze samples in cases where NIR spectroscopy results were not accepted (for example, because of persistent outliers) [19]. The acceptance criterion for water content was ≤ 1.7%.