Plasma-derived medicinal products are prepared industrially from human plasma by pharmaceutical companies and include products such as immunoglobulins, albumin, and coagulation factors, which are life-saving therapeutics for several chronic and acute life-threatening diseases.1) Among these, human immunoglobulin is used for the treatment of hypogammaglobulinemia, agammaglobulinemia and viral diseases. Human immunoglobulin is biologics that can be distributed on the market after verification through the national lot release conducted by the government to determine the quality of the final product manufactured by the manufacturer.2) The Korean Minimum Requirements for Biological Products (KMRBP) specifies that the tests including immunoglobulin protein composition, anticomplementary activity, and the molecular size distribution using size-exclusion highperformance liquid chromatography (SE-HPLC), etc. shall be conducted for the final product of human immunoglobulin.3,4) Molecular size distribution test is used for the quantification of unwanted aggregated forms in human immunoglobulin products.5) The analytical procedures of molecular size distribution tests were similar in the KMRBP and European Pharmacopeia (Ph. Eur.), however, there were differences in some operating conditions.3,6-8) Accordingly, the National Institute of Food and Drug Safety Evaluation (NIFDS) established the test method by harmonizing with the Ph. Eur. and performing method validation.9) A reference standard for peak identification of each immunoglobulin molecule is required for the molecular size distribution test, and the human immunoglobulin (molecular size) biological reference preparation (BRP), established by the European Directorate for the Quality of Medicines & HealthCare (EDQM), is being used.10) However, the number of national lot release of human immunoglobulin products is increasing every year, and as many reference standards are required for the national lot release assays, it was attempted to establish the national reference standard (NRS) through the collaborative study led by the NIFDS.
Four laboratories, including the NIFDS and manufacturers of human immunoglobulin products, participated in the collaborative study. Three of these four laboratories participated in the shortterm stability study. Participating laboratories were randomly allocated from Lab 1 to Lab 4. Participants performed ten assays using two vials of the candidate and one vial of human immunoglobulin (molecular size) BRP. Participants were asked to return data of retention time and peak areas for the monomer, dimer, polymers/aggregates, and fragments.
The candidate materials are from the human normal immunoglobulin product for intramuscular administration and were manufactured by Green Cross Corp. through the same manu-facturing process as the approved drug product. All units of human plasma used to manufacture the final product were tested by the manufacturer and found to be non-reactive for human immunodeficiency virus, hepatitis C virus, and hepatitis B virus. In addition, quality control tests according to the KMRBP were performed on candidate, and it was confirmed that all test items met the specifications. The human immunoglobulin (molecular size) BRP, which was established by the EDQM, was used as a reference standard for each assay.
SE-HPLC analysis for the molecular size distribution test was performed using a hydrophilic silica gel column (7.5×600 mm, 10 μm) (TOSOH, Tokyo, Japan) connected to HPLC system (Waters, MA, USA). Other operating conditions for the molecular size distribution test was applied as described previously.9)
The real-time stability of the candidate was assessed based on the appearance, pH, and molecular size distribution during the 24 months. A real-time stability test was initiated when the candidate was manufactured in 2019. Appearance and pH tests were performed according to the KMRBP. For the short-term stability study, reconstituted vials from collaborative study (Day 0) were stored at 2-8°C and retested for molecular size distribution on Day 7 and 14. Participants performed three assays using two vials of the candidate and one vial of human immunoglobulin (molecular size) BRP.
The data for the molecular size distribution test are representative of ten injections for collaborative study, and of three injections for real-time stability and short-term stability study. Appearance and pH are the results of one test. The results were analyzed statistically and the mean, standard deviation (SD), and relative standard deviation (RSD) were determined using Microsoft Excel 2016 (Microsoft, WA, USA).
For each vial of the candidate and BRP, the relative retention of dimer to monomer was calculated (Table 1). For the candidates, the relative retention of the four laboratories was 0.847-0.854, mean±SD was 0.851±0.002, and RSD was 0.286%. The relative retention of BRP was 0.852±0.002 in mean±SD and 0.279% in RSD. This result was in line with the monographs of the (Ph. Eur., which indicated that the relative retention of dimer to monomer of the reference standard was about 0.85 when performing the molecular distribution test.6-8) In addition, the values were calculated by comparing the retention time for the monomer and dimer of the candidate with the retention time for the monomer and dimer of BRP, respectively (Table 1). The results of the four laboratories were 0.999-1.002, and the mean± SD of each monomer and dimer was 1.001±4×10−4 and 1.000± 3×10−4, respectively and the RSDs were 0.043 and 0.035%. It is also indicated in the monographs of the Ph. Eur. that when performing a molecular size distribution test, the retention time for the monomer and dimer of the sample is required to be 1±0.02 compared to the retention time for the monomer and dimer of the reference standard.6-8) This candidate was found to satisfy all of these requirements for monomers and dimers. Through this, it was confirmed that this candidate exhibits almost the same characteristics as human immunoglobulin (molecular size) BRP for the two values described above. SE-HPLC separates protein molecules based on their hydrodynamic radii, the resolving power being dictated by the nature of the column packing material, including its particle size, pore size, and mobile phase characteristics.11) In the monographs of the Ph. Eur. for human immunoglobulin products, it is described to use only two columns with different lengths and diameters, and particle size or pore size, which is the main feature of the size exclusion chromatography column, is not considered. For this reason, it has been reported that when a molecular size distribution test is performed on the same product using columns with different resolutions, oligomers could appear between the dimer and polymers/aggregates, resulting in a different peak identity assignment in the chromatogram.5) Currently, the NIFDS and manufacturers of human immunoglobulin have harmonized the test methods to use the size exclusion chromatography column under the same conditions, so that the NRS established through this collaborative study can be used without additional consideration for this part. However, further studies and international harmonization efforts are needed to reflect the most recent column technology in the molecular size distribution test.
Secondary reference standards, such as the national reference standard, should be tested for the assigned value compared to the primary reference standard at appropriate intervals to ensure stability for long-term use.12) The candidates are stored at storage temperature (2-8°C). As a result of the stability test performed for up to 24 months during storage, the appearance, pH, and molecular size distribution were all suitable for the specifications of the KMRBP (Table 2). The stability tests for the NRS for biologics in the NIFDS are performed regularly following manufacturing and the stability is evaluated by comparing the results with the control limit.13) This NRS is different from the reference standard used in the potency and content tests of the final product. It is used for the purpose of confirming the retention time of each molecule of human immunoglobulin in the chromatogram obtained through SE-HPLC.10) For this reason, relative retention of dimer to monomer was selected as a control limit setting item, and the control limit was set to 0.830-0.872 by applying the test limit ±5% and the test accuracy ratio 1/2 for the variability of the test method using this reference standard to the geometric mean value of the relative retention calculated through collaborative study. Until 24 months, all relative retentions of dimer to monomer were distributed within the control limit, indicating that the long-term stability was very good.
Since the volume of the NRS in one vial was larger than the volume required for one test, molecular size distribution tests were performed on the 7th and 14th days to confirm the stability when the candidate was reconstituted and stored at storage temperature again. The mean of the relative retention of dimer to monomer in the three laboratories for the candidate was 0.853, 0.852, and 0.852 on days 0, 7, and 14, respectively (Table 3). In addition, the retention time for the monomer and dimer of the candidate was compared with the retention time for the monomer and dimer of BRP, respectively, and the values were 0.997-1.002 until 14 days. In general, reference standards are used for test immediately after reconstitution, but it was found that this candidate can be used after storage at normal storage temperature for up to 14 days if necessary.
According to the WHO, the aims of collaborative study for the establishment of biological reference standard is to demonstrate that the candidate is suitable for its intended use, and it includes confirmation that the candidate has the properties expected of it, assignment of parameter to the contents, and assessment of the stability of the candidate, etc.14) The study results demonstrated that the NRS candidate is suitable as a reference standard for the molecular size distribution test of human immunoglobulin products as required by the Ph. Eur., so it seems that it can replace the human immunoglobulin (molecular size) BRP that has been used so far. In addition, through this study, the stability of the candidate up to 24 months following manufacture was confirmed when stored at a storage temperature (2-8°C). Nevertheless, we plan to continuously monitor the stability by conducting a real-time stability test on the relative retention of dimer to monomer once a year. The NIFDS expects to be able to perform thorough quality control for the supply of safe and effective human immunoglobulin products by using this NRS candidate as a reference standard for the national lot release test in the future.
This work was supported by the Ministry of Food and Drug Safety (MFDS) (grant number 20201MFDS288). The authors gratefully acknowledge the participants of the study, Dohyun Kim, Eunji Lee (Green Cross), Jinwoo Im, Subin Oh (SK Plasma), and Tae-Jun Park, Jayoung Hong (Biologics Research Division, NIFDS) for participating in the collaborative study.
All authors declare that they have no conflict of interest.