Sub Topic | Secondary Topic: Analytical Methodology | HPLC (Biopharmaceutical Molecule)
Authors: Mariolina De Luca, Patheon Inc. (Presenting Author); Paolo Cavarischia, Patheon Inc. (Main Author); Roberta Bucci, Patheon Inc.
Presenting Author: Mariolina De Luca
Purpose: An unknown peak has been observed during HPLC analysis of In Process Controls (IPC) related to a GMP Drug Product batch of a sterile liquid product. The materials in contact with the bulk solution were investigated by RP-HPLC under several sample preparations to identify the source of the contamination.
Methods: The related substances determination by RP-HPLC foresees a two steps dilution for preparation of the reference standard. The standard solution preparation is performed using a disposable PolyPropylene (PP) flask due to the nature of the molecule and the low concentration of the working solution. The sample solutions analyzed were injected as is without any additional manipulation/dilution. The unknown peak was observed on different samples that also showed unexpected results for impurities. The registered level of the unknown peak in the samples was about 7% (% Area on total area). The above was noted on selected IPC steps of the same batch as well as on the standard solution. Disposable PP flasks were also used to contain the bulk solution withdrawn directly during the manufacturing step (IPC samples) but only for a certain step flasks from a different vendor/brand were used having a different brand (i.e. Brand 1) respect to the usual PP flasks (i.e. Brand 2). Samples related to the IPC step of filter flushing and other samples related to a different manufacturing step were analysed in the same analytical sequence. Only the IPC samples withdrawn from the filter flushing step showed the unknown peak.The hypothesis was made that the type of PP flasks used to withdraw the bulk solution during the filter flushing step might release material that caused the sample contamination. Alternatively the accidental contamination of the samples with the sanitizing agent may have happened. Confirmatory testing was performed on the samples with and without the contaminant peak. The latter were considered as being the “monitor samples”. Solution free from contaminant was put in contact with the sanitizing agent employed in the equipment/filter cleaning procedure as well as with Brand n.1 PP flasks used in both the laboratory and production area. New PP flasks and filter were investigated as well, to check for potential leachables and extractables.
Results: The contaminant peak was detected in the filter flushing step samples while it was not present in the monitor sample. The monitor sample was put in direct contact with the Brand 1 PP flask for about 1 hour and found to contain the contaminant peak. In addition to this a further experiment was performed by spiking the monitor sample with the sanitizing agent previously collected into a Brand 1 PP flask: a lower level of the contaminant peak was noted in this case. Considering that the contaminant peak is not significantly higher in the sample put into contact with the sanitizing agent, the conclusion can be drawn that the source of contamination is the PP flask. The experiment was repeated with the reference standard solution that showed the same behavior thus confirming the hypothesis made. Considering the new brand of PP flask introduced (although the declared material composition for both the falcon type used is PP) and the data collected during the investigation, it is demonstrated that the peak detection was due to the contact between the solutions and the new PP tube type.
Conclusion: The unknown peak comes from an external contamination. The PP tube investigated was from a new vendor/brand and, contrary to Brand n.2, was supplied as sterilized material. The chemical sterilization of Brand 1 is performed by the vendor and is the only difference identified vs Brand 2 previously used. Therefore the sterilization treatment is the major root cause for the contamination observed.
See attached abstract pdf for images.