Manquements sur l'intégrité des données pour ce fabricant chinois...
Des remarques d'inspection de la US FDA mettent encore en cause les pratiques de "data and record management" pour ce fabricant chinois d'API :
1. Failure to have laboratory control records that include complete data derived from all laboratory tests conducted to ensure compliance with established specifications and standards.
Your quality control laboratory failed to record and maintain complete data from analyses of your (b)(4) ((b)(4)) API. For example:
- Prior to conducting official analyses,your quality control laboratory performed “experimental” analyses on product batches to assess whether your API met specifications, but failed to document these “experimental” tests in official laboratory records or to justify their exclusion. Our investigator found the results of 2,404 high performance liquid chromatography (HPLC) injections in a folder titled “Experimental” on instrument SZG-002-006l. Your quality unit indicated that these “experimental” injections were being conducted in all (b)(4)chromatographic units in your quality control laboratory. Your management provided different explanations in an attempt to justify the practice, including “fear” that the sample results would not pass.
- Our review of the audit trails of chromatographic systems SZG-002-009, -010, -011, and -012 documented that your laboratory analysts deleted raw chromatographic data on multiple occasions. Your firm indicated that analysts may have been testing the system and may have deleted associated files. You also indicated that the deleted files may represent aborted analyses. However, we documented that some audit trail entries of deleted raw data files contained batch numbers for actual batch samples being tested. There is no assurance that laboratory records and raw data are accurate and valid...
2. Failure to follow and document laboratory controls at the time of performance, and failure to document and explain any departures from laboratory procedures.
During the inspection, your firm provided our investigator a chromatogram for an assay analysis of (b)(4) batch(b)(4) dated August 30, 2014, at 9:46:39 a.m. Your firm later submitted to FDA a different chromatogram corresponding to the same analysis, instrument, date, time, and batch. The second chromatogram appears exactly the same as the one provided during the inspection, but it includes a different method file name, column type and serial number, and system temperature. Both versions of these documents cannot represent the actual assay analysis that you conducted for batch (b)(4) on August 30, 2014, at 9:46:39 a.m...
Plus d'informations ici.