Warning Letter of the Week: "trial analysis" edition

A letter from the Center for Drug Evaluation and Research to the President and CEO of Yuki Gosei Kogyo Co., Ltd. of Tokyo: 

1.      Failure to maintain complete data derived from all laboratory tests conducted to ensure your API complies with established specifications and standards. Your firm does not ensure that complete data from testing of your API are included in the official batch record and reviewed by your quality unit. For example, you reported passing results for related substances testing of [redacted] lot #[redacted] analyzed starting at [redacted] on July 28, 2015. However, our investigator found unreported analyses including out-of-specification (OOS) results for the same lot acquired earlier on the same date, and on the next day as the reported results. You failed to include this data to be reviewed by your quality unit prior to the release of the lot. Our investigator documented the same pattern with other products not intended for the U.S. market.

In your response, you explained that this “trial analysis” was performed on the sample solution for conditioning the high-performance liquid chromatography (HPLC) column. However, your explanation did not address why the “trial analysis” was performed using a sample solution instead of a standard solution, or why you ran this extra analysis in addition to the system suitability test, which verifies that a chromatographic system is adequate as set forth in USP 621.

You also acknowledged that a retrospective review conducted after the inspection found additional instances of unreported electronic data in original batch records. Your review only assessed laboratory data and did not assess all parts of your facility’s operation where CGMP information is generated and maintained. In addition, you failed to provide details of your review criteria and methodology. 

These FDA inspectors, picky-picky!