Warning Letter of the Week: inadequate investigation edition

A friendly note to the Executive Vice President of Zhejiang Huahai Pharmaceutical Co., Ltd. from the Center for Drug Evaluation and Research:

1.      Failure of your quality unit to ensure that quality-related complaints are investigated and resolved. 

Your firm received a complaint from a customer on June 6, 2018, after an unknown peak was detected during residual solvents testing for valsartan API manufactured at your facility. The unknown peak was identified as the probable human carcinogen N-nitrosodimethylamine (NDMA). Your investigation (DCE-18001) determined that the presence of NDMA was caused by the convergence of three process-related factors, one factor being the use of the solvent [redacted]). Your investigation concluded that only one valsartan manufacturing process (referred to as the [redacted] process in your investigation) was impacted by the presence of NDMA.

However, FDA analyses of samples of your API, and finished drug product manufactured with your API, identified NDMA in multiple batches manufactured with a different process, namely the [redacted] process, which did not use the solvent [redacted]. These data demonstrate that your investigation was inadequate and failed to resolve the control and presence of NDMA in valsartan API distributed to customers.... 

...Your response states that NDMA was difficult to detect. However, if you had investigated further, you may have found indicators in your residual solvent chromatograms alerting you to the presence of NDMA. For example, you told our investigators you were aware of a peak that eluted after the [redacted] peak in valsartan API residual solvent chromatograms where the presence of NDMA was suspected to elute. At the time of testing, you considered this unidentified peak to be noise and investigated no further. Additionally, residual solvent chromatograms for valsartan API validation batches manufactured using your [redacted] process, with [redacted] in 2012 ([redacted], and [redacted]) show at least one unidentified peak eluting after the [redacted] peak in the area where the presence of NDMA was suspected to elute....

It's never good when other people can find evidence in your own data that you missed....