In previous GMP News (27 May 2010, 25 March 2011 and 24 October 2012), we presented FDA "Findings" in the complex topics validation / qualification / calibration in manufacturing with regard to medicinal products. In the following, you will find an analysis of the 37 Warning Letters issued in fiscal year 2013 (from 1st October 2012 to 30 September 2013) on that topic.
In 2013, process validation was at the top of the critised deviations. Missing or insufficient process validation was quoted 15 times. The most frequent critics were made to dosage forms (5 times) as well as to the manufacture of creams (3 times). The inexistent validation of a filtration time and of a filter replacement as well as insufficient packaging validation (filling, sealing and labelling) have been criticised too. More and more terms from the "new" FDA Guidance of 2011 appear among the corrective measures; for example "control strategy" and "scientific evidence that your manufacturing process are capable of consistently manufacturing quality products" are required.
Deviations regarding missing or insufficient cleaning validation have been observed 7 times; more precisely: the inexistent use of worst case products, handling errors in "swabbing" and missing swab recovery rates were observed and criticised.
Media Fills have also been often criticised (4 times) Missing hold times of vials before incubation, inexistent rational numbers for the non-incubating of filled vials, no documentation about the reasons for rejecting vials, and insufficient failure investigation were among the points of criticism. Interestingly, the (non-validated) hold time was necessary as the company hadn't enough resources to evaluate the Media Fills.
Deficiencies of qualification in manufacturing were identified 11 times and concerned 5 times missing "smoke studies" for clean room qualification: the insufficient qualification of a water facility and poor clean room design were respectively criticised once. An expired balance calibration was observed once.
Also one observation was made to the lack of validation for the Computer software Microsoft Dynamics.
Conclusion: The analysis is similar to that of the previous years and the top deficiencies haven't changed that much. The most "Findings" concerned the process validation topic, and again, the majority had to do with solid forms. The number of insufficient Media Fills was also comparable to that of the previous years. Nevertheless, one can notice the increase of qualification deficiencies - whereby also in the past, missing or poor "smoke studies" often appeared.