Out of sight, out of date
"We were preparing for an FDA inspection at an overseas parenteral plant that hadn't produced for the US previously," our GMP Agent-in-Place explains. "We had started our walkthrough earlier that morning, and so far it had been going quite well. The plant manager, production and engineering manager were all part of the group walking through the WFI [water for injection] system. We had followed the water system from the incoming city water to the WFI holding tank. All was in order with the pumps, piping and controls, and so forth. The WFI holding tank was a large vessel with the agitator positioned between one of the major trusses supporting the building's roof. We climbed to the mezzanine to view the return lines into headspace of the tank. The QA [quality assurance] lead from headquarters was the first to notice that there was apparently no vent filter on the vessel.
"When this was noted, the site representatives' body language spoke volumes. Their dismay was evident in the awkward silence that followed. Because the entire visit had gone so well to that point, I decided to look more closely. Ducking under the support beam to look at the rest of the vessel's upper portion, I was pleased see the vent filter housing close to the beam out of sight from everyone else."'I've got good news,' I said, 'the filter's over on this side.' The sighs of relief from everyone on the mezzanine were clearly audible. The audit was going to be okay. Then I noticed a tag on the housing, it had the date the filter had last been changed. Then I said, 'I think there's some bad news as well. The last date on the tag is almost 7 years ago!' The filter housing was so hard to find that it had been forgotten all that time."
"When validation was still a relatively new activity, I was involved in the start up of a new parenteral facility,"says our GMP Agent-in-Place. "The sterilizers, tunnels, freeze driers, and other items of equipment got detailed qualification documents. However, it was agreed that for utilities, the engineering staff merely sign off that the systems had been built according to the approved specifications because they would be tested directly and extensively. The PQ [performance qualification] for the utilities had the usual number of deviations, but eventually all of the systems were acceptably validated. The last stages of the PQ included the production of a limited number of batches. Shortly afterward, FDA approved the facility for commercial use.
"Production at a much larger scale began immediately afterward. With the increase in usage of WFI, a major problem soon ensued. The very large system had difficulty meeting the WFI endotoxin limit. A massive investigation was undertaken, and after several weeks, we determined that the feedwater to the stills experienced wide swings in the incoming endotoxin levels. The mixed bed deionization system was being sanitized (actually regenerated with strong acid and base flushes) frequently (every other day at one point!) in an effort to eliminate the problem. Water was being released on a batch basis for all uses.
"Everyone was feeling pressured to correct the situation immediately. Finally, we decided to open the beds and see whether something was amiss. The problem was immediately evident. The subcontractor charged with putting the resins in the beds had put the resin in the beds along with the bags it was supplied in! They had merely punched large holes in the plastic bags and dropped them into the deionization vessels. The process team was considered visionaries after that."
Pharmaceutical Technology's monthly "Agent-in-Place" column distills true-life cautionary tales from the secret files of Control, a senior compliance officer. If you have a story of clueless operators, oblivious management, inopportune lapses of judgment, or Murphy's Law in action, please send it to Control at [email protected]