The walls have ears
"She was a sneaky inspector is all I have to say," grouses our GMP Agent-in-Place. "She wouldn't accept a cup of coffee in the inspection room as it was 'unethical' and instead wanted to go to our cafeteria and pay for her own. So we let her. Unfortunately, she overheard several conversations during her cafeteria coffee break that lead to new avenues of inspection. And FDA Form 483 observations."
Our own worst enemy"We were validating the configuration and installation of our global enterprise management system," says our GMP Agent-in-Place. "The validation plan required that an audit be performed at specific times during the validation process. Apparently, the plan author and approvers never saw fit to tell the quality department, as they didn't know this was a requirement and the audits were not completed.
"It was only at the end of the project when a third-party expert was brought in to determine the project compliance profile was this noted. Now we are tracking corrective actions for this audit finding," sighs our Agent.
"We were an old line-manufacturing site with a lot of history. Some things we did just because we always did them that way," reports our Agent-in-Place. "One example was sampling the sterile bulks for sterility tests. At one point we had four bulks in a row that failed the sterility test. However, all sterility tests done downstream of that point, including the filled vials, passed. We sampled additional filled vials and obtained passing sterility-test results. Due to the labile nature of the product, it was filled immediately after the bulk was sampled, without waiting for the sterility-test results. There was no additional filtration after that point.
"Our conclusion was that it was a sampling issue, and that our old-fashioned method of sampling by sticking a syringe needle through the filtration tubing was not the best approach, and a vessel redesign was necessary. At a cost of around $8000 each, we installed an aseptic sample port in more than 20 vessels," says our Agent.
"Our product was an emulsion that had been manufactured for years without a problem," said our GMP Agent-in-Place. "It turns out that since there were only a couple of batches a year, one person had been performing all the manufacture, and we had no stability or separation problems. One year, he took a vacation just as the semi-annual production was scheduled, so a different employee made the emulsion. Three months later, we had separation showing up in the stability samples.
"The investigation showed there were no changes anywhere—in chemicals, suppliers, batch records, or analytical methods. Nowhere! The only difference was the employee who made the product. The batch records appeared to be identically completed, all meeting the requirements. What could possibly be wrong?
"We ended up interviewing the two employees, separately. Both claimed to have followed the directions exactly. So we discussed each line of the batch record with them, and there was only one small difference. The new employee said he had set the RPMs to 3000, started the mixer, set the timer for 60 min, then shut the machine down. The old employee said he set the RPMs to 3000, started the mixer, waited for the mixer to come up to speed, set the timer for 60 min, then shut the machine down. When he was asked how long he it takes the mixer to come up to speed, he said, 'About an hour, that stuff is really thick!'"
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]