"Our animal facility began as a partnership with a local university's large animal research facility. Our offices were situated next door," our GMP Agent-In-Place explained. "Although we had stopped doing research on large animals, they had not. One of the species they maintained was American Bison, and on one occasion, the bison escaped and ended up less than 10 feet from our laboratory windows."
Completely full"We had several batches of sterile water for injection that were designated to accompany products to be distributed in the United States and European Union," our GMP Agent-In-Place grimaced. "We tested the batches to ensure compliance with the US Pharmacopeia and European Pharmacopoeia (Ph.Eur.). The batches failed the Ph.Eur. extractable-volume test.
"During the investigation, we realized that the Ph.Eur. test method had changed since the last time the sterile water for injection had been manufactured," continued our Agent. "We ended up rejecting these batches and revising the filling volume upward to ensure that we would meet the extractable-volume requirement.
"For some tests, we refer directly to the appropriate pharmacopeia test method rather than extracting and writing the method in our own format. This process is meant to keep our written methods in sync with the pharmacopeia. However, when the pharmacopeia changes and we don't recognize the ramifications of those changes, this type of case can occur."
"Our problem started with an limulus amebocyte lysate (LAL, endotoxin) assay that produced highly variable results on a product that had never had any LAL problems," sighed our GMP Agent-In-Place. "Three batches were rejected and destroyed, resulting in a loss of about $1.5 million. We tried another LAL methodology in hopes of isolating a testing problem, but that method also produced atypical results. Upon reviewing the manufacturing records, we realized that all of the batches with especially high LAL results involved an early manufacturing step that was performed by a specific operator.
"It turned out that the operator had large hands and wore extra-large gloves," our Agent said. "The extra-large gloves were the only ones not purchased as sterile and pyrogen free. The processing step involved contact between the gloves and the product and equipment, which leached a pyrogen-like material out of the gloves. When this pyrogen-like material bound to the product, it produced exponentially higher LAL test readings compared with a water extract of the same gloves. The fix was easy."
"We normally don't schedule Saturday deliveries because our customers aren't present to receive refrigerated products," our GMP Agent-In-Place said. "But when one customer explained they had an emergency and promised to have someone there to accept the overnight delivery on Saturday, we agreed. Of course, that Monday, the same customer called to refuse the product because it had sat on their dock all weekend and was now hot and unsuitable for use."
"Computers have greatly increased our productivity," bragged our GMP Agent-In-Place, "especially in the laboratory where we have several autosamplers running hundreds of samples overnight and logging the results automatically into the system. However, when a manager puts a query into the test database late in the workday and then leaves when the query doesn't return a result quickly, the computer lags and no longer accepts data. As a result, a full night's work is lost, as are hundreds of results. I'll never do that again."
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]