Pharmaceutical mulligans

The story “Foreign Drug Suppliers Caught in Data Dragnet” has a subtitle that reads, “Critics say FDA’s focus on data integrity is blocking drugs from otherwise good manufacturers” (C&EN, Sept. 19, page 26).

Apparently, some “proficient drug manufacturers” in India and China are complaining that they have received warning letters from the U.S. Food & Drug Administration or have been banned from the U.S. because they “have failed inspections for undocumented deletions of test results or for not controlling who has access to data history.”

These active pharmaceutical ingredient (API) companies suggest that it is okay to discard data because of incorrect sampling, incorrect standard preparation, incorrect sample weighing, etc. They want FDA to give them “mulligans” because they believe that “it’s common sense to discard meaningless data.”

Nonsense! If a manufacturer’s analytical group is making multiple “innocent” errors such as those described on the “mulligan” list and then the group discards the data, it is not only violating current Good Manufacturing Practice (cGMP) regulations, it is avoiding corrective actions for incidents that may be part of a systemic problem. Are the multiple errors because a company’s analysts are not properly trained and/or not following the appropriate standard operating procedures? Do such companies even have SOPs?

Let’s, for the sake of argument, say that an analytical group for a cGMP manufacturer had only one innocent error and it discarded the data. If, under a “mulligan” approach, a manufacturer is allowed to discard data, how would FDA know whether there was just a single incident or thousands of events?

The cGMP regulations were established and are enforced to prevent injury and death to the ultimate consumer, the patient. Although there are numerous examples where analytical error and/or outright fraud has led to serious consequences, I’ll cite just one for brevity:

Between 2007 and 2008, Scientific Protein Laboratories, a manufacturer with facilities in China, substituted oversulfated chondroitin sulfate for heparin, possibly because the cost of chondroitin was one-hundredth the cost of heparin. The “error” went undetected by Scientific Protein Laboratories’ quality-control lab; the company then sold the drug substance to Baxter International. Baxter’s manufacturing analytical group failed to detect the contamination both in the raw material and in the final drug product. This resulted in hundreds of deaths and injuries.

I’ve worked in the pharmaceutical industry for several decades. I don’t believe that cGMP suppliers in China, India, or elsewhere should be given “mulligans” for what they perceive are trivial problems. API suppliers need to have a system to report all analytical problems and, when appropriate, implement and report corrective actions.

David Allen Marsh
Bonita Springs, Fla.