Maintaining Control on the Learning Curve
FDA is opening the door to continuous learning in process development.
Learning starts with failure; the first failure is the beginning of education.
John Hersey, The Child Buyer
To make a long commute seem shorter, I've taken to listening to recorded books on American history. The current selection is David McCullough's 1776, which chronicles George Washington's growth as a military leader in the crucial opening year of the American Revolution. McCullough shows that Washington was not, at first, a great military leader. Rather, he became one by learning from a series of near-fatal mistakes in the New York campaign, failures that might have broken another man—leaving important roads undefended, repeatedly dividing his army in the face of the enemy, and holding on too long to undefendable Manhattan Island.
Douglas McCormick
But the best of us learn from failure, and Washington learned from his, understanding both his own shortcomings and those of his officers and men—and learning, too, their strengths.
In fact, the freedom to learn is the freedom to fail—and to come back stronger from that failure. That's worth keeping in mind as the US Food and Drug Administration puts into practice its long-stated objective to embed continuous process learning into the regulatory protocols. Drug manufacturing processes do not, after all, spring complete from the developers' foreheads. We learn how things work by watching them work. It takes time, trial, and error.
The agency has taken that pro-learning approach in amending 21 CFR 210 ("Current Good Manufacturing Practice in Manufacturing, Processing, Packing, or Holding of Drugs; General") to exempt Phase I drug manufacturing from the requirements of 21 CFR 211 ("Current Good Manufacturing Practice for Finished Pharmaceuticals"). The rule followed, by a few days, an accompanying draft guidance, INDs—Approaches to Complying with CGMP during Phase 1.
The new guidance will ultimately replace 1991's Guideline on the Preparation of Investigational New Drug Products (Human and Animal). Although the old document paid lip-service to recognizing "that manufacturing procedures and specifications will change as clinical trials advance," it went on to require IND adherence to commercial-scale CGMP standards. Or, in the words of the new guidance, "the 1991 document did not address fully the Agency's expectation that an incremental approach to manufacturing controls would be taken during investigational drug development."
It might be tempting to see the new Phase I recommendations as "CGMP Lite," but that would be a mistake. It emphasizes the responsibilities of the quality control unit (lapses in which are now the leading cause of Form 483 citations). It seems clear that the agency expects full compliance with the spirit of CGMP, if not with the prescriptive letter of Part 211. That said, the guidance does ease the burdens of compliance, and offers some practical suggestions for reducing the effort needed to comply. (Use disposable equipment, for example, or outsource manufacturing.)
The objective is to help re-fill the pipeline of new medicines by widening it at the first clinical stages. Reducing some of the regulatory load—in time, money, and resources—will allow large organizations to evaluate more candidates, and will permit small organizations (small research institutions in particular) to do trials they couldn't dream of affording under the old rules.
"These requirements are so burdensome for early phase 1 studies that many leading medical research institutions have not been able to conduct these studies of discoveries made in their laboratories," said FDA Deputy Commissioner for Operations Janet Woodcock when the guidance was released.
With this freedom, of course, comes responsibility: able now to admit that they are learning how to control their processes as they go along, developers must learn how to learn. Will there be mistakes along the way? Of course. But they will be the foundations of education, and the beginnings of leadership.
Douglas McCormick is the editor in chief of Pharmaceutical Technology, dmccormick@advanstar.com