United States of America v. Barr Labs, Inc. 812 F. Supp 458, 3/30/93
GMP Lessons From a Federal Judge

Briefed by: A. Longwell, Esq.


This was a landmark decision because it lent a lot of legal strength to the old aphorism, “you can’t test a product into compliance.” It also established once and for all that FDA could get a recall order for product without any evidence that patients were being harmed, on the basis of non-compliance with GMP alone. Finally, it attempted to answer the age old question, “what do you do when the regulations allow conflicting but plausible views of the precise requirement?”


Barr Labs, Inc., is a generic drug manufacturer headquartered in New Jersey. In the years preceding 1992, they disagreed several times with their FDA district on whether or not certain of their practices were compliant with Drug GMP. In April ’92, Barr sued FDA for “ad hoc” drug regulation. FDA applied to the court for a “stop ship” injunction, claiming that Barr products were adulterated. The cases were combined and heard by Alfred Wolin, Federal District Judge in the New Jersey District. By the time a decision was reached in February ’93, the judge had reviewed 2300 pages of testimony and viewed 400 exhibits. Aside from a little impatience with both sides for continuing to litigate what the judge thought were issues capable of settlement by rational discussion, this opinion is an example of reasoned, impartial interpretation of the way GMP requirements should be applied to a specific case.


Any out-of-spec result requires a failure investigation.

  1. The extent of the investigation will depend on the type of error — least investigation for a lab error, most investigation for a process error — but all must be investigated, and the investigation documented.
  2. Barr’s practice of responding to an out-of-spec result by testing twice more and taking “best two out of three” is insupportable and unscientific, and is not GMP.
  3. FDA’s claim that a single out-of-spec result invalidates a lot is unreasonable, and is not GMP.


When there is more than one reasonable interpretation of GMP…

  1. Role of “industry practice:” “the Court cannot rely on industry practice alone to determine whether an individual firm meets the statutory requirements.”
  2. “Industry can look for guidance to literature from seminars and pharmaceutical firms, textbooks and reference books, and FDA letters to manufacturers.”
  3. Role of good science: “Industry can… employ scientific judgment where appropriate.”
  4. Interpretation must be “reasonable and consistent with the spirit and intent of the cGMP regulations.”


User/Patient injury or the potential for such injury is NOT a prerequisite to granting a stop ship order.

  1. A firm can be forced to stop ship and even recall product simply on the basis of GMP failure alone.
  2. Injury to the company (e.g., bankruptcy) is not a sufficient reason to deter the court from issuing an order.


Results *

  • Barr forced to recall 12 products in commercial distribution.
  • Barr forced to cease shipment of 24 products.
  • Barr logs loss of $2.4 million in quarter ending 3/31/93. (Author’s note: The Feds had 4 attorneys, while Barr fielded 12.)
  • FDA develops “good guidance practices,” improves inspector training, improves industry communication.


Don’t think you can persuade a U.S. court to accept a procedure not specified by regulation as GMP if it has no (or weak) scientific/engineering rationale.

You can win against FDA by going to court, if you have a good case. However, if your case is that good, you should first explore an agreement with FDA that precludes going to court, and is much less expensive.

At times suits against the FDA have been successful: This year courts have found FDA can’t classify one product as a device and three similar products a