Executive Summary

FOREST PHARMACEUTICALS SHOULD AUDIT ALL ITS ANDAs FOR DISCREPANCIES between raw data and submissions to FDA, FDA's Cincinnati district office recommended in a March 16 warning letter to the company. The letter specifically cites discrepancies between raw data and ANDA submissions for Bancap and Esgic CF discovered during a Dec. 3 - Feb. 13 inspection. "We recommend you audit all of your ANDA submissions to assure that the raw data on which they are based is available and reliable," the warning letter says. "If in your self-audit review you find significant discrepancies between data, errors and/or deficiencies, you may wish to voluntarily withdraw the defective ANDA submissions," FDA stated. Forest said that its reply to FDA will be that the firm has had an auditing program for years. Forest says that the discrepancies are due to inadequate practices on the part of personnel who joined Forest when it purchased the Cincinnati, Ohio manufacturing facility from small drug firm O'Neil, Jones, and Feldman in 1985. Forest said it will respond to FDA's warning letter in the next several days, adding that it has updated and improved recordkeeping procedures following the acquisition. The inspection that uncovered the discrepancies was a preapproval inspection for Esgic CF, an analgesic/muscle relaxer, which Forest said it is no longer interested in marketing. The analgesic Bancap, which was acquired from O'Neil, Jones and Feldman, is a small volume product with annual sales of $50,000, the company noted. Describing particular problems, FDA stated that "stability information, submitted in a Jan. 15, 1987 annual report in support of the ANDA for Bancap, contains false data." The tests "were only performed on the 500 tablet bottles, yet the same results were also submitted as documentation for stability on 100 tablet bottles." Forest said that the problem was just an oversight. FDA investigators also found that "finished product test records, submitted in support of a July 12, 1985 ANDA amendment for Bancap, were altered to show uniformity of dosage results dated Feb. 20, 1985. However, raw data show this test was not actually performed until July 11, 1985." In addition, FDA's letter says, "the original (unaltered) finished product test record shows no uniformity of dosage results." FDA said that because of the discrepancies it will recommend that the ANDA for Bancap capsules be withdrawn. In the case of the Esgic CF ANDA, accelerated stability information "indicates the 100 tablet samples passed, whereas raw data indicates the dissolution did not pass the 's1' phase," the warning letter notes. "Furthermore, the testing was not carried through to the 's2' phase, and no explanation exists as to the reason." "Based upon our observations made during this and other recent inspections, we may also recommend withdrawal of approved ANDAs for other products at a future date," FDA said. "This determination will be made on a product-by-product basis, pending additional inspectional coverage." During the inspection, Forest ceased production to address deficiencies in good manufacturing practices observed by FDA inspectors. The warning letter states that "validation of manufacturing processes and controls is inadequate for all drug products." Forest said that the problem is that FDA is now requiring new validation standards. The company believes that its existing validation standards are adequate. Forest said it will adopt the new standards and expects production will begin again sometime in April.