ROCKVILLE, MarylandThe FDAgoing against a recommendation of its Oncologic Drugs Advisory Committee (ODAC)has denied the supplemental new drug application for the use of Gliadel Wafer (polifeprosan 20 with carmustine(Drug information on carmustine) implant, Guilford Pharmaceuticals) to treat newly diagnosed malignant glioma.
In its nonapproval letter to Guilford, the FDA said its analysis of data from the pivotal phase III, multicenter trial submitted by the company "failed to reach statistical significance by the protocol-specified log rank test." The FDA did not question the safety of the drug delivery system, in which time-release wafers containing the active drug are laid over the site of a removed brain tumor.
Guilford expressed disappointment at the FDA decision, but said that it intended to pursue expanded labeling for its product. "We have requested a meeting with the agency to resolve these issues and will amend our application in an effort to address their concerns," said Craig R. Smith, MD, the company’s chairman and chief executive officer.
The FDA approved Gliadel Wafer in 1996 as an adjunct to surgery in patients with recurrent glioblastoma multiforme. The product is now approved in 23 other countries for this indication, and in Canada for recurrent and newly diagnosed malignant glioma.
In December, ODAC members voted 8 to 5 that the company’s pivotal study provided "substantial evidence of survival benefit" in newly diagnosed malignant glioma patients. However, they also voted 7 to 6 that the study was not adequate. Federal law requires that FDA approval must be supported by adequate and well-controlled studies.
At the ODAC meeting, Guilford presented two randomized, double-blind, placebo-controlled, phase III trials in support of its application. Its pivotal study involved 240 patients treated with either Gliadel Wafer or placebo in 14 countries, including 12 patients accrued in the United States. The company interpreted the 2.3-month increase in survival in the Gliadel group as significant. The FDA’s analysis, however, put the P value of the increased survival at .08, a finding that it reaffirmed in its nonapproval letter.