ODAC Favors FDA Approval of Xeloda and New Indication for Gemzar

BETHESDA, Md--The Oncologic Drugs Advisory Committee (ODAC) has recommended approval by the Food and Drug Administration of a new anticancer agent and favors a new indication for a previously approved agent.

The committee recommended accelerated approval for Hoffmann-La Roche’s Xeloda tablets (capecitabine) for the treatment of patients with advanced or metastatic breast cancer who have failed paclitaxel (Taxol) and an anthracycline-containing chemotherapy regimen, or who have failed paclitaxel and are not candidates for an anthracycline.

Xeloda is the first tumor-activated drug aimed at malignant breast cancer, and the first agent shown to benefit patients with metastatic breast cancer for whom standard therapies have failed, the company said.

It belongs to a new class of pharmaceuticals known as fluoropyrimidine carbamates, and it acts preferentially against tumor cells. According to the company, 1,200 patients worldwide have received the drug. It is currently in phase III testing for the treatment of colon cancer.

Hoffmann-La Roche presented data from a multicenter noncomparative pivotal phase II study in 162 postmenopausal women with breast cancer whose metastatic tumors were no longer responding to a standard chemotherapy regimen of paclitaxel and an anthracycline drug. The primary efficacy endpoint was the objective response rate in the 135 women with measurable disease. Forty-three of these women were resistant to both pac-litaxel and their anthracycline.

The company’s analysis showed tumors shrinkage of more than 50% in 27 women (20%); the response rate among the women resistant to both drugs was 25.6%. The FDA review team put the overall response rate at 18.5% and agreed with the 25.6% response rate in the doubly resistant women.

The FDA analysis found an identical median duration of response for both groups of 154 days. "I think we have demonstrated the drug has activity in the most refractory patents," Thomas Griffin, MD, medical director for oncology at Hoffmann-La Roche, told the advisory committee.

In the pivotal study, approximately 15% of patients experienced grade 3/4 diarrhea, the company reported; 10% had grade 3 hand-foot syndrome, and 7% had grade 3 stomatitis. These adverse effects were the most frequently reported, along with fatigue, but reversed after discon-tinuation of treatment.

In its presentation, Hoffmann-La Roche also argued that Xeloda improved patients’ quality of life, noting a reduction in pain scores and an increase in performance functioning, both statistically significant. The company called Xeloda well-tolerated, and noted that it caused minimal hair loss and limited bone-marrow depression.

ODAC members gave Eli Lilly’s Gemzar (gemcitabine) a favorable vote for approval as palliative treatment of locally advanced and metastatic (stage III and IV) non-small-cell lung cancer (NSCLC), both as a single agent and in combination with cisplatin (Platinol). The drug was approved by the FDA in May 1996 for the treatment of patients with locally advanced or metastatic adenocarcinoma of the pancreas.

In support of its supplemental new drug application to treat NSCLC, the company presented data from a multicenter trial, designated JHEX, that randomized 522 NSCLC patients to either Gemzar plus cisplatin or cisplatin alone. Much of its presentation, however, relied on an interim analysis that involved 155 patients on Gemzar-cisplatin group and 154 on cisplatin alone.

Overall, the 262 Gemzar-cisplatin patients had a significantly better median survival than the 260 cisplatin patients (9.1 months vs 7.6 months). They also showed a trend to a greater 1-year survival rate (39% vs 28%). The FDA staff review put the median survival time at 8.7 months vs 7.5.

The interim analysis by Eli Lilly showed a significant difference in median time to disease progression of 5.8 months in the Gemzar arm vs 3.7 months for the cisplatin-only patients.

Hematologic toxicity proved significantly greater in the Gemzar-cisplatin patients: 26% suffered grade 3/4 anemia vs 5% in the cisplatin arm; 23% had grade 3 and 35% had grade 4 neutropenia (vs 4% and 1%); 23% experienced grade 3 and 28% grade 4 thrombocytopenia (vs 2% and 1%).

Gemzar patients also required statistically significant more peripheral red blood cell and platelet transfusions.

"Bone marrow suppression is more pronounced with gemcitabine and cisplatin that with cisplatin alone," Alan Sandler, assistant professor of medicine, Indiana University Medical Center, concluded in presenting the JHEX data. "Nonhematologic toxicities occur at approximately the same frequency in both treatment arms." Of these, nausea and vomiting were most common.

Lilly supported its application with two smaller European studies. JHBR, a multicenter trial, randomized 69 patients to Gemzar-cisplatin and 64 to cisplatin-etoposide. The tumor response rate (41% vs 22%) and median duration of response (8.4 months vs 6.1 months) were significantly better in the Gemzar group.

In JHEZ, a phase II multicenter study that tested Gemzar alone (72 patients) against cisplatin-etoposide (75 patients), no patient in either arm showed a complete response; 18% of the Gemzar arm and 15% of the cisplatin-etoposide group had a partial response.

After some discussion, the committee voted 9-to-0 with one abstention that the efficacy of the Gemzar-cisplatin combination shown in the JHEX study outweighed its increased toxicity.