Encouraging Results With Ixabepilone in Metastatic Breast Ca

NEW YORK—Recent clinical studies from the National Cancer Institute (NCI) show that the macrolide analog ixabepilone (BMS-247550) is effective in treating metastatic breast cancer, is less susceptible to resistance than paclitaxel, and is associated with much lower rates of peripheral neuropathy than the taxanes. Sandra Swain, MD, chief of NCI's Cancer Therapeutics Branch, discussed current trials of ixabepilone in these patients at the Chemotherapy Foundation Symposium XXIII (abstract 62).

Ixabepilone (aza-epothilone B) is a semisynthetic analog of epothilone B, "a natural macrolide produced by the myxobacterium Sorangium cellulosum, discovered in the soil of the Zambezi River in Africa about 20 years ago," Dr. Swain said. Epothilones bind at a separate site on beta-tubulin from paclitaxel, are strong promoters of tubulin polymerization, and produce mitotic arrest at the G2/M phase of cell division, she noted. Epothilone B has tubulin polymerization activity 2 to 10 times stronger than paclitaxel. Another advantage of this drug class, she said, is that it is less susceptible than paclitaxel to multidrug resistance (MDR) via P-glycoprotein.

Efficacy, Low Neurotoxicity

Ixabepilone has strong antitumor activity in paclitaxel-resistant cell lines expressing MDR protein; and, in a mouse model of PAT-21 breast cancer xenografts, which are paclitaxel-resistant, it has been shown to more strongly suppress tumor growth than paclitaxel, docetaxel (Taxotere), and vinorelbine (Navelbine). It is also well tolerated, Dr. Swain emphasized, with low rates of grade 3 neuropathy, and unlike taxane treatment, patients do not require antiemetic medication or premedication with steroids.

Several early clinical trials have evaluated different schedules of ixabepilone, she said. In an effort to maintain efficacy but reduce neurotoxicity, a phase I trial from NCI assessed a regimen of ixabepilone at 6 mg/m2/d X 5 in patients with advanced tumors. There were two partial responses (PRs) among four patients with breast cancer in this study, both of whom were heavily pretreated.

In a subsequent phase II trial, NCI-0229, a total of 37 patients who had received prior taxane therapy and a group of 19 taxane-naive patients were treated with ixabepilone at 6 mg/m2/d X 5 until disease progression. The overall objective response rate "was really very good" in women treated with prior taxanes, at 22% (one complete response, after 11 cycles, and seven PRs), Dr. Swain said (J Clin Oncol 23:2726-2734, 2005).

The agent was well tolerated, with grade 3-4 neutropenia seen in 16% of patients, and diarrhea and fatigue in 11% and 13%, respectively. About 3% of the women experienced grade 3 sensory neuropathy, and 22% had grade 2.

"It was disappointing that the incidence of grade 3 neuropathy wasn't zero, because we really wanted it to be lower, but it was still lower than what you would see with weekly paclitaxel, which could be up to 20% grade 3," she said. The incidence of grade 3 peripheral neuropathy during ixabepilone treatment in the taxane-naive group was about 5%, and the grade 2 incidence was 16%.

To predict which patients would develop treatment-related neuropathy, Dr. Swain and her colleagues applied several tools normally used to assess neurologic function in diabetic patients.

Matched-pair testing of women who had received at least two cycles of ixabepilone revealed that baseline performance on the grooved peg board test (using the dominant hand) and the Jebsen test of hand function (on the nondominant hand) "most often correlated with and predicted ixabepilone-induced grades 2 and 3 neuropathy and should be explored further in larger trials," she reported. For example, she said, for women with baseline scores of 55 or higher on the Jebsen test, median time to neuropathy was 4 months from cycle 3.

Given that increased microtubule assembly (greater stabilization) appears to be predictive of improved sensitivity to stabilizing agents, the investigators conducted pre- and post-ixabepilone histologic studies of biopsies from five patients in the taxane-pretreated group, to see whether post-translational acetylated tubulin levels correlated with response.

They found that patients with higher levels of acetylated tubulin at baseline or at cycle 2 of ixabepilone, indicating microtubule stabilization, had a better response to treatment than those with low levels at baseline and during treatment. Although the small sample size yields only hypothesis-generating data, Dr. Swain said, acetylated tubulin levels should be investigated further in this setting. "We are expanding [this assessment] in our taxane-naive group," she said.

Phase III Studies

Ixabepilone is now being evaluated in two large randomized phase III trials in metastatic breast cancer, given alone or in combination with capecitabine (Xeloda). CA163046 will study the agent with or without capecitabine in 750 anthracycline- and taxane-resistant patients with measurable tumor. CA163048 will investigate single-agent and combination therapy in 1,200 anthracycline- and taxane-pretreated women both with and without measurable disease. Patients are still being enrolled in both trials.

Other clinical studies investigating ixabepilone in combination with capecitabine, carboplatin (Paraplatin), or trastuzumab (Herceptin) are ongoing.