Ixabepilone and the Narrow Path to Developing New Cytotoxic Drugs

During the last decade, the focus of attention in the field of new drug development has drifted away from the uninterrupted quest of new and better cytotoxic drugs to the newer and more glamorous grounds of targeted therapy, proof of principle and proof of concept, and sophisticated pharmacodynamic investigations. After years of relentless work behind the scenes, ixabepilone is emerging from the somewhat neglected list of cytotoxic compounds to stand on stage as something more than an apprentice, as reported in this issue of the Journal of Clinical Oncology (JCO). There are reports of five phase II studies of the drug, four in metastatic breast cancer (MBC) and one in non–small-cell lung cancer (NSCLC). Ixabepilone (BMS-247550) is an analog of epothilone B and a reference compound of a novel class of tubulin-active drugs that blocks tubulin polymerization in a way similar to that of the taxanes. Ixabepilone binds to a different site of the tubulin molecule than the taxanes; such a feature may be relevant to the reported lack of complete cross resistance with paclitaxel and docetaxel in preclinical models. The data presented in this issue deserve attention for multiple reasons, forcing us to reconsider the role of the development of old-fashioned chemotherapy to complement the current armamentarium of drugs for cancer treatment in the era of targeted therapies. Ixabepilone enters with the almost mandatory ambition of challenging the still undisputed role of the taxanes. The data presented in this issue of JCO, as summarized in Table 1, support such ambition: ixabepilone worked well as first-line therapy in women with MBC, and had clear antitumor activity in tumors resistant to or even refractory to multiple other classes of cytotoxic drugs and—most important—to paclitaxel or docetaxel. For NSCLC, the activity was observed despite progression within 4 months from treatment with cisplatin or carboplatin. Such therapeutic activity is an encouraging start to widening the range of therapeutic options after failure of the taxanes, prompting the testing of ixabepilone in early stages of disease. But ixabepilone treatment comes at a price: neurotoxicity. Tubulin-active drugs almost invariably cause neurologic adverse effects. Ixabepilone, which was well tolerated regarding bone marrow and other organ toxicities, is not an exception, causing a paclitaxel-like neuropathy that was primarily sensory in nature, cumulative, and slowly reversible in the majority of patients. The neuropathy associated with ixabepilone had much to do with the uneven and slower than expected clinical development of the drug. Table 1 reveals a situation that points to the differences of dose and schedule of ixabepilone in the different trials reported in this issue of JCO. The difference is indeed larger than it appears from the table, due to the extent and severity of the neuropathy initially observed in four of the studies that forced a lowering of the dose from 50 to 40 or even to 32 mg/m per cycle to make treatment feasible with the schedule of administration once every 3 weeks. The only exception to the remedy of amending the dose was the study in women who never received taxanes before ixabepilone, in which daily administration for 5 consecutive days was adopted. One may be tempted to conclude that the daily schedule was less neurotoxic. However, after dose titration to feasibility of the once-per-cycle dosing, there was no obvious pattern of different