She noted that Dr. Rugo must “remain strict to the statistical analysis of the study” and cannot state that “the two novel agents were worse” than paclitaxel. But she, as the discussant, could claim “they were not better, they were more toxic, and they were substantially more expensive,” Dr. Miller said. “With the exception of patients with major issues with steroids, I simply don’t see any reason for [nab-paclitaxel or ixabepilone] to be a preferred choice in the first-line setting.”

CALGB 40502 Implications

It is unclear why the novel microtubules do not improve activity over paclitaxel, she added. This is unlikely due to chance or to a differential interaction with bevacizumab, she said. The drugs appeared more potent, not less, in preclinical studies, and the doses used in this trial had been previously established as effective.

“We saw that the toxicity of these regimens attenuates their exposure, with nearly half the nab-paclitaxel arm requiring dose reduction … and a substantially greater number discontinuing therapy…. So, in effect, this was a trial of lower-dose ixabepilone and lower-dose nab-paclitaxel,” she pointed out. “There is no reason to repeat this effort with a lower dose.”

The study also suggests that preclinical potency studies can be misleading, and randomized phase II trials “get us excited but also let us down,” she added. “There is no substitute for well-done phase III trials as seen here.” ■

Disclosure: Dr. Miller has served in a consulting or advisory role for Clovis, Genentech, and Nektar, and has received research funding from Genentech, Geron, Merrimack, and Roche.