The negative trials don't get a lot of discussion, but the ZEPHYR trial, a phase III study that directly compared Zactima (vandetanib), an oral inhibitor of EGFR and angiogenesis, vs. placebo, was one that merits some follow-up after my reporting that it failed to show a survival benefit, which was essentially the only thing we learned about the trial prior to ASCO this year. A more complete report of the ZEPHYR trial was presented at ASCO, and though it suffered the indignity of being the rare phase III trial that didn't get presented during an oral session, I think the results are important enough in trying to assess the real value of Zactima that the results merit being reviewed.

ZEPHYR was the last of four major lung cancer trials with Zactima to be completed and reported. The others are summarized in a prior post, and they showed at best equivocal results. The others are discussed in a point-counterpoint fashion by GRACE member and moderator Neil Berch (arguing that the Zactima glass is half full) and me (taking the less beloved view that the benefits are extremely marginal and probably not enough) -- let the record show that, at the present time, his post has been more favored by the people who have added a rating.

But with the overall results of these Zactima trials being so... debatable, I guess you could say... the results of a trial that compares Zactima to placebo alone is especially important. The ZEPHYR trial did that, enrolling 924 patients who had received 1-2 prior lines of chemotherapy, possibly with Avastin (bevacizumab) as well, and had also received and eventually progressed on an oral EGFR inhibitor such as Tarceva (erlotinib) or Iressa (gefitinib); patients were randomized 2:1 to receive Zactima (300 mg by mouth daily, a dose that is believed to have both anti-angiogenic activity and block EGFR), or placebo. What was especially notable is that this was a very unusual NSCLC population, comprised of 53% never-smokers, 53% women, with 80% of patients having an adenocarcinoma, and a median duration of prior EGFR inhibitors of 5 months. These results strongly suggest that the patients who went on this trials were heavily over-representing patients with an EGFR mutation and/or other demographic characteristics associated with unusually long survival with NSCLC.

As shown in the figures below, the trial showed while Zactima improved progression-free survival by nearly 40% (though essentially identical at the median, where half of the patients have progressed), it showed only the slightest hint of a more favorable overall survival with Zactima.

(click on image to enlarge)

Overall survival was the primary endpoint of the study, so this was considered a negative trial.

Importantly, the study looked at many clinical and molecular biomarkers to see if it was possible to identify one or more subgroups likely to benefit more with Zactima, but none of these analyses yielded any results to support the idea that we could focus our efforts with Zactima on a particular population.

Zactima was also associated with several side effects that were far greater than seen in the placebo arm. As in prior studies with Zactima, diarrhea, rash, high blood pressure, and EKG changes associated with altering the heart's electrical system were seen with it far more than with placebo.

Taken together, the results indicate that Zactima has some activity in this population, but with an improvement in progression-free survival but not overall survival, and with several associated side effects, this agent has lost momentum. I think with several other more encouraging leads out there for new agents, our attention, as well as opportunities for patients are better focused on agents that show the promise of doing far better.