STOCKHOLM The failure of the largest randomized phase III trial ever conducted in stage IV melanoma has left investigators around the globe with a question for which they have no easy answer: What next?
Temozolomide did not improve upon the dismal survival rates achieved by standard therapy with dacarbazine (DTIC) in the disappointing multinational study presented at the European Society for Medical Oncology Congress. And some experimental treatments for advanced melanoma produced objective responses in less than 10% of patients reported upon by phase II investigators at the same meeting.
Moreover, DTIC may not be better than best supportive carethe two have never been tested against each other, according to Dr. Lorenz Jost, who painted a glum picture of melanoma research to date in his discussion of the temozolomide study.
"We don't have any proof that dacarbazine extends survival. Even worse, we don't have any proof that DTIC doesn't shorten survival," Dr. Jost of Kantonsspital Bruderholz in Switzerland told Congress attendees.
Except for the historical failure to compare DTIC to best supportive care, Dr. Jost found nothing wrong with the conduct of the European Organisation for Research and Treatment of Cancer (EORTC) 18032 trial comparing oral temozolomide with DTIC injection. At a median 18-month follow-up, median overall survival was little better than 9 months in both arms of the 859-patient study.
No chemotherapy combination has shown a significant advantage over DTIC, Dr. Jost said, citing a study comparing DTIC with vinblastine, bleomycin, and cisplatin (J. Clin. Oncol. 1984;2:164-8), nor have more aggressive regimens, such as one augmenting DTIC with cisplatin and carmustine (J. Clin. Oncol. 1999;17:2745-51).
Vaccines likewise have produced similar survival rates to DTIC (Ann. Onc. 2006;17:563-70), he continued, and a metaanalysis of 18 trials involving 2,621 patients randomized to biochemotherapies versus chemotherapy yielded an odds ratio of 0.99 for overall survival (J. Clin. Oncol. 2007;25:5426-34).
If DTIC cannot clear large tumors, Dr. Jost asked, can it be better used as an adjuvant therapy? No benefit has been seen with that approach, and boosting DTIC with sorafenib also produced no advantage in a study published earlier this year (J. Clin. Oncol. 2008;26:2178-85).
Maybe investigators should move on to a new class of drugs targeting cytotoxic T lymphocyte-associated antigen 4 (CTLA-4), he suggested, showing a list of about a dozen studies, among which only three trials reported patient responses (J. Trans. Med. 2008;6:22).
Despite all these red lights, Dr. Jost urged his audience not to despair. "Keep putting patients onto trials," he said.
But what trials? Even before Dr. Jost discussed the temozolomide results, an audience member suggested that perhaps it was time to stop doing phase III chemotherapy trials in melanoma.
Indeed, so many phase III trials have failed to improve survival rates in patients with late-stage melanoma that asking "What is your next phase III?" is to pose "a below-the-belt question," according to Dr. John M. Kirkwood, director of the Melanoma Center at the University of Pittsburgh Cancer Institute.
"We don't have an idea in the cooperative groups in the United States. Both the Southwest [Oncology Group] and the ECOG [Eastern Cooperative Oncology Group] are totally waiting for something in a blinding streak of brilliance," said Dr. Kirkwood, an ECOG investigator.
He reported on the investigational anti-CTLA-4 agent tremelimumab in a Pfizer Inc.-sponsored trial during the same session (see related story above). The bottom line seen in trials of tremelimumab and another experimental anti-CTLA-4 agent, ipilimumab, is that "objective response rates are very similar and there are people alive who wouldn't have been without these agents," he said in an interview. The main obstacle is anti-CTLA-4 therapy is "not efficient," with only a small number of people responding out of hundreds so far treated. The next step is to study proinflammatory cytokines and proteomics to identify factors predictive of response to these agents, Dr. Kirkwood said.
For Dr. Poulam M. Patel of the University of Nottingham (England), and investigator of the temozolomide trial, finding ways to identify molecular targets and subtype patients is likely to be the next direction taken by collaborative groups in melanoma. In his presentation, he also noted that no therapy has been proved more effective than DTIC in 2 decades.