SAN FRANCISCO—Randomized controlled trial data from earlier this year show that azacitidine prolonged overall survival in patients with high-risk myelodysplastic syndrome (MDS) compared with three standard therapeutic regimens. A question left open in that publication, though, was whether patients with stable disease but no evidence of hematologic response derived a survival benefit from azacitidine. During a review of epigenetic therapies at the Oncology Congress here, one researcher revealed that as yet unpublished data indicate that these patients do not show a survival benefit and can be taken off therapy.
“I can tell you—between you, me, and the walls—that it is my impression of the data, that if a patient doesn't have hematologic improvement by six months and you decide to stop because you don't see it benefiting the patient, you can sleep well that night,” said Steven Gore, MD, Professor of Oncology at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins University. “It is probably a reasonable decision. In terms of survival, you are not selling the patient short.”
He made those comments based on his posthoc analysis of the large randomized AZA001 trial. He said the full analysis will be formally presented at the American Society of Clinical Oncology Annual Meeting this spring.
The AZA001 trial met its primary endpoint of overall survival, with the 179 patients in the control arm having a median survival of 24.5 months compared with 15.0 months for the 179 patients in the control arm. The investigators undertook the posthoc analysis to discover whether the survival benefit extended to patients who had stable disease but did not show a hematologic response. According to Dr. Gore, those patients who did not show evidence of hematologic response by the end of six cycles (six months) did not gain a survival advantage with azacitidine therapy compared with standard treatment.
“People who are not showing hematologic improvements after six months, I am personally stopping therapy or switching therapy,” Dr. Gore said. “If the survival of the people who are stable on supportive care is the same or better, then I don't think there is any reason to spend a lot of money and treat the patient in a way that is not being beneficial to them.”
Wendy Stock, MD, Professor of Medicine at the University of Chicago Medical Center, who chaired the session in which Dr. Gore spoke, declined to comment specifically on Dr. Gore's unpublished data, but said “What the data suggest, I think, is that if there is no major response by four to six cycles, then there is not going to be. If there is clear evidence of progression, I would take them off before that. There is some discussion that there may be some benefit to stabilization of disease in terms of quality of life, but that remains to be determined.”
A previous analysis of AZA001 data, presented by Lewis Silverman, MD, of Mount Sinai School of Medicine in New York City, at the 2008 American Society of Hematology Annual Meeting, looked at the time to first evidence of hematologic improvement.
Unlike standard chemotherapy regimens, which work quickly, epigenetic therapies take time to work. Fifty percent of the patients who ultimately developed an objective response showed some hematologic improvement by two cycles; by six cycles 87% had done so.
“So if you do what I do and stop drug after six cycles if you don't see a response, there are a few people who might have gotten a response,” Dr. Gore said. “But if cut off before six cycles you are certainly going to miss some responders.” Based on that analysis and the unpublished analysis, he concluded “I think six cycles is a fair trial.”
That said, the time to best response was even longer, based on Dr. Silverman's analysis. Sixty percent of patients had achieved their best response by four cycles, but other patients required 12 cycles. Similar data are true of decitabine, Dr. Gore noted, so the agents must be used chronically.
“The worst mistake you can make is to give patients two cycles and they end up with neutropenic fever in the hospital and you say ‘okay this is too much for you.’ Then you've given them all pain and no gain,” Dr. Gore said. “It is important to explain that it will get worse before it gets better, and they need to tough it out for six cycles.”
Are Azacitidine & Decitabine Equivalent?
The data from AZA001 clearly show that azacitidine provides a survival advantage for patients with high-risk MDS over standard care. But how does that compare with decitabine, the other FDA-approved epigenetic therapy?
That question, said Dr. Gore, is hard to answer because there have been no head-to-head comparisons, and the trials testing the agents have been very different in design.
It is clear that both agents are active in the disease, which is not surprising, noted Dr. Gore, because the two drugs are chemically very similar and mechanistically probably pretty similar.
The dosing schedules in the trials, though, have been very different. “The azacitidine trials in general never capped the number of cycles you could have,” Dr. Gore said. “If you have a complete response, you get three more cycles and then stopped in [the AZA001] trial. But otherwise, if patients have anything less than a complete response, they were treated until progression; so people who had less than a complete response but were responding, essentially got maintenance therapy and that's been true of most of the azacitidine trials.
“Whereas in the decitabine trials, you get two cycles past complete response and six cycles total if you did not develop a partial response, but no more than eight cycles.”
Additionally, the FDA-approved dosing schedule for decitabine is very intense and difficult to give, he explained, so patients have tended to tolerate fewer cycles in the decitabine trials than in the azacitidine trials. (The approved dose is 15 mg/m2 every eight hours intravenously for three days.)
For example, the median number of cycles patients received in the registration trials for each of these agents was nine for azacitidine and three for decitabine. “I think this reflects not only the design of the trial, but also the fact that that 15 mg/m2 dose IV every eight hours for decitabine is a more toxic dose schedule than the FDA-approved dose schedule for azacitidine,” Dr. Gore said.
The FDA-approved dose schedule of decitabine is more myelosuppressive and leads to longer and deeper suppression, than the azacitidine dosing schedule.
He points out that although the decitabine registration trial did not show an overall survival advantage, the overall survival curves initially showed a difference similar to that seen in the azacitidine registration trial, but then came back together and the difference was not statistically significant.
“That is not surprising given that these patients got only a median of three cycles of drug” Dr. Gore said. “If anything, I think it is really spectacular that there was any difference seen at all. But the trial didn't hit its survival endpoint, so we can't say that decitabine at that dosing schedule improves survival in this population.”
Despite the fact that the FDA-approved dosing schedule is not the one used by most clinicians currently, it is the one that has been most broadly tested in clinical trials. Thus, it is the one that researchers and clinicians can evaluate using evidence-based medicine criteria.
“So at the end of the day…I think we have a very well-done azacitidine trial in high-risk MDS that shows a doubling of survival at two years, and an increase in median survival in two randomized trials, and decitabine, albeit using a dose schedule most of us don't use, doesn't show a survival benefit,” Dr. Gore concluded.
Dr. Gore emphasized that while his conclusions are based on clinical trial data, audience members should be aware that he has consulted for both Celgene, which makes azacitidine, and MGI Pharma, which developed decitabine before Eisai took over, and that he owns stock in and received research funding from Celgene.