A positive trend in median overall survival was also observed favoring pixantrone (10.2 months) over the comparator treatments (6.9 months.)
Would these additional efficacy data sway the FDA to approve pixantrone even if the primary endpoint of the EXTEND study -- the response rate measured by CR/CRus -- was not met? Not likely, for when a study misses its primary endpoint, all secondary endpoints, including survival, become exploratory in nature, only.
4. Were the patients in the comparator arm of the study treated with the best drugs available?
The 7.1% CR/CRu rate amongst patients treated with one of six different single-agent chemotherapy drugs in the comparator arm of the study seems low, especially compared to the response rates found in other similar trials. (The FDA compiled a chart of single-agent drug trials in aggressive NHL patients as part of its unflattering 2004 review of Marqibo, another lymphoma drug. You can find it
If the FDA finds that patients were treated with substandard drugs in the comparator arm of the study, or if the FDA believes that more efficacious drugs should have been used, it could, like the point above, raise questions about the benefit of treating patients with pixantrone.
5. How sick really were the NHL patients in the EXTEND study?
One of the central arguments that Cell Therapeutics makes in support of pixantrone's approval is that the NHL patients in the EXTEND study benefitted from the drug despite the advanced stage of their disease. These patients, to put it simply, were the sickest of the sick and were no longer responding to other available treatments. Yet, they responded well to pixantrone.
But is that really true? The median age of pixantrone patients enrolled in the study was 60, in the comparator arm it was 58. Patients with aggressive NHL under the age of 65 often undergo stem-cell transplants as part of the normal course of therapy. If the EXTEND patients were, in fact, the sickest of the sick, you'd expect the rate of stem-cell transplantation to be relatively high.