FDA Panel Supports Anemia Drugs for Cancer Patients

[gdpawel]

Increased death rates among cancer patients taking erythropoiesis-stimulating agents were supported in a large meta-analysis.

Among nearly 14,000 patients in 53 studies, those taking the anemia drugs were 17% (95% CI 6% to 30%) more likely to die during the study, and their overall survival chances were reduced by 6% (95% CI 0% to 12%), reported Julia Bohlius, M.D., M.Sc.P.H., of the University of Bern in Switzerland.

The findings, disclosed at the American Society of Hematology meeting, confirmed results reported earlier in individual studies.

The trials included Epogen, Procrit, NeoRecormon and Aranesp. The analysis was notable for relying on full data on each patient, contributed by the trial sponsors and individual investigators, not on outcome data as published.

And in fact, the on-study mortality results were higher than previously published. On-study mortality was defined as death from any cause occurring from randomization to four weeks after the active study's end. Overall survival was measured from randomization to the end of available follow-up.

When the patient pool was limited to about 10,000 patients treated with chemotherapy, the relationship between erythropoiesis-stimulating agents and mortality fell just short of statistical significance.

On-study mortality in the chemotherapy population was increased by 10% for those taking the anemia drugs (95% CI -2% to 24%), and the overall death rate was increased by 4% (95% CI -3% to 11%), the researchers found.

The results changed little when the researchers controlled for known risk factors.

Age, sex, hemoglobin, hematocrit at baseline, type and stage of tumor, and type of study were among the factors researchers took into account.

Patients getting no cancer treatment showed a 33% increase in on-study mortality if they were receiving erythropoiesis agents (95% CI 7% to 67%).

There were even higher increases in patients receiving other forms of cancer treatment (52% for chemoradiation, 52% for radiation alone, and 53% for "other" non-chemotherapies), but these did not reach statistical significance.

Moreover, a test for interaction among all the non-chemotherapy patient groups yielded a P value of 0.42.

Co-author Andreas Engert, M.D., of the University of Cologne in Germany, said the specific reasons for the increased deaths associated with erythropoiesis agents remained unclear, but on the basis of more recent studies, most of the excess deaths are probably from thromboembolic events.

The American Society of Hematology and the American Society of Clinical Oncology would convene a new guideline panel early in 2009 to consider revisions derived from these findings and other developments since the existing version was published.

Since the current guideline was produced, the FDA had required a boxed warning on epoetin and darbopoetin about increased mortality risks for cancer patients.

Source:

Bohlius J, et al., "Recombinant human erythropoiesis stimulating agents in cancer patients: individual patient data meta-analysis on behalf of the EPO IPD Meta-Analysis Collaborative Group" Blood 2008; abstract LBA-6.