This study represents the first of ten prospective clinical trials in oncology patients, which fails to show a benefit for the addition of intravenous iron to erythropoiesis-stimulating agents in improving hemoglobin responses. The patient population was the largest reported to date (502), the trial was well done, and prospective, multicenter with a placebo control. No difference in hemoglobin response or time to target hemoglobin levels was seen among the intravenous iron, oral iron or placebo arms. Further, the study was closed prematurely due to an imbalance of grade III and IV adverse events in the treatment arm (55%) compared to oral iron (45%) and oral placebo (47%). Why, then, was the outcome so dramatically different from the other published trials?
The most glaring difference from other published trials was the dosing of IV iron. In this trial, 187.5 mg (three vials) of ferric gluconate (Ferrlecit, then Watson Pharma, Morristown, New Jersey, USA, now Sanofi-Aventis, Paris, France) were given every three weeks. This was the lowest weekly dose ever administered and it is possible that this dose is not able to overcome the iron-restricted erythropoiesis that is nearly ubiquitous in the cancer patient. However, a more likely explanation was that the mean total dose was only 650 mg which was 40% less than the lowest mean dose administered in the other trials. Whereas the planned dose of intravenous iron was considerabaly higher, there is no explanation why so many doses of an agent with such little toxicity were held. A planned sub-analysis of these data are planned and is eagerly awaited.
For those of us with clinical practice experience with ferric gluconate, a grade III or IV adverse event rate of 55% is simply not seen. However, in this trial the grade III or IV adverse event rate with placebo was 47%. One can only speculate what the actual adverse event rate was. Premedication with diphenhydramine was allowed at the discrimination of the principal investigators. Patients receiving oral iron or placebo could not have received diphenhydramine and based on published literature it is likely that the increase in adverse events seen with intravenous ferric gluconate was due to the premedication.
These data must be respected for what they are. The design and conduct of this trial was carefully done and executed. Post-hoc stratification of the data based on actual dose of iron received and pretreatment hepcidin levels is anticipated. Based on the quality of this trial, we can expect an exciting update to be forthcoming shortly.
– Michael Auerbach