Real World Outcomes of Ferric Carboxymaltose and Ferumoxytol Iron Replacement Therapy: A Retrospective Analysis of Electronic Health Records

Introduction:

Iron deficiency anemia is the most common cause of anemia in the US and it is often observed in patients with chronic diseases and inflammation. Although iron deficiency anemia is often treated with oral iron supplementation, parenteral iron replacement therapy is needed for patients who cannot tolerate or are nonresponsive to oral iron, or who have chronic kidney disease. This study examined the effectiveness of two parenteral iron formulations (ferric carboxymaltose [FCM] and ferumoxytol [FM]) in the real world setting.

Methods:

A retrospective cohort analysis was conducted using the QuintilesIMS Real-World Data Electronic Medical Records - US database. Adult patients treated with FCM and FM from August 2014 to June 2016 were identified using Current Procedural Terminology (CPT) codes. Index date was defined as the first date FCM or FM was received. Patients who received any parenteral iron therapy six weeks prior to index date were excluded. Baseline hemogloblin [Hgb] level was determined by the latest Hgb value within the 30-day period before and including the index date. Only patients who had lower than normal Hgb (<12 g/dL [female] or <13.5g/dL [male]) at baseline, and Hgb data recorded in the database during the 3 months post-index were included in the analysis. Generalized linear model was used to compare mean change from baseline Hgb value at 3 months between treatments controlling for baseline Hgb level and gender. Logistic regression was used to test for difference in the proportion of patients achieving normal Hgb level between treatment groups controlling for baseline Hgb level and gender.

Results:

121 FCM patients and 134 FM patients were included in the analysis. Age distributions were similar between FCM (58.7% >65 years) and FM (54.5% >65 years) groups ( P =0.499), but there were more female patients in the FCM group (72.7%) than the FM group (54.5%) ( P =0.003). The FCM group had fewer patients with chronic kidney disease (24.8% vs. 86.6%, p<0.001) and congestive heart failure (2.48% vs. 8.21%, p=0.045), more patients with cancer (28.1% vs. 9.70%, P <0.001) and similar proportion of patients with inflammatory bowel disease (2.48% vs. 0.75%, P =0.266) compared to FM group. The proportions of patients receiving prescription oral iron therapy in the six months prior to index date (24.0% FCM, 18.7% FM, P =0.300) and follow-up (9.09% FCM, 7.46% FM, P =0.637) were similar between treatment groups. Utilization of FCM and FM was similar, with 28.1% FCM and 27.6% FM patients receiving one dose of index iron injection/infusion, 66.9% FCM and 65.7% FM patients receiving two doses, and 4.96% FCM and 6.72% FM patients receiving 3 doses or more during the 3-month follow-up period ( P =0.84). Mean (SD) Hgb at baseline was 9.86 (1.37) g/dL for FCM and 9.67 (1.01) g/dL for FM ( P =0.189). At 3 months post-index, more FCM patients (45.5%) than FM patients (14.9%) achieved Hgb normalization (adjusted p<0.001). Mean (SD) change from baseline Hgb was 2.01 (1.73) for FCM, compared to 1.28 (1.48) for FM (adjusted P -value <0.001).

Conclusion:

This retrospective study of electronic health records provided real world evidence suggesting ferric carboxymaltose may provide better Hgb improvement compared to ferumoxytol in a diverse group of patients with various medical conditions. Future research using a larger sample size that allows for sufficient adjustment of baseline characteristics and potential confounders is needed to confirm current study findings.