Efficacy of intravenous iron treatment for chemotherapy-induced anemia: A prospective Phase II pilot clinical trial in South Korea

Abstract

Background: Anemia is the most common and serious cancer-related complication. This study aimed to evaluate the efficacy of administration of ferric carboxymaltose without erythropoiesis-stimulating agents for treating anemia in cancer patients. Moreover, we identified the biomarkers of hemoglobin response to predict the need for iron therapy.

Methods and findings: We enrolled patients with solid cancers who were treated at a single institute (Samsung Medical Center, South Korea), from April 2015 to July 2017, in this prospective single-arm Phase II clinical trial. Patients received intravenous ferric carboxymaltose (1,000 mg) infusion on the first day (visit 1) of treatment. The primary end point was the number of hemoglobin responders, defined as patients with an increase in hemoglobin level ≥ 1.0 g/dL from the baseline, a hemoglobin level ≥ 11.0 g/dL, or both, within an 8-week observation period (week 3, 6, or 8). Secondary end points included changes in transferrin saturation and levels of soluble transferrin receptors, hepcidin, erythropoietin, interleukin-6, and C-reactive protein (CRP) at each visit. Of the 103 recruited patients, 92 were eligible for analysis. The mean patient age was 57.3 ± 12.5 years, and 54.3% of the patients were women. The most common diagnoses were breast cancer (n = 23, 25.1%), lung cancer (n = 21, 22.9%), gastrointestinal cancer (n = 20, 20.9%), and lymphoma (n = 16, 17.7%). A hemoglobin response was observed in 36 (39.1%), 53 (57.6%), and 61 (66.3%) patients in the third, fifth, and eighth weeks, respectively. The mean increase in hemoglobin levels from the baseline to the end of treatment was 1.77 ± 1.30 g/dL. Baseline values of hepcidin (p = 0.008), total iron binding capacity (p = 0.014), ferritin (p = 0.048), and CRP (p = 0.044) were significantly different between the responder and nonresponder groups. Multiple logistic regression analysis for baseline anemia-related biochemical variable significantly associated with the hemoglobin response showed that only baseline hepcidin level was a significant factor for hemoglobin response (odds ratio = 0.95, 95% confidence interval 0.90-1.0, p = 0.045). Hemoglobin responders had significantly lower hepcidin levels than nonresponders (mean [±standard deviation], 13.45 [±14.71] versus 35.22 [±40.470 ng/ml]; p = 0.007). However, our analysis had some limitations such as the different patient characteristics in the studies that were included, institutional differences in the measurement of hepcidin level, and missing data on long-term safety. Therefore, our findings need further validation.

Conclusions: Intravenous ferric carboxymaltose (1,000 mg) monotherapy increases hemoglobin levels without serious adverse events in patients with cancer. Hepcidin is a useful biomarker for predicting iron requirement in cancer patients.

Trial registration: Clinicaltrials.gov NCT02599012.

Fig 1. Patient screening and enrollment.

ITT*, intention-to-treat; PP**, per-protocol; RBC, red blood cell; TSAT, transferrin saturation.

Fig 2. Hemoglobin response according to baseline hepcidin levels and the changes in mean sTfR.

(A) Number of patients who achieved hemoglobin response according to baseline hepcidin levels. (B) Analysis of the changes in mean sTfR levels of responders and nonresponders from baseline to 8W. sTfR, soluble transferrin receptor; W, weeks.