Current practice guidelines for the management of chemotherapy-induced anemia (CIA) recommend initiating erythropoietic intervention at hemoglobin (Hb) ≤ 10 g/dL (American Society of Hematology [ASH]/American Society of Clinical Oncology [ASCO]), or at Hb ≤ 11 g/dL (National Comprehensive Cancer Network [NCCN]). This comprehensive literature review summarizes published findings of randomized controlled trials (RCTs) that evaluated the effects of initiating erythropoietic treatment at Hb > 10 g/dL on transfusion incidence, Hb concentration, and/or patient-reported outcomes, relative to control. A search of the Medline database and conference proceedings (ASH and ASCO) from 1999 to 15 June 2004 was conducted to identify RCTs that evaluated the ability of erythropoietic agents to prevent the onset or worsening of CIA. Mantel-Haenszel weighted summary estimates of the relative risk were calculated to evaluate outcomes in patients treated with epoetin alfa compared to control patients. Using predefined selection criteria, 4 full-length papers (summarized in the table) and 5 abstracts (Chang ASCO 2003; Savonije ASCO 2004; Straus ASH 2003; Crawford ASCO 2003; Rearden ASCO 2004) were identified and reviewed. Epoetin alfa effectively decreased transfusion incidence and the percentage of patients with Hb decline to < 10 g/dL vs no treatment in 4 RCTs reported from 1997 onward (table). The estimated summary relative risk (95% CI) for transfusion and Hb < 10 g/dL are 0.39 (0.26–0.57; P < 0.0001) and 0.48 (0.25–0.89), respectively. These results were consistent with early results reported in 2 additional RCTs (Chang ASCO 2003; Savonije ASCO 2004), which were of similar design. The preliminary results of 3 additional RCTs that directly compared outcomes following early (Hb > 10 g/dL) and late (Hb ≤ 10 g/dL) erythropoietic intervention were presented at recent ASH and ASCO conferences (Straus ASH 2003; Crawford ASCO 2003; Rearden ASCO 2004). All 3 studies provided evidence of lower incidence of transfusion, higher Hb levels over time, and/or better patient-reported outcomes among patients treated early compared with those treated late. The results of the studies examined demonstrate the clinical benefits of initiating erythropoietic treatment at a Hb concentration > 10 g/dL, with respect to reducing transfusion requirements and improving Hb levels, in patients with cancer undergoing chemotherapy. Although the findings are preliminary and have yet to appear in peer-reviewed journals, the comparative RCTs of early vs late intervention point to a trend in which patients who received erythropoietic treatment early experienced better clinical outcomes overall than patients treated late.

Bamias 2003Thatcher 1999ten Bokkel Huinink 1998Del Mastro 1997
E=epoetin alfa; C=no treatment control; reported p < 0.05 (*), 0.01 (**), 0.001 (***) vs control. ªHb ≤10 g/dL in Del Mastro study 
Hb Eligibility Criterion (g/dL) ≤13 ≥10.5 < 13 ≥ 12 
Baseline Hb (g/dL) 11.5 (E) 13.7 (E) 12.0 (E) 13.0 (E) 
 11.5 (C) 13.4 (C) 11.8 (C) 13.1 (C) 
Transfusion Incidence (%) 15* (E) 45* (E) 4**(E) 0 (E) 
 33 (C) 59 (C) 39 (C) 6 (C) 
% Patients with Hb < 10 g/dLª 17*** (E) 48* (E) 18* (E) 0*** (E) 
 46 (C) 66 (C) 50 (C) 52 (C) 
Dosing Regimen 10,000 U TIW 150 U/kg TIW 150 U/kg TIW 150 U/kg TIW 
 No treatment (control) No treatment (control) No treatment (control) No treatment (control) 
Bamias 2003Thatcher 1999ten Bokkel Huinink 1998Del Mastro 1997
E=epoetin alfa; C=no treatment control; reported p < 0.05 (*), 0.01 (**), 0.001 (***) vs control. ªHb ≤10 g/dL in Del Mastro study 
Hb Eligibility Criterion (g/dL) ≤13 ≥10.5 < 13 ≥ 12 
Baseline Hb (g/dL) 11.5 (E) 13.7 (E) 12.0 (E) 13.0 (E) 
 11.5 (C) 13.4 (C) 11.8 (C) 13.1 (C) 
Transfusion Incidence (%) 15* (E) 45* (E) 4**(E) 0 (E) 
 33 (C) 59 (C) 39 (C) 6 (C) 
% Patients with Hb < 10 g/dLª 17*** (E) 48* (E) 18* (E) 0*** (E) 
 46 (C) 66 (C) 50 (C) 52 (C) 
Dosing Regimen 10,000 U TIW 150 U/kg TIW 150 U/kg TIW 150 U/kg TIW 
 No treatment (control) No treatment (control) No treatment (control) No treatment (control) 

Author notes

Corresponding author

Sign in via your Institution