Table 5

Summary of findings from studies evaluating thresholds for initiating treatment from the AHRQ CER

StudyRandomly assigned (No.)
MalignancyDrug and treatment durationHb when EPO/DARB initiated
Patients given EPO/DARB (%)
Transfused (%)
Between arm differences in Δ (FACT measures) from baselineThromboembolic events (%)
IDIDIDIDID
Straus 200611  135 134 Hematologic Epoetin ≤ 16 wk 11.1 ± 0.7* <9 100 19.4 17.8 26.1 FACT-An 8.22 (2.62) unadjusted mean, 6.67 (2.25) estimated mean; FACT-fatigue 3.13 (1.16) unadjusted mean, 2.67 (0.94) estimated mean 11.1 3 
Rearden 200412  102 102 Mixed solid or hematologic Darbepoetin 12 wk§ 11.1 ± 0.07 <10 100 62.7 17.2 26.5 NS 
Crawford 200710  109 107 Non–small-cell lung cancer Epoetin 16 wk (D group from first administration to end of study) 13.1 ± 1.0* ≤ 10 100 46.0 12.3 20.0 NS 12 15 
StudyRandomly assigned (No.)
MalignancyDrug and treatment durationHb when EPO/DARB initiated
Patients given EPO/DARB (%)
Transfused (%)
Between arm differences in Δ (FACT measures) from baselineThromboembolic events (%)
IDIDIDIDID
Straus 200611  135 134 Hematologic Epoetin ≤ 16 wk 11.1 ± 0.7* <9 100 19.4 17.8 26.1 FACT-An 8.22 (2.62) unadjusted mean, 6.67 (2.25) estimated mean; FACT-fatigue 3.13 (1.16) unadjusted mean, 2.67 (0.94) estimated mean 11.1 3 
Rearden 200412  102 102 Mixed solid or hematologic Darbepoetin 12 wk§ 11.1 ± 0.07 <10 100 62.7 17.2 26.5 NS 
Crawford 200710  109 107 Non–small-cell lung cancer Epoetin 16 wk (D group from first administration to end of study) 13.1 ± 1.0* ≤ 10 100 46.0 12.3 20.0 NS 12 15 

Adapted from original work done by Seidenfeld et al.

Abbreviations: AHRQ, Agency for Healthcare Research and Quality; CER, comparative effectiveness review; I, erythropoietic stimulant therapy begun immediately after random assignment; D, erythropoietic stimulant therapy delayed until Hb falls to threshold; Hb, hemoglobin; DARB, darbepoetin; EPO, epoetin; FACT, Functional Assessment of Cancer Therapy; NS, no significant difference; NR, not reported.

*

Mean ± standard deviation.

Statistically significant differences.

Estimated mean based on the treatment effect using a random coefficient linear growth curve model.

§

Transfusion data include 22 weeks follow-up as patients received chemotherapy throughout.

Mean ± standard error.

Included seven participants who recived EPO despite having Hb > 10 g/dL.

Close Modal

or Create an Account

Close Modal
Close Modal