Evidence-based management of anemia in cancer patients: Validity of RESPOND, a web-based clinical guidance system based on the EORTC guidelines

Abstract

20573 Background: The EORTC has published evidence-based guidelines for the management of anemia using erythropoiesis- stimulating agents (ESA). We previously described background/validation methodology (Van Erps et al, ASCO 2007 abstr 19633) and content validity results (Foubert et al, ASCO 2007 abstr 19676) for the RESPOND system. Concurrent, discriminant, and predictive validity are reported here. Methods: 68 patients (2x34) matched by gender, cancer type, and treatment, and managed prior to (pre-cohort) and after introduction of the EORTC guidelines (post-cohort). Endpoint: congruence score (CS) with EORTC guidelines. Results: See Table 1 for CS results. Mean CS for the pre-cohort was 3.0 (SD=1.5; 95% CI 2.5–3.5), for the post-cohort 8.2 (SD=1.4, 95% CI 7.7–8.7). Concurrent validity was inferred from the high mean CS for post-cohort. Discriminant validity was inferred from the statistically significant difference between M±SD CSs for both cohorts (p<.00001) and higher likelihood of the post-cohort to have Hb ≥11g/dL (OR=3.6, 95% CI 1.1–11.8) and ≥12g/dL (OR=2.9, 95% CI 1.1–8.1). Predictive validity was inferred from significant Hb differences at each of the 4 visits and HbΔ from visit 1 to 4 (all at least p<.03), interaction effect for time and cohort (p=.006), regression slope coefficient of 0.3g/dL HbΔ increase for every 1 CS point increase, and CS predicting likelihood of Hb ≥11g/dL (OR=1.5, 95% CI 1.2–2.0) and ≥12g/dL (OR=1.5, 95% CI 1.2–1.8). Conclusions: RESPOND has now been shown to have content, concurrent, discriminant, and predictive validity. This validation study identified areas where web-based clinical guidance can improve guideline-adherent anemia management: ruling out or treating other causes of anemia, initiating ESA treatment, setting Hb target range, fixed initial dosing per ESA evidence, and achieving target range at 4–8wks; and where good practice is reinforced in terms of managing starting Hb levels at <9g/dL or between 11.0–11.9g/dL. Congruence scores (0-10) for pre-cohort and post- cohort Rule out/treat iron deficiency, bleeding, nutritional deficits, hemolysis 0.25 each 0% 85–100% p<.0001 ESA Rx initiated at Hb 9–11g/dL 1 6% 100% p<.0001 If Hb<9g/dL: patient considered for blood transfusion; or: not applicable 1 80% 88% p=ns If Hb 11.0–11.9g/dL: pt individually considered for ESA Rx; or: not applicable 1 100% 100% p=ns Target Hb range set at 12–13g/dL 1 0% 100% p<.0001 Initial ESA dose: 40K alfa Q1W; 30K beta Q1W; darbepoetin Q1W or Q3W 1 41% 100% p<.0001 Initial ESA dose fixed 1 3% 100% p<.0001 Target achieved at 4–8wks; if not: individualized dose escalation 1 21% 53% p=.006 Hb maintained at 12–13g/dL up to 8 weeks .125/wk 21% 38% p=ns ESA discontinued if Hb≥13g/dL (or 12–23g/dL at end of study) 1 29% 50% p=ns Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration Roche Roche Roche