**Background:** Erythropoiesis-stimulating agent (ESA) hyporesponsiveness affects 10–15% of chronic dialysis patients and is associated with higher mortality and poor clinical outcomes. While several predictors have been identified—including iron deficiency, inflammation, and nutritional deficiencies—the influence of demographic, socioeconomic, and geographic factors is poorly understood. Hypoxia-inducible factor prolyl hydroxylase inhibitors (HIF-PHIs) represent a new class of drugs that may be effective in ESA-hyporesponsive patients by upregulating erythropoietin and improving iron delivery. This study investigated baseline characteristics and predictors of ESA hyporesponsiveness in the global ASCEND-D trial, which compared daprodustat (an HIF-PHI) with conventional ESA treatment.
**Methods:** ASCEND-D (NCT02879305) recruited 2,964 chronic dialysis patients from September 2016 to June 2018 across four geographic regions: North America (NAm), Latin America (LA), Europe/Middle East/Africa (EMEA), and Asia Pacific (APAC). All patients had been on dialysis for >90 days, were receiving maintenance ESA, and were iron replete at baseline (ferritin >100 ng/mL, TSAT >20%). The primary definition of ESA hyporesponsiveness was an ESA Resistance Index (ERI) ≥2 U/kg/week/g/L or a standardized ESA dose ≥450 U/kg/week. An alternate definition (ERI ≥1.5 U/kg/week/g/L) was also explored. Baseline ERI was calculated as dry weight-adjusted standardized weekly ESA dose (U/kg/week) divided by hemoglobin (g/L). Predictors were identified using stepwise logistic regression (p<0.10 to enter, p<0.20 to stay). Baseline parameters included in the model were region, BMI, TSAT, age, albumin, IV iron dose, sex, heart failure history, dialysis vintage, smoking status, aspirin use, and ACEi/ARB use. ESA dose, Hb, and dry weight were excluded as they were part of the hyporesponder definition.
**Key Results:** Of 2,964 patients, 354 (12%) were ESA hyporesponsive by the primary definition. ESA hyporesponsive patients had lower Hb (median 10.0 vs. 10.5 g/dL), higher hsCRP (6.0 vs. 3.8 mg/L), lower TSAT (29% vs. 33%), and lower albumin (3.8 vs. 3.9 g/dL) compared to responsive patients. No difference in ferritin or hepcidin was observed. Geographic region was the strongest predictor (p<0.0001). Compared to NAm, patients in EMEA were less likely to be hyporesponsive (OR 0.32, 95% CI 0.23–0.44), while LA patients were more likely (OR 1.81, 95% CI 1.30–2.51). APAC was similar to NAm (OR 0.99, 95% CI 0.68–1.42). The proportion of ESA hyporesponsive patients varied dramatically by region: LA 24% (103/426), APAC 16% (63/385), NAm 13% (112/857), and EMEA 6% (76/1,296). Country-level analysis showed the highest rates in India (30%), Mexico (27%), and Brazil (26%), while several European countries had 0% hyporesponsiveness (France, Netherlands, Norway, Poland, Romania, Sweden). Other significant predictors included lower BMI (OR 0.93 per kg/m², p<0.0001), lower TSAT (OR 0.98 per %, p<0.0001), younger age (OR 0.98 per year, p<0.0001), lower albumin (OR 0.41 per g/dL, p<0.0001), higher IV iron dose (OR 1.05 per 50 mg/month, p=0.0004), female sex (OR 1.38, p=0.010), and history of heart failure (OR 0.74, p=0.035). Longer dialysis vintage showed moderate association (p=0.077). Using the alternate definition, 564 (20%) patients were hyporesponsive, with region again the strongest predictor.
**Clinical Implications:** This is the first global HIF-PHI study to report prespecified definitions and predictors of ESA hyporesponsiveness. The striking regional variation—particularly the 4-fold difference between LA (24%) and EMEA (6%)—is a novel finding that merits further investigation. Potential explanations include differences in ESA type (all LA patients used short-acting epoetin, possibly including less effective biosimilars), variations in medical management, healthcare access, and reimbursement policies. The finding that lower TSAT predicted hyporesponsiveness despite protocol-mandated iron repletion suggests that conventional TSAT targets may be insufficient for optimal erythropoiesis, consistent with the PIVOTAL trial. The study's strengths include its large sample size, global representation, prespecified definitions, and inclusion of hepcidin measurements. Limitations include the restrictive eligibility criteria (iron replete at baseline), potential false-positive findings, and the observational nature of associations. These findings highlight the need for further research into geographic and practice-pattern factors contributing to ESA hyporesponsiveness.