TREAT

Background

Phase III, double-blind, placebo-controlled RCT. 4,038 patients with type 2 diabetes, chronic kidney disease (CKD stages 3–4, eGFR 15–59), and anemia (hemoglobin 9.0–11.0 g/dL). Patients not on dialysis. Target Hb ~13 g/dL in darbepoetin arm vs placebo with rescue at Hb <9.0 g/dL. Two co-primary composite endpoints assessed.

Interventions and follow up

Arm A: Darbepoetin alfa dosed to target hemoglobin ~13 g/dL
Arm B: Placebo (rescue darbepoetin if Hb <9.0 g/dL)
Primary endpoint: (1) Death or cardiovascular event (nonfatal MI, CHF, stroke, hospitalization for ischemia); (2) Death or ESRD
mFollow up: Median 29.1 month

Results

CV composite (death or CV event): 632 vs 602, HR 1.05, 95% CI 0.94–1.17, P=.41 — not significant
Renal composite (death or ESRD): 652 vs 618, HR 1.06, 95% CI 0.95–1.19, P=.29 — not significant
Stroke: 101 vs 53, HR 1.92, 95% CI 1.38–2.68, P<.001
Red cell transfusions: 297 vs 496 patients required transfusions, P<.001

Adverse events

Main adverse events: Stroke occurred in 5.0% vs 2.6% (HR 1.92, P<.001) — the trial's primary safety finding. Cancer-related mortality numerically higher in darbepoetin arm. Hypertension and edema more common. No significant difference in overall mortality.

Conclusions

Targeting Hb ~13 g/dL with darbepoetin in T2DM+CKD patients did not reduce the risk of CV or renal endpoints and was associated with a 92% relative increase in stroke risk. These results, along with CHOIR and CREATE, established the basis for FDA black box warnings on ESAs and the recommendation to use the lowest effective dose.

Key Limitations

Key Limitations: Hb target of 13 g/dL was supraphysiologic for this population — the harm may be target-specific rather than drug-class specific. ESA dose required to maintain this Hb was high, and ESA dose itself may drive CV harm via VEGF-mediated effects or prothrombotic mechanisms. The trial enrolled a high-risk CKD+DM population not generalizable to non-diabetic CKD. Rescue darbepoetin in the placebo arm complicated pure placebo comparison.

Clinical Context

TREAT, combined with CHOIR (Hb 13.5 target) and CREATE (Hb 13-15 target), established that normalizing hemoglobin with ESAs in CKD increases stroke and possibly mortality. Current guidelines recommend ESA use at the lowest dose sufficient to avoid transfusion, targeting Hb 10–11.5 g/dL. ESAs carry FDA black box warnings for cardiovascular events, thrombosis, and mortality when targeting Hb >11 g/dL.

References

References: Pfeffer MA et al, NEJM 2009 (TREAT)