Abstract BACKGROUND AND AIMS Increased HIF-1 levels are associated with mortality in some cancer patients, and levels of HIF-1 may also impact response to cancer treatments. Use of recombinant human erythropoietin or its analogs [erythropoiesis-stimulating agents (ESA)] in clinical trials in patients with cancer has been associated with increased risk of cancer-related morbidity and mortality, though the mechanism(s) for these effects is not clear. Daprodustat is an oral hypoxia-inducible factor (HIF)-prolyl hydroxylase inhibitor that increases the secretion of endogenous erythropoietin and the production of red cells. Cardiovascular outcome trials in patients with anemia of chronic kidney disease (CKD) requiring dialysis (ASCEND-D [1]; NCT02879305) and those not on dialysis (ASCEND-ND [2]; NCT028768355) have recently been published. Given the long latency with cancer events, we explored the effect of daprodustat as compared with ESA on the risk of cancer development utilizing a post-hoc modified intention-to-treat (mITT) approach, which includes events on and off treatment and provides greater protection to randomization than on-treatment approaches. METHOD Cancer-related adverse events (AEs) from the ASCEND studies were identified based on a predefined list of terms that included new cancer events or tumor progression and recurrence for patients receiving at least one dose of daprodustat or ESA (Safety Population). Patients were included into these trials as long as they had no history of cancer within 2 years prior to screening, were not currently receiving treatment for cancer and did not have a complicated kidney cyst. In addition to reporting the number and percentage of patients with cancer AEs, overall and by types, the rate per 100 person-year (PY) is also provided given differential follow-up for patients in these trials. RESULTS A total of 2964 patients were randomized in ASCEND-D and 3872 in ASCEND-ND, with ∼14 200 PY of follow-up. At baseline, patients reported a past history of cancer in 5.0% and 4.9% of patients receiving daprodustat or ESA, respectively, in ASCEND-D; and in 5.2% and 4.4% of patients receiving daprodustat or darbepoetin alfa, respectively, in ASCEND-ND. In the Safety Population, both the rate of cancer-related AEs and the rate per 100 PY were balanced for each study, with no pattern emerging regarding location or type (Table 1). Of the patients with overall cancer AEs in the Safety Population (Table 1), fatal AEs in ASCEND-D occurred in 11 patients receiving daprodustat and in 15 patients receiving ESA, which was <1% and 1% of the Safety Population, respectively. In ASCEND-ND, fatal AEs occurred in 6 patients receiving daprodustat and 11 patients receiving darbepoetin alfa, which was <1% of the Safety Population in both groups. Pooling of data from the D and ND studies also demonstrated balance in cancer events between the dapro and ESA arms. Further analysis of the characteristics of patients with cancer AEs demonstrated that the number of patients with multiple events, the time to first onset and outcome of the event and event seriousness and severity were also similar between treatment groups for both trials. CONCLUSION In cardiovascular outcome trials comprised of patients requiring dialysis (ASCEND-D) and not requiring dialysis (ASCEND-ND) with anemia of CKD, daprodustat was not associated with an increased risk of cancer, or cancer mortality, relative to ESA. Collectively, across these trials, cancer AEs are balanced.