Hydralazine Associated With Reduced Therapeutic Phlebotomy Frequency in a Nationwide Cohort Study: Real-World...

Background: Therapeutic phlebotomy, known as scheduled bloodletting, has been the main method for managing erythrocytosis symptoms and thrombocytosis-associated complications in various blood disorders. One of the major indications for phlebotomy is polycythemia vera (PV). The main goal of current treatment strategies for patients who require phlebotomy is to prevent thrombohemorrhagic complications rather than to prolong survival or lessen the risk of myelofibrotic or leukemic progression. Additional cytoreductive therapy is recommended for high-risk PV, for which the common first-line drug is hydroxyurea. However, recent evidence suggests that phlebotomy may not reduce the risk of thrombosis in patients with PV. Further evidence suggests that patients with PV treated with hydroxyurea who require three or more phlebotomy procedures per year have a higher risk of thrombotic complications. Methods: We hypothesized that a drug-repurposing strategy of utilizing antineoplastic drugs for patients who require phlebotomy would result in greater benefits than would phlebotomy. The antihypertensive hydralazine and the anticonvulsant valproate, which have both been reported to have antineoplastic activity that mimics cytoreductive agents, were selected as candidates for the drug-repositioning strategy in a retrospective cohort study. We measured the hazard ratios (HR) and the frequencies of phlebotomy in patients with prescriptions for hydralazine or valproate or the two drugs in combination by using data from Taiwan's National Health Insurance Research Database from 2000 to 2015 (n = 1,936,512). Results: The HRs of undergoing phlebotomy in groups with hydralazine, valproate, and combination hydralazine-valproate prescriptions were reduced to 0.729 (p = 0.047), 0.887 (p = 0.196), and 0.621 (p = 0.022), respectively. The frequency of undergoing phlebotomy decreased from 2.27 to 1.99, 2.01, and 1.86 per person-year (p = 0.015), respectively. However, no significant differences were observed for the hydralazine group or the hydralazine-valproate combination group. Conclusion: Whether a repurposed drug can serve as a cytoreductive agent for patients who require phlebotomy depends on its risk-benefit balance. We suggest that hydralazine, instead of the hydralazine-valproate combination, is a reasonable alternative for patients who require regular phlebotomy.