Association of polycythemia vera treatments with reduced mortality rates among US veterans.

Abstract

e23298 Background: Polycythemia vera (PV) is a myeloproliferative neoplasm characterized by excessive red blood cell production due to genetic mutations, mainly in the Janus kinase 2 ( JAK2 ) gene. Phlebotomy (PHL) is often conducted to manage thromboembolic risk. Cytoreductive treatment (CT) is used for high-risk patients or low-risk patients who need CT. We aimed to examine the impact of treatments on mortality rates among US veterans with PV. Methods: We identified patients using International Classification of Disease (ICD-10) code (D45) in the US Veterans Affairs Healthcare System between Oct 2017 to Dec 2023. Patients were included if they had no PV diagnosis in the prior 2-year cleaning period before the first PV diagnosis date and have either two outpatient codes (separated by 30 or more days) or one inpatient code. The first PV diagnosis date was defined as the index date. Patients with myelofibrosis or acute myeloid leukemia before the index date were excluded. PV treatments were grouped into three categories, CT, PHL only, and no treatment as the reference group. CT included hydroxyurea, ruxolitinib, peginterferon, ropeginterferon, and interferon-a. PHL was identified using the Current Procedural Terminology (CPT) code. Using the adjusted Cox proportional hazard model, we examined the association between PV treatments and risk of mortality from the index date. The Covariates in the Cox model included age, sex, race, ethnicity, and the following factors prior to the index date: thrombosis and cancer, phlebotomy, risk of thrombosis (age > = 60 or had prior thrombosis event), and medications. Results: We identified 11,809 qualified PV patients per criteria and observed 3,066 deaths, equivalent to a 26.0% mortality rate. The average age of the 11,809 patients was 67.85 (Median = 69; standard deviation = 11.18). The majority were male (n = 11,394, 96.5%), 9,488 (80.7%) were white, and 9,205 (77.9%) were at high-risk. Of the patients, 25.14% (n = 2,969) had CT, 12.15% (n = 1,435) had PHL only, and 62.70% (n = 7,405) did not receive any PV treatments. After controlling for potential confounders, patients undergoing CT and PHL had a 39.6% (HR = 0.604, 95%CI = 0.540-0.675) and 23.4% lower risk of mortality (HR = 0.766, 95%CI = 0.669 - 0.878), respectfully, compared to patients without treatment. Nevertheless, the high-risk PV (HR = 1.368, 95% CI = 1.037 – 1.803), native Hawaiian or pacific islander race (HR = 1.585, 95%CI = 1.167 - 2.151; white as the reference group), and prior cancer history (HR = 1.375, 95%CI = 1.265 – 1.494) were associated with an increased risk of mortality. Conclusions: Patients with PV who received CT or PHL were associated with a reduction in mortality and CT showed the most significant effect. Considering the unique characteristics in patient demographics, comorbidities, and healthcare resources among veterans, a well-defined patient treatment plan is the key to improve clinical outcomes in patients with PV.