Observational Data Highlight Factors for Polycythemia Vera Progression

"Mean [standard deviation] JAK2 p. V617F VAF was higher in patients with vs without PV progression," according to the study authors.

Factors that appear to significantly correlate with an increased disease progression risk in polycythemia vera (PV) included increased white blood cell (WBC) counts and variant allele frequency (VAF), according to data from the REVEAL study (NCT02252159) presented during the 2024 European Hematology Association (EHA) Congress.

“Five predictors of PV progression were identified: disease duration, thrombotic event [TE] history, WBC count of greater than 11 × 109/L, hematocrit [HCT] level of 0.45 L/L or lower, and VAF. However, HCT [level] of 0.45 L/L or lower may be confounded by disease duration and cytoreductive treatment covariates. These results provide additional support for the use of disease duration and elevated WBC and VAF as risk factors for disease progression, and identify history of TEs as a potential novel risk factor,” Michael R. Grunwald, MD, and coauthors wrote in a poster presentation of the findings. Grunwald is chief of the Leukemia Division at Atrium Health’s Levine Cancer Institute and director of the Transplantation and Cellular Therapy Program at Levine Cancer Institute in Charlotte, North Carolina.

At a median follow-up of 3.7 years, findings from REVEAL, the largest prospective, observational clinical study in patients with PV to date (n = 2023), showed that 6.7% of patients progressed to myelofibrosis (MF). Results from a univariate analysis of patients with vs without progression revealed that significant covariates consisted of time from PV diagnosis to enrollment (OR, 1.065; 95% CI, 1.040-1.090; P < .0001), history of TEs (yes vs no; OR, 1.722; 95% CI, 1.170-2.534; P = .0059), HCT levels of 0.45 L/L or lower (>0.45 vs ≤0.45 L/L; OR, 0.593; 95% CI, 0.410-0.858; P = .0056), and white blood cell (WBC) count of greater than 11 × 109/L at enrollment (>11 vs ≤11 × 109/L; OR, 2.053; 95% CI, 1.445-2.918; P < .0001).

Investigators noted that each covariate remained significant in multivariate analyses with step wise model selection, including time from PV diagnosis to enrollment (OR, 1.059; 95% CI, 1.032-1.086; P < .0001), history of TEs (OR, 1.597; 95% CI, 1.062-2.401; P = .0246), HCT levels (OR, 0.579; 95% CI, 0.392-0.857; P = .0062), and WBC count (OR, 2.144; 95% CI, 1.475-3.117; P < .0001).

“At enrollment, patients with HCT ≤0.45 L/L vs >0.45 L/L had a significantly longer duration from PV diagnosis, had higher PV risk, and were more likely to be receiving hydroxyurea or other cytoreductive treatment, which potentially confounds the HCT ≤0.45 L/L covariate. These confounders were not observed in patients with WBC >11×109/L vs ≤11×109/L, disease duration, TE history, WBC count of greater than 11 × 109/L, HCT of 0.45 L/L or lower, and VAF,” Grunwald and coauthors noted.

The real-world, multicenter, REVEAL study examined data from patients treated in US community and academic centers between 2014 and 2019. Of the 2510 patients included in the study, 2023 received a confirmed PV diagnosis; of these patients, 1524 had JAK2 p. V617F-positive disease confirmed by droplet digital polymerase chain reaction, 365 had a physician-reported JAK2 mutation, and 134 had PV confirmed by bone marrow aspirate/biopsy.

Among all patients with a confirmed diagnosis of PV, the median age at enrollment was 68.0 years (range, 22-95); most patients were male (51.2%), White (89.1%), and treated in a community practice setting (82.6%). Patients who experienced disease progression (n = 135) experienced a median time from diagnosis to enrollment of 7.2 years (range, 0.0-34.1) and the median follow-up for these patients was 3.4 years (range, 0.0-4.9). Patients who did not experience disease progression (n = 1888) had a median time from diagnosis to enrollment of 3.7 years (range, 0.0-38.5) and the median follow-up for these patients was 3.7 years (range, 0.0-5.0).

Among patients with disease progression vs without, PV risk at enrollment was high in a majority of patients (85.9% vs 78.7%). PV treatments received at enrollment were watchful waiting (6.7% vs 5.5%), phlebotomy (17.8% vs 27.8%), hydroxyurea (38.5% vs 31.1%), hydroxyurea plus phlebotomy (17.8% vs 27.3%), or other (19.3% vs 8.4%), respectively. Further, 29.6% vs 19.7% of patients had a history of TEs at enrollment, respectively.

Additional findings from a subset analysis revealed that of patients with quantitative allele burden data available, VAF was a significant covariate (P < .0001), and VAF remained a significant covariate in multivariate analyses with stepwise model. WBC count was excluded from the multivariate analysis models because VAF and VWBC count of greater than 11 vs less than or equal to 11 × 109/L are confounded variables.

“Mean [standard deviation] JAK2 p. V617F VAF was higher in patients with vs without PV progression [P < .0001],” investigators also noted.

Reference

Grunwald M, Zwicker J, Gerds A, et al. A real-world evaluation of risk factors for disease progression in patients with polycythemia vera (PV) enrolled in REVEAL. Presented at: 2024 EHA Congress; June 13-16, 2024; Madrid, Spain. Abstract P1047.