Age-Related Differences in Disease Characteristics and Clinical Outcomes in Polycythemia Vera

Advanced age is an important prognostic indicator in Polycythemia Vera (PV), since this variable influences thrombosis risk as well as myelofibrotic and leukemic transformation. Whether younger patients are at less risk for these complications is unknown, but longer disease duration may offer more opportunity to experience these cardinal complications. To address this question, we retrospectively analyzed 134 PV patients, comparing those diagnosed before age 45 (N=70), to those diagnosed after age 65 (N=64), from the Johns Hopkins Center for Myeloproliferative Disorders.

The parent cohort consisted of 223 PV patients; those diagnosed between the ages of 46 and 64 were excluded. Clinical and demographic variables were ascertained at the last intake visit, including disease duration, spleen size, blood counts, JAK2 V617F allele burden as well as a history of vascular complications, myelofibrotic and leukemic transformation. The Mann-Whitney Rank Sum test was used to test age group differences for continuous variables; testing for age group differences in proportions was performed using the z test, or Fisher's exact test.

PV in the young significantly differed from PV in the old with respect to gender (83% women in young vs. 55% in the old, p=0.001), and median disease duration at last follow-up (8 years in the young vs. 4 years in the old; p=<0.001). The proportion of patients with a Myeloproliferative Neoplasm (MPN) family history did not differ between younger and older patients (13% vs. 9%; p=0.80), nor did a history of antecedent ET (21.4 vs. 21.9%). Both groups were predominantly JAK2 V617F-positive (98.6 and 98.4%), but the younger group had a significantly lower median JAK2 V617F allele burden (52% vs. 66%; p=0.03). The younger and older cohorts did not differ with respect to hydroxyurea use (36% vs. 45%; p=0.19), but younger patients were more likely to have received interferon (17% vs. 3.3%; p=0.04), and marginally less likely to take aspirin (32% vs. 50%; p=0.05).

The proportion with splenomegaly was greater in the younger cohort (50% vs. 25.8%, p=0.007), but the leukocyte count was significantly lower when compared to the older cohort (9.2 × 10⁹/L vs. 14.3 × 10⁹/L; p=0.001). The proportion of patients with a vascular complication did not differ between the younger and older cohorts (30.4% vs. 30.2%, p=0.99); however, 6 patients in the younger cohort had multiple events. Overall, a dependence between age group and the type of vascular complication was identified (p=0.009); venous events were more common in younger patients vs. older patients (19/29 (65.5%) vs. 4/19 (21%); p=0.0025), the majority of which involved the abdominal venous system (11 vs. 2 events). Acute coronary syndromes (5 events vs. 1 event) and transient ischemic attacks (8 events vs. 3 events) were more common in the older cohort.

The proportion developing post-PV myelofibrosis, usually marked by the onset of anemia, was also similar by age group (14.3% (young) vs. 8%; p=0.28), but the median time to transformation was longer in the young (17 years vs. 4 years; p=0.04). AML transformation occurred in 1 young cohort patient, at 28 years from diagnosis, and in 4 older cohort patients, at a median of 6 years from diagnosis. There were 6 deaths (8.6%) in the younger cohort (1 AML, 1 CNS hemorrhage, 3 considered to be disease-related, 1 unknown), and 6 deaths (9.4%) in the older cohort (3 AML, 2 unknown, 1 renal failure).

In conclusion, in one of the larger retrospective studies of PV patients diagnosed before age 45, this group was predominantly female and had a comparable history of vascular complications, though events more frequently involved the abdominal venous system compared to patients diagnosed after age 65. The proportion with myelofibrotic transformation was similar in the two groups, but occurred after a longer disease course in the younger cohort. Leukemic transformation was rare, and of comparable proportions, but contributed to more deaths in the older age group. Thus, younger patients do not appear to be protected from cardinal MPN complications, possibly due to a longer disease duration, which offers more opportunity for time-dependent complications. If disease-modifying effects of pegylated interferon and JAK-inhibitors are identified, then these agents might also be considered in young PV patients to offset time-dependent MPN complications.

No relevant conflicts of interest to declare.