Young Versus Old Age at Diagnosis Confers Distinct Genomic Profiles in Patients with Polycythemia Vera

Introduction: Polycythemia vera (PV) is a myeloproliferative neoplasm (MPN) characterized by overproduction of mature erythrocytes and a propensity for transformation to myelofibrosis (MF) and acute myeloid leukemia (AML). The defining genetic feature of PV is the JAK2 V617F mutation, present in 95-97% of patients, resulting in overactive JAK-STAT signaling. Secondary mutations have been identified which may modulate PV disease phenotype. Although typically diagnosed in older individuals, PV occurs in young patients (≤ 45 years old) as well. Compared with old patients (≥ 65 years old), a significant association with young PV and female gender, lower white blood cell count, lower JAK2 V617F allele burden, palpable splenomegaly, and splanchnic vein thrombosis has been reported. In the same study, the frequency of transformation to MF or AML was found to be similar, although the median time to transformation was shorter in old PV. The reasons for these age-related associations are not clearly understood. Since somatic mutations have been shown to accumulate with age, we hypothesized that the mutational profile of young PV patients may differ from that of old PV, which could contribute to the differences in clinical phenotypes observed.

Methods: DNA from tumor (bulk peripheral blood mononuclear cells or granulocytes) and matched normal tissue (skin or sorted CD3+ T cells) was isolated from samples from 10 young (≤ 45 years old) and 11 old (≥ 65 years old) PV patients. Enhanced exome sequencing utilizing the standard IDT exome capture panel plus additional probes for genes known to be recurrently mutated in MPN and/or AML was performed. To determine order of mutation acquisition (and to rule out the possibility of biclonal disease) qPCR-based genotyping was performed on single cell-derived colonies from patients with ≥ 1 secondary mutation (in addition to JAK2 V617F). Plasma concentrations of 10 cytokines were assayed in 15 young and 23 old PV patient samples compared with age-matched healthy controls.

Results: Comparative analysis of 7 young and 10 old patients (four samples were removed due to sample contamination/artifact issues) identified a total of 133 somatic mutations. A significantly higher overall mutational load was observed in the old group (mean 10.9 vs 3.4; p = 0.0025). A trend toward higher JAK2 V617F variant allele frequency (VAF) was observed in old versus young PV in the sequenced patients, and in a separate clinical cohort the JAK2 V617F VAF was significantly higher in old PV patients (p = 0.0067). Notably, putative secondary driver mutations were identified in 9/10 old PV patients (including ASXL1, TET2, and DNMT3A), whereas JAK2 appeared to be the only driver mutation present in each of the young PV patients. Secondary mutations were validated as being acquired before, after, or coincident with JAK2 V617F, with no specific predominant pattern observed. MPN-predisposing germline JAK2 46/1 haplotype occurrence did not differ with age, and additional germline mutations in MPN genes were not enriched in either cohort. Supranormal levels of multiple cytokines (e.g. IFNγ, IL-2Rα, IP-10, MIP-1α) were observed in both young and old PV, with a slight enhancement observed with increased age.

Conclusions: These data indicate a striking age-based distinction in the near uniform absence (young) or presence (old) of secondary non-JAK2 mutations in PV patients. The finding that putative secondary driver mutations were acquired before JAK2 V617F in only a minority of old PV suggests that the overall increase of secondary mutations in this cohort may not merely be a function of aging, implying that these mutations do in fact contribute to PV development and/or progression. As transformation can be induced by accrual of disease-modifying mutations, the presence of secondary mutations in old PV patients is consistent with the reported shorter time to transformation in this age group, since young PV patients would likely need additional time to acquire secondary mutations. Abnormal inflammatory cytokine production was common to both young and old PV, suggesting this is not the predominant factor distinguishing these two cohorts. The contribution of additional non-genomic factors (e.g. epigenetic, hormonal) will be the subject of future study.