Introduction: Polycythemia vera (PV) is a primary etiology of erythrocytosis most often caused by mutations in JAK2. Other primary causes of erythrocytosis include congenital erythrocytosis caused by EPOR mutations and Chuvash polycythemia caused by mutations in VHL (Von-Hippel Lindau). PV is associated with increased thrombotic risk, but the clinical implications of rarer mutations remain unclear. The advent of myeloid next generation sequencing (NGS) has expanded the landscape of detectable mutations in patients investigated for suspected myeloproliferative neoplasms. SH2B3 encodes a lymphocyte-specific adaptor protein (LNK) which acts as a negative regulator of hematopoiesis through inhibition of JAK-STAT signaling. It is hypothesized that mutations in SH2B3 mayattenuate LNK production and cause hematopoietic dysregulation, contributing to a clinical phenotype of erythrocytosis and increased thrombotic risk. To elucidate the clinical impact of mutations in SH2B3,we conducted a single-centre retrospective cohort study to characterize patients with erythrocytosis and mutations in SH2B3, comparing clinical presentation and thrombotic outcomes to patients with JAK2-mutated PV and JAK2- and SH2B3-wild type erythrocytosis.

Methods: This single-centre retrospective cohort study included adult patients (≥18 years) evaluated for erythrocytosis (hemoglobin ≥160 g/L in women, ≥165 g/L in men) with myeloid NGS panel (Oncomine Myeloid Research Assay, ThermoFisher Scientific, Waltham, MA) at a tertiary centre in Southwestern Ontario, Canada. Patients were divided into three groups: erythrocytosis associated with mutations in SH2B3 (SH2B3-mutated erythrocytosis), JAK2-mutated PV, and secondary erythrocytosis (SE, JAK2- and SH2B3-wild type). Patients with hematologic malignancies other than PV were excluded. Data on clinical characteristics, secondary causes of erythrocytosis, hematologic treatments (phlebotomy, cytoreduction, antithrombotic therapy), venous/arterial thrombosis, and mortality were extracted.

Results: We identified 16 patients with SH2B3-mutated erythrocytosis, 103 with PV, and 566 patients with SE. Baseline hemoglobin and hematocrit at diagnosis were similar between SH2B3-mutated, SE, and PV groups (176g/L, range 160-217; 0.53L/L, range 0.48-0.66 vs. 175g/L, range 160-246; 0.52L/L, range 0.45-0.74, vs. 176g/L range 157-230; 0.53L/L, range 0.48-0.66). Median age was similar between patients with SH2B3-mutated erythrocytosis and SE (59 years, range 24-79, vs. 59 years, range 18-91), whereas patients with PV were older (median age 72 years, range 29-97). Median follow-up duration in SH2B3-mutated, SE, and PV groups was similar (49.6 months, range 27-83 vs. 54.7 months, range 21-85 vs. 49.6 months, range 21-84). Seventy-five percent (n=12) of SH2B3-mutated patients were on either aspirin or an anticoagulant, compared to 48% (n=270) of SE and 97% (n=100) of PV patients. No arterial thrombotic events were observed in patients with SH2B3-mutated erythrocytosis. Rates of arterial thrombosis in the PV and SE groups were 1.9% (n=2) and 4.6% (n=26), respectively. Venous thrombosis was observed in one patient with SH2B3-mutated erythrocytosis. Rates of venous thrombosis in the PV and SE groups were 1.9% (n=2) and 3.0% (n=17), respectively. Death occurred in one patient with SH2B3-mutated erythrocytosis. All cause mortality was 10.7% (n=11) and 3.5% (n=20) in PV and SE groups, respectively.

Discussion: This single-centre retrospective cohort study describes a small subset of patients with SH2B3-mutated erythrocytosis, representing a molecularly distinct group with uncertain clinical implications. Thrombotic events were infrequent in this group, with one case of venous thrombosis and no arterial events observed. However, the small sample size limits the ability to draw definitive conclusions about thrombotic risk or other clinical outcomes in comparison to JAK2-mutated PV and SE.

The identification of SH2B3 mutations in patients with erythrocytosis underscores the evolving molecular landscape of erythrocytosis and the potential role of SH2B3 as a contributing factor to this clinical presentation. While the clinical impact of SH2B3 mutations remains unclear, their presence in a subset of patients with erythrocytosis warrants further investigation. These preliminary findings support the need for larger cohort studies with longer follow-up to better define the clinical phenotype associated with SH2B3-mutated erythrocytosis.

This content is only available as a PDF.
2025
Sign in via your Institution