Abstract
PURPOSE: Essential thrombocythemia (ET) manifests substantial interpatient heterogeneity in rates of thrombosis, hemorrhage, and disease transformation. Bone marrow histology reflects underlying disease activity in ET but many morphological features show poor reproducibility. PATIENTS AND METHODS: We evaluated the clinical significance of bone marrow reticulin, a measure previously shown to have relatively high interobserver reliability, in a large, prospectively-studied cohort of ET patients. RESULTS: Reticulin grade positively correlated with white blood cell (P = .05) and platelet counts (P = .0001) at diagnosis. Elevated reticulin levels at presentation predicted higher rates of arterial thrombosis (hazard ratio [HR], 1.8; 95% CI, 1.1 to 2.9; P = .01), major hemorrhage (HR, 2.0; 95% CI, 1.0 to 3.9; P = .05), and myelofibrotic transformation (HR, 5.5; 95% CI, 1.7 to 18.4; P = .0007) independently of known risk factors. Higher reticulin levels at diagnosis were associated with greater subsequent falls in hemoglobin levels in patients treated with anagrelide (P < .0001), but not in those receiving hydroxyurea (P = .9). Moreover, serial trephine specimens in patients randomly assigned to anagrelide showed significantly greater increases in reticulin grade compared with those allocated to hydroxyurea (P = .0003), and four patients who developed increased bone marrow reticulin on anagrelide showed regression of fibrosis when switched to hydroxyurea. These data suggest that patients receiving anagrelide therapy should undergo surveillance bone marrow biopsy every 2 to 3 years and that those who show substantially increasing reticulin levels are at risk of myelofibrotic transformation and may benefit from changing therapy before adverse clinical features develop. CONCLUSION: Our results demonstrate that bone marrow reticulin grade at diagnosis represents an independent prognostic marker in ET, reflecting activity and/or duration of disease, with implications for the monitoring of patients receiving anagrelide.