Myelofibrotic transformation in essential thrombocythemia

The exact incidence of myelofibrotic transformation in essential thrombocythemia (ET) is still a critical issue1,2 and is certainly influenced by risk status and diagnostic criteria. In a recently published study by Passamonti and co-workers in this journal,3 a cohort of 605 patients with ET is described recruited during a lengthy period (1975 to 2008) from one center. Unfortunately, as explicitly stated, diagnosis of ET was performed using criteria at the time of the first observation, implicating that a large fraction, probably the majority of patients, were diagnosed by following the cited criteria of the Polycythemia Vera Study Group (PVSG),4 contrasting the more recently entered fraction to whom the WHO classification was applied.5

Consequently, the diagnostic guidelines are inconsistent regarding a strict differentiation of ET from early stages of primary myelofibrosis (PMF) with associated thrombocytosis. It has been demonstrated by different groups that significant differences exist between the PVSG versus the WHO classification schemes for ET: depending on risk factors and selection of patients only about 20–40% of ET cases diagnosed according to the PVSG4 are validated by the WHO criteria.6–8 The diagnosis of post-ET myelofibrosis was made by the authors using the criteria of the International Working Group on Myelofibrosis (IWG-MRT).9

However, these criteria do explicitly include as first major requirement the documentation of a previous diagnosis of WHO-defined essential thrombocythemia . This postulate can not be fulfilled by the obviously predominant PVSG-diagnosed series of patients incorporated into this investigation. Moreover, following the IWG-MRT criteria, bone marrow fibrosis grade 2–3 consistent with the cited European Consensus Grading System10 is the second major requirement. This important feature implicates the evaluation of sequential bone marrow biopsy specimens - has this been performed in all patients under study? Concerning the frequency of post-ET myelofibrosis, the transformation rate was 2.8% at a median follow-up of 9.1 years (or a 10-year risk of 3.9% and a 15-year risk of 6%).3 This relatively low incidence may possibly be created by including an unknown number of WHO-diagnosed ET cases into this study. When strictly following the diagnostic criteria of the PVSG4 applied in the UK-PT1 Study for high risk ET,11 in our retrospective evaluation of 539 patients under standard therapy, after three years overall incidence of myelofibrosis proved to be significantly different depending on diagnostic criteria (Table 1).

Table 1.Risk of myelofibrotic transformation (post-essential thrombocythemia myelofibrosis) within 36 months in 539 high-risk patients with essential thrombocythemia strictly diagnosed according to the UK-PT1 Study criteria11 receiving standard therapy regimens.

This difference was generated by discrimination of the total cohort according to the WHO classification5 into (true) ET and PMF with accompanying thrombocytosis mimicking this condition. On the other hand, when maintaining the PVSG guidelines,4 a comparable rate (2.8% vs. 2.6%)11 of progression into myelofibrosis was revealed during an observation time of about three years. Regarding leukemic transformation, although no systemic study has yet been published, similar discrepancies associated with applied classification schemes may be encountered. In conclusion, a thorough re-evaluation of this series of 605 patients with distinction into those diagnosed as ET cases according to PVSG criteria4 versus those in whom diagnosis was based on the WHO classification seems to be warranted, including a repeatedly performed calculation of progression rates into myelofibrosis and leukemia. As criteria for myelofibrotic transformation, the corresponding guidelines of the UK-PT1 Study11 should be used for the presumably prevalent ET group diagnosed after PVSG, and for the WHO classified series the IWG-MRT parameters can be applied. As has been reviewed for myelofibrosis in chronic myeloproliferative neoplasms,12 a significant difference between both groups of patients can be predicted consistent with the diagnostic criteria applied.