Prefibrotic myelofibrosis: is this diagnosis valid?

Primary myelofibrosis (PMF) is a clonal myeloproliferative disorder (MPD) characterized by a proliferation of abnormal megakaryocytes, reactive bone marrow fibrosis, and extramedullary hematopoiesis (see figure). In 2001, the World Health Organization (WHO) published criteria for diagnosing MPDs and included the concept of “prefibrotic myelofibrosis,” which is meant to represent the early stage of PMF.¹  The diagnostic criteria in the WHO classification emphasize bone marrow histology and are based largely on multiple publications from a group that has studied trephine biopsy specimens from patients with MPDs.²  This group suggested that many patients diagnosed with ET using the Polycythemia Vera Study Group (PVSG) criteria actually have prefibrotic myelofibrosis, and further suggested that bone marrow biopsies from patients with thrombocythemia can be reliably subdivided into prefibrotic MF and ET and that there is an apparent reduction in life expectancy in the patients with prefibrotic MF compared with true ET.³ 

The findings from Wilkins and colleagues are important because diagnosis of prefibrotic myelofibrosis has been controversial and has generated much discussion among those interested in MPDs. Prefibrotic MF is also included in the revised WHO criteria,⁴  but there is little prospective data available to validate the diagnostic criteria for distinguishing prefibrotic MF from ET.

The current study by Wilkins and colleagues is large and well done. Three experienced hematopathologists independently reviewed bone marrow trephine biopsies from 370 patients with ET that was diagnosed using the PVSG criteria. Biopsies were evaluated for 16 morphologic features reported to be of value in distinguishing prefibrotic MF from ET. Cases were categorized into ET or prefibrotic MF according to the WHO classification. The researchers found substantial interobserver variability in the classification of these cases, even though there was reasonable agreement about several of the histological features. No difference was found in clinical outcome between those cases classified as ET versus those classified as prefibrotic MF.

The findings of this report suggest that the published histological criteria for distinguishing ET from prefibrotic MF are difficult to apply reproducibly. This does not mean that the criteria are invalid, however. The study could be criticized on the grounds that the investigators were relying on biopsy morphology with limited clinical information (age and sex only), with no laboratory data or blood films to examine. It is possible that evaluating other clinical and laboratory features along with histopathology would have increased the reproducibility of the classification. It is interesting that the rate of transformation to myelofibrosis in the patient population studied was extremely low even though the median follow-up was 68 months. This low rate of progression raises the possibility that the population had a very small number of prefibrotic MF patients, making the detection of the subgroup difficult. As the authors indicate, even experienced hematopathologists may need specialized training to identify prefibrotic MF, but the study addressed the utility of the criteria in a real practice setting.

This paper will undoubtedly stimulate additional discussion regarding the classification of the MPDs. It provides further insight into our understanding of the classification MPDs at a time when our knowledge is being enhanced and diagnostic criteria revised by the recent and ongoing discoveries of JAK2 and other mutations^5,6  in these neoplasms.