prefibrotic myelofibrosis: Why does it matter

Best Practice & Research Clinical Haematology 27 (2014) 129e140

Contents lists available at ScienceDirect

Best Practice & Research Clinical Haematology journal homepage: www.elsevier.com/locate/beha

5

Essential thrombocythemia vs. early/prefibrotic myelofibrosis: Why does it matter Giovanni Barosi, MD * Center for the Study of Myelofibrosis, IRCCS Policlinico S. Matteo Foundation, Viale Golgi 19, 27100 Pavia, Italy

Keywords: primary myelofibrosis essential throbocythemia early/prefibrotic myelofibrosis bone marrow biopsy

Essential thrombocythemia (ET) and primary myelofibrosis (PMF), together with polycythemia vera (PV) are Phildelphia-negative (Ph-neg) classical myeloproliferative neoplasms (MPN). ET has been traditionally identified by thrombocytosis and absence of relevant bone marrow (BM) fibrosis, while PMF by BM reticulin or collagen fibrosis with megakaryocyte hyperplasia and dysplasia, and extramedullary hematopoiesis. These diagnostic profiles have been challenged since 2001 when the World Health Organization (WHO) has included in the domain of PMF a new category of patients, namely early/prefibrotic MF, characterized by the absence of relevant reticulin fibrosis in BM, dual megakaryocyte and granulocyte proliferation, and megakaryocyte dysplasia. This review is focused on summarizing the diagnostic uncertainties of early/ prefibrotic myelofibrosis, recent advances in our understanding of the biology of the variant, and the accompanying translational implications. © 2014 Elsevier Ltd. All rights reserved.

Introduction Essential thrombocythemia (ET), primary myelofibrosis (PMF), and polycythemia vera (PV) are Phildelphia-negative (Ph-neg) classical myeloproliferative neoplasms (MPN) [1]. ET has been traditionally identified by thrombocytosis and absence of relevant bone marrow (BM) fibrosis [2], while PMF by BM reticulin or collagen fibrosis with megakaryocyte hyperplasia and dysplasia, and extramedullary hematopoiesis [3]. These diagnostic profiles have been challenged since 2001 [1] when the World Health Organization (WHO) has included in the domain of PMF a new category of patients,

* Tel.: þ39 0382 503636; Fax: þ39 0382 503917. E-mail address: [email protected].

http://dx.doi.org/10.1016/j.beha.2014.07.004 1521-6926/© 2014 Elsevier Ltd. All rights reserved.

130

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

namely early/prefibrotic MF, characterized by the absence of relevant reticulin fibrosis in BM, dual megakaryocyte and granulocyte proliferation, and megakaryocyte dysplasia. Early/prefibrotic myelofibrosis: the history of an idea The idea that PMF could be diagnosed in the presence of a non-fibrotic BM, first came in the late twentieth century, when a group of pathologists from Hannover, led by Andreas Georgii, described an entity among chronic myeloproliferative disorders (CMPD), they called chronic megakaryocytic granulocytic myelosis (CMGM) [4]. The idea was functional to the simple contention that CMPD could be classified in four categories according the dominance of proliferating myeloid lineage: granulocytic for chronic myeloid leukemia (CML), erythroid for polycythemia vera (PV), megakaryocytic for ET, and granulocytic-megakaryocytic for CMGM. As a matter of fact, they defined CMGM by the proliferation of megakaryocytic and granulocytic cell lines, atypical megakaryocyte maturation, nuclear-cytoplasmic asynchrony, nuclear inclusions and production of micromegakaryocytes. In the first reports, CMGM was considered to be merely a variant of chronic myeloid leukemia (CML) [5]. However, CMGM was subsequently separated from CML, and positioned as an early phase of PMF. In 1996, when the group established the histopathological classification and staging of Ph-negative MPD (Hannover system), CMGM appeared as a synonym of PMF [6]. The consolidation of the seminal intuition of a non-fibrotic early stage of PMF, however, has to be attributed to Jurgen Thiele and co-workers from Cologne. From 1989 to 1999 they produced multiple retrospective analyses of an expanding and well-characterized archive of trephine biopsy specimens from patients with CMPD and thrombocytosis where the name of CMGM disappeared, but the concept of an early hyperplastic stage of PMF was set off [7e9]. Moreover, they clearly delineated the distinction between a category of patients with thrombocytosis and a BM with a proliferation of a not severely dysplastic megakaryopoiesis and a normal content of reticulin fibers, compatible with ET, and another characterized by a decline of the initially elevated thrombocyte count, and conspicuous abnormalities of megakaryocytes accompanied by a slight to moderate increase in argyrophilic fibers and a leftshifted neutrophilic granulopoiesis. The name “prefibrotic myelofibrosis” appeared, at the best of our knowledge, for the first time in 1999 in a paper signed by a high number of European pathologists and clinicians (first author Jean Jacques Michiels from Amsterdam) [10]. Diagnosis of early/prefibrotic MF The diagnostic histological hallmarks that allow to distinguish early/prefibrotic MF from ET [11] are those delineated from the beginning by the Hannover group, refined by the Cologne group, and formalized in the WHO histological criteria in 2001 [12] and then revised in 2008 [1] (Table 1, Fig. 1). Table 1 Histopathology of bone marrow biopsy in early/prefibrotic myelofibrosis and essential thrombocythemia according to the WHO description [13]. Early/prefibrotic myelofibrosis

Essential thrombocythemia

Hypercellularity by a prominent neutrophil granulocytic and megakaryocytic proliferation often associated with a concomitant reduction of nucleated red cell precursors in the absence or only minor reticulin MF, consistent with MF-0 and MF-1. Abnormalities in the megakaryocytopoietic cell lineage: these include first of all the distribution within the marrow space showing often extensive and dense clustering and translocation towards the endosteal borders. Moreover, there are markedly expressed anomalies of megakaryocyte maturation and differentiation detectable which consist of a high variability in size ranging from small to giant megakaryocytes. There are prominent aberrations of the nuclei (marked hypolobulation, condensed chromatin, and irregular foldings crating a bulbous, cloud-like aspect) and marked elevation of the nuclear-cytoplasmic ratio, as well as an increased frequency of bare (denuded) nuclei. Age-matched cellularity and a predominant megakaryopoiesis without a significant erythroid or neutrophilic myeloproliferation. In ET megakaryocytes reveal a more or less random distribution or very loose groupings within the BM space. An important feature is the large to giant cell forms with extensively folded (hyperlobulated) nuclei, surrounded by well differentiated (mature) cytoplasm. An increase in reticulin is not compatible with early stages of ET.

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

131

Fig. 1. Bone marrow histological picture in essential thrombocythemia (A) and in early/prefibrotic myelofibrosis (B).

Even though the category of early/prefibrotic MF has joined the ranks of the diagnoses used by pathologists for classifying the Ph-neg MPNs in the wider part of the world, the problem of the differential diagnosis between ET and early/prefibrotic MF continues to spark lively discussions. Serious concerns have been expressed arguing against the reliability/repeatability of the criteria as they were set up by the WHO classification. The major criticism derived from a highly esteemed group of researchers in UK who designed a blinded study by which three pathologists had to assess the reproducibility and validity of distinguishing so called ET from early/prefibrotic MF according the WHO classification [13]. The result was that the reproducibility of the major diagnostic features were profoundly questioned, in particular megakaryocyte morphology, size, and clustering. Even though a number of inconsistencies have been denounced by the WHO proposers that would impair the results of this study (no intra-observer evaluation, non canonical definition of parameters, no standardization of morphological parameters, presence of BM fibrosis unacceptable for ET) [11], following reports revealed the same difficulty in the reproducibility of the WHO classification [14e16]. The most relevant observations concerning the reliability of the WHO classification consisted in the non infrequent examples of BM histology with some of the morphological features said to reflect true ET coexistent with changes thought to imply early/prefibrotic MF or even overt MF [13]. Moreover, the absence of guidance on the importance of different morphological aspects was claimed to leave the pathologist uncertain on the weight to pose to different morphologic criteria. For example, in a study [15], it was claimed that granulocytic proliferation may not be considered a prerequisite for the diagnosis of early/ prefibrotic MF. The strongest argument of the WHO promoters against the irreproducibility of the WHO criteria was that in well educated and specifically trained pathologists, the criteria proved to be reproducible. As a matter of fact, several groups confirmed their ability to discriminate between the thrombocythemic manifestations of early/prefibrotic MF versus ET [17e21]. A hypothetical interpretation that could reconcile the discrepancies in the diagnostic performance between the sustainers of the WHO

132

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

classification and the opponents is that the formers possess a tacit knowledge that allows them to move from the pattern recognition process, that is advocated by the WHO criteria, towards a rulebased procedural one. This seems documented by the fact that a rule-based morphometric approach to BM biopsies is sometimes used in the practice (Orazi A, personal communication). The need of better precision of the WHO definition of early/prefibrotic MF was recently argued by a group of experts in the field of MPN who claimed that a major issue for the uncertainty of the distinction between early/prefibrotic MF from ET was the measure of the details by which BM megakaryocytes features are expressed [22]. The experts agreed that a more precise definition of early/ prefibrotic MF could be obtained by moving from a qualitative morphological description and a pattern recognition diagnostic process to a quantitative assessment of the key morphological features in BM and a standardized procedural diagnostic pathway. They finally convened that the unmet clinical need should be addressed by a consensus project among pathologists that should categorize which of the BM morphological features need a morphometric evaluation and provide details about the procedure to employ for assessing those morphometric characteristics that better discriminate between ET and early/prefibrotic MF. The biology of early/prefibrotic MF A distinctive characterization of the biological basis of early/prefibrotic MF would be the strongest evidence that a prodromic phase of PMF indeed exists as a separate diagnostic entity amongst those patients with an apparent diagnosis of ET. A clear-cut biological distinction between ET and early/ prefibrotic MF has not yet been provided. However, an array of initial evidences is on the way for this distinction. Clonality of hematopoiesis is a constant feature only of early/prefibrotic MF Biological difference between ET and early/prefibrotic MF has been described with regard to clonality of hematopoiesis. One study has reported the inactivation patterns of X-chromosomal genes in 12 female patients with PMF classified as early stages, i.e. with 0 or 1 BM fibrosis [23]. In all heterozygous granulocyte samples obtained from different stages of MF, a monoclonal inactivation pattern of X-chromosomal genes was shown, as demonstrated by restriction with methylation sensitive HpaII and hybridization with either the PGK or the HPRT gene. Clonal granulocytes could be found in cases with mild BM fibrosis as well as in cases with advanced MF. At variance, studies in ET revealed either monoclonality or polyclonality [24,25]. For example, in a study in 46 patients with ET, in which Xchromosome inactivation patterns results were obtained on neutrophils and T cells, either HUMARA, PGK, or both documented monoclonal hematopoiesis in only 10 patients [25]. These results must be interpreted with appropriate reference to both the individual constitutive X-chromosome inactivation patterns, patients' age and certainty of the diagnosis. Early/prefibrotic MF has higher JAK2V617F and MPLW515L allele burden than ET In the recent years, extensive molecular and genome wide screening have resulted in high number of somatic mutations in PMF. In particular JAK2V617F [26e29], mutations of MPL exon 10 [30], and frameshift mutations in exon 9 of CALR [31,32], the gene encoding calreticulin, were found to be mutually exclusive, disease-initiating mutations. In addition, a high number of different concurrent gene mutations has been reported, some having prognostic relevance, like EZH2, ASXL1 and SRS2 [33]. No study has reported a systematic difference profile in the incidence of both diseaseinitiating mutations and concurrent mutations in early/prefibrotic MF and in ET patients. However, in a study of 490 patients, a significantly lower JAK2V617F allele burden was detectable in ET at disease presentation when compared to early/prefibrotic PMF [34]. One-quarter of prefibrotic PMF cases exhibited an allele burden exceeding 50% (38% median). In ET not a single case was found with >40% JAK2V617F alleles (median, 24%; P < 0.001). MPLW515L was found in 3% of ET and 8% of PMF, with a significantly higher percentage of mutated alleles in fibrotic than early/prefibrotic MF. The authors concluded that in newly diagnosed Ph-neg MPN with thrombocytosis and no or initial BM

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

133

fibrosis, >50% JAK2V617F allele burden will most likely represent early/prefibrotic MF, whereas ET does not appear to be probable. This result however, was not confirmed in a work in which the allele burden of JAK2V617F was assessed in BM biopsies [35]. The JAK2V617F allele burden was significantly higher in patients with PV (median: 50.9%) and PMF (median: 44.1%) compared with a low allele burden in ET (median: 23.4%) and early PMF (median: 25.6%) respectively. However, by correlating these findings with the amount of phosphorylated STAT3 (P-STAT3) and STAT5 (P-STAT5) in megakaryocyte nuclei in the BM, the authors found a significantly higher phosphorylation of STAT5 and STAT3 in the JAK2V617F positive group than in the negative group, and they found low values of P-STAT5 in BM biopsies from patients with JAK2V617F mutated ET as compared with JAK2V617F mutated early/prefibrotic MF. NFE2 is aberrantly expressed in early/prefibrotic MF patients The transcription factor “nuclear factor-erythroid 2” (NF-E2) is aberrantly expressed in MPN patients [36]. In ET and PMF, NF-E2 overexpression is independent of the presence or absence of the JAK2V617F mutation [37]. By studying a cohort of 163 BM biopsies of patients with different types of MPNs, Aumann and coworkers demonstrated that NF-E2 staining was highly significantly increased in the nucleus of erythroid cells of patients with PMF with respect to those with PV or ET [38]. This marked increase was apparent even in patients with grade 0 or 1 BM fibrosis, suggesting that increased nuclear staining of the transcription factor is inherent to disease development, rather than being a feature of disease progression. A threshold of 20% nuclear NF-E2 staining was cross-validated. Moreover, this cutoff correctly classified diagnostic BM biopsies of MPN patients specified upon follow-up as ET or PMF with 92% accuracy. Because inter-observer concordance between independent pathologists was high (Spearman's rank correlation coefficient, 0.73), the authors proposed that quantitative NF-E2 immunohistochemistry could represents a diagnostic tool that can reliably support a differential diagnosis between ET and early/prefibrotic MF. In vitro megakaryopoiesis distinguishes early/prefibrotic MF from ET patients The study of in vitro patterns of differentiation of megakaryocytes from circulating hematopoietic progenitors has proven to be a reliable model for studying the biological diversity of MPNs. Balduini and coworkers studied megakaryocyte differentiation and proplatelets formation in vitro in 30 PMF patients (13 with early/prefibrotic MF and 17 with fibrotic MF), 8 ET and 8 PV patients [39]. Megakaryocytes were differentiated from peripheral blood CD34-postive or CD45-posive cells in the presence of thrombopoietin. The median output of CD41-positive megakaryocytes at day 14 was not statistically significantly different between early/prefibrotic MF and ET. However, analysis of megakaryocyte morphology revealed significant differences in the maturation profile of PMF compared to ET, indicating a peculiar defect of megakaryocyte development in PMF. Consistently, a lower percentage of megakaryocytes derived from PMF was polyploid (>8N), presented bulbous nuclei, and displayed a decreased diameter than those from ET (Fig. 2). These characteristics were maintained both in early/prefibrotic MF and in fibrotic type MF. In order to exclude that differences in megakaryocyte morphology were dependent on the maturation stage of progenitors derived from the different MPNs, a time course analysis was performed in PMF derived cultures. Progenitors derived from early/prefibrotic MF were maintained in culture for 18 days and megakaryocyte morphology was analyzed. Results demonstrated that, even prolonging the culture incubation time, megakaryocytes derived from PMF remained smaller than ET and showed the same characteristics of immaturity observed when cultures were performed for shorter period of time. In order to explore whether defects in megakaryocyte development were associated to altered megakaryocyte function, the authors investigated the generation of proplatelets by MPN-derived megakaryocytes. Mature megakaryocytes, at the end of the culture, were reseeded and proplatelet formation was evaluated after 16 h. In control samples, a median of 7.5% of megakaryocytes formed proplatelets, compared to 3.8% of PMF-derived megakaryocytes, 8.65% of ET-derived megakaryocytes and 9.15% of PV-derived megakaryocytes (P ¼ 0.001 for all the comparisons). No differences were observed between early/prefibrotic MF (median: 3.3%) and fibrotic PMF (median: 4.3%). The authors

134

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

Fig. 2. Characteristics of in vitro megakaryocyte morphology in myeloproliferative neoplasms and controls. Megakaryocytes derived from patients with primary myelofibrosis (PMF), even in prefibrotic stage (pre-PMF), presented lower polyploidy with bulbous and hyposegmented nuclei, with respect to those derived from controls (CRL) and essential thrombocythemia (ET) patients. Derived from our publication in PLoS One 2011; 6:e21015 (Balduini et al.) [39].

found that the proplatelets extended by PMF megakaryocytes had a very variable numbers of bifurcations that frequently did not present any tips at the terminal end (Fig. 3). In contrast, ET- and PVderived proplatelets displayed a striking increase in bifurcations and tips compared to PMF. These results suggest that early/prefibrotic MF and ET displayed peculiar alterations of megakaryocyte differentiation and function in vitro. High mobilization of ECFCs is characteristic of early/prefibrotic MF Mobilization of circulating endothelial progenitor cells (EPCs) represents a biological hallmark of MPN [40]. The majority of EPCs reside in the BM, in close association with hematopoietic stem cells and BM stromal cells that provide the microenvironment for hematopoiesis, representing only 0.02% of circulating mononuclear cells in peripheral blood. Studies on EPCs mobilization must contrast with the different nature of EPC and different assay methods. Up to date the proper definition of EPCs is still a matter of debate, and EPCs characterization is performed according to the definition of human EPC as a circulating cell that promotes neovascularization at sites of ischemia, hypoxia, injury, or tumor formation [41]. We measured circulating endothelial colony forming cells (ECFCs) in 106 patients with primary myelofibrosis, fibrotic stage, 49 with early/prefibrotic MF, 59 with ET or PV, and 43 normal controls [40]. We found that increased frequency of ECFCs resulted independently associated with diagnosis of early/prefibrotic MF, history of splanchnic vein thrombosis, and a summary measure of non-active disease, i.e. hemoglobin of 13.8 g/dL or lower, white blood cells count of 7.8  109/L or lower, and platelet count of 400  109/L or lower. Thirteen patients with splanchnic vein thrombosis non associated with myeloproliferative neoplasms were recruited as controls. We excluded a causal role of splanchnic vein thrombosis in ECFCs increase, since no control had elevated ECFCs. We concluded that

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

135

Fig. 3. Analysis of proplatelet architecture of in vitro derived megakaryocytes. Patients with primary myelofibrosis (PMF) presented differently organized proplatelets (lower bifurcations and tips) with respect to controls, ET and polycythemia vera (PV) patients. Derived from our publication in PLoS One 2011; 6:e21015 (Balduini et al.) [39].

increased frequency of ECFCs represents the biological hallmark of a non-active MPN with the histological characteristics of early/prefibrotic MF and with high risk of splanchnic vein thrombosis. This result points to different BM microenvironment between ET and early/prefibrotic MF, being the latter more prone to EPC mobilization. Clinical differences between ET and early/prefibrotic MF Clinical presentation The largest series of cases in which the clinical characteristics at presentation of WHO-classified ET and early/prefibrotic MF were compared derives from a clinicopathologic data-base of patients previously diagnosed as having ET according to non-WHO based criteria (N ¼ 1104) and created by representatives from seven international centers of excellence for MPN [42]. All BM biopsies taken at diagnosis underwent a central re-review by Jurgen Thiele, one of the authors of the paper as well as the author of the WHO chapter on diagnostic criteria for early/prefibrotic MF and ET. The reviewer had access to patient information that was necessary for routine interpretation of BM morphology. However, the review process was completely blinded to outcome data, which were analyzed after the completion of the histopathology review. Diagnosis was confirmed as ET in 891 patients (81%) and was revised to early/prefibrotic MF in 180 (16%). A comparative analysis of presenting clinical and laboratory features for patients with ET and those with early/prefibrotic MF documented that age and sex distribution and JAK2V617F mutational frequencies were similar between the two groups. Significant differences were seen for leukocyte count (greater in early/prefibrotic MF than ET), hemoglobin level

136

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

(greater in ET than early/prefibrotic MF), platelet count (greater in early/prefibrotic MF than ET), serum lactate dehydrogenase (LDH) level (greater in early/prefibrotic MF than ET), circulating CD34 cell count (greater in early/prefibrotic MF than ET), incidence of palpable splenomegaly (greater in early/prefibrotic MF than ET), and frequency of grade 1 BM fibrosis (greater in early/prefibrotic MF than ET). In a substudy of the same population of patients, the disease characteristics in young patients (<40 years) were presented (n ¼ 213) including 178 patients (84%) with WHO-defined ET and 35 patients (16%) showing early/prefibrotic MF [43]. The platelet count was higher in patients with early PMF, who also showed a trend for lower hemoglobin value, higher leukocyte counts, and higher LDH levels. This pattern was consistent with the findings documented in older patients. In an additional work, 536 patients were extracted for studying sensitivity and specificity of the hematological and clinical features (hemoglobin, white blood cells (WBC), and LDH) that distinguished ET from early/prefibrotic MF [44]. The best performance was found for LDH. WBC and hemoglobin had superimposable curves. A diagnostic algorithm was generated by applying these parameters in a stepwise fashion: nearly half of the patients could be correctly allocated to WHO-defined ET or early/ prefibrotic MF. The authors claimed that, although BM biopsy represents the gold standard for differentiating WHO-ET from early/prefibrotic MF, in clinical practice it might be useful to use the above resulting parameters to get more suspicious of early/prefibrotic MF in a patient presumably presenting with ET. The above reported results delineate the hematological characteristics of patients with early/prefibrotic MF diagnosed with thrombocytosis and their differences from ET. However, due to the nature of the data-base that selected patients with thrombocytosis, the reported series of cases does not represent the whole population of patients that may receive a diagnosis of early/prefibrotic MF in the clinical setting. We recently reported our series of cases consecutively diagnosed with PMF [45]. In the cases with BM fibrosis grade 0 (prefibrotic MF), we revealed that they (n ¼ 132) had several epidemiological and clinical features not previously come to light. In comparison with patients with PMFfibrotic type (BM fibrosis 1e3), the patients with prefibrotic MF were more likely to be female, and to be under the age of 50 years at diagnosis. Moreover, with the diagnosis of prefibrotic MF we selected a category of patients with very high frequency (30%) of coincidental thrombotic events, mainly in splanchnic area. This latter presentation coincides with that of masked MPNs associated with splanchnic vein thrombosis [46e50], and in particular of the variant “atypical myeloproliferative disorder with high thrombotic risk and slow disease progression” we first described in 1991 [51]. Despite the narrow criteria used for the diagnosis, the cases of prefibrotic MF were aligned along a continuum of cases with different degrees of BM fibrosis, and displayed different hematological presentations (Table 2). Seven percent female patients had hemoglobin greater than 16.5 g/dL, the value WHO system uses for classifying PV patients. These patients received an original diagnosis of PV, but polyglobulia spontaneously remitted early. Sixty-nine percent of patients had more than 450  109/L platelets, the value the WHO system uses for classifying ET patients. One patient had thrombocytopenia (platelet count less than 150  109/L), but the patient at the same time suffered by portal vein

Table 2 Hematological and biological parameters of 683 cases with primary myelofibrosis stratified according to the bone marrow fibrosis grade.

Number Female Age, yrs, mean Hemoglobin (g/dL), mean White blood cell count (109/L), mean Platelet count (109/L), mean Spleen index (cm2), mean LDH U/mL, mean CD34þ in PB (106/L), mean

Prefibrotic myelofibrosis (BM fibrosis grade 0)

Myelofibrosis-fibrotic type (BM fibrosis grade 1)

Myelofibrosis-fibrotic type (BM fibrosis grade 2e3)

132 (19.3%) 60.3% 38.3 14.2 9.2

206 (30.2%) 42.2% 50.9 13.3 11.5

345 (50.5%) 28.1% 57.4 11.1 10.1

647 118 451 3.24

553 152 597 17.5

359 199 826 34.3

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

137

thrombosis that could be the cause of low platelet count. Three patients had white blood cell count equal or greater than 25  109/L, and 2 patients had white blood cell count equal or lower than 4  109/L. An observation that could open new research perspective on the clinical phenotype of early/ prefibrotic MF arose from a series of cases from Asian patients (42 Chinese patients), with early/ prefibrotic MF who had clinical and laboratory features that differed significantly from those of the predominantly white patients in Western countries [52]. Chinese patients with early/prefibrotic MF were more often male (1.28:1) and younger, less likely to have higher median hemoglobin concentrations (12.6 g/dL), and less frequently had palpable spleens (35.7%). Moreover, on bone marrow trephine sections, Chinese patients were more likely to have increased granulopoiesis (78.6%) and less frequently had balloon-like megakaryocytes (61.9%), giant and staghorn megakaryocytes (35.7%), or megakaryocytes with hyperchromatic and dysplastic nuclei (40.4%). More clinicopathologic and molecular studies, particularly constitutional factors predisposing to MPN, involving different ethnic populations and geographic regions of the world should help unfold the characteristics of this disease. Natural history and outcome prediction The clinical course of early/prefibrotic MF has been studied in the largest series of cases published by Barbui and coworkers, and was compared with that of ET [42]. It resulted that the survival rates, leukemic transformation rates, and rates of progression to overt myelofibrosis were significantly worse in early/prefibrotic MF than in ET. The respective death, leukemia, and overt myelofibrosis incidence rates per 100 patient-years for early/prefibrotic MF compared with ET were 2.7% and 1.3% (relative risk [RR], 2.1; P < 0.001), 0.6% and 0.1% (RR, 5.2; P ¼ 0.001), and 1% and 0.5% (RR, 2.0; P ¼ 0.04). The results of this study indicate that the distinction between early/prefibrotic MF and ET is not trivial and that accurate diagnosis is prognostically relevant. At the same time, they underline the fact that event-free survival is still relatively long in patients with early/prefibrotic MF. Multivariable analyses in this study confirmed the prognostic relevance of BM histopathology for survival, leukemic transformation, and overt fibrotic progression. BM histopathologyeindependent risk factors for survival included older age (older than 60 years), leukocytosis (leukocyte count > 11  109/L), anemia (hemoglobin < 12 g/dL), and thrombosis history. Taken together, it is reasonable to conclude that anemia, increased serum LDH, leukocytosis, extreme thrombocytosis in patients with high platelet number might be markers of occult early/prefibrotic MF, and that could explain their association with inferior survival or higher risk of disease progression. The incidence of thrombotic events was not different between ET and early/prefibrotic MF [42]. On the contrary, major bleeding during follow-up occurred in 55 (6%) WHO-ET and 21 (12%) PMF patients (P ¼ 0.009), at a rate of 0.79 and 1.39% patients per year, respectively, (P ¼ 0.039) [53]. In a multivariable analysis, predictors of bleeding included diagnosis of PMF (P ¼ 0.05; hazard ratio (HR) 1.74), leukocytosis (P ¼ 0.04; HR 1.74), previous hemorrhage (P ¼ 0.025; HR 2.35) and aspirin therapy (P ¼ 0.001; HR 3.16). The analysis restricted to patients with WHO-ET confirmed previous hemorrhage (P ¼ 0.043; HR 1.92) and aspirin (P ¼ 0.027; HR 2.24) as independent risk factors. The study revealed that major bleeding associated with thrombocytosis might be relatively specific to PMF, as opposed to WHOdefined ET. Furthermore, it shows that low-dose aspirin exacerbates these hemorrhagic events of PMF. In contrast, thrombocytosis per se was not a risk factor for bleeding; however, low-dose aspirin had a synergistic hemorrhagic effect unmasking the bleeding tendency of patients with extreme thrombocytosis. These observations carry significant therapeutic implications in these two WHO entities. Risk factors for thrombotic events in early/prefibrotic MF were calculated on a total number of 264 patients [54]. After a median follow-up of 6.28 years, 42 (15.9%) patients experienced arterial (n ¼ 31) or venous thrombosis (n ¼ 11). A higher leukocyte count correlated with an increased risk for total thrombosis and in particular, with an increased risk for arterial thrombosis (P ¼ 0.005, HR 1.15 and P ¼ 0.047, HR 1.12, respectively). A platelet count above 870  10⁹/L was associated with a lower risk for total thrombosis and also for venous thrombosis (P ¼ 0.022, HR 0.44 and P ¼ 0.027, HR 0.19). Moreover, a lower hemoglobin level was associated with an increased risk for venous thrombosis (P ¼ 0.007, HR

138

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

0.59). These data indicate that leukocytosis is a prominent risk factor for thrombosis in early/prefibrotic MF. Therapeutic considerations In the recommendations on therapy for PMF issued by European Leukemia Net (ELN) in 2012, early/prefibrotic MF was not separately considered [55]. By classifying patients with early/prefibrotic MF according to the standard prognostic system for PMF, such as IPSS [56], they mostly fall in the lowrisk category, thus they do not deserve treatment, but monitoring. However, patients with early/ prefibrotic MF have a risk of vascular events that is not different form that of ET [42]. Thus, it seems rationale to approach treatment decision making in these patients like in those with ET, in which treatment recommendations are shaped according to the thrombosis risk categories. So, only patients with an age older than 60 years and previous thrombotic event, are deemed to need cytoreductive treatment. The era of molecularly targeted agents in MPNs has brought forth new rationales for treatment. Reducing or abolishing the diseased clone or its malignant activity, thus relenting or preventing disease progression, has been advocated as a new targeted therapy for low-risk PMF patients or those with an early disease, such as early/prefibrotic MF [22]. A Panel of experts has recognized the decision of treatment with new agents, like JAK2 inhibitors, an unmet clinical need [22], claiming that documentation of slow-down of disease progression requires a controlled clinical trial with a new molecularly targeted therapy in comparison with the best supportive therapy. Conclusions and perspectives An array of clinical and biological evidences point to bear out the seminal idea that early/prefibrotic MF is a prodromal presentation of PMF. Even though thrombocytosis is the most common presenting characteristic, clinical studies have provided evidence that early/prefibrotic MF may present with different phenotypes, like that of PV or masked MPN associated with splanchnic vein thrombosis. The natural history of early/prefibrotic MF is significantly different from that of true ET; in particular for higher rates of evolution to overt MF and blast transformation, thus higher mortality. These contentions signify the importance of recognizing this variant among patients with signs of MPNs. Even though WHO has provided minor criteria for the diagnosis of early/prefibrotic MF (leuko-erythroblastosis, increase in serum LDH level, anemia, and splenomegaly) the main clue for the diagnosis remains the BM picture. In this regard, the issue is not as much about reproducibility of morphologic interpretation but more about increased awareness by pathologists. Early/prefibrotic MF is an evolving field of research. Many claims have been raised to move towards a scientific project for providing sound, objective and reproducible criteria for the diagnosis [22,57]. Moreover, the diagnostic boundaries of the disease are a matter of debate. The inclusion of patients having either early or prefibrotic phase of MF, i.e. grade 0 or grade 1 BM fibrosis, may be criticized since it dumps the great biological value of prefibrotic MF which has modified our pathogenic and clinical thinking on the disease. On the contrary, early phase MF (BM fibrosis grade 1) reflects a conventional idea by which the diagnostic criteria of the disease have always envisaged the presence of any degree, even moderate, of BM fibrosis (3). Moreover, the inclusion of both prefibrotic and early BM fibrosis in the prodromal stage of PMF, may be also criticized from a nominal point of view. As a matter of fact, it has been estimated that between 40% and 50% of patients present with a prefibrotic or early MF (13) making the early/prefibrotic MF variant a misnomer. The greatest research question in the evolving concept of early/prefibrotic MF is the understanding of the biological and molecular reasons that can explain why in the prefibrotic stage of MF the unique dysplastic characteristics of megakaryocytes confer different evolution pattern to a disease that is in no other way different from ET. Conflict of interest statement None.

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

139

References [1] Swerdlow SH, Campo E, Harris NL, et al., editors. WHO classification of tumours of hematopoietic and lymphoid tissues. Lyon: IARC; 2008. [2] Laszlo J. Myeloproliferative disorders (MPD): myelofibrosis, myelosclerosis, extramedullary hematopoiesis, undifferentiated MPD, and hemorrhagic thrombocythemia. Semin Hematol 1975;2:409e32. [3] Barosi G, Ambrosetti A, Finelli C, et al. The Italian consensus conference on diagnostic criteria for myelofibrosis with myeloid metaplasia. Br J Haematol 1999;104:730e7. [4] Thiele J, Georgii A, Vykoupil KF. Ultrastructure of chronic megakaryocytic-granulocytic myelosis. Blut 1976;32:433e8. [5] Georgii A, Vykoupil KF, Thiele J. Chronic megakaryocytic granulocytic myelosis-CMGM. A subtype of chronic myeloid leukemia. Virchows Arch A Pathol Anat Histol 1980;389:253e68. [6] Georgii A. Histopathology of bone marrow in human chronic leukemias. Haematol Blood Transfus 1979;23:59e70. [7] Thiele J, Zankovich R, Steinberg T, et al. Primary (essential) thrombocythemia versus initial (hyperplastic) stages of agnogenic myeloid metaplasia with thrombocytosisea critical evaluation of clinical and histomorphological data. Acta Haematol 1989;81:192e202. [8] Thiele J, Steinberg T, Zankovich R, et al. Primary myelofibrosis- osteomyelosclerosis (agnogenic myeloid metaplasia): correlation of clinical findings with bone marrow histopathology and prognosis. Anticancer Res 1989;9:429e35. [9] Thiele J, Kvasnicka HM, Diehl V, et al. Clinicopathological diagnosis and differential criteria of thrombocythemias in various myeloproliferative disorders by histopathology, histochemistry and immunostaining from bone marrow biopsies. Leuk Lymphoma 1999;33:207e18. [10] Michiels JJ, Kutti J, Stark P, et al. Diagnosis, pathogenesis and treatment of the myeloproliferative disorders essential thrombocythemia, polycythemia vera and essential megakaryocytic granulocytic metaplasia and myelofibrosis. Neth J Med 1999;54:46e62. [11] Kvasnicka HM, Thiele J. Prodromal myeloproliferative neoplasms: the 2008 WHO classification. Am J Hematol 2010;85:62e9. [12] Thiele J, Vardiman JW, Pierre R, et al. Chronic idiopathic myelofibrosis. In: Jaffe E, Harris NL, Stein HS, et al., editors. Pathology and genetics of tumours of haematopoietic and lymphoid tissue. World Health Organization classification of tumours. Lyon: IARC Press; 2001. pp. 35e8. [13] Wilkins BS, Erber WN, Bareford D, et al. Bone marrow pathology in essential thrombocythemia: interobserver reliability and utility for identifying disease subtypes. Blood 2008;111:60e70. [14] Buhr T, Hebeda K, Kaloutsi V, et al. European bone marrow working group trial on reproducibility of World Health Organization criteria to discriminate essential thrombocythemia from prefibrotic primary myelofibrosis. Haematologica 2012;97:360e5. [15] Koopmans SM, Bot FJ, Lam KH, et al. Reproducibility of histologic classification in nonfibrotic myeloproliferative neoplasia. Am J Clin Pathol 2011;136:618e24. [16] Brousseau M, Parot-Schinkel E, Moles MP, et al. Practical application and clinical impact of the WHO histopathological criteria on bone marrow biopsy for the diagnosis of essential thrombocythemia versus prefibrotic primary myelofibrosis. Histopathology 2010;56:758e67. [17] Florena AM, Tripodo C, Iannitto E, et al. Value of bone marrow biopsy in the diagnosis of essential thrombocythemia. Haematologica 2004;89:911e9. [18] Gianelli U, Vener C, Raviele PR, et al. Essential thrombocythemia or chronic idiopathic myelofibrosis? A single-center study based on hematopoietic bone marrow histology. Leuk Lymphoma 2006;47:1774e81. [19] Buhr T, Georgii A, Schuppan O, et al. Histologic findings in bone marrow biopsies of patients with thrombocythemic cell counts. Ann Hematol 1992;64:286e91. [20] Kreft A, Büche G, Ghalibafian M, et al. The incidence of myelofibrosis in essential thrombocythaemia, polycythaemia vera and chronic idiopathic myelofibrosis: a retrospective evaluation of sequential bone marrow biopsies. Acta Haematol 2005; 113:137e43. [21] Thiele J, Kvasnicka HM, Müllauer L, et al. Essential thrombocythemia versus early primary myelofibrosis: a multicenter study to validate the WHO classification. Blood 2011;117:5710e8. [22] Barosi G, Vannucchi AM, De Stefano V, et al. Identifying and addressing unmet clinical needs in Ph-neg classical myeloproliferative neoplasms: a consensus-based SIE, SIES, GITMO position paper. Leuk Res 2014;38:155e60. [23] Kreipe H, Jaquet K, Felgner J, et al. Clonal granulocytes and bone marrow cells in the cellular phase of agnogenic myeloid metaplasia. Blood 1991;78:1814e7. [24] Shih LY, Lin TL, Dunn P, et al. Clonality analysis using X-chromosome inactivation patterns by HUMARA-PCR assay in female controls and patients with idiopathic thrombocytosis in Taiwan. Exp Hematol 2001;29:202e8. [25] Harrison CN, Gale RE, Machin SJ, et al. A large proportion of patients with a diagnosis of essential thrombocythemia do not have a clonal disorder and may be at lower risk of thrombotic complications. Blood 1999;93:417e24. [26] Kralovics R, Passamonti F, Buser AS, et al. A gain-of-function mutation of JAK2 in myeloproliferative disorders. N Engl J Med 2005;352:1779e90. [27] Levine RL, Wadleigh M, Cools J, et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell 2005;7:387e97. [28] James C, Ugo V, Le Couedic JP, et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature 2005;434:1144e8. [29] Baxter EJ, Scott LM, Campbell PJ, et al. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet 2005;365:1054e61. [30] Rumi E, Pietra D, Guglielmelli P, et al. Acquired copy-neutral loss of heterozygosity of chromosome 1p as a molecular event associated with marrow fibrosis in MPL-mutated myeloproliferative neoplasms. Blood 2013;121:4388e95. [31] Klampfl T, Gisslinger H, Harutyunyan AS, et al. Somatic mutations of calreticulin in myeloproliferative neoplasms. N Engl J Med 2013;369:2379e90. [32] Nangalia J, Massie CE, Baxter EJ, et al. Somatic CALR mutations in myeloproliferative neoplasms with nonmutated JAK2. N Engl J Med 2013;369:2391e405.

140

G. Barosi / Best Practice & Research Clinical Haematology 27 (2014) 129e140

[33] Vannucchi AM, Lasho TL, Guglielmelli P, et al. Mutations and prognosis in primary myelofibrosis. Leukemia 2013;27: 1861e9. [34] Hussein K, Bock O, Theophile K, et al. JAK2(V617F) allele burden discriminates essential thrombocythemia from a subset of prefibrotic-stage primary myelofibrosis. Exp Hematol 2009;37:1186e93. [35] Risum M, Madelung A, Bondo H, et al. The JAK2V617F allele burden and STAT3- and STAT5 phosphorylation in myeloproliferative neoplasms: early prefibrotic myelofibrosis compared with essential thrombocythemia, polycythemia vera and myelofibrosis. Acta Pathol Microbiol Immunol Scand 2011;119:498e504. [36] Goerttler PS, Kreutz C, Donauer J, et al. Gene expression profiling in polycythaemia vera: overexpression of transcription factor NF-E2. Br J Haematol 2005;129:138e50. [37] Wang W, Schwemmers S, Hexner EO, et al. AML1 is overexpressed in patients with myeloproliferative neoplasms and mediates JAK2V617F-independent overexpression of NF-E2. Blood 2010;116:254e66. [38] Aumann K, Frey AV, May AM, et al. Subcellular mislocalization of the transcription factor NF-E2 in erythroid cells discriminates prefibrotic primary myelofibrosis from essential thrombocythemia. Blood 2013;122:93e9. [39] Balduini A, Badalucco S, Pugliano MT, et al. In vitro megakaryocyte differentiation and proplatelet formation in Phnegative classical myeloproliferative neoplasms: distinct patterns in the different clinical phenotypes. PLoS One 2011;6: e21015. [40] Rosti V, Bonetti E, Bergamaschi G, et al. High frequency of endothelial colony forming cells marks a non-active myeloproliferative neoplasm with high risk of splanchnic vein thrombosis. PLoS One 2010;5:e15277. [41] Critser PJ, Yoder MC. Endothelial colony-forming cell role in neoangiogenesis and tissue repair. Curr Opin Organ Transplant 2010;15:68e72. [42] Barbui T, Thiele J, Passamonti F, et al. Survival and disease progression in essential thrombocythemia are significantly influenced by accurate morphologic diagnosis: an international study. J Clin Oncol 2011;29:3179e84. [43] Barbui T, Thiele J, Carobbio A, et al. Disease characteristics and clinical outcome in young adults with essential thrombocythemia versus early/prefibrotic primary myelofibrosis. Blood 2012;120:569e71. [44] Carobbio A, Finazzi G, Thiele J, et al. Blood tests may predict early primary myelofibrosis in patients presenting with essential thrombocythemia. Am J Hematol 2012;87:203e4. [45] Barosi G, Rosti V, Bonetti E, et al. Evidence that prefibrotic myelofibrosis is aligned along a clinical and biological continuum featuring primary myelofibrosis. PLoS One 2012;7:e35631. [46] De Stefano V, Fiorini A, Rossi E, et al. Incidence of the JAK2 V617F mutation among patients with splanchnic or cerebral venous thrombosis and without overt chronic myeloproliferative disorders. J Thromb Haemost 2007;5:708e14. [47] Primignani M, Barosi G, Bergamaschi G, et al. Role of the JAK2 mutation in the diagnosis of chronic myeloproliferative disorders in splanchnic vein thrombosis. Hepatology 2006;44:1528e34. [48] Allegra A, Alonci A, Penna G, et al. JAK2 V617F-positive latent essential thrombocythemia and splanchnic vein thrombosis: the role of bone marrow biopsy for the diagnosis of myeloproliferative disease. Acta Haematol 2009;121:218e20. [49] Pagliuca A, Mufti GJ, Janossa-Tahernia M, et al. In vitro colony culture and chromosomal studies in hepatic and portal vein thrombosis-possible evidence of an occult myeloproliferative state. Q J Med 1990;76:981e9. [50] Orr DW, Patel RK, Lea NC, et al. The prevalence of the activating JAK2 tyrosine kinase mutation in chronic portosplenomesenteric venous thrombosis. Aliment Pharmacol Ther 2010;31:1330e6. [51] Barosi G, Buratti A, Costa A, et al. An atypical myeloproliferative disorder with high thrombotic risk and slow disease progression. Cancer 1991;68:2310e8. [52] Gong X, Lu X, Xiao X, et al. Clinicopathologic characteristics of prefibrotic-early primary myelofibrosis in Chinese patients. Hum Pathol 2014;45:498e503. [53] Finazzi G, Carobbio A, Thiele J, et al. Incidence and risk factors for bleeding in 1104 patients with essential thrombocythemia or prefibrotic myelofibrosis diagnosed according to the 2008 WHO criteria. Leukemia 2012;26:716e9. [54] Buxhofer-Ausch V, Gisslinger H, Thiele J, et al. Leukocytosis as an important risk factor for arterial thrombosis in WHOdefined early/prefibrotic myelofibrosis: an international study of 264 patients. Am J Hematol 2012;87:669e72. [55] Barbui T, Barosi G, Birgegard G, et al. Philadelphia-negative classical myeloproliferative neoplasms: critical concepts and management recommendations from European LeukemiaNet. J Clin Oncol 2011;29:761e70. [56] Cervantes F, Dupriez B, Pereira A, et al. New prognostic scoring system for primary myelofibrosis based on a study of the International Working Group for Myelofibrosis Research and Treatment. Blood 2009;113:2895e901. [57] Barbui T, Thiele J, Vannucchi AM, et al. Problems and pitfalls regarding WHO-defined diagnosis of early/prefibrotic primary myelofibrosis versus essential thrombocythemia. Leukemia 2013;27:1953e8.