FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2021336654> ?p ?o ?g. }

• W2021336654 endingPage "808" @default.
• W2021336654 startingPage "796" @default.
• W2021336654 abstract "Essential or primary thrombocythaemia (ET) is one of the myeloproliferative disorders (MPDs) which includes polycythaemia vera (PV), chronic idiopathic myelofibrosis (MF) and chronic myeloid leukaemia (CML). Epstein & Goedel (1934) first used the term ‘haemorrhagic thrombocythaemia’ to describe a disorder characterized by ‘permanent elevation of the platelet count to more than three times normal values, hyperplasia of megakaryocytes and the tendency for venous thrombosis and spontaneous haemorrhage’. Much later, Ozer et al (1960) and Gunz (1960) described two series of patients, reviewed the world literature and defined ET as an independent clinical entity. In the last 40 years some progress has been made in our understanding of ET and its pathogenesis; as expected there are more treatment options and we may be wiser regarding their best use. In this review, which relates to adult ET, common themes have been discerned and then applied first to diagnostic tests and to pathogenesis, which are merged together because of their considerable overlap, then to clinical features and what is known about the factors influencing their occurrence, and finally to consider prognosis and treatment. The gold standard diagnostic criteria for ET remain those proposed by the Polycythaemia Vera Study Group (PVSG), initially in 1976, most recently modified in 1997 and shown in Table I (Murphy et al, 1997). The major hurdle in making a diagnosis of ET is the lack of a hallmark feature analogous to the presence of bcr/abl transcripts in CML. During the past decade several possible diagnostic features have been suggested and will now be examined in turn. Many of these potential diagnostic features have important implications for our understanding and further investigation of the pathogenesis of ET, and therefore this is also discussed. Numerous authors (reviewed in Westwood & Pearson, 1996) have espoused the use of spontaneous megakaryocyte colonies to diagnose ET. However, these techniques are fraught with many potential pitfalls including hypersensitivity to trace cytokines, lack of standardization, a requirement for positive identification and not least the reported occurrence of such colonies in reactive thrombocytosis (RT) (Westwood & Pearson, 1996). A recent report focused upon the hypersensitivity of mononuclear cells derived from the blood of ET patients to thrombopoietin (TPO) and proposed this as a potential positive diagnostic hallmark (Axelrad et al, 2000). However, as only 18/26 ET patients displayed such hypersensitivity to TPO, which was also observed in 1/9 patients with RT, it would be difficult to support the use of this strategy to confirm a diagnosis of ET. The pathogenic insights provided from this data remain unclear as unselected mononuclear cells, and not CD34+ cells, were evaluated and many of the patients were treated with hydroxyurea (HU), which is likely to confound the results. The role of the trephine biopsy in establishing the diagnosis of ET has traditionally been to exclude MF; however, a number of workers would now propose its inclusion to positively establish the presence of ET. Furthermore, some (Lengfelder et al, 1998; Sacchi et al, 2000) have suggested that interpretation of the trephine could be used to facilitate earlier diagnosis in those patients with a platelet count of 400–600 × 109/l. However, the morphological criteria used in such circumstances have neither been clearly defined (an increase in megakaryopoiesis appears most consistently reported) nor compared with those patients with a reactive thrombocytosis in a blinded manner. Two recent reports have also suggested using trephine features to identify a disease subtype associated with an inferior prognosis, as discussed later (Annaloro et al, 1999; Thiele et al, 2000). These proposals, although under review, have yet to gain widespread credence, with many haematologists being reluctant to assign a diagnosis of ET on the basis of a trephine biopsy in a patient not otherwise fulfilling the PVSG criteria. Clonal analysis using X chromosome inactivation patterns (XCIPs) has been investigated as a diagnostic tool in a variety of malignant disorders. Fialkow et al (1981) first reported results of XCIP analysis in three women with ET who were heterozygous for the glucose-6-phosphate dehydrogenase (G6PD) (A/B) polymorphism; only a single enzyme type was found in the neutrophils, red cells and platelets, whereas both types were found in skin samples. These results were interpreted as demonstrating the clonal nature of ET and suggesting that the disease is likely to arise in a multipotent stem cell. Subsequent studies upon patients with ET performed on unfractionated blood cells demonstrated imbalanced or skewed XCIP in most cases. However, as these techniques were increasingly applied, several important technical and biological factors have emerged which influence their interpretation. These are, first, the occurrence of constitutive skewing which necessitates appropriate reference to both the individuals' own constitutive XCIP and, second, the patients' age. XCIP analyses of haematological cells have revealed that about 20–25% of haematologically normal women have a constitutively skewed XCIP defined as > 75% expression of one allele (Gale et al, 1994, 1997). The most favoured explanation for the occurrence of this phenomenon is the small number of stem cells present at the time of Lyonization (Gale & Linch, 1994). As a consequence, the interpretation of neutrophil (or platelet) XCIPs must be by comparison with a tissue of close embryological origin, routinely T lymphocytes (Gale & Linch, 1994; El-Kassar et al, 1997). The phenomenon of age-related skewing of myeloid cells has important implications both for the interpretation of XCIPs in the elderly and the usefulness of these techniques in older women (Gale et al, 1997; Champion et al, 1997). For example, an imbalanced myeloid XCIP in the presence of a balanced T-cell XCIP cannot be interpreted in elderly women (arbitrarily > 65 years) as evidence of clonal haemopoiesis. Conversely, balanced myeloid XCIPs can be used to indicate the presence of polyclonal haemopoiesis. There are a restricted number of reports of the use of XCIP as a diagnostic tool in ET in which all these technical factors were taken into account. Our own study of 46 patients with ET diagnosed according to the PVSG criteria revealed some surprising results. Overall, 23/46 patients were not assessable due to constitutive skewing in 17 and possible age-related skewing in 6 patients. Less than half of the assessable patients had monoclonal XCIPs and 13 (28%) of the total group had polyclonal XCIP results (Harrison et al, 1999a). This finding of a significant proportion of patients with ET, an apparently malignant condition, having polyclonal XCIPs was unexpected. However, studies from French (El-Kassar et al, 1997) and Italian groups (Chiusolo et al, 2001) have reported similar results with 14/46 and 15/40, respectively, of ET patients having polyclonal XCIPs. What does this mean? There are many possible interpretations of these results which must first include the technical limitations of XCIP analysis, a relatively crude tool, which might be unable to detect a small population of clonal cells against a polyclonal background. Second, it is possible to interpret oligoclonal disease as polyclonal and this certainly cannot be excluded. Third, do these patients really have ET? In our own patient cohort, the follow up of these patients extended beyond 15 years and an occult cause of a reactive thrombocytosis should have declared itself during this period. Familial thrombocytosis exists and in one kindred has been shown to be polyclonal (Jorgensen et al, 1998), but this is a rare condition with only a limited number of families reported in the world literature. The potential clinical significance of the results of clonality assays became apparent when age and platelet count at diagnosis, length of follow up, incidence of hepato- or splenomegaly, and occurrence of thrombotic or haemorrhagic complications for clonal and polyclonal patients were compared. Thrombotic events were more common in the clonal patients and this difference was statistically significant (P = 0·039, Fisher's exact test) (Harrison et al, 1999a). The report from Rome (Chiusolo et al, 2001) also documented more frequent thrombosis in the monoclonal group (P = 0·04). As the primary aim of treatment in ET is to reduce the incidence of such complications, these results, if confirmed in future studies, could have an important impact in individually tailoring patient therapy. Clonal analysis using XCIP is thus of limited application in the diagnosis of ET by virtue of the difficulty in interpreting results in patients over 65 years and, furthermore, the presence of polyclonal haemopoiesis in a younger woman cannot be used to exclude a diagnosis of ET. Conversely, in a young woman with thrombosis in an unusual site and accompanying borderline thrombocytosis, the presence of clonal haemopoiesis could be used to support a diagnosis of ET. XCIP assays may also have a role in risk assessment for thrombosis and demonstrate a link between biological and clinical features, if preliminary results are confirmed. This biological heterogeneity will also have important impact upon the direction and interpretation of future investigation into the pathogenesis of ET. The increase in megakaryocytes and platelets, a characteristic feature of ET, could be due to perturbation of normal thrombopoiesis by a number of different mechanisms, including either increased sensitivity to stimulatory or reduced response to inhibitory cytokines, or growth factor independence. Most recently, it has been suggested that hypersensitivity is in fact limited to TPO (Axelrad et al, 2000). The successful isolation of TPO was made possible by the earlier description of its receptor c-Mpl (Souyri et al, 1990). This receptor was identified from a virus, which caused a myeloproliferative-like condition in infected mice and expressed on the viral envelope a cellular gene (v-Mpl for myeloproliferative leukaemia). This viral receptor v-Mpl is a truncated version of the cellular protein c-Mpl; it lacks the ligand-binding domain of its cellular homologue c-Mpl and is constitutionally active. Thrombopoietin is the primary regulator of thrombopoiesis and is therefore of potential interest in the pathogenesis of ET. The most widely supported mechanism of TPO regulation is the so called ‘Kuter and Rosenberg’ theory which proposes that TPO is downregulated by binding to its receptor c-Mpl that is expressed by CD34+ cells, megakaryocytes and platelets (de Gabrielle & Pennington, 1967; Kuter & Rosenberg, 1995). The measurement of TPO levels in a range of clinical conditions associated with thrombocytopenia lends strong support to this model of TPO regulation (Emmons et al, 1996; Tahara et al, 1996). However, Tahara et al (1996) first reported that in six patients with ET TPO levels (2·8 ± 1·55 fmol/l) appeared to be either normal or elevated, compared with 50 normal controls (0·79 ± 0·07 fmol/l). This finding has been confirmed in subsequent studies of ET and other MPD (Cerutti et al, 1997; Pitcher et al, 1997; Harrison et al, 1999b). These results do not support the favoured model of TPO regulation but neither do they exclude the possibility of a coexisting regulatory mechanism. Furthermore, as TPO levels are also elevated in other MPDs as well as patients with a reactive thrombocytosis, their measurement is not useful diagnostically (Cerutti et al, 1997; Harrison et al, 1999b). The descriptions (Jorgensen et al, 1998; Kondo et al, 1998; Wiestner et al, 1998; Ghilardi et al, 1999) of TPO mutations causing elevated TPO levels in familial thrombocytosis contribute to our knowledge of potential mechanisms of TPO regulation, as well as its dysregulation in disease states. Similar mutations, although not hitherto demonstrated (Harrison et al, 1998a; Allen et al, 2000), might be present in ‘sporadic’ ET, which could therefore account for the apparently paradoxical elevation of TPO levels in some patients. All these mutations enhance the likelihood of the initiation complex binding to the true start site for the TPO protein by removing the inhibitory affect of the seventh AUG codon. These are the first cases of gain-of-function mutations leading to systemic overproduction of a cytokine and hence disease. Furthermore, they also suggest the possibility that differential splicing of the 5′-UTR is a physiological means of TPO regulation, which certainly merits further analysis. The thrombopoietin receptor c-Mpl is a candidate oncogene in ET for a number of reasons including the occurrence of v-Mpl, which causes a fatal myeloproliferative condition in mice as well as its role in TPO homeostasis. A search for mutations in c-Mpl in ET by a number of groups has merely identified a number of common polymorphisms as well as new isoforms (Debili et al, 1996; Taylor et al, 1996; Kiladjian et al, 1997; Li et al, 2000). Thus far, it would appear that mutations of c-Mpl are unlikely to contribute to the pathogenesis of ET, but what about quantitative change? Reduced expression of c-Mpl in platelets from patients with ET was first reported by Li et al (1996). These preliminary findings suggested an explanation for the elevated TPO levels that might possibly relate to the pathogenesis of ET. Subsequently, several further studies have been published using different techniques which support these results (Horikawa et al, 1997; Harrison et al, 1999b; Li et al, 2000). In contrast, however, Moliterno et al (1998) reported marked reduction of c-Mpl expression in 34 PV patients and 13/14 MF patients, but not in nine ET patients. The reasons for these conflicting findings in ET remain unclear. Most studies rely upon rabbit anti-c-Mpl polyclonal antibodies, and potential differences between different sources may explain these discrepant results. An additional explanation may lie in the apparently heterogeneous nature of ET as defined by clonal analysis, although platelet c-Mpl expression did not appear to correlate with clonality (Harrison et al, 1999b). The difference between RT and ET patients in terms of their platelet c-Mpl expression was statistically significant, but there was sufficient overlap between the two groups which would reduce the usefulness of this measurement as a diagnostic tool (Harrison et al, 1999b). This requires confirmation in a larger prospective study. Reduced platelet c-Mpl expression in ET is unlikely to be a primary pathogenic event as it is also evident in other MPDs. However, it also seems unlikely to occur solely as a direct consequence of downregulation due to thrombocytosis per se, as patients with RT had platelet c-Mpl expression which was only slightly less than that of the normal controls. Furthermore, Horikawa et al (1997) demonstrated reduced c-Mpl mRNA levels in ET platelets, thus implying that the lower levels of the corresponding protein may not necessarily simply reflect ligand binding and downregulation. Perhaps whatever abnormalities underlie the myeloproliferative process could influence c-Mpl expression by virtue of some more widespread cellular effect. For example, post-translational processing of c-Mpl in PV is reportedly abnormal, with differential glycosylation correlating with stage of the disorder (Moliterno & Spivak, 1999). Most negative regulators of megakaryopoiesis are produced by megakaryocytes themselves. TGF β 1 is thought to be the most potent inhibitor of megakaryopoiesis, but platelet factor 4, β-thromboglobulin and connective tissue activating peptide III, all members of the chemokine family, also have inhibitory actions (Gewirtz et al, 1995). Han et al (1990) reported that platelet factor 4 was able to inhibit megakaryocyte colony growth to a normal degree in ET. Zauli et al (1993), using fibrin clot assays, found that the normal suppression of megakaryocyte growth in response to TGF β was reduced in ET, and that this reduction in inhibition was reversed by the addition of antibodies to TGF β. It is therefore possible that a blunted response to TGF β might be responsible for the hypersensitivity to other cytokines in ET. TGF β levels are known to be elevated in ET. However, this appears likely to be as a direct result of elevated platelet and megakaryocyte mass, as platelet lysates were shown to contain similar amounts of TGF β in samples from normal individuals as in those from ET patients (Zauli et al, 1993). TGF β RII is the principle TGF β receptor in haemopoietic cells and there was a provisional report of mutations in this receptor in two ET patients (Niitsu et al, 1998). However, our own work has subsequently suggested this is probably a polymorphism (Sud et al, 2000). Recently, it has been suggested that there may be an important interaction between TGF β and TPO in the bone marrow microenvironment, with TGF β stimulating local TPO production which in turn stimulates expression of TGF β receptor and hence potential susceptibility to growth inhibition (Sakamaki et al, 1999). Downregulation of Smad 4, a pivotal transcriptional activator in response to TGF β, in ET has also been provisionally reported (Kuroda et al, 1999) but not yet confirmed. Further studies of TGF β and downstream cellular events consequent upon its interacting with receptors in ET are warranted. Only a minority of ET patients have cytogenetic abnormalities; when present, however, these abnormalities are useful in confirming a diagnosis. Their overall significance in terms of their impact upon complications and prognosis remains unclear (Groupe Français de Cytogenetique Hematologique, 1998; Krsnik et al, 2001), although abnormalities of chromosome 17p are proposed to relate to both hydroxyurea therapy and secondary myeloid malignancy (Sterkers et al, 1998), as discussed later. There have only been limited studies of oncogenes and tumour suppresser genes in patients with ET. Neri et al (1996) studied 51 patients with ET for mutations in the Ras proto-oncogene and the p53 tumour suppresser gene. No Ras mutations were identified; mutations in p53 were found in three cases, one in an untreated patient, one in a patient treated with cytoreductive therapy and one in a patient who had transformed to acute myeloid leukaemia (AML). These results imply that p53 mutations may occur sporadically in patients with ET and that Ras genes are probably not involved in the pathogenesis of ET. Gaidano et al (1997), in contrast, only found p53 mutations in 4/10 patients in blast crisis and not in any chronic phase samples, nor in samples preceding those from the blastic phase. It is therefore possible that p53 mutations may be a feature of leukaemic transformation in ET. The p53 locus is located at 17p and, interestingly, abnormalities in this region have been documented in ET and PV patients treated with HU (Sterkers et al, 1998), as discussed later. The main clinical features of ET are thrombosis affecting micro- and major vessels (arterial and, less frequently, venous), haemorrhage and, in a proportion, transformation to either a myelofibrotic or leukaemic phase. A significant number of patients are asymptomatic at presentation and apparently remain so throughout the duration of their disease. The reported incidence of haemostatic complications varies; these differences probably relate to the fact that ET is a rare disorder, and the literature consists of relatively small patient cohorts and retrospective data. Symptoms are present at diagnosis in 40–90% of cases. Haemostatic complications, in particular thrombotic episodes, are the major cause of morbidity in patients with ET and also have a significant impact on survival (Lengfelder et al, 1998). Haemorrhage appears to be less common than thrombosis and has thus proved difficult to study. Most reports have failed to find a correlation between a broad array of specific structural, biochemical and metabolic platelet defects, and bleeding in ET (Schafer, 1984; Finazzi et al, 1996). Platelet function tests are therefore not generally considered to be useful as indicators of haemorrhagic risk in ET. In some ET patients with markedly elevated platelet counts in excess of 1500 × 109/l, a reduction in higher molecular weight multimers of von Willebrand factor may be detected and have been associated with bleeding symptoms (Budde et al, 1984). Major features associated with risk of bleeding therefore are platelet counts > 1000–1500 × 109/l, acquired von Willebrand disease and aspirin at doses in excess of 300 mg/d, as discussed later. Thrombotic events in ET are more common and, in a study of a 100 patient cohort (Cortelazzo et al, 1990), 20 thrombotic episodes were observed [6·6%/patient time at risk (patient year)]. When the incidence of thrombosis was analysed according to patient age, a rate of 1·7%/patient year was found in those < 40 years, 6·3%/patient year in the 40–60 year group, and 15·1%/patient year in those > 60 years (P < 0·001, compared with age < 40 years). Cortelazzo et al (1990) also found that the likelihood of thrombosis increased progressively in relation to previous thrombotic episodes, from 3·4%/patient year in patients without a previous thrombotic event to 31·4%/patient year in those with a history of thrombosis. This study therefore identified patients aged > 60 years with previous thrombotic episodes as a subset of patients at greater risk of future thrombotic complications; other studies have subsequently verified this (Lengfelder et al, 1998; Besses et al, 1999). However, it must be remembered that life-threatening thrombotic events do still occur in young patients without prior thrombosis, probably at greater incidence than the general population, although this is debated (Ruggeri et al, 1998; Pearson et al, 1999). A randomized controlled trial (Cortelazzo et al, 1995) demonstrated the benefit of cytoreductive therapy with HU in high-risk ET patients (> 60 years, prior thrombosis) with a target platelet value of < 600 × 109/l. These results were subsequently reported with longer follow up as incidence rates of 8%/patient-year in the untreated group and 1·5% in the HU-treated group (Finazzi et al, 2000). Other studies have suggested that there is an association between degree of thrombocytosis and risk of thrombosis (Lahuerta-Palacios et al, 1988), as well as demonstrating the benefit of cytoreductive treatment (Lengfelder et al, 1998). A recent report that thrombotic complications only occurred in young ET patients receiving cytoreductive therapy when the platelet count exceeded 400 × 109/l (Storen & Tefferi, 2001) supports controlling the platelet count to below this threshold. However, it is clear that thrombotic risk is not directly related to platelet count per se, and thrombosis does occur at relatively low platelet counts (Regev et al, 1997; Lengfelder et al, 1998; Saachi et al, 2000). Furthermore, HU affects other parameters which might influence thrombotic tendency, for example the neutrophil count and perhaps adhesion molecule expression. Nevertheless, in ET in which the sole abnormality may be thrombocytosis it would seem irrational not to ascribe the majority of thrombotic predisposition to this abnormality. The degree of thrombocytosis which confers greatest risk has not been established; the Medical Research Council (MRC) study has selected a threshold of 1000 × 109/l (vide infra), although elsewhere a less conservative 1500 × 109/l was occasionally suggested (Tefferi et al, 2000). A variety of risk factors for vascular disease, including hyperlipidaemia and hypertension, have been assessed in a number of studies with variable results. For example, in a subsequent study by Cortelazzo et al (1995) only cigarette smoking was associated with an increased risk in ET (P = 0·01). Diabetes was too infrequent to assess its contribution to thrombotic hazard, although all three diabetics had thrombotic events. Hypercholesterolaemia has subsequently been significantly associated with thrombosis in one study (Besses et al, 1999). Venous thromboembolism is increased in a variety of inherited alterations in proteins involved in the coagulation cascade or its regulation, including factor V Leiden, prothrombin G20210A gene mutation and methylene tetrahydrofolate reductase gene polymorphism. There appears to be little or no contribution to thrombotic risk by these factors in the few ET patients studied (Rintelen et al, 1996; Harrison et al, 1998b). However, venous thromboembolism is less common than arterial or microvascular thrombosis in ET. Furthermore, although hyperhomocysteinaemia has been documented in patients with ET and PV, this appears neither to be associated with thrombosis nor the MTHFR polymorphism (Gisslinger et al, 1999; Faurschou et al, 2000). Provisional data suggests an increased prevalence of antiphospholipid antibodies in ET associated with thrombosis providing a link between another known acquired risk factor for arterial and venous thrombosis and ET (Harrison & Machin, 1997). Other factors proposed to correlate with thrombotic complications include spontaneous megakaryocyte and/or endogenous erythroid colonies (Juvonen et al, 1993) and the presence of monoclonal haemopoiesis (Harrison et al, 1999a; Chiusolo et al, 2001). There may also be a correlation between histological features and occurrence of thrombotic complications (Annaloro et al, 1999); this is discussed in more detail later and remains to be confirmed. Risk factors for thrombosis that have been reported in patients with ET are shown in Table II. Whether the presence of one or more of these risk factors for thrombosis should alter the management of asymptomatic young patients is unclear. The natural history of ET also includes an innate tendency for the development of acute leukaemia, usually AML and myelofibrosis (MF). Indeed, as discussed later, it has been suggested that some ET patients may have a form of ‘prefibrotic’ MF rather than ‘true’ ET. The differential predisposition to undergo transformation to MF of these proposed entities remains undefined. The risk of AML and MF in untreated ET is hard to assess although there have been several reports of the occurrence of AML in untreated patients with ET (Andersson et al, 2000), but there is little information about myelofibrotic transformation. This may relate in part to the nature of this transition being more gradual than that to AML, but the reluctance to perform serial trephine biopsies compounds the problem. The development of AML in ET is significantly associated with cytogenetic abnormalities (Lofvenberg et al, 1990a) and treatment (see later). The pathogenesis of MF is thought to be due to platelet derived growth factor secreted by the megakaryocytes stimulating fibroblast proliferation (Groopman, 1980). The role of cytoreductive therapy in reducing the risk of myelofibrosis and potentially reversing its progression is controversial (Manoharan & Pitney, 1984; Lofvenberg et al, 1990b). There is a clear need for large longitudinal studies of patients with ET to contribute further to our knowledge in this area. For those of us looking after patients with ET it is apparent that many remain well with minimal treatment; some will, however, have a more aggressive disease course with frequent disabling complications or early transformation to MF or AML. Overall, there is little data pertaining directly to life expectancy in ET. This type of analysis was performed on 1067 MPD patients (247 with ET); actuarial survival probability was calculated from the expected survival of control population derived from age- and sex-specific Spanish life tables (Rozman et al, 1991). Over 10 years, 37 ET patients died compared with 30·31 expected from the life tables (O/E ratio 1·22, Χ2 0·17, P = 0·22). The conclusion from this study was that a diagnosis of ET did not result in significant loss of life expectancy. However, 10 years is a short time for many patients with this disease; the latency period for the development of myelofibrosis and leukaemia is likely to be 10 years or more, and this study also demonstrated that patients with myelofibrosis have far from a normal life expectancy. In contrast, the Olmsted County study suggested than survival for patients with ET was significantly worse than age- and sex-matched healthy controls (Mesa et al, 1999). Two studies have specifically studied younger ET patients and also report conflicting results. Tefferi et al (2001) suggested those < 48 years had a survival equivalent to that of an age- and sex-matched group. Bazzan et al (1999) suggested that the relative risk of mortality for ET patients < 55 years was fourfold that of healthy controls. There is a definite need for further studies with more substantial patient numbers and follow up. Unfortunately there are very few studies which contribute to our knowledge in this vital area. Lengfelder et al (1998) collected clinical data pertaining to 143 patients with ET. Multivariate analysis of factors influencing survival in this study revealed that age (P < 0·0001), absence of complications (P < 0·013) and cytoreductive therapy (P < 0·03) were significant, although the latter was only beneficial for patients < 60 years. Lack of benefit for those older than 60 years was suggested to relate to the higher median age in this study compared with those patients reported by Cortelazzo et al (1995) and possible higher incidence of age-related deaths. The trephine biopsy has recently been proposed to be of prognostic value in ET. Annaloro et al (1999) reviewed bone marrow biopsies for diagnostic as well as prognostic parameters. Patients with reticulin abnormalities, increased and dysplastic granulopoiesis, hypoplastic erythropoiesis, trapped but not large megakaryoc" @default.
• W2021336654 created "2016-06-24" @default.
• W2021336654 creator A5046827092 @default.
• W2021336654 date "2002-06-01" @default.
• W2021336654 modified "2023-10-17" @default.
• W2021336654 title "Current trends in essential thrombocythaemia" @default.
• W2021336654 cites W118940326 @default.
• W2021336654 cites W131887615 @default.
• W2021336654 cites W144517407 @default.
• W2021336654 cites W1493897517 @default.
• W2021336654 cites W1570611984 @default.
• W2021336654 cites W1571376117 @default.
• W2021336654 cites W166660488 @default.
• W2021336654 cites W167119892 @default.
• W2021336654 cites W1969637800 @default.
• W2021336654 cites W1972476718 @default.
• W2021336654 cites W1973034846 @default.
• W2021336654 cites W1974301854 @default.
• W2021336654 cites W1994668564 @default.
• W2021336654 cites W1996093792 @default.
• W2021336654 cites W1997218725 @default.
• W2021336654 cites W1997944331 @default.
• W2021336654 cites W1998325169 @default.
• W2021336654 cites W1998365820 @default.
• W2021336654 cites W2003011062 @default.
• W2021336654 cites W2003597939 @default.
• W2021336654 cites W2007353459 @default.
• W2021336654 cites W2010527821 @default.
• W2021336654 cites W2016417342 @default.
• W2021336654 cites W2016865426 @default.
• W2021336654 cites W2018868908 @default.
• W2021336654 cites W2024839136 @default.
• W2021336654 cites W2025975637 @default.
• W2021336654 cites W2026938652 @default.
• W2021336654 cites W2027201363 @default.
• W2021336654 cites W2027676334 @default.
• W2021336654 cites W2028835250 @default.
• W2021336654 cites W2029353341 @default.
• W2021336654 cites W2030371755 @default.
• W2021336654 cites W2030391513 @default.
• W2021336654 cites W2032738172 @default.
• W2021336654 cites W2038107955 @default.
• W2021336654 cites W2051540735 @default.
• W2021336654 cites W2052501792 @default.
• W2021336654 cites W2052905336 @default.
• W2021336654 cites W2059548130 @default.
• W2021336654 cites W2059903100 @default.
• W2021336654 cites W2059979402 @default.
• W2021336654 cites W2061071054 @default.
• W2021336654 cites W2066249318 @default.
• W2021336654 cites W2067426157 @default.
• W2021336654 cites W2068741393 @default.
• W2021336654 cites W2069996679 @default.
• W2021336654 cites W2072002296 @default.
• W2021336654 cites W2072367160 @default.
• W2021336654 cites W2073999714 @default.
• W2021336654 cites W2075207025 @default.
• W2021336654 cites W2075295588 @default.
• W2021336654 cites W2076145755 @default.
• W2021336654 cites W2076231724 @default.
• W2021336654 cites W2076689289 @default.
• W2021336654 cites W2077812003 @default.
• W2021336654 cites W2080968939 @default.
• W2021336654 cites W2092484653 @default.
• W2021336654 cites W2092622881 @default.
• W2021336654 cites W2093478866 @default.
• W2021336654 cites W2093493529 @default.
• W2021336654 cites W2093839397 @default.
• W2021336654 cites W2095188600 @default.
• W2021336654 cites W2095454219 @default.
• W2021336654 cites W2104843924 @default.
• W2021336654 cites W2105697793 @default.
• W2021336654 cites W2121310716 @default.
• W2021336654 cites W2123657626 @default.
• W2021336654 cites W2132430173 @default.
• W2021336654 cites W2134923454 @default.
• W2021336654 cites W2140818834 @default.
• W2021336654 cites W2148931009 @default.
• W2021336654 cites W2149377204 @default.
• W2021336654 cites W2151664419 @default.
• W2021336654 cites W2164702061 @default.
• W2021336654 cites W2167655217 @default.
• W2021336654 cites W2223901856 @default.
• W2021336654 cites W2229971338 @default.
• W2021336654 cites W2240871144 @default.
• W2021336654 cites W2281127542 @default.
• W2021336654 cites W2282276226 @default.
• W2021336654 cites W2313755278 @default.
• W2021336654 cites W2320032845 @default.
• W2021336654 cites W2321503226 @default.
• W2021336654 cites W2334781899 @default.
• W2021336654 cites W2462238653 @default.
• W2021336654 cites W2507320858 @default.
• W2021336654 cites W29651041 @default.
• W2021336654 cites W4229734315 @default.
• W2021336654 cites W4234642826 @default.
• W2021336654 cites W4235322334 @default.
• W2021336654 cites W4238316183 @default.

• next