Guideline for investigation and management of adults and children presenting with a thrombocytosis
Year: 2010
Type: article
Abstract: The guideline group was selected to be representative of UK-based medical experts. MEDLINE and EMBASE were searched systematically for publications in English from 1966 until October 2009 using a variety of key words. The writing group produced the draft guideline which was subsequently revised by consensus by members of the Haemostasis and Thrombosis and Haemato-oncology Task Forces of the British Committee for Standards in Haematology. The guideline was then reviewed by a sounding board of approximately 50 UK haematologists, the British Committee for Standards in Haematology (BCSH) and the British Society for Haematology Committee and comments incorporated where appropriate. Criteria used to quote levels and grades of evidence are as outlined in the Procedure for Guidelines Commissioned by the BCSH (Table I). The objective of this guideline is to provide healthcare professionals with clear guidance on the investigation and management of thrombocytosis in both adult and paediatric patients. The guidance may not be appropriate to all patients with thrombocytosis and in all cases individual patient circumstances may dictate an alternative approach. There is no previous guideline for this topic. The purpose of this guideline is to provide an approach to the diagnosis, investigation and management of patients with a thrombocytosis (i.e., a platelet count >450 × 109/l). This will include advice on how to distinguish reactive thrombocytosis from true haematological disease and how to distinguish essential thrombocythaemia (ET) from other myeloproliferative neoplasms (MPN), myelodysplastic syndromes (MDS) or overlap syndromes. Recommendations for the management of complications of ET, such as splanchnic vein thrombosis, blast crisis and post-ET myelofibrosis are given, as well as advice on management of pregnancy and children with ET. Thrombocytosis is a common finding and is a frequent cause of referral for further investigation. There is a wide range of primary and secondary causes as well as false or ‘spurious’ conditions mimicking thrombocytosis (Table II). Establishing the cause therefore requires consideration of clinical features, haematological parameters, bone marrow aspirate and trephine biopsy morphological features and the presence or absence of clonal genetic abnormalities. The definitions of specific neoplastic entities are guided by diagnostic criteria and algorithms within the world Health Organization (WHO) Classification of Tumours of Haematopoietic and Lymphoid Tissues (Swerdlow, 2008). These definitions are based on combinations of clinical and pathological characteristics which may be major (required) or minor (supportive). This most recent iteration of the WHO classification emphasizes the neoplastic nature of the previously termed myeloproliferative diseases and renamed them MPN. The guideline group however discussed and agreed specific diagnostic criteria for ET (Table III) that are subtly different from those of the WHO 2008 classification based upon evidence from the analysis of data from a randomized controlled trial (Wilkins et al, 2008). The diagnostic features, including detailed morphological descriptions of the common myeloid entities associated with thrombocytosis, are provided below and are intended to act as a useful reference to supplement, but not replace, the WHO monograph. The guideline group also developed a diagnostic algorithm to synthesize practice in undertaking the investigation of thrombocytosis (Fig 1). Diagnostic pathway for the investigation of thrombocytosis. *BCR-ABL1 testing is only recommended if there are atypical features on the blood count and/or film (e.g. basophilia). †Bone marrow examination is recommended to confirm the diagnosis, as per the WHO Classification. In some circumstances this may not be clinically indicated. ‡The need for Cytogenetics on the bone marrow should be guided by the blood and bone marrow morphology. The quantification of stromal reticulin fibres and detection of collagen fibrosis are fundamental to the classification and assessment of progression in MPN. Several different semi-quantitative methods for grading reticulin and collagen fibrosis have been developed, ranging from 4 to 6 grades (reviewed in Kuter et al, 2007). The WHO 2008 classification has adopted the European consensus grading scheme (Thiele et al, 2005), developed specifically in the context of MPN trephine histology assessment. This defines four grades, grade 0 being normal and grades 1–3 representing progressive increments above normal. Both grade 2 and grade 3 allow the presence of varying amounts of collagen fibrosis. While the various systems have not been compared in prospective studies, this scheme has the potential advantage of allowing separation of cases with small foci of collagen from those with more extensive fibrosis. A description of this scoring system is illustrated in Fig 2 and in Table IV. Reticulin grading: examples of reticulin stains. A. Grade 0/3 scattered linear reticulin with no intersections (cross-overs) corresponding to normal one marrow; B. Grade 1/3 loose network of reticulin with many intersections, especially in perivascular areas; C. Grade 2/3 widespread and dense increase in reticulin with extensive intersections; D. Grade 3/3 diffuse and dense increase in reticulin with extensive intersections; E. Grade 3/3 coarse bundles of collagen demonstrated; F. Collagen demonstrated using MSB trichrome stain (martius yellow, brilliant crystal scarlet 6R, and soluble blue). All panels original magnification ×40. The most common secondary (or reactive) causes of thrombocytosis are infection, inflammation, iron deficiency, tissue damage, haemolysis, severe exercise, malignancy, hyposplenism and other causes of an acute phase response (Table II). These are usually, but not always, characterized by an elevated C-reactive protein, or erythrocyte sedimentation rate. The platelets are mostly small with a normal mean platelet volume. The blood film may show other features to indicate an underlying cause, including acute infective, or inflammatory, processes. A bone marrow aspirate or trephine is not usually required for reactive thrombocytosis. If one has been performed due to diagnostic uncertainty, this will show megakaryocytic hyperplasia with normal mature and left-shifted megakaryocyte morphology. The megakaryocytes will have a normal interstitial distribution and not show clustering. Reticulin is typically not increased. In chronic infective or inflammatory processes there may also be granulocytic hyperplasia and features of the anaemia of chronic disease. The diagnosis of ET requires a sustained thrombocytosis of >450 × 109/l and the exclusion of reactive causes. The blood film shows a thrombocytosis with varying degrees of platelet anisocytosis. Platelet morphology can vary from those of normal size and granulation to larger atypical forms that may be hypogranular. Features of iron deficiency may be seen if there has been an associated chronic blood loss: while iron deficiency may also mask an underlying polycythaemia vera (PV). Leucoerythroblastosis and poikilocytosis are not seen in ET. An elevated white blood cell count may be detected. Bone marrow examination (aspirate and trephine biopsy) is required according to the WHO classification (Swerdlow, 2008) to make the diagnosis of ET. However, in elderly patients where a clonal marker, such as JAK2 V617F or MPL 515L/K has been detected, without features suspicious of MDS or primary myelofibrosis (PMF), a bone marrow examination may not be necessary. This represents a departure from the WHO classification. However the guidelines group agreed that this was justified for at least two reasons; first, in view of data from the PT-1 histology review demonstrating poor reproducibility in interpreting some trephine biopsy features (Wilkins et al, 2008) and second, that the most recent WHO diagnostic criteria for all stages of PMF requires the presence of additional features such as splenomegaly, leucoerythroblastic film, or significant constitutional symptoms, that bone marrow examination was not always necessary. Nevertheless it is recommended to perform a bone marrow biopsy where there are atypical features, or if during the course of treatment a change in management is planned, such as change of cytoreductive therapy, or if transformation is suspected. The guideline group therefore proposed a modification of the WHO diagnostic criteria for ET (Table III). In ET, the bone marrow is normocellular for age or mildly hypercellular. Megakaryocytes are increased in number and mostly present as single cells; they may form occasional small, loose clusters within the interstitium of the marrow. Large or giant megakaryocytes with hyperlobated (‘staghorn’) nuclei are prominent and these may show increased emperipolesis. The WHO Classification states that large or giant megakaryocytes predominate but smaller forms may also be seen, particularly if immunohistochemistry is used to aid megakaryocyte identification in trephine biopsy sections. Megakaryocytes with normal morphology, including nearly-bare end-stage variants with pyknotic nuclei are also present. This spectrum of megakaryocyte morphology is typical of ET. Erythropoiesis and granulopoiesis are generally normal. Significant dyserythropoiesis and/or dysgranulopoiesis are not features of ET and favour myelodysplasia. Iron stores may be reduced and siderotic granulation is normal. Reticulin is generally not increased (grades 0–2/4 or grade 0/3, depending on the grading system used, see above). A clonal genetic abnormality can be demonstrated in approximately 60% of cases of ET. The JAK2 V617F mutation is detectable in 50% and the MPL mutation in up to 10% (Pardanani et al, 2006; Pikman et al, 2006; Beer et al, 2008; Vannucchi et al, 2008). The presence of the BCR-ABL1 rearrangement (diagnostic of chronic myeloid leukaemia) excludes ET and testing for this should be performed if atypical features are present, such as basophilia, left shift of neutrophils or atypical trephine features. Routine karyotyping is not always required but a karyotypic abnormality where present may be a useful future marker for disease progression. PV is characterized by an elevated haemoglobin, haematocrit and red cell mass, although up to 15% of patients may present with a marked thrombocytosis and clinical features such as pruritis would add weight to a diagnosis of PV. Bone marrow examination (though not always essential) shows a pan-myelosis with normal erythroid and granulocytic differentiation but with spatial disorganisation. In contrast to ET, the megakaryocytes show marked pleomorphism with a higher than normal nuclear:cytoplasmic ratio plus a characteristic mixture of large and small variants; loose clustering of megakaryocyte is usual. The giant megakaryocytes with hyperlobated nuclei that are a feature of ET are not seen in PV. Iron deficiency is common in those patients whose presentation overlaps with ET and should be investigated to exclude a second pathology. The JAK2 V617F mutation is present in more than 97% of PV patients and a further 1–2% will have an exon 12 mutation of JAK2 (Scott et al, 2007),(Pietra et al, 2008). Exon 12 mutations have not been documented in the context of ET or PMF. The ability to discriminate between ET and PV on the basis of trephine evaluation has never been tested, or evaluated in a formal setting. Primary myelofibrosis can present with a marked thrombocytosis, with or without anaemia, teardrop poikilocytosis, leucocytosis and a leuco-erythroblastic blood film. Clinical features again are similar to ET. Platelets tend to show marked anisocytosis with large forms. Bone marrow aspiration may yield a ‘dry tap’ or show hypercellular particles with hypocellular trails. Bone marrow trephine histology shows a spectrum of findings including increased overall cellularity, particularly of granulocytes and megakaryocytes: commonly, there is also reduced erythropoiesis. Granulopoiesis may have a disordered distribution and may be left-shifted but is usually not dysplastic. In PMF the megakaryocytes are markedly abnormal. They have an abnormal distribution and pattern, including adjacency to bone trabeculae and within sinuses. Megakaryocytes may form dense and often large clusters, sometimes forming sheets and have atypical morphology, with the majority being large and having distorted elongated and angular shapes. They have a high nuclear:cytoplasmic ratio and their nuclei are commonly enlarged, hyperchromatic and poorly lobated. Reticulin fibrosis is increased (grade 2/4, grade 1/3 or above) and may be accompanied by overt collagen and/or new bone formation. At its most extreme, there may be markedly reduced haemopoietic cellularity and osteosclerosis. As for ET, the diagnosis of primary myelofibrosis requires the demonstration of a clonal marker (e.g. JAK2 V617F which is present in 50–60% of cases or MPL mutations in 10% (Pardanani et al, 2006; Pikman et al, 2006; Lasho et al, 2006) and the exclusion of PV, chronic myeloid leukaemia (CML) and MDS. Although the 2008 WHO classification includes ‘pre-fibrotic/early stage myelofibrosis’, there is contradictory evidence as to whether this is a distinct entity which requires distinction from ET (Thiele & Kvasnicka, 2003; Gianelli et al, 2006; Wilkins et al, 2008). In the new WHO classification (Swerdlow, 2008) a diagnosis of PMF, whatever the grade, requires the presence of other features, such as splenomegaly, a leucoerythroblastic blood film and constitutional symptoms. This new classification has resolved some uncertainty regarding the classification of patients with thrombocytosis who have increased reticulin fibrosis in the marrow but no other features of PMF – provided such patients have other features consistent with ET they should be managed according to protocols for this condition. The existence of diseases that can demonstrate the clinical and laboratory features of both MDS and a chronic MPN, have been recognized within the WHO Classification of Tumours (Swerdlow, 2008). These entities have been termed the myelodysplastic syndromes/myeloproliferative neoplasms (MDS/MPN). Care should be taken to ensure that those patients who have a clear history of a preceding myeloproliferative disorder with evidence of subsequent dysplastic transformation, are not included in this group, as the evidence base for their clinical management is different. The MDS/MPN overlap category has been divided into four separate sub-groups: chronic myelomonocytic leukaemia (CMML); atypical CML (aCML); juvenile myelomonocytic leukaemia (JMML); and MDS/MPN, unclassifiable (MDS/MPN-U). Within the latter designation, there has been recognition of the provisional entity of refractory anaemia with ringed sideroblasts associated with marked thrombocytosis (RARS-T). With the exception of RARS-T, where the presence of thrombocytosis is a diagnostic criterion, the other MDS/MPN entities are more commonly associated with thrombocytopenia. JMML is a rare leukaemic disorder of young children, and will not be discussed further in these guidelines. Chromosomal anomalies involving the 3q21–26 locus may be associated with thrombocytosis and should be categorized as MDS unless the blast count is >20%, when the disease is classed as AML (Swerdlow, 2008). Myelodysplastic/myeloproliferative neoplasm; provisional entity – refractory anaemia with ring sideroblasts associated with marked thrombocytosis (RARS-T). This is a provisional entity which has features of both ET and refractory anaemia with ring sideroblasts. Platelet morphology is normal. Red cells are dimorphic and although the majority of cells are normochromic and normocytic, there are small numbers of hypochromic and markedly microcytic cells. The bone marrow (aspirate and trephine biopsy) is hypercellular and shows increased megakaryocytes with morphological features similar to those seen in ET or PMF. Megakaryocyte clustering can be present. There is also increased erythropoiesis with dys-erythropoiesis and ring sideroblasts accounting for more than 15% of late normoblasts. Reticulin varies, from normal to moderately increased. The JAK2 V617F mutation is present in more than 50% of patients (Boissinot et al, 2006; Szpurka et al, 2006). Myelodysplastic Syndrome associated with isolated del(5q). Myelodysplastic syndrome with isolated del(5q) is commonly associated with thrombocytosis (30–50% of patients) and macrocytic anaemia. The platelet morphology is unremarkable whilst the red cells are macrocytic, with mild poikilocytosis and minimal polychromasia. The bone marrow is normocellular, or mildly hypercellular, with dyserythropoiesis. Megakaryocytes are increased in number and are present in the marrow interstitium; they are small or of normal size and have monolobed, or hypolobated nuclei that are characteristically eccentrically placed. Cytogenetic demonstration of del(5q), with loss of bands q31–q33 and without additional complex cytogenetic abnormalities, is required to make the diagnosis. A small number of patients may also have the JAK2 V617F mutation (Ingram et al, 2006). Thrombocytosis is a common finding with a wide range of primary and secondary causes as well as false or ‘spurious’ conditions mimicking thrombocytosis, evaluation of these patients therefore requires a comprehensive approach involving clinical and laboratory parameters The guideline group also developed a diagnostic algorithm to synthesize practice in undertaking the investigation of thrombocytosis (Fig 1). Limitation of life expectancy is a major concern for many patients with ET, especially at the time of diagnosis and unfortunately data on this is scarce and sometimes conflicting (Rozman et al, 1991; Jensen et al, 2000; Passamonti et al, 2004; Wolanskyj et al, 2006). Life expectancy for the first 10 years after a diagnosis of ET is not affected, beyond 10 years the data is less clear. For example, a study from the Mayo clinic involving 322 ET patients followed for a median of 13·6 years, suggested that survival became significantly worse after the first decade (Wolanskyj et al, 2006). However, this evidence from the Mayo clinic study is likely to be influenced by limited knowledge regarding appropriate treatment at that time with use of agents known to increase the risk of leukaemic transformation (Campbell & Green, 2006). In the study by Passamonti et al (2004), ET patients at all ages had a standardized mortality ratio of 1, which was no different from an age- and sex-matched normal population. However, other authors suggest that young patients, who have a longer duration of disease, will accumulate more risk of transformation to myelofibrosis or AML (Passamonti et al, 2004; varez-Larran et al, 2007). Emerging evidence suggests that some patients with ET have a familial component to their disease, although they still appear to have the same risk of vascular complications and evolution to myelofibrosis and AML as classical ET (Rumi et al, 2007). The proposal that various factors may influence prognosis is attractive as this facilitates identification of patients who might be suitable for more aggressive therapies. The major disease-related events that impact upon survival and quality of life for patients with ET are the occurrence of thrombosis, and transformation to leukaemia or myelofibrosis. Fatigue and other constitutional symptoms can also impact upon quality of life and there is data to suggest that these symptoms may be relatively common amongst the MPN patient population (Mesa et al, 2007a), and are therefore worthy of further study. The most frequent complication of ET is thrombosis. For example, in the study reported by Passamonti et al (2004), the rates of thrombosis for ET patients were 12 per 1000 patients years, compared with 1·6 and 1·2 per 1000 patient years for PMF and AML, respectively. Consequently, risk stratification based upon the estimated chance of has been a useful to the management of ET. risk stratification for thrombosis in ET includes an assessment of history of thrombosis, medical conditions and the platelet suggest that patients years and with no thrombosis not show an increased of thrombosis compared to et al, in contrast to the findings of other et al, Clinical suggested that cytoreductive reduced the of thrombosis, which to the that this was secondary to a in the platelet count et al, & et al, platelet count not with the of thrombosis et al, the in the occurrence of events in the two of the PT-1 of the platelet suggests that additional such as of count or such as in the of thrombosis et al, The impact of risk factors for including and have been in MPN with et al, et al, has been specifically performed in ET. for the management of suggest that this patient group may from aggressive risk management with the use of and a where appropriate. The of in patients with but without an MPN has been as it not usually management et al, A single suggests that may be more common in MPN patients with et al, although more are as is not a risk for in patients without an MPN. if this finding is the presence of a genetic is to clinical management of in this An increased of syndrome has been in ET and is associated with an increased risk of both and thrombosis et al, et al, et al, 2007). However for is not indicated. with or should be managed according to guidelines for this et al, The identification of risk for thrombosis is an and may to further risk by which to However such should be and data is discussed blood cell In a study of patients, a normal white blood cell count at presentation to a high was associated with a influence on survival in but not analysis et al, recent that a white blood cell count above × 109/l was associated with complications (Wolanskyj et al, 2006; et al, In latter the of leucocytosis and thrombosis was more in the group, as age years with no et al, The potential of the count to patients is attractive but requires a large prospective study. JAK2 A large number of have the impact of the JAK2 V617F mutation upon the clinical of ET patients. These that JAK2 ET patients had a higher risk of thrombosis. A recent suggests that there was a significant increase in the ratio for thrombosis and transformation to PV in JAK2 ET patients et al, a in a second analysis et al, In a study of ET patients, Vannucchi et al that and symptoms were significantly more common the small number of ET patients with and V617F respectively. also with a higher of thrombosis at diagnosis and this was in analysis with a risk of in patients having a However, other have not this and that this have to be MPL In the PT-1 patients with MPL mutations had an increased risk of when compared to JAK2 patients although this on analysis et al, 2008). In a subsequent study et al, 2008) the presence of MPL mutations was associated with but not with other At present, it that MPL which are relatively in ET, not add with to or should be according to their risk of complications Grade The most risk stratification is as who are years of age have had an or who have a platelet count of × For patients who have no high risk features. This group may be further by age patients years of age with no high risk features, and patients years with no high risk features symptoms are not generally as events for the purpose of risk classification but if they are severe or not to the patient be as Grade Platelet count per not well with a platelet count × 109/l has been used as an for cytoreductive in view of the increased risk Grade In young patients it may be to use a higher platelet for risk classification in the absence of symptoms. Grade The impact of risk factors on risk assessment in or risk patients years and no Grade Emerging risk factors include the count and JAK2 both of which prospective with Grade prospective randomized that potential treatment for patients with ET have been An study randomized patients years or between treatment with and no cytoreductive et al, With a median up duration of patients on This study demonstrated that for patients with ET, cytoreductive with vascular The second randomized study is the primary trial et al, In this patients thrombosis, or age years, or or or platelet count × with ET according to the criteria et al, were randomized to plus or plus to plus patients with plus higher rates of thrombosis, major and transformation, but a of These of the platelet count from 6 after trial These can be compared with the randomized study. The of first thrombosis at 2 years was for those in both studies, the two of patients were For the the rates of first thrombosis at 2 years were and for patients plus or no cytoreductive respectively. Although such have their these data suggest for high risk patients, plus at least thrombosis, but this is not as as plus for patients. randomized will further the management of ET. The first is the study for risk patients no high risk This is a randomized of plus with is to in the of this study are treatment guidelines for this risk group of patients poorly The second trial is the in study which has its This study is as a between and in
Source: British Journal of Haematology
Authors: Claire N. Harrison, David Bareford, Nauman Butt, Peter Campbell, Eibhlean Conneally, Mark Drummond, Wendy Erber, Tamara Everington, Anthony R. Green, Georgina W. Hall, Beverley J. Hunt, Christopher A. Ludlam, Richard Murrin, Catherine Nelson‐Piercy, Deepti H. Radia, John T. Reilly, Jon Van der Walt, Bridget Wilkins, Mary F. McMullin
Institutions: Guy's and St Thomas' NHS Foundation Trust, Russells Hall Hospital, Arrowe Park Hospital, Wellcome Sanger Institute, St. James's Hospital, Gartnavel General Hospital, Addenbrooke's Hospital, Salisbury NHS Foundation Trust, University of Cambridge, John Radcliffe Hospital, Edinburgh Royal Infirmary, Sandwell & West Birmingham Hospitals NHS Trust, Royal Hallamshire Hospital, Belfast City Hospital
Language: English
Subfield: Hematology
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2010992436