Investigation and management of thrombocytosis without <scp> <i>JAK2</i> </scp> , <scp> <i>CALR</i> </scp> or <scp> <i>MPL</i> </scp> mutations: A British Society for Haematology Guideline

This guideline was compiled according to the BSH process at (https://b-s-h.org.uk/media/16732/bsh-guidance-development-process-dec-5-18.pdf). The Grading of Recommendations Assessment, Development and Evaluation (GRADE) nomenclature was used to evaluate the levels of evidence and to assess the strength of recommendations. The GRADE criteria can be found at http://www.gradeworkinggroup.org and are summarised in appendix 3 of the guidance document linked above. Literature review details are in the Supplementary Appendix. Review of the manuscript was performed by the British Society for Haematology (BSH) General Haematology Task Force and BSH Guidelines Committee. It was also on the members' section of the BSH website for comment. It has been reviewed by members of the UK Blood Cancer Research Network MPN Group, the NHS England Haematological Malignancies Genomics Working Group and patient representatives from MPN Voice; these organisations do not necessarily approve or endorse the contents. Since previous BSH guidelines covering investigation and management of thrombocytosis,1-3 there have been considerable advances in the biology, diagnosis and management of essential thrombocythaemia (ET), particularly relating to myeloproliferative neoplasm (MPN) driver mutations in JAK2, CALR and MPL. However, many patients with unexplained thrombocytosis are negative for these mutations. Optimal application and interpretation of bone marrow biopsy and broader genomic testing have not been defined in this context and real-world practice varies.4 This guideline provides expert consensus on investigation and management of patients with thrombocytosis lacking classical MPN driver mutations, focusing on practical diagnostic and management strategies appropriate for UK haematology practice. Table 1 shows the main primary and secondary causes of thrombocytosis. Over 80% of thrombocytosis is secondary to causes such as iron deficiency, infection, cancer or inflammation (‘reactive’ thrombocytosis, Table 1).6 Thrombopoietin (TPO), the primary regulator of megakaryopoiesis, is predominantly produced by the liver and cleared from plasma via receptors (thrombopoietin receptor, TPOR, encoded by the MPL gene) on megakaryocytes and platelets.7 Plasma TPO concentration is therefore regulated by platelet mass, although hepatic production is also increased by acute phase reactants (e.g. interleukin-6). Additionally, other hormones and cytokines can modulate megakaryopoiesis independently of TPO. There is an association between platelet count and increased body mass index, most significant in pre-menopausal women8, 9 and with metabolic syndrome,10, 11 although there are no defined ‘normal ranges’ for these contexts. Thrombocytosis is a cardinal feature of ET and can occur in polycythaemia vera (PV) and primary myelofibrosis (PMF). Constitutive activation of MPL/TPOR signalling is driven by recurrent somatic mutations in JAK2, CALR and MPL,12 most often JAK2 c.1849G>T (p.Val617Phe, JAK2 V617F; 50%–60% of ET), frameshift variants in CALR exon 9 (25%–35% of ET) or MPL exon 10 variants (particularly Trp515/Ser505; 5%–10% of ET). Non-hotspot JAK2 or MPL mutations occur in a small minority.4, 13, 14 ET cases lacking JAK2, CALR and MPL mutations (‘triple-negative’) comprise 10%–15% of cases13, 15 and are diagnosed on histology.16, 17 Most of these patients, who are predominantly young females, lack definitive evidence of a clonal origin to their thrombocytosis.18-20 Compared to other ET genotypes, triple-negative patients exhibit lower risks of myelofibrotic transformation and thrombosis, with highly favourable survival.13, 15, 21, 22 Less common clonal causes of thrombocytosis include BCR::ABL1-positive chronic myeloid leukaemia (CML), myelodysplastic neoplasms (MDS) with chromosome 5q deletion and MDS/MPN overlap syndromes including chronic myelomonocytic leukaemia and MDS/MPN with SF3B1 mutation and thrombocytosis. Very rarely thrombocytosis is a feature of myeloid/lymphoid neoplasms with gene rearrangements, such as ETV6::ABL1 and JAK2 rearrangements,16 usually associated with additional blood count abnormalities. Clonal haematopoiesis (CH) refers to the presence of clonal blood cell populations in otherwise healthy individuals, often driven by somatic mutations in a set of established genes, for example, DNMT3A, TET2 and ASXL1.16, 23-26 CH increases with age in both individuals with and without thrombocytosis and is commonly detected in older patients undergoing investigation for thrombocytosis.4, 14, 27, 28 Initial studies estimated the prevalence of CH at ~10% of healthy individuals aged over 70.23 With more sensitive sequencing techniques detecting mutations at variant allele fractions (VAFs) <1%, prevalence increases to over two-thirds, if not all, elderly individuals.29, 30 There are no clear platelet count differences between individuals with and without CH (excluding JAK2)31 and whether common CH-associated mutations directly drive thrombocytosis in some patients is currently unknown. However, a finding of CH-associated mutations may indicate emergent haematological malignancy, particularly those present at higher VAF (e.g. >10%), in multiple genes, in less commonly mutated genes, in younger individuals and/or in the presence of additional raised counts and/or cytopenias. Of note, some genetic alterations, in particular SH2B3 mutations and chromosome 20 deletions, while not typical primary MPN drivers, have been reported in triple-negative ET and share mechanisms that could drive thrombocytosis.13, 32, 33 High penetrance germline variants in several genes (e.g. SH2B3, EGLN1/EPAS1, 14q32.2 duplication) can predispose to the development of acquired MPN in rare cases.34, 35 Individuals often have a strong family history of MPN, haematological malignancy or other cancers. Germline genetic factors also contribute to the natural platelet count variation between individuals, through largely polygenic and complex mechanisms.36, 37 Hereditary thrombocytosis (HT), with autosomal dominant or recessive inheritance, comprises a rare, heterogeneous group of disorders most commonly driven by pathogenic germline variants in JAK2, MPL or THPO38-42 (Table 2). Individual familial cohorts have been reported where germline SH2B343 or GSN44 variants clustered with the HT phenotype, although these are likely to be very rare.45 The clinical phenotype in HT appears variable; most patients are asymptomatic and identified incidentally.46 Some mutations are associated with extra-haematological effects such as distal limb defects (THPO variants),47 autoimmune manifestations and hepatosplenomegaly (biallelic SH2B3 variants).43 A minority of reports have described increased thrombotic and/or haemorrhagic risk; however, studies are limited by small sample size, short follow-up and insufficient data on additional risk factors. During investigation of thrombocytosis, uncommon germline variants in JAK2, MPL, THPO or SH2B3 may be identified by next-generation sequencing (NGS). Lack of high-quality variant-specific data on associated clinical phenotypes and long-term outcomes can make interpretation of their clinical significance challenging. There are also no defined criteria to discriminate between a thrombocytosis ‘trait’ (a genetic characteristic, e.g. platelet count slightly above normal range) and a hereditary ‘disorder’. Positive family history or long history of stable thrombocytosis may increase suspicion for HT. Patients with persistent isolated thrombocytosis (platelet count >450 × 109/L) should be reviewed clinically and secondary causes (Table 1) considered with a thorough history and examination. Assessment should include symptoms of headaches, erythromelalgia or other microvascular disturbances, cardiovascular risk factors, thrombotic or haemorrhagic events, family history of thrombocytosis, body mass index, palpable splenomegaly and targeted assessment of any symptoms or signs suggesting non-haematological malignancy.65, 66 First-line investigations (Figure 1) include blood film examination for spurious thrombocytosis (e.g. cryoglobulinaemia) or features of other myeloid diseases (e.g. myeloid left shift, basophilia, monocytosis, dysplasia), iron studies and inflammatory markers (e.g. C-reactive protein). Imaging should be considered, including chest X-ray for patients aged over 40, abdominal ultrasound if splenomegaly cannot be excluded clinically, and other targeted investigations such as ultrasound or computed tomography (CT) imaging if non-haematological malignancy or other systemic disease is suspected clinically.67 Serum thrombopoietin levels do not distinguish between reactive and clonal thrombocytosis68, 69 and are not routinely recommended. If no secondary cause is found, peripheral blood assays for canonical mutations in JAK2, CALR and MPL should be undertaken. Recommended minimum limits of detection are 1%–3% VAF for JAK2 V617F and 5% VAF for CALR exon 9 and MPL exon 10 variants; methods are discussed more fully in the BSH Good Practice Paper.71 Laboratories are increasingly moving to peripheral blood screening with NGS mini-panels, which often sequence larger gene regions than recognised hotspots and may detect non-canonical mutations outside JAK2 V617 and MPL W515/S505, in a minority of patients investigated.4 Clinicians should be aware of their laboratory's peripheral blood screening techniques, including gene regions covered and sensitivity, since these may influence subsequent testing strategies. BCR::ABL1 rearrangement should be excluded if JAK2, CALR and MPL screening is negative, especially if there are other features suspicious for CML, using assays that detect typical and atypical transcripts.2, 72 Histological assessment of bone marrow (BM), requiring an adequate trephine biopsy specimen with high-quality sections and staining, is a mandatory feature of the World Health Organization (WHO) and International Consensus Classification (ICC) diagnostic criteria for ET regardless of molecular results, specifically the finding of megakaryocyte proliferation and increased enlarged, mature megakaryocytes with hyperlobated nuclei.16, 17 Most patients with isolated thrombocytosis, an acquired JAK2, CALR or MPL mutation and no other blood film abnormalities will have ET, but where a trephine biopsy is performed, this allows exclusion of differential diagnoses such as prefibrotic myelofibrosis and baseline fibrosis assessment. BM histology is also recommended where a highly sensitive peripheral blood screening assay has detected JAK2 V617F at low VAF (e.g. 0.5%–1.0%); co-existent CALR or MPL mutation should be excluded and additional genomic studies considered, as discussed below. Histology can be subtle in such cases and the reporting pathologist could be influenced by knowledge of molecular results. A diagnosis of CH rather than ET should be carefully considered if histology is inconclusive and/or there is an alternative reactive cause; interval trephine sampling may be helpful together with clinical monitoring for emergent MPN features. In investigation of triple-negative thrombocytosis, BM aspirate and trephine biopsy may identify features suspicious for either MPN or another disorder (e.g. dysplasia, increased blasts, ring sideroblasts, eosinophilia, reduced storage iron on Perls stain). BM assessment should be performed routinely in individuals with unexplained cytopenias, monocytosis or other atypical blood film findings. Very occasionally a non-myeloid malignancy such as a plasma cell neoplasm may be identified, typically in older individuals. In patients investigated for triple-negative thrombocytosis, BM histology reports are more likely to describe ET-like morphological features in those with higher platelet counts.4 Many patients with modest thrombocytosis show normal megakaryocyte numbers, morphology and distribution or increased megakaryocytes without the characteristic morphological atypia (large, hyperlobated megakaryocytes) of ET. A proportion of patients, often young females with marked (e.g. >800 × 109/L) but unexplained thrombocytosis, show more striking morphology with increased megakaryocytes and ET-like atypical forms, albeit with negative genomic investigations.4, 14 Definitive histological interpretation can, however, be challenging due to borderline features with only occasional atypical megakaryocytes. Furthermore, there is inter-individual variability in interpretation; one series found 18% of diagnoses expert most often a diagnosis of the clinical diagnosis of patients investigated for triple-negative thrombocytosis in practice is by histology BM histology should therefore expert in with molecular and clinical features a Assessment for a myeloid neoplasm on BM examination can be with NGS myeloid gene testing and assessment of abnormalities by or (Figure is negative in of patients investigated for triple-negative thrombocytosis, but a minority have pathogenic acquired variants in JAK2 and MPL outside pathogenic germline variants as discussed above and/or variants in genes found in other myeloid 13, 14, The England a minimum for MPN investigation including DNMT3A, JAK2, MPL, SH2B3, and This also allows patients who have of investigations to sequencing and/or could be considered, although clear diagnostic for these has not been Positive genomic be in the context of clinical phenotype and BM variants in myeloid genes (e.g. DNMT3A, particularly at low are commonly found in older individuals with CH and normal blood counts and may not be for to may be associated with CH variants in or In while some JAK2 and MPL variants are recurrent in ET and are established MPN (e.g. MPL the significance of other rare acquired or germline variants in JAK2, MPL and SH2B3 be or genomic BM histology and clinical in and such variants with their associated phenotypes may interpretation in the through a Patients with a germline variant in an MPN gene on BM which is or by appropriate should testing of a germline or there is a family history of thrombocytosis, may be performed on peripheral However, may not genes (e.g. and a thrombocytosis may be investigations for patients, the diagnostic of and (e.g. BM should be considered the likely of on 1 shows a including criteria for testing that occasionally a canonical driver mutation or may be of if are BM histology and complex genomic investigations are to influence the management of patients with negative peripheral blood screening aged without significant cardiovascular or and with thrombocytosis (e.g. × It should be that this platelet is on patients may be that a investigation not the that features of ET be identified if a BM biopsy was patients with low but higher platelet counts (e.g. × 109/L) or symptoms (e.g. BM biopsy and genomic studies may be particularly if peripheral blood screening assays for JAK2, CALR and MPL limited that the from genomic studies is low and patients with atypical histology will often not BM histology and genomic studies are also often negative in patients aged over these should be considered where appropriate to assess for or JAK2 or MPL mutations that alternative myeloid neoplasms or non-myeloid It is to on peripheral blood or bone and some a for peripheral blood screening for thrombocytosis. However, any usually interpretation in the context of BM genomic investigation is (Figure this should usually be performed in with BM testing the BM sample especially if megakaryocyte morphology is If a patient is or to this and/or histological assessment is very to their management peripheral blood could be considered as an additional for an clonal Over of patients diagnosed with ET are and negative studies have been reported in a histological features to ET described in a small are and in the of events, symptoms or thrombocytosis, most patients no count increase in the of A higher for investigation of thrombocytosis can therefore be considered appropriate in young (e.g. × at to × by In with thrombocytosis, exclusion of secondary testing of blood counts is JAK2, CALR and MPL mutations, BCR::ABL1 and an thrombocytosis The guidance for BM biopsy and genomic investigations in 1 can be although the low of molecular or for a higher platelet (e.g. × 109/L) may be appropriate in without thrombosis, or A for diagnostic histological and genomic is in with features in be persistent for at 3 a and a of may be for or thrombocytosis. for any reactive cause should Patients with an acquired molecular or outside of JAK2, CALR and MPL (Figure have evidence of although most variants are not a clonal is a a diagnosis of ET by the histological megakaryocyte atypia of a clonal genomic should thorough clinical and BM for alternative myeloid This is particularly for patients with multiple pathogenic variants and/or variants at VAF (e.g. an alternative myeloid neoplasm is and megakaryocyte atypia in with ET is found, as ET with clonal to be to distinguish from the without clonal A minimum VAF is in with other diagnostic such as clonal of of triple-negative thrombocytosis with an acquired genomic outside and without classical megakaryocyte morphology outside diagnostic the natural history and a diagnosis thrombocytosis of is and It should be that there is no evidence that such patients have an increased risk of to classical ET, although CH in is associated with an increased risk of haematological of this group is patients without reactive causes in myeloid genomic abnormalities have been excluded by and a diagnosis of ET can be on histological features in the 17 This has several histological is 14, and the of these criteria has not been in triple-negative many triple-negative cases have borderline histological particularly with modest ET driven by canonical driver mutations has a and natural history to patients with 21, 22 but patient and are typically the and there is no clinical diagnosis for patients with significant thrombocytosis but reactive causes typical megakaryocyte atypia of ET. these alternative from expert consensus that helpful in clinical practice. patients lacking the characteristic morphological atypia (large, hyperlobated megakaryocytes) of ET, the diagnosis thrombocytosis of and Patients with classical ET-like megakaryocyte atypia the diagnostic criteria for ET if reactive thrombocytosis is If a diagnosis of ET, as ET without clonal with the patient the and likely However, for such individuals, that an alternative diagnosis thrombocytosis with atypical and may be more appropriate than ET, for the discussed above. In the between and both lack a clonal but in whether megakaryocyte atypia is and between and triple-negative ET with clonal both have a clonal but in whether megakaryocyte atypia is that histological is and should be review and The diagnostic is to that patients diagnosed with triple-negative ET are and that cases can be carefully in studies of risk and of the clinical of these will be young patients with thrombocytosis (e.g. × 109/L) and low risk may not a BM biopsy that discriminate between and as discussed above. A of may be helpful in such The management of is of the or rare but studies are and cases for example, is associated with an increased thrombotic risk There is no evidence that such as in guidelines for should be used in or patients, the and other risk factors. A blood count should be to the thrombocytosis as the reactive is Patients with iron deficiency, thrombocytosis and normal should also have a to an increase in iron which investigation for or hereditary There is no consensus on to the management of HT. some variants have been associated with an increased risk of have favourable and most cohorts lack the long-term follow-up data to data from ET to management may in the of patients at low risk of could be considered as primary in a patient with a significant family history of associated with the and also in those at risk of cardiovascular due to other Patients who unexplained should be using these patients be to have long-term should not be routinely but has been in patients with symptoms associated with thrombocytosis or recurrent the long-term risks may be for patients, clinical monitoring including and blood film for disease and monitoring of cardiovascular risks in a to ET is patients with HT should be reviewed by an appropriate which may include in a with from clinical and on if management for the diagnostic is summarised in cases with thrombocytosis lacking secondary causes and with no clonal or atypical megakaryocyte considered there is a lack of evidence to There is no evidence to in the of another such as cardiovascular In guidance should be for example, for of patients with thrombocytosis (e.g. × 109/L) with no evidence of a myeloid neoplasm or disorder should be on any symptoms or signs that may and monitoring could be considered (e.g. if platelet count younger patients, particularly those in investigations are limited to peripheral blood could be considered significant in blood at the age of or in the of a as these where a in management occur if ET patients with evidence of a myeloid particularly borderline histological haematology follow-up may be appropriate to at a or in the of in Patients this have the typical megakaryocyte atypia of ET but lack an clonal cases have been diagnosed with ET, and and/or appropriate for ET have been used routinely in with although are BM and genomic assessment should be considered in patients of where the assessment was some (e.g. There is no evidence to any particular although in the of molecular of a clonal without such as may be where is platelet count should the for for example, for of symptoms where these the primary × in those for × in those of age and/or risk factors without the normal in those with otherwise unexplained patients also have been as ET and the guidance for above should be A low for BM and genomic assessment is in patients atypical features (e.g. unexplained or on or Patients with not histological criteria for ET, and the natural history and risks are from In the of for above could be considered, but to cases where there is that a thrombocytosis is to symptoms or an otherwise unexplained for cardiovascular risk factors including are especially since increases the risk of CH and although not all, studies have identified CH to be associated with increased risk of cardiovascular or is particularly in this group to patients that the of these on risk are not criteria for these management are also for patients with thrombocytosis in a very JAK2 V617F mutation (e.g. VAF is identified in the of histology MPN or other molecular findings. testing is recommended if platelet count and could be considered if counts are stable but a diagnosis of ET influence to the and review of the to the The to for in the The BSH General Haematology members at the of this guideline and The to the BSH and the BSH guidelines for their in this as as and for their on the was by the British Society for The BSH the the of this have a of to the which may be on and from for from from and from from and from from and from from and for from and for from from from and for and from from and from and for and from and from and for and from and from and and in a from the of and from and from The members of the group have no of to of the group will the group if any evidence that the strength of the in this document or The document will be reviewed by the and the will be to for any evidence that may have been The document will be and from the BSH guidelines website if the BSH guidelines for any that may be produced the the and in this guidance are to be and at the of to the the BSH the any for the of this can be found for the group to not to this as no or the The is not for the or of any by the than should be to the for the