guideline Spanish Guidelines for the use of targeted deep sequencing in myelodysplastic syndromes and chronic myelomonocytic leu- kaemia Laura Palomo,1 Mariam Iban  ~ ez,2,3,4 Marıa Abaigar,5 Iria Vazquez,6,7 Sara Alvarez, 8 Marta Cabez on,9 10,11 12,13,14 1 2,3,15 Barbara Tazon-Vega, Inmaculada Rapado, Francisco Fuster-Tormo, Jose Cervera, Rocıo Benito,5 6,7 8 9 Marıa J. Larrayoz, Juan C. Cigudosa, Lurdes Zamora, David Valcarcel, 10,11 Marıa T. Cedena, 12,13,14 Pamela Acha,1 5,16 6,17,18 2,3 Jes us M. Hernandez-Sanchez, Marta Fernandez-Mercado, Guillermo Sanz, Jes us M. Hernandez-Rivas,5,16,19 6,7 1 2,3,4 Marıa J. Calasanz, Francesc Sole, Esperanza Such, On behalf of the Spanish Group of MDS (GESMD) 1 Josep Carreras Leukaemia Research Institute, ICO Badalona-Hospital Germans Trias i Pujol, Universitat Autonoma de Barcelona, Sada- lona, 2Department of Haematology, Hospital Universitari i Politecnic La Fe, Valencia, 3Centro de Investigacion Biomedica en Red de Cancer (CIBERONC), Instituto Carlos III, Madrid, 4Departamento de Ciencias Biomedicas, Facultad de Ciencias de la Salud, Universidad CEU Cardenal Herrera, Valencia, 5Institute of Biomedical Research of Salamanca (IBSAL), Cancer Research Centre (IBMCC-CIC; Univ. of Salamanca-CSIC), Salamanca, 6Haematological Diseases Laboratory, CIMA LAB Diagnostics, University of Navarra, 7Navarra Institute for Health Research (IdiSNA), Pamplona, 8NIMGenetics, Genomica y Medicina, S.L., Madrid, 9Haematology Service, ICO Badalona- Hospital Germans Trias i Pujol, Josep Carreras Leukaemia Research Institute, Badalona, 10Department of Haematology, Vall d’Hebron Institute of Oncology, Vall d’Hebron University Hospital, 11Department of Medicine, Universitat Autonoma de Barcelona, Barcelona, 12 Haematology Department, Hospital Universitario 12 de Octubre, 13Haematological Malignancies Clinical Research Unit, CNIO, 14Centro de investigacion Biomedica en Red Cancer (CIBERONC), Madrid, 15Genetics Unit, Hospital Universitario y Politecnico La Fe, Valencia, 16 University of Salamanca (USAL), Salamanca, 17Advanced Genomics Laboratory, Centre for Applied Medical Research (CIMA), Univer- sity of Navarra, Haemato-Oncology, Pamplona, 18Biomedical Engineering Department, School of Engineering, University of Navarra, San Sebastian and 19Hospital Universitario de Salamanca, Salamanca, Spain and reporting of NGS data. To that end, the Spanish MDS Summary Group (GESMD) has expanded the present set of guidelines, The landscape of medical sequencing has rapidly changed aiming to establish common quality standards for the ade- with the evolution of next generation sequencing (NGS). quate implementation of NGS and clinical interpretation of These technologies have contributed to the molecular charac- the results, hoping that this effort will ultimately contribute terization of the myelodysplastic syndromes (MDS) and to the benefit of patients with myeloid malignancies. chronic myelomonocytic leukaemia (CMML), through the identification of recurrent gene mutations, which are present Keywords: myelodysplastic syndromes, chronic myelomono- in >80% of patients. These mutations contribute to a better cytic leukaemia, next generation sequencing, guidelines, classification and risk stratification of the patients. Currently, molecular genetics. clinical laboratories include NGS genomic analyses in their routine clinical practice, in an effort to personalize the diag- The development of next generation sequencing (NGS) tech- nosis, prognosis and treatment of MDS and CMML. NGS niques during the last decade has contributed to the molecu- technologies have reduced the cost of large-scale sequencing, lar characterization of haematological malignancies, including but there are additional challenges involving the clinical vali- myeloid neoplasms, such as the myelodysplastic syndromes dation of these technologies, as continuous advances are con- (MDS) and chronic myelomonocytic leukaemia (CMML) stantly being made. In this context, it is of major importance (Arber et al, 2016). According to the World Health Organi- to standardize the generation, analysis, clinical interpretation zation (WHO) classification (Arber et al, 2016), MDS com- prise a group of clonal haematopoietic stem cell disorders characterized by ineffective haematopoiesis in one or more Correspondence: Francesc Sole, Institut de Recerca Contra la myeloid cell lines, abnormal dysplastic cell morphology and Leucemia Josep Carreras (IJC), Crta. de Can Ruti, Cami de les potential for clonal evolution. On the other hand, CMML is Escoles, Edifici IJC, Badalona 08916, Barcelona, Spain. included in the WHO group of myelodysplastic/myeloprolif- E-mail: fsole@carrerasresearch.org erative neoplasms (MDS/MPN) and is characterized by ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology First published online 16 October 2019 and John Wiley & Sons Ltd. British Journal of Haematology, 2020, 188, 605–622 doi: 10.1111/bjh.16175 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. Guideline absolute peripheral monocytosis in conjunction with both 2014). Moreover, some mutations are also associated with effective (myeloproliferative variant, MP-CMML ≥ 13 9 109/ shorter survival and a higher risk of progression to AML and l leucocytes) and ineffective (myelodysplastic variant, MD- therefore can be used for prognostic stratification (Table I) CMML < 13 9 109/l leucocytes) haematopoiesis. MDS and (Bejar et al, 2011, 2012; Papaemmanuil et al, 2013; Haferlach CMML are very heterogeneous at the clinical, morphological et al, 2014; Elena et al, 2016; Makishima et al, 2017). Thus, and genetic level and both and have the risk of progressing an MDS/CMML targeted sequencing panel should at least to acute myeloid leukaemia (AML). The natural history of include all those genes that have been proven relevant for the these diseases is highly variable and a risk-adapted treatment disease (Malcovati et al, 2013; Arber et al, 2016; Greenberg strategy is mandatory. et al, 2017) (Table II). As NGS is a high throughput tech- Myeloid neoplasms arise through the serial acquisition of nique, it can be complemented with genes that are altered in chromosomal alterations and somatic mutations that affect other myeloid neoplasms to generate a pan-myeloid sequenc- genes involved in several mechanisms. These mutations drive ing panel, in view of a wider utility of the tool (e.g. for disease evolution from asymptomatic clonal haematopoiesis MDS, MPN and AML) (Table III). In this regard, the Associ- to overt disease and, eventually, progression to AML (Ken- ation for Molecular Pathology recently reviewed the clinical nedy & Ebert, 2017). Recurrent mutations in more than 40 relevance of small DNA variants in chronic myeloid neo- different genes have been identified in MDS and CMML and plasms, summarizing key findings that support clinical util- at least one mutation is detected at diagnosis in >80% of ity, and defining the need for gene inclusion in high- patients (Papaemmanuil et al, 2013; Haferlach et al, 2014; throughput sequencing testing panels (McClure et al, 2018). Elena et al, 2016). These mutations are related to the patho- Driver genes in MDS and CMML, which are common physiological features of these diseases and play a role in among the spectrum of myeloid malignancies, affect specific their clinical heterogeneity. These molecular markers can cellular processes and can be categorized according to their complement current diagnostic criteria in MDS and CMML, function (Greenberg et al, 2017; Kennedy & Ebert, 2017; as well as contribute to the risk assessment of these patients McClure et al, 2018) (Fig 1). (Bejar et al, 2011; Haferlach et al, 2014; Elena et al, 2016). Mutations in RNA-splicing machinery are the most fre- Therefore, diagnostic laboratories are including NGS geno- quent type of somatic mutations detected in MDS. Recurrent mic analyses in their routine clinical practice, in an effort to mutations have been reported in the spliceosome compo- personalize the diagnosis, prognosis and treatment of MDS nents of the SF3B1, SRSF2, U2AF1 and ZRSR2 genes, while and CMML. Thorough evaluation of technical aspects, data mutations in PRPF40B, SF3A1, SF1 and U2AF2 (U2AF65) analysis and variant interpretation is required for the correct genes have been described in only 1–2% of MDS patients implementation of NGS in routine diagnosis. Thus, the aim (Yoshida et al, 2011; Makishima et al, 2012; Papaemmanuil of this collaborative project within the Spanish MDS Group et al, 2013). SF3B1 mutations are associated with a higher (GESMD) was to develop a set of guidelines, agreed by overall survival rate and an inferior risk of AML progression, expert consensus panels, to standardize the use of targeted as they are the only ones that are clearly associated with a deep sequencing in the routine genetic testing of MDS and better prognosis, especially as a single mutation (Papaem- CMML patients, for the detection of single nucleotide vari- manuil et al, 2011; Malcovati et al, 2014). SRSF2 mutations ants (SNVs) and small insertions/deletions (indels). Our are associated with monocytosis and marked thrombo- main objective is to harmonize analyses and variant interpre- cythaemia and are especially frequent in CMML (Yoshida tation, ensuring high standards in clinical reporting of et al, 2011; Meggendorfer et al, 2012). Mutations in SRSF2 genetic data. These consensually proposed NGS guidelines and U2AF1 have been associated with less favorable out- involve the following major sections, which will be further comes in some studies (Graubert et al, 2011; Thol et al, discussed in detail: (i) mutational landscape in MDS and 2012; Meggendorfer et al, 2012; Haferlach et al, 2014; Mak- CMML, (ii) sequencing workflow and quality control, (iii) ishima et al, 2017). data analysis and variant filtering, (iv) variant categorization/ Epigenetic regulators are the second most common class interpretation and (v) clinical reports. of mutation in MDS, and can be divided into DNA methylation enzymes (TET2, DNMT3A and IDH1/2 genes) and chromatin modifiers (ASXL1 and EZH2 genes) (Del- Mutational landscape of MDS and CMML hommeau et al, 2009; Walter et al, 2011; Gelsi-Boyer et al, Over 40 different genes are found recurrently mutated in 2012; Shih et al, 2012; Ganguly & Kadam, 2016). TET2 MDS and CMML (Papaemmanuil et al, 2013; Haferlach mutations are very common in CMML and the combina- et al, 2014; Elena et al, 2016). However, in contrast to other tion of TET2 and SRSF2 is highly associated with this dis- diseases, such as myeloproliferative neoplasms (MPN), none order (Malcovati et al, 2014). TET2 mutations do not have of them are specific to these disorders. Still, a few of these a prognostic impact but they are associated with an genes have been proved to be useful for diagnosis or to pre- increased response rate to hypomethylating agents (Itzykson dict response to specific treatments (Papaemmanuil et al, et al, 2011; Bejar et al, 2014) (Table III). DNMT3A muta- 2011; Itzykson et al, 2011; Bejar et al, 2014; Traina et al, tions are associated with a more aggressive clinical course 606 ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd British Journal of Haematology, 2020, 188, 605–622 Guideline Table I. Clinical relevance of mutated genes in MDS and CMML (Malcovati et al, 2013; Arber et al, 2016; Greenberg et al, 2017). MDS CMML Gene Incidence Clinical impact Incidence Clinical impact ASXL1 5–25% Unfavourable 40–50% Unfavourable CSNK1A1 <1% Uncertain <1% Unknown 5–15% MDS with del(5q) Associated with del(5q) DNMT3A 12–18% Unfavourable in patients without SF3B1 mutations 2–10% Uncertain EZH2 5–10% Unfavourable 5–12% Unfavourable IDH1 <5% Uncertain <1% Uncertain IDH2 <5% Uncertain 5–10% Unfavourable JAK2 <5% MDS with del(5q), 5–7% 2–10% Associated with MP-CMML KRAS 5–10% Uncertain 10–20% Unfavourable Associated with MP-CMML NRAS 5–10% Unfavourable 10–20% Unfavourable Associated with MP-CMML RUNX1 10–15% Unfavourable 10–30% Unfavourable Can be of germline origin SETBP1 <5% Unfavourable 5–10% Unfavourable SF3B1 20–30% Associated with RS 5–10% Unknown 80% MDS-RS Favourable SRSF2 10–15% Unfavourable 30–50% Uncertain STAG2 5–10% Unfavourable 5–10% Unfavourable TET2 20–25% Uncertain 45–60% Uncertain TP53 8–12% Unfavourable <5% Unfavourable Associated with CK (50%), del(5q) (15–20%) Lower response rate to lenalidomide Can be of germline U2AF1 8–12% Unfavourable 5–10% Unfavourable ZRSR2 5–10% Unfavourable 5–10% Uncertain CK, complex karyotype; CMML, chronic myelomonocytic leukaemia; HMA, hypomethylating agents; MDS, myelodysplastic syndrome; MDS-RS, MDS with ring sideroblasts; MP-CMML; myeloproliferative CMML; NK, normal karyotype; RS, ring sideroblasts. and rapid progression to AML in both MDS and CMML high risk phenotype, thus being more frequent in high risk patients (Walter et al, 2011; Bejar et al, 2014; Patnaik MDS and CMML and in secondary AML (Bejar et al, 2011; et al, 2017). ASXL1 mutations are independently associated Cazzola et al, 2013; Elena et al, 2016). with a poor outcome in the spectrum of myeloid diseases, MDS and CMML patients frequently harbour mutations including a shorter overall survival and a higher risk of in signal transduction genes, which constitutionally activate progression to AML (Boultwood et al, 2010; Gelsi-Boyer a signalling pathway involved in cell proliferation, apoptosis et al, 2012; Itzykson et al, 2013a). EZH2 mutations are and cell differentiation (CBL, CSNK1A1, JAK2, N/KRAS, associated with worsened overall survival in both low and MPL, NF1, PTPN11, KIT, and FLT3 genes). These muta- high risk MDS and CMML (Bejar et al, 2011; Grossmann tions are more frequent in myeloproliferative CMML cases et al, 2011; Bejar et al, 2012). and usually correspond to secondary events that may lead The multiprotein cohesin complex is involved in the cohe- to AML progression (Pich et al, 2009; Kohlmann et al, sion of sister chromatids and the post-replicative DNA repair 2010; Ricci et al, 2010; Schwaab et al, 2012; Itzykson et al, system, and it is encoded by the genes STAG1/2, SMC1A, 2013b; Smith et al, 2015; Elena et al, 2016). SMC3 and RAD21, which are mutated in up to 11% of low Other genes recurrently mutated in MDS and CMML risk and 17% of high risk MDS patients (Kon et al, 2013; that are involved in other cellular processes, include TP53, Thota et al, 2014). Mutations in STAG2 are more frequent in SETBP1, BCOR and BCORL1 (Bejar et al, 2011; J€adersten patients with multilineage dysplasia and excess of blasts and et al, 2011; Damm et al, 2013a,2013b; Bejar et al, 2014; seem to be associated with shorter overall survival (Thota Elena et al, 2016). TP53 mutations are more frequent in et al, 2014). high risk patients and are associated with adverse prognosis Loss-of-function mutations in a transcription factor gene, features, including advanced disease, complex karyotypes, such as RUNX1, GATA2 or ETV6, are detected in approxi- monosomal karyotypes, excess of blasts, severe thrombocy- mately 20% of MDS (Chen et al, 2007; Bejar et al, 2011; topenia and therapy-related MDS (Kita-Sasai et al, 2001; Wall et al, 2012). RUNX1 mutations are associated with a Bejar et al, 2011; Kulasekararaj et al, 2013; Ok et al, 2015). ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology 607 and John Wiley & Sons Ltd. British Journal of Haematology, 2020, 188, 605–622 Guideline Table II. List of GESMD-recommended genes for studying the clinical management of MDS and CMML. Frequency Gene Region Type of mutation MDS CMML ASXL1 Exons 10–13 Nonsense, frameshift 5–25% 40–50% Codons: all CSNK1A1 Exons 2–4 Missense 5–10% <1% Codons: all DNMT3A Complete coding region All 12–18% 2–10% Hotspot codon: R882 EZH2 Complete coding region Nonsense, frameshift 5–10% 5–12% Codons: all IDH1 Exon 4 Missense <5% <1% Hotspot codon: R132 IDH2 Exon 4 Missense <5% 5–10% Hotspot codons: R140 and R172 JAK2 Complete coding region Missense <5% 2–10% Hotspot codon: V617F KRAS Exons 2 and 3 Missense 5–10% 10–20% Hotspot codons: G12, G13, Q61 and G146 NRAS Exons 2 and 3 Missense 5–10% 10–20% Hotspot codons: G12, G13 and Q61 RUNX1 Complete coding region Nonsense, frameshift 10–15% 10–30% Codons: all SETBP1 Exon 4 Missense <5% 5–10% Codons: 858–870 SF3B1 Exons 10–16 Missense 20–30% 5–10% 80% RS Codons: 622–781 SRSF2 Complete coding region Missense 10–15% 30–50% Hotspot codon: P95 STAG2 Complete coding region Nonsense, frameshift, splicing 5–10% 5–10% Codons: all TET2 Complete coding region All 20–25% 45–60% Codons: 1134–1444 or 1842–1921 TP53 Complete coding region All 8–12% <5% Codons: all U2AF1 Exons 2–6 Missense 8–12% 5–10% Hotspot codons: S34 and Q157 ZRSR2 Complete coding region Nonsense, frameshift 5–10% 5–10% Codons: all GESMD, Spanish Group of MDS; CMML, Chronic Myelomonocytic Leukaemia; MDS, Myelodysplastic Syndrome; RS, ring sideroblasts. TP53 mutations are associated with a very poor prognosis has led the WHO classification to include a new section on in all MDS subtypes, including patients who have under- myeloid neoplasms with germline predisposition. This sec- gone an haematopoietic stem cell transplantation (Bejar tion comprises cases of MDS, MDS/MPN and AML that et al, 2011; Kulasekararaj et al, 2013; Bejar et al, 2014). arise on the background of a predisposing germline mutation TP53 mutations are also associated with the presence of del (Arber et al, 2016). Several genes have been associated with a (5q) and with a lower response rate to lenalidomide treat- genetic predisposition to myeloid malignancies, including ment in these patients (J€adersten et al, 2011; Meggendorfer ANKRD26, CBL, CEBPA, DDX41, ELANE, ETV6, GATA2, et al, 2017). GSKIP, MPL, NF1, SAMD9, SAMD9L, PTPN11, RUNX1, SRP72 and TP53, among others (Kennedy & Shimamura, 2019). The majority of these genes are included in myeloid- Myeloid neoplasms with germline predisposition related targeted panels focused on somatic mutations. In Although most cases of MDS or AML are sporadic diseases, addition, recent studies suggest that the frequency of germ- a number of cases arise as the result of genetic predisposition line variants in some of these genes, like DDX41, may be due to the presence of inherited germline mutations, which underappreciated, suggesting that they should also be 608 ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd British Journal of Haematology, 2020, 188, 605–622 Guideline Table III. Other myeloid-related genes more frequent in AML, MPN and other MDS/MPN. Frequency Region Gene Frequent mutations Type of mutation MDS (%) CMML (%) Frequent in myeloproliferative neoplasms CALR Exon 9 Frameshift <1 <1 Codons: all CBL Exons 8 and 9 Missense <5 8–18 Codons: 366–420 CSF3R Complete coding region Missense <1 3–4 Hotspot codons: 618 GATA2 Exons 2 and 6 Missense, frameshift <5 <1 Codons: all MPL Complete coding region Missense <1 <1 Codons 505 and 515 NF1 Complete coding region Nonsense, frameshift, splicing <5 <5 Codons: all PTPN11 Exons 3 and 7 Missense <1 4 Codons: all Frequent in acute myeloid leukaemia BCOR Complete coding region All <5 <5 Codon: all BCORL1 Complete coding region Nonsense, frameshift <1 <1 Codons: all CEBPA Complete coding region Missense, frameshift <5 <5 Codons: all ETV6 Complete coding region Nonsense, frameshift <5 <1 Codons: all FLT3 Exons 13–15 and 20 Missense, frameshift <5 <5 Hotspot codons: FLT3-ITD and D835 KIT Exons 2, 8–11, 13 and 17 Missense, frameshift <3 <1 Codons: all NPM1 Exon 11 Frameshift <5 <5 Hotspot codons: W288 WT1 Exons 7 and 9 Missense, frameshift <3 <3 Codons: all CMML, chronic myelomonocytic leukaemia; ITD, internal tandem duplication; MDS, myelodysplastic syndrome. Splicing factors SF3B1, SRSF2, U2AF1, ZRSR2 Epigenetic regulators DNMT3A, TET2, IDH1/2, ASXL1, EZH2 Cohesins SMC1A, SMC3, RAD21, STAG1, STAG2 Transcription factors RUNX1, GATA2, ETV6 Signal transduction CBL, JAK2, NRAS, KRAS, MPL, NF1, PTPN11, KIT, FLT3 DNA damage response TP53 0 20 40 60 80 100 Mutation frequency (%) Fig 1. Classification of frequently mutated genes in MDS and CMML according to their functional category (modified from Kennedy & Ebert, 2017; Kennedy & Ebert, 2017). ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology 609 and John Wiley & Sons Ltd. British Journal of Haematology, 2020, 188, 605–622 Guideline considered for inclusion in a panel of recommended genes of cardiovascular disease and a higher risk of developing an (Berger et al, 2017; Tawana et al, 2018). This may be nota- haematological malignancy, although thankfully the rate of bly relevant in the context of the search for a related donor. malignant transformation is very low (05–1% per year) Considering all the above, when performing targeted (Jaiswal et al, 2014, 2017). This risk increases when the VAFs sequencing in MDS patients, the possibility of detecting a are higher (>10%) or more than one mutation is detected. germline mutation must be considered. Although a variant In addition, recent studies have shown that patients with allele frequency (VAF) close to 50% or 100% may be sug- solid tumours who have CHIP before being exposed to treat- gestive of a germline mutation, its nature should be con- ment, are at increased risk of developing a therapy-related firmed using a control germline sample before acting on it, myeloid neoplasm (TRMN) (Takahashi et al, 2013; Gillis and should always be evaluated in the clinical and familial et al, 2017). Factors that influence the natural history of context of each patient. Even though germline genetic pre- CHIP are currently under investigation, but it is known that disposition is frequently associated with MDS with early the acquisition of further mutations drives the progression of onset (children and young adults), it is increasingly recog- CHIP to overt malignancy. nized also in MDS cases diagnosed in advanced ages. The Apart from healthy individuals, mutations in these MDS- prevalence of hereditary myeloid malignancies has not been associated genes can also be detected in 35% of patients with clearly established but is suspected to concern at least 5% persistent blood cytopenias for which no explanation is of adult patients with MDS/AML (Tawana et al, 2018). In apparent, usually called idiopathic cytopenias of undeter- addition, currently known MDS/AML predisposition genes mined significance (ICUS) (Table IV). These cases, that pre- account for only 25% of familial cases, so other risk loci sent with both clonal haematopoiesis and cytopenias, are remain to be discovered (Tawana et al, 2018). When a designated clonal cytopenias of uncertain significance germline mutation is confirmed, referral of patients to pro- (CCUS). Patients with CCUS are at a higher risk of develop- fessionals with expertise in cancer predisposing syndromes ing a myeloid malignancy compared to those individuals and in genetic counselling is recommended, due to the need with CHIP not presenting any cytopenia, especially in the for highly specialized and rapidly advancing evaluations and presence of two or more mutations or if a single mutation in tailored treatments for these patients (Godley & Shimamura, a splicing gene is detected (Arber et al, 2016; Bejar, 2017; 2017). Jaiswal et al, 2017) (Table IV). In addition, average VAFs in patients with CCUS (30–40%) are higher compared to those found in CHIP. Clonal haematopoiesis of uncertain significance Taking all these observations into account, when perform- Recurrent somatic mutations in MDS-associated genes, such ing NGS routine tests in a patient with an indication of as DNMT3A, TET2, ASLX1, TP53 and JAK2 genes, have been MDS, diagnosis should always be performed in the context identified in the peripheral blood of healthy individuals with of other laboratory and clinical data. advanced age and normal blood counts (Genovese et al, 2014; Jaiswal et al, 2014; Xie et al, 2014). These mutations, Sequencing workflow and quality criteria usually detected at very low VAFs (<10%), provide a selective advantage to the haematopoietic stem cells in which they Targeted NGS panels commonly include a selected number occur, but cells still maintain the ability to differentiate into of specific genes, or coding regions within genes that are circulating leucocytes (Steensma et al, 2015). This phe- clinically-actionable in the disease of interest, or that are nomenon has been defined as clonal haematopoiesis of inde- known to harbour mutations that contribute to the patho- terminate potential (CHIP) (Table IV) (Steensma et al, genesis of the disease. Focusing on a limited set of genes 2015). The incidence of CHIP increases with age, with more allows greater depth of coverage for increased analytical sen- than 10% of healthy people older than 70 years of age carry- sitivity and specificity. Thus, targeted sequencing is often ing a mutation in one of those genes (Genovese et al, 2014; used in clinical care because, while keeping the cost relatively Jaiswal et al, 2014). These individuals have an increased risk low, it provides greater confidence and facilitates the Table IV. Characteristics of CHIP, ICUS, CCUS and MDS. CHIP ICUS CCUS MDS Cytopenia No Yes (≥1) Yes (≥1) Yes (≥1) Dysplasia No No, or minimum No, or minimum Yes Mutations Yes (n = 1) No Yes (n ~ 1, 6) Yes >85% (n ~ 2, 6) VAF 2–12% – 30–40% 30–50% Progression Risk 0.5–1% per year >10% in 5 years >85% in 5 years CCUS, clonal cytopenia of uncertain significance; CHIP, clonal haematopoiesis of uncertain significance; ICUS, idiopathic cytopenia of uncertain significance; MDS, myelodysplastic syndrome; VAF, variant allele frequency. 610 ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd British Journal of Haematology, 2020, 188, 605–622 Guideline interpretation of the findings in their clinical context. The of different patient samples in the same sequencing run choice of sequencing strategy will depend on several factors, (Table S1). In addition, some strategies use unique molecular including size of the panel, sequencing platform, turnaround barcodes that tag each original DNA fragment and allow time (TAT) for reporting the results and cost per sample. identification and removal of PCR duplicates during bioin- formatic processing of sequencing data. Panel design strategy Sequencing platforms When designing a custom panel, it is recommended to select targets and transcripts only from genes with potential clinical Sequencers for NGS, based on the ability to perform an ele- relevance. There are several available online tools that can be vated number of chemical reactions in parallel, are increas- used to design the specific primers/probes to enrich the tar- ingly being used in the clinical setting. Sequencing gets of interest. Although custom panels are usually flexible instruments can be classified according to the mean read and can be modified over time, they require a long process length (short or long), the type of sequencing (single-end of optimization and validation, which is skipped when using and paired-end sequencing) and the sequencing chemistry off-the-shelf panels. On the other hand, commercial panels (mainly by synthesis, ligation or hybridization). Multiple fac- offer a closed design, but do not require a previous design tors should be considered when choosing a sequencing plat- phase and are already optimized. Of note, these panels are form, such as run time, size of sequenced region, required commercially available as ready-to-use, but quite often they depth of coverage, read length, TAT requirements and cost still require validation in each laboratory. per sample. According to the GESMD, Illumina (San Diego, CA, USA) and Thermo-Fisher Scientific, short-read sequenc- ing platforms are ideal for the implementation of targeted Library preparation deep sequencing in diagnostic laboratories (Table S2). In MDS and CMML, as in most haematological neoplasms, There is a number of sequencing parameters that define it is recommended to perform molecular analyses on bone the scope and quality of data output of a given NGS technol- marrow-derived cells. Some studies suggest that, due to the ogy: sequencing capacity, type and length of the reads, high sensitivity of NGS, there is little difference between sequencing output, depth of coverage, percentage of reads bone marrow and peripheral blood samples (Duncavage mapped correctly and uniformity of the reads (Table S3). et al, 2017; Martin et al, 2018). Therefore, if bone marrow is Some of these parameters can be influenced by other vari- not available, peripheral blood can be useful, especially for ables, such as input DNA quality, adequate library prepara- molecular monitoring or in myeloproliferative disorders like tion, correct design of the primers/probes, choice of CMML. Although DNA from fresh samples is the optimal sequencing platform, and the initial estimation of the num- input material, targeted NGS can be performed on any DNA ber of required reads. Regarding the depth of coverage, we sample as long as the quality and quantity are sufficient. recommend that ≥95% of the bases targeted in the panel DNA purity can be measured by spectrophotometry (i.e. design are covered by at least 100 reads (≥1009), and that Nanodrop device, Thermo Fisher Scientific, Waltham, MA, the mean coverage for each sample is ≥10009. This will USA), while DNA integrity can be assessed by capillary elec- enable the identification of variants with a sensitivity of at trophoresis (i.e. Tape Station or Bioanalyzer devices, Agilent, least 5% in most regions, as long as variant quality criteria Santa Clara, CA, USA). Input DNA and DNA libraries are fulfilled (we recommend the presence of at least 25 reads should be quantified using a fluorometric assay (i.e. Qubit or for the alternative allele). Of note, even though this is our Quant-iT PicoGreen, Thermo Fisher Scientific) that can dis- recommendation, less conservative approaches are also likely tinguish and measure the concentration of double-stranded to be highly valid, especially for certain hotspot locations, DNA. The manufacturer’s recommendations should be and should be validated in each laboratory. The minimum strictly followed during library preparation. output required per library can be estimated by taking into Two main strategies for library preparation are widely account the intended depth of coverage together with the used in targeted NGS: amplicon-based and hybrid-capture- size of the panel (including on- and off-target regions). It is based. Amplicon panels are based on the use of specific pri- recommended that the percentage of on-target mapped reads mers and the amplification of the regions of interest by poly- is higher than 70%. merase chain reaction (PCR)-based methods (single or multiplex PCR, or emulsion PCR). On the other hand, target Test validation enrichment is performed in capture panels using probe hybridization-based methods. In both cases, DNA fragments During test development, iterative cycles of performance generated during library preparation contain adapter optimization should be executed until all the different com- sequences on both ends, which are complementary to plat- ponents of the assay are optimized, including panel design, form-specific sequencing primers. Samples can be tagged library preparation conditions, sequencing parameters, num- with a unique oligonucleotide barcode that enables pooling ber of samples that should be pooled in order to achieve the ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology 611 and John Wiley & Sons Ltd. British Journal of Haematology, 2020, 188, 605–622 Guideline desired coverage level, as well as analysis settings and pipe- coming from the sequencer into a list of variants that can be line. TAT for data reports need to be estimated and should visualized, filtered and interpreted. The challenge of data be clinically appropriate. Once the entire panel conditions analysis includes the huge number of available tools, the con- have been established, the assay needs to be validated, and stantly evolving analysis pipelines and the lack of consensus performance characteristics (such as sensitivity, specificity regarding which tools to use. Typical NGS data analysis and reproducibility) should be determined. workflows include base calling, read alignment variant calling Two rounds of validation should be consecutively per- and variant annotation. Data analysis steps and tools that are formed. Firstly, a technical validation round aimed at evaluat- recommended, and most commonly used, by the GESMD ing the quality of the design and identifying the regions with are summarized in (Data S1, Tables S4, S5). It should be poor coverage. For technical validation, input DNA require- noted that several open-source and commercial tools have ments, library generation and sequencing should be per- been developed to facilitate NGS data analysis in the clinical formed, strictly following manufacturer’s recommendations. setting, in the form of user-friendly interfaces. Some of these During this process, the percentage of bases covered at differ- tools, used by the GESMD, include MiSeq Reporter (Illu- ent depths of coverage should be determined, as well as the mina), Variant Studio (Illumina), IonReporter (Thermo- regions with low or no coverage, which should be then rede- Fisher Scientific), DNAnexus (DNAnexus, Santa Clara, CA, signed or removed from the design. Biological replicates USA) and Sophia DDM (Sophia Genetics, Lausanne, Vaud, should also be assessed, including sequencing two libraries Switzerland). generated from the same input DNA and sequencing two libraries from two different DNA extractions of the same Variant filtering sample. Recurrent artefacts should also be identified. For example, variants that appear at greater frequency than After data analysis, variant filtering should then be per- expected in a large cohort or that have reads supporting them formed in order to obtain a list of candidate variants that in a large number of samples, especially if they all cluster at will later be categorized, interpreted and reported. Of note, similar VAFs, are likely to be technical artefacts. Generation variant filtering criteria should always be updated according of internal databases with known technical artefacts can be to the advances made in NGS technologies and in the disease helpful in order to filter these artefacts out during the analy- of interest. The variant filtering workflow proposed by the sis-filtering steps. Secondly, a clinical validation round should GESMD is described in Table V and Fig 2. also be performed with the aim of determining if the designed test is able to detect known alterations associated with the Variant interpretation and categorization disease of interest. In order to do so, previously characterized samples with known SNVs and/or indels, should be analysed Interpretation systems for classifying variants are useful to with the new test. Whenever possible, previously undescribed standardize the way in which variants are reported to clini- variants and complex indels should be validated by other cians. In response to the classification discrepancies that technologies, such as Sanger sequencing, amplification refrac- often exist between laboratories, the use of unified guidelines tory mutation system (ARMS)-PCR, allele-specific oligonu- is highly recommended. On one hand, germline variants clas- cleotide (ASO)-PCR, digital PCR or other NGS platforms. sification systems aim to classify variants according to their Finally, sequencing of serial dilutions from the same sample pathogenicity in a given disease (Richards et al, 2015; Mat- will help to determine the sensitivity of the test. thijs et al, 2016). A joint consensus between the American In NGS, standards for TAT should be established by each College of Medical Genetics and Genomics and the Associa- laboratory based on the clinically proper indication for test- tion for Molecular Pathology, for example, proposes updated ing. Multiplexing samples from different diseases and using standards for variant assessment and promotes the systematic small-scale sequencing systems can help reduce TAT. In real evidence-based classification of variants found in Mendelian world scenario of diagnosis of MDS and CMML, these TAT disorders (Richards et al, 2015). On the other hand, in can- should be consistent with other cytogenetic tests (karyotype cer, somatic variants are assessed for diagnostic, prognostic, and fluorescence in situ hybridisation), and should not predictive and/or therapeutic impact in the context of exceed, for most cases, 15 working days (Rack et al, 2019). tumour site and histology (Sukhai et al, 2016). Given their Urgent referrals should be prioritized and, for these cases, relevance, a system to prioritize variants detected by whole results should try to be reported within 10 days. exome sequencing (WES) as well as two somatic tier-based variant classification systems have been reported in the litera- ture (Van Allen et al, 2014; Sukhai et al, 2016; Li et al, Data analysis and variant filtering 2017). Sukhai et al (2016) propose a classification scheme in which variants are categorized from Class 1 (highest priority) Data analysis to Class 5 depending on the influence that the variant shows The goal of data analysis is to use bioinformatic tools within on diagnosis, prognosis or treatment of the specific tumour an analysis pipeline in order to transform the raw data being evaluated. Similarly, Li et al (2017) classify variants 612 ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd British Journal of Haematology, 2020, 188, 605–622 Guideline Table V. Variant filtering workflow detailed information proposed by the GESMD. Step Description Variant pre-filtering Filter according to variant region • Preserve variants located in: ◦ Coding regions (exonic) ◦ Splicing sites (12 bp) • Remove variants located at: intergenic regions, downstream, upstream, non-coding RNAs, intronic regions far from splicing sites. Remove sequencing errors • Remove sequencing errors non-detected previously by the software of analysis. Some commonly detected errors include: ◦ Non-uniform coverage in the region flanking the variant ◦ Strand bias: variants only covered by forward or reverse reads ◦ Small indels located at repetitive and homopolymeric regions ◦ Edge effect: variants located at the end of the amplicon • A genomic viewer that can open alignment files (BAM or SAM), such as IGV (Broad Institute), should be used to visualize the data and identify the errors (Robinson et al., 2011). Variant filtering Polymorphisms • MAF refers to the frequency at which a variant occurs in a given population. Variants with a MAF ≥ 1% are considered as polymorphisms in somatic mutation analysis. The clinical implication of polymorphisms in MDS and CMML is currently not known; therefore, we recommend not reporting these variants for now. • Population databases provide comprehensive information about frequencies of alternative (minor) alle- les at a given locus in a large cohort of individuals. The following population databases can be used to identify and filter out polymorphisms: 1000 Genomes Project, Exome Variant Server, ExAC, dbSNP, dbVar and gnomAD. Synonymous variants • Synonymous (or silent) variants are SNVs that do not alter the encoded amino acid sequence. Their clinical relevance in MDS and CMML is unknown. Based on this current knowledge, we recommend that these variants should be filtered out for now, except well annotated pathogenic variants (i.e. TP53 synonymous variants that are known to affect splicing). Variants in UTR regions • The clinical relevance of variants located at 3’UTR or 5’UTR regions is currently unknown. Based on this we recommend that these variants should be filtered out for now. Quality criteria Coverage • Coverage of the locus: ≥1009. We recommend that the genomic position in which the variant is located is covered by at least 100 reads. • Coverage of the variant: ≥259. We recommend the presence of at least 25 reads for the alternative allele. • If the variant does not fulfil the quality criteria, sequencing should be repeated or the variant should be validated by a different technique. BAM, binary alignment map; bp, base pairs; CMML, chronic myelomonocytic leukaemia; GESMD, Spanish MDS Group; IGV, Integrative Geno- mics Viewer; MAF, minor allele frequency; MDS, myelodysplastic syndrome; SAM, sequence alignment map; SNV, single nucleotide variant; UTR, untranslated region. from Tier I (Variants with Strong Clinical Significance) to In light of this, and based on the same tier-based classifi- Tier IV (Benign or Likely Benign Variants) based on the clin- cation system of the American Association for Molecular ical impact of a given variant, which is determined according Pathology, American Society of Clinical Oncology and Col- to currently available evidence. It is of high importance to lege of American Pathologists (Li et al, 2017), we propose a include a separate category for benign variants, to inform system to classify somatic variants detected in myeloid neo- clinicians and patients and to reduce the burden on laborato- plasms (Fig 2). ries (Hoskinson et al, 2017). Based on our experience, the use of specific standard terminology (pathogenic, likely Tools for variant interpretation pathogenic, uncertain significance, likely benign and benign) facilitates the use of classification systems, especially for clini- In order to properly classify and annotate the detected cians. somatic variants, several currently available repositories and ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology 613 and John Wiley & Sons Ltd. British Journal of Haematology, 2020, 188, 605–622 Guideline CALLED VARIANTS - Remove variants outside regions of interest Pre-filtering - Remove sequencing errors TRUE VARIANTS AT REGIONS Variant filtering OF INTEREST - Polymorphisms Filtering: identify and filter out - Synonymous variants - Variants at UTR regions MYELOID-RELATED CANDIDATE VARIANTS - ≥100x locus coverage Does the variant fulfill the QC? - ≥25 reads for the alternative allele Quality control (qc) No Yes Repeat or validate the Categorize variant according to the variant with another proposed variant classification) technique Is the variant reported as PATHOGENIC in the available clinical databases? Yes No Is it reported in the disease of interest? Is the variant reported as BENIGN? Variant categorization Yes No No Yes Pathogenic Likely Variant of Is it reported in the disease of interest? (Tier I) Pathogenic Uncertain (Tier II) Significance (Tier III) No Yes Not reported variants (unknown pathogenicity) Likely benign Benign (Tier IV) (Tier V) Fig 2. Proposed workflow for variant filtering and categorization. UTR, untranslated region. genomic databases can be used. Before using these data- nih.gov/refseq] and Locus Reference Genomic (LRG, NCBI bases, clinical laboratories should make an effort to under- and EMBL-EBI; https://www.lrg-sequence.org/). This infor- stand their content and note their limitations (Li et al, mation is essential for the correct nomenclature of the vari- 2017). Likewise, establishing and consulting an in-house ants according to the Human Genome Variation Society database of variants identified in the laboratory is recom- (HGVS) (Li et al, 2017). mended. The tools recommended by the GESMD are listed in Table VI. Population databases. Population databases are useful for obtaining the frequency of variants at a given locus in a vari- Reference sequence databases. Reference sequence databases ety of geographically distinct populations (Cottrell et al, encompass the human reference genome and related infor- 2014; Richards et al, 2015). They are commonly used to filter mation for the unequivocal localization of a variant in the out polymorphic variants based on an established cut-off of genome. These databases annotate variant location (coding, minor allele frequency (MAF). Although a standardized MAF non-coding, splicing site or untranslated region) and the to discriminate between polymorphic or benign variants has strand representation (positive or negative) within the corre- not been established yet, current practice within GESMD sets sponding gene. Frequently consulted resources include 1% as the cut-off point in studies of somatic variants. Addi- Ensembl [European Molecular Biology Laboratory- European tionally, an ethnic specificity of the variant should be consid- Bioinformatics Institute (EMBL-EBI) and the Wellcome ered based on the ethnic background of the patient. Finally, Trust Sanger Institute; http://www.ensembl.org/index.htm], we would emphasise a cautionary note when consulting these RefSeq [National Center for Biotechnology Information databases, as they were built under the assumption that the (NCBI) Reference Sequence Database; https://www.ncbi.nlm. population included in them is healthy, but some somatic 614 ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd British Journal of Haematology, 2020, 188, 605–622 Guideline Table VI. List of web resources useful for variant interpretation. Database Website URL Reference sequence NCBI reference sequence database https://www.ncbi.nlm.nih.gov/genome Ensembl genome browser http://www.ensembl.org/index.htm Locus reference genomic https://www.lrg-sequence.org/ RefSeqGene https://www.ncbi.nlm.nih.gov/refseq MitoMap https://www.mitomap.org/MITOMAP/HumanMitoSeq UCSC genome browser https://genome.ucsc.edu/cgi-bin/hgTables Population databases dbSNP https://www.ncbi.nlm.nih.gov/snp/ Exome variant server http://evs.gs.washington.edu/EVS/ 1000 genomes project http://browser.1000genomes.org dbVar https://www.ncbi.nlm.nih.gov/dbvar Exome Aggregation Consortium (ExAC) http://exac.broadinstitute.org/ Genome Aggregation Database (gnomAD) http://gnomad.broadinstitute.org/ Short genetic variation https://www.ncbi.nlm.nih.gov/snp Somatic and constitutional variants databases National Cancer Institute’s Genome Data Commons https://gdc.cancer.gov Catalogue of Somatic Mutations in Cancer (COSMIC) http://cancer.sanger.ac.uk/cosmic IARC TP53 mutation database http://p53.iarc.fr/TP53GeneVariations.aspx TP53 website http://p53.fr/ Personalized cancer therapy https://pct.mdanderson.org cBioPortal http://www.cbioportal.org Intogen https://www.intogen.org/search ClinicalTrials.gov https://clinicaltrials.gov Pediatric Cancer Genome Project http://explorepcgp.org My Cancer Genome http://www.mycancergenome.org International Cancer Genome Consortium (ICGC) https://dcc.icgc.org The Cancer Genome Atlas (TCGA) https://cancergenome.nih.gov/ VarSome The Human Genomic Variant Search Engine https://varsome.com DECIPHER https://decipher.sanger.ac.uk/ ClinVar http://www.ncbi.nlm.nih.gov/clinvar Online Mendelian Inheritance in Man (OMIM) https://www.omim.org/ Human Gene Mutation Database (HGMD) http://www.hgmd.cf.ac.uk/ac/index.php Leiden Open Variation Database (LOVD) http://www.lovd.nl In-house laboratory databases RESMDmol https://www.gesmd.es/carrerasresearch/index.php Algorithms for in silico prediction PolyPhen2 http://genetics.bwh.harvard.edu/pph2 SIFT http://sift.jcvi.org Mutation Assessor http://mutationassessor.org Mutation Taster http://www.mutationtaster.org PROVEAN http://provean.jcvi.org/index.php CoVEC https://sourceforge.net/projects/covec/files CADD http://cadd.gs.washington.edu GERP++ http://mendel.stanford.edu/sidowlab/downloads/gerp/index.html PhyloP and PhastCons http://compgen.bscb.cornell.edu/phast ConSurf http://consurftest.tau.ac.il FATHMM http://fathmm.biocompute.org.uk PANTHER http://www.pantherdb.org/tools/csnpScoreForm.jsp PhD-SNP http://snps.biofold.org/phd-snp/phd-snp.html SNPs & GO http://snps-and-go.biocomp.unibo.it/snps-and-go Align GVGD http://agvgd.iarc.fr/agvgd_input.php MAPP http://mendel.stanford.edu/SidowLab/downloads/MAPP/index.html MutPred http://mutpred.mutdb.org nsSNPAnalyzer http://snpanalyzer.uthsc.edu ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology 615 and John Wiley & Sons Ltd. British Journal of Haematology, 2020, 188, 605–622 Guideline Table VI. (Continued) Database Website URL Condel http://bg.upf.edu/condel/home LRT http://www.genetics.wustl.edu/jflab/lrt_query.html DANN https://omictools.com/dann-tool Splice site prediction Human splicing finder http://www.umd.be/HSF3 MaxEntScan http://genes.mit.edu/burgelab/maxent/Xmaxentscan_scoreseq.html NetGene2 http://www.cbs.dtu.dk/services/NetGene2 NN plice http://www.fruitfly.org/seq_tools/splice.html GeneSplicer http://www.cbcb.umd.edu/software/GeneSplicer/gene_spl.shtml NNSplice http://www.fruitfly.org/seq_tools/splice.html FSPLICE http://www.softberry.com/berry.phtml?topic=fsplice&group=programs&subgroup=gfind variants may accidentally have been included as well. For Sanger Institute, which contains millions of somatic varia- example, common hotspot mutations, such as DNMT3A tions described in numerous types of tumours and has also R882 or JAK2 V617F, can be found in some of these data- recently included intron variants. However, the information bases and therefore a whitelist of known hotspot mutations provided in these databases is not always up-to-date. In addi- could be useful to rescue these variants. tion, a poor representation of pathological diagnostic stan- Recommended population databases include The Database dards is detected as well as a lack of an exhaustive validation of Short Genetic Variation (dbSNP, the NCBI database of of the findings and/or the sources from which the included genetic variation; https://www.ncbi.nlm.nih.gov/snp/), Exome variants were extracted. Variant Server (EVS, NHLBI GO Exome Sequencing Project; Germline variants are also usually detected upon tumour http://evs.gs.washington.edu/EVS/), 1000 Genomes Project sequencing. Such variants may be associated with cancer pre- (International Genome Sample Resource, IGSR; http://browse disposition syndromes. To evaluate them, there are several r.1000genomes.org), Exome Aggregation Consortium (ExAC, germline mutations databases, such as Human Gene Muta- Broad Institute; http://exac.broadinstitute.org/) and Genome tion Database (Institute of Medical Genetics; http://www. Aggregation Database (gnomAD, Broad Institute; http://gno hgmd.cf.ac.uk/ac/index.php) or ClinVar (NCBI; http://www. mad.broadinstitute.org/). Databases such as ExAC or gno- ncbi.nlm.nih.gov/clinvar). mAD are more accurate than dbSNP when it comes to Finally, for TP53 gene variant interpretation, the Interna- filtering polymorphisms and neutral variants. tional Agency for Research on Cancer (IARC) TP53 Database (WHO; http://p53.iarc.fr/TP53GeneVariations.aspx) and Somatic and germline variants in clinical databases. Both TP53 Website (Centre de Recherche des Cordeliers and somatic and germline variants are included in clinical data- Karolinska Institute; http://p53.fr/) are useful resources for bases. The incidence and prevalence of a variant is often evaluating their impact in haematological malignancies, detailed in different subtypes of cancers and functional pre- because they include the frequencies of the variants observed diction algorithms are provided to categorize the variants in both the somatic and germinal contexts. according to their pathogenicity. Frequently, clinical data- bases also include information for accurate annotation and In-house laboratory databases. We recommend that each prioritization of somatic variants; bibliographic references clinical laboratory should establish an in-house database to with or without systematic review; additional information provide consistent variant annotation and categorization, to about the tissue in which the variant has been described; determine the frequency of mutations and to identify plat- outcome disease data or targeted therapies, among others. form-specific artefacts or potential false-positive variants. A An example is the National Cancer Institute’s Genome Data continuous review of new publications, clinical trials and Commons (https://gdc.cancer.gov), which contains all the databases used in variant interpretation, which are regularly genomic data generated at the National Cancer Institute, and updated, should be carried out to ensure that variant annota- which includes the well-known databases of The Cancer Gen- tion is up-to-date. To increase the quality of interpretation, ome Atlas (TCGA; https://cancergenome.nih.gov/), the Ther- we recommend that in-house databases are complemented apeutically Applicable Research to Generate Effective with further relevant annotation, including results from other Therapies (TARGET; https://ocg.cancer.gov/programs/target), techniques such as morphology, cytogenetic and and the Cancer Genome Characterization Initiative (CGCI; immunophenotyping, because NGS results should always be https://ocg.cancer.gov/programs/cgci). Another public data- combined and interpreted in the context of other laboratory base is the Catalog of Somatic Mutations in Cancer (COS- and clinical diagnostic data. Moreover, it would be extremely MIC; http://cancer.sanger.ac.uk/cosmic) from the Wellcome useful to share those internal databases with other centres/ 616 ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd British Journal of Haematology, 2020, 188, 605–622 Guideline laboratories that perform similar determinations. Indeed, the clear clinical relevance, such as TP53 or JAK2. However, in GESMD is making that effort by building the RESMDmol order to report variants with a VAF below 5%, the depth of database (https://www.gesmd.es/carrerasresearch/index.php), coverage must be increased to maintain, at least, the minor in view of facilitating the homogenization of results informed allele coverage over 25 reads. in the clinical reports. The GESMD proposes that variants should be categorised according to the following criteria: (i) relevance of the Functional consequence of the variant. In silico predictive altered gene based on its actionability, defined as being rele- algorithms estimate the possible deleterious consequences of a vant for the diagnosis/classification, prognosis and/or treat- variant on the encoded protein of a given gene. The criteria ment (therapeutic target or related to sensitivity, toxicity or used to determine this effect varies across different tools. resistance to therapy) of the disease; (ii) presence of the vari- Algorithms designed to predict the functional consequence of ant in clinical databases and published literature that indi- a missense variant are commonly based on the evolutionary cates the recurrence or pathogenicity of the identified conservation of an amino acid or nucleotide, the location and variant; (iii) tissue and/or tumour histology in which the context within the protein sequence and the biochemical con- variant has been described; and (iv) in silico predictive algo- sequence of the amino acid substitution (Brunak et al, 1991; rithms and functional studies. Vreeswijk et al, 2009; Thusberg et al, 2011). In silico tools have Therefore, the categorization of the variant must be based also been developed to evaluate the consequence of splicing on its clinical impact on the specific disease and tissue under variants related to the creation or loss of splice sites at the study. Due to the importance of the definition of the action- exonic or intronic regions (Thusberg et al, 2011; Houdayer ability of a gene, we strongly recommend following the evi- et al, 2012). However, in the context of cancer, the interpreta- dence defined by Sukhai et al (2016). Of note, determining tion of these predictions is usually not easy, especially for the actionability of a gene will be particularly relevant when gain-of-function mutations. Therefore, it is recommended using pan-haematological or oncological targeted NGS panels that least four different prediction algorithms should be used that include genes which have not been previously described to analyse each variant, although the obtained output should in the disease of interest, as well as for WES and whole gen- never be used as sole evidence for clinical decision making, as, ome sequencing. to date, there is no consensus about the relevance of each pre- dictor, nor about how to combine outputs from different pre- Variant classification system diction algorithms (Li et al, 2017). In the case of variants with available information about According to our categorization criteria, variants will be clas- their functional effect, two scenarios can be considered. On sified into one of the following classes (Fig 2, Table VII): one hand, when the pathogenicity of the affected gene is associated with a loss of function, frameshift variants must Pathogenic (Tier I). Described in the literature as relevant be interpreted as probably pathogenic, as long as the variant for the clinical management of MDS/CMML or other mye- does not affect the last exon of the gene, unless the loid neoplasms. The variant affects an actionable gene and pathogenicity of those variants affecting the last exon has has been established as a pathogenic variant in myeloid neo- been validated (i.e. variants in ASXL1 and CALR). Loss of plasms. function variants located at the last exon of a gene will need evaluation according to other parameters, such as the type of Likely pathogenic (Tier II). Described in the literature but protein domain that is altered and the effect of other prox- with no established relevance in the clinical management of imity variants. Accordingly, these will be classified as variants MDS or other myeloid neoplasms. The variant affects an of uncertain significance or probably pathogenic. On the actionable gene but has been established as a pathogenic other hand, when the pathogenicity of the affected gene is variant in solid tumours or in non-myeloid haematological associated with a gain-of-function, and a frameshift variant is neoplasms. This category also includes variants previously detected, this will be interpreted as a variant of uncertain sig- not described for which predictive algorithms strongly clas- nificance unless there is strong evidence of pathogenicity sify them as pathogenic. based on functional studies in the literature. Variant of uncertain significance (VUS) (Tier II). Previously described but with no sufficient evidence of its pathogenicity Criteria for variant classification on actionable MDS genes, other haematological neoplasms We recommend to interpret and report variants that fulfil all or solid tumours. This category also includes variants not three following criteria: (i) MAF < 1%, (ii) VAF ≥ 5% and previously described for which predictive algorithms cannot (iii) minor allele coverage ≥25 reads. Although we recom- strongly classify them into pathogenic or benign. mend that cut-off of VAF ≥ 5% should be used, the analyst should consider informing variants with lower allele frequen- Likely benign (Tier IV) or benign (Tier V). Described in the cies in particular cases, such as hotspot regions in genes with literature as clinically insignificant because it has no effect on ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology 617 and John Wiley & Sons Ltd. British Journal of Haematology, 2020, 188, 605–622 Guideline the protein. Frequently, these variants present MAF > 1% Predictive algorithms and and, sometimes, VAF of approximately 50% or 100%. These levels of VAF may be indicative of the germline nature of the functional studies Likely pathogenic Probably benign Probably benign variant and, therefore, are neither acquired nor tumour-spe- cific. However, the VAF criterion should be considered with caution, as there are somatic variants present in major clones with high VAF. This category also includes variants previ- NA NA ously not described for which predictive algorithms strongly classify them as benign. We recommend that, to avoid confu- Other haematological neoplasms Other haematological neoplasms Tissue and/or tumour histology sion, benign and likely benign variants should not be Disease or tissue of interest Disease or tissue of interest included in the clinical report. Clinical report or solid tumors or solid tumors Next generation sequencing clinical reports should follow the general professional organizations’ recommendations and guidelines (Richardson, 2002; Rack et al, 2019), such as NA avoiding long reports for the benefit of clarity, inclusion of pagination, and reviewing and signing by, preferably, two Variant recurrence in databases geneticists with the relevant expertise on haematological Described and confirmed Described and confirmed malignancies. We recommend that the report includes (at Described as pathogenic Unknown pathogenicity least) the information detailed below. Described as benign as pathogenic Patient identification as benign As in current standard tests, proper patient identification should be incorporated in the clinical report, including two unique patient identifiers, demographic data and referral information (reason for referral, suspected or confirmed treatment significance in in the disease of interest (Biomarker) haematological neoplasms or Diagnostic, prognostic and/or diagnosis and clinical time point). Sample details should also Clinical significance in other Variant clinical significance be provided (type, source and received date) as well as hospi- Of uncertain significance tal and referring physician. A statement of signed informed Clinically irrelevant Clinically irrelevant consent should also be made. solid tumors Methodology Details on the methodology used should be specified, includ- ing library preparation chemistry and sequencing instrument. Regions included in the panel (genes according to the Actionable/not actionable Actionable/not actionable Actionable/not actionable Human Genome Organisation nomenclature, exons and hot- spots) and low covered regions should be detailed. Analysis Gene actionability settings, filters and cut-off values (such as minimum cover- age or VAF) should be specified. The version of the human Actionable Actionable Criterion reference sequence to which sequence reads are aligned Table VII. Classification variant system. should also be included, as well as the clinical databases used for annotation. Key quality control metrics and limitations need also be listed. significance (VUS) (Tier III) Likely pathogenic (Tier II) Likely benign (Tier IV) Results and interpretation Variant of uncertain NA, not applicable. Pathogenic (Tier I) A list of identified variants should be included in the results Benign (Tier V) section, annotated in accordance to the HGVS mutation nomenclature. For each variant, we recommend including Category the following information: gene, exon, nucleotide change, protein change, type of variant (missense, frameshift, 618 ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd British Journal of Haematology, 2020, 188, 605–622 Guideline nonsense, splicing), transcript RefSeq ID, variant ID from is supported by a grant from 2017 SGR288 (GRC) Generali- clinical databases, VAF and position depth of coverage. tat de Catalunya, economical support from CERCA Pro- When using a variant classification system, the categorization gramme/Generalitat de Catalunya, Fundaci o Internacional of each variant should be included and the used classification Josep Carreras and from Celgene International. LP and system should be described. Regarding variant interpretation, JMHS are supported by a research grant from FEHH (Fun- we recommend including a critical summary of the clinical dacion Espa~ nola de Hematologıa y Hemoterapia, 2017). IV relevance of the variant in the disease of interest, including acknowledges support from Pethema. MC and LZ research treatment or referenced clinical trials information, if applica- is supported by a grant from Instituto de Salud Carlos III, ble. For the categorization of the disease, the WHO 2017 Ministerio de Sanidad y Consumo, Spain (PI 11/02519). nomenclature should be used (Arber et al, 2016). Further MFM and her research is supported by the Spanish Associa- studies to validate the significance of the results obtained tion against Cancer (AECC, AIO2014), and the Ministerio should be conducted when appropriate. de Economıa y Competitividad of Spanish Central Govern- ment (PI16/00159). Conclusion Author contribution We have long awaited the introduction of NGS into the rou- tine diagnostic armamentarium of myeloid malignancies. All authors contributed in the development of these guideli- Currently, an increasing number of laboratories are imple- nes and in the revision of the manuscript. menting NGS procedures into the diagnostic algorithms of patients with haematological malignancies, opening new Supporting Information horizons for individualized clinical management of these patients. It is therefore of major importance to standardize Additional supporting information may be found online in the generation, analysis and clinical interpretation of NGS the Supporting Information section at the end of the article. data. To that end, the GESMD has expanded the present set of guidelines, aiming at stablishing common quality stan- Table S1. Main characteristics and differences between dards for the adequate clinical interpretation of NGS results, amplicon- and capture-based panels. hoping that this effort will ultimately contribute to the bene- Table S2. Comparison of Illumina and Thermo Fisher fit of patients with myeloid malignancies. NGS platforms. Table S3. Main sequencing parameters. Table S4. List of most commonly used alignment tools. Acknowledgements Table S5. List of most commonly used somatic variant This work was supported by a grant from the Spanish callers. Group of MDS (GESMD, 2017). LP, FF, PA and FS research Data S1. Supplementary methods. References Bejar, R., Stevenson, K., Abdel-Wahab, O., Galili, American Society of Clinical Oncology, 32, N., Nilsson, B., Garcia-Manero, G., Kantarjian, 2691–2698. Van Allen, E.M., Wagle, N., Stojanov, P., Perrin, H., Raza, A., Levine, R.L., Neuberg, D. & Ebert, Berger, G., van den Berg, E., Sikkema-Raddatz, B., D.L., Cibulskis, K., Marlow, S., Jane-Valbuena, J., B.L. (2011) Clinical effect of point mutations in Abbott, K.M., Sinke, R.J., Bungener, L.B., Mulder, Friedrich, D.C., Kryukov, G., Carter, S.L., myelodysplastic syndromes. The New England A.B. & Vellenga, E. (2017) Re-emergence of acute McKenna, A., Sivachenko, A., Rosenberg, M., Journal of Medicine, 364, 2496–2506. myeloid leukemia in donor cells following allo- Kiezun, A., Voet, D., Lawrence, M., Lichtenstein, Bejar, R., Stevenson, K.E., Caughey, B.A., Abdel- geneic transplantation in a family with a germline L.T., Gentry, J.G., Huang, F.W., Fostel, J., Farlow, Wahab, O., Steensma, D.P., Galili, N., Raza, A., DDX41 mutation. Leukemia, 31, 520–522. D., Barbie, D., Gandhi, L., Lander, E.S., Gray, Kantarjian, H., Levine, R.L., Neuberg, D., Gar- Boultwood, J., Perry, J., Pellagatti, A., Fernandez- S.W., Joffe, S., Janne, P., Garber, J., MacConaill, cia-Manero, G. & Ebert, B.L. (2012) Validation Mercado, M., Fernandez-Santamaria, C., Cala- L., Lindeman, N., Rollins, B., Kantoff, P., Fisher, of a prognostic model and the impact of muta- sanz, M.J., Larrayoz, M.J., Garcia-Delgado, M., S.A., Gabriel, S., Getz, G. & Garraway, L.A. tions in patients with lower-risk myelodysplastic Giagounidis, A., Malcovati, L., Della Porta, M.G., (2014) Whole-exome sequencing and clinical syndromes. Journal of Clinical Oncology: Official J€adersten, M., Killick, S., Hellstr€ om-Lindberg, E., interpretation of formalin-fixed, paraffin-embed- Journal of the American Society of Clinical Oncol- Cazzola, M. & Wainscoat, J.S. (2010) Frequent ded tumor samples to guide precision cancer ogy, 30, 3376–3382. mutation of the polycomb-associated gene ASXL1 medicine. Nature Medicine, 20, 682–688. Bejar, R., Stevenson, K.E., Caughey, B., Lindsley, in the myelodysplastic syndromes and in acute Arber, D.A., Orazi, A., Hasserjian, R., Thiele, J., R.C., Mar, B.G., Stojanov, P., Getz, G., myeloid leukemia. Leukemia, 24, 1062–1065. Borowitz, M.J., Le Beau, M.M., Bloomfield, Steensma, D.P., Ritz, J., Soiffer, R., Antin, J.H., Brunak, S., Engelbrecht, J. & Knudsen, S. (1991) C.D., Cazzola, M. & Vardiman, J.W. (2016) The Alyea, E., Armand, P., Ho, V., Koreth, J., Neu- Prediction of human mRNA donor and acceptor 2016 revision to the World Health Organization berg, D., Cutler, C.S. & Ebert, B.L. (2014) sites from the DNA sequence. Journal of Molecu- classification of myeloid neoplasms and acute Somatic mutations predict poor outcome in lar Biology, 220, 49–65. leukemia. Blood, 127, 2391–2405. patients with myelodysplastic syndrome after Cazzola, M., Della Porta, M.G. & Malcovati, L. Bejar, R. (2017) CHIP, ICUS, CCUS and other hematopoietic stem-cell transplantation. Journal (2013) The genetic basis of myelodysplasia and four-letter words. Leukemia, 31, 1869–1871. of Clinical Oncology: Official Journal of the its clinical relevance. Blood, 122, 4021–4034. ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology 619 and John Wiley & Sons Ltd. British Journal of Haematology, 2020, 188, 605–622 Guideline Chen, C.-Y., Lin, L.-I., Tang, J.-L., Ko, B.-S., Tsay, Sancho-Tello, R., Catricala, S., Vicente, A.I., Schnittger, S. (2011) Molecular profiling of W., Chou, W.-C., Yao, M., Wu, S.-J., Tseng, Haferlach, T., Haferlach, C., Sanz, G.F., Malcov- chronic myelomonocytic leukemia reveals M.-H. & Tien, H.-F. (2007) RUNX1 gene muta- ati, L. & Cazzola, M. (2016) Integrating clinical diverse mutations in >80% of patients with tion in primary myelodysplastic syndrome–the features and genetic lesions in the risk assess- TET2 and EZH2 being of high prognostic rele- mutation can be detected early at diagnosis or ment of patients with chronic myelomonocytic vance. Leukemia, 25, 877–879. acquired during disease progression and is asso- leukemia. Blood, 128, 1408–1417. Haferlach, T., Nagata, Y., Grossmann, V., Okuno, ciated with poor outcome. British Journal of Ganguly, B.B. & Kadam, N.N. (2016) Mutations of Y., Bacher, U., Nagae, G., Schnittger, S., Sanada, Haematology, 139, 405–414. myelodysplastic syndromes (MDS): an update. M., Kon, A., Alpermann, T., Yoshida, K., Roller, Cottrell, C.E., Al-Kateb, H., Bredemeyer, A.J., Dun- Mutation Research. Reviews in Mutation A., Nadarajah, N., Shiraishi, Y., Shiozawa, Y., cavage, E.J., Spencer, D.H., Abel, H.J., Lockwood, Research, 769, 47–62. Chiba, K., Tanaka, H., Koeffler, H.P., Klein, H.- C.M., Hagemann, I.S., O’Guin, S.M., Burcea, L.C., Gelsi-Boyer, V., Brecqueville, M., Devillier, R., U., Dugas, M., Aburatani, H., Kohlmann, A., Sawyer, C.S., Oschwald, D.M., Stratman, J.L., Sher, Murati, A., Mozziconacci, M.-J. & Birnbaum, D. Miyano, S., Haferlach, C., Kern, W. & Ogawa, S. D.A., Johnson, M.R., Brown, J.T., Cliften, P.F., (2012) Mutations in ASXL1 are associated with (2014) Landscape of genetic lesions in 944 George, B., McIntosh, L.D., Shrivastava, S., poor prognosis across the spectrum of malig- patients with myelodysplastic syndromes. Leuke- Nguyen, T.T., Payton, J.E., Watson, M.A., Crosby, nant myeloid diseases. Journal of Hematology & mia, 28, 241–247. S.D., Head, R.D., Mitra, R.D., Nagarajan, R., Oncology, 5, 12. Hoskinson, D.C., Dubuc, A.M. & Mason-Suares, Kulkarni, S., Seibert, K., Virgin, H.W., Milbrandt, Genovese, G., K€ahler, A.K., Handsaker, R.E., Lind- H. (2017) The current state of clinical interpre- J. & Pfeifer, J.D. (2014) Validation of a next-gener- berg, J., Rose, S.A., Bakhoum, S.F., Chambert, tation of sequence variants. Current Opinion in ation sequencing assay for clinical molecular K., Mick, E., Neale, B.M., Fromer, M., Purcell, Genetics & Development, 42, 33–39. oncology. The Journal of Molecular Diagnostics: S.M., Svantesson, O., Landen, M., H€ oglund, M., Houdayer, C., Caux-Moncoutier, V., Krieger, S., JMD, 16, 89–105. Lehmann, S., Gabriel, S.B., Moran, J.L., Lander, Barrois, M., Bonnet, F., Bourdon, V., Bronner, Damm, F., Chesnais, V., Nagata, Y., Yoshida, K., E.S., Sullivan, P.F., Sklar, P., Gr€ onberg, H., M., Buisson, M., Coulet, F., Gaildrat, P., Lefol, Scourzic, L., Okuno, Y., Itzykson, R., Sanada, Hultman, C.M. & McCarroll, S.A. (2014) Clonal C., Leone, M., Mazoyer, S., Muller, D., Reme- M., Shiraishi, Y., Gelsi-Boyer, V., Renneville, A., hematopoiesis and blood-cancer risk inferred nieras, A., Revillion, F., Rouleau, E., Sokolowska, Miyano, S., Mori, H., Shih, L.-Y., Park, S., Drey- from blood DNA sequence. The New England J., Vert, J.-P., Lidereau, R., Soubrier, F., Sobol, fus, F., Guerci-Bresler, A., Solary, E., Rose, C., Journal of Medicine, 371, 2477–2487. H., Sevenet, N., Bressac-de Paillerets, B., Har- Cheze, S., Prebet, T., Vey, N., Legentil, M., Duf- Gillis, N.K., Ball, M., Zhang, Q., Ma, Z., Zhao, Y., douin, A., Tosi, M., Sinilnikova, O.M. & fourd, Y., de Botton, S., Preudhomme, C., Birn- Yoder, S.J., Balasis, M.E., Mesa, T.E., Sallman, Stoppa-Lyonnet, D. (2012) Guidelines for splic- baum, D., Bernard, O.A., Ogawa, S., Fontenay, D.A., Lancet, J.E., Komrokji, R.S., List, A.F., ing analysis in molecular diagnosis derived from M. & Kosmider, O. (2013a) BCOR and McLeod, H.L., Alsina, M., Baz, R., Shain, K.H., a set of 327 combined in silico/in vitro studies BCORL1 mutations in myelodysplastic syn- Rollison, D.E. & Padron, E. (2017) Clonal hae- on BRCA1 and BRCA2 variants. Human Muta- dromes and related disorders. Blood, 122, 3169– mopoiesis and therapy-related myeloid malig- tion, 33, 1228–1238. 3177. nancies in elderly patients: a proof-of-concept, Itzykson, R., Kosmider, O., Renneville, A., Gelsi- Damm, F., Itzykson, R., Kosmider, O., Droin, N., case-control study. The Lancet. Oncology, 18, Boyer, V., Meggendorfer, M., Morabito, M., Renneville, A., Chesnais, V., Gelsi-Boyer, V., de 112–121. Berthon, C., Ades, L., Fenaux, P., Beyne-Rauzy, Botton, S., Vey, N., Preudhomme, C., Clavert, Godley, L.A. & Shimamura, A. (2017) Genetic pre- O., Vey, N., Braun, T., Haferlach, T., Dreyfus, A., Delabesse, E., Park, S., Birnbaum, D., Fonte- disposition to hematologic malignancies: man- F., Cross, N.C.P., Preudhomme, C., Bernard, nay, M., Bernard, O.A. & Solary, E. (2013b) agement and surveillance. Blood, 130, 424–432. O.A., Fontenay, M., Vainchenker, W., Sch- SETBP1 mutations in 658 patients with Graubert, T.A., Shen, D., Ding, L., Okeyo-Owuor, nittger, S., Birnbaum, D., Droin, N. & Solary, E. myelodysplastic syndromes, chronic T., Lunn, C.L., Shao, J., Krysiak, K., Harris, C.C., (2013a) Prognostic score including gene muta- myelomonocytic leukemia and secondary acute Koboldt, D.C., Larson, D.E., McLellan, M.D., tions in chronic myelomonocytic leukemia. myeloid leukemias. Leukemia, 27, 1401–1403. Dooling, D.J., Abbott, R.M., Fulton, R.S., Sch- Journal of Clinical Oncology: Official Journal of Delhommeau, F., Dupont, S., Della Valle, V., midt, H., Kalicki-Veizer, J., O’Laughlin, M., Gril- the American Society of Clinical Oncology, 31, James, C., Trannoy, S., Masse, A., Kosmider, O., lot, M., Baty, J., Heath, S., Frater, J.L., Nasim, T., 2428–2436. Le Couedic, J.-P., Robert, F., Alberdi, A., Link, D.C., Tomasson, M.H., Westervelt, P., Itzykson, R., Itzkson, R., Fenaux, P. & Solary, E. Lecluse, Y., Plo, I., Dreyfus, F.J., Marzac, C., DiPersio, J.F., Mardis, E.R., Ley, T.J., Wilson, (2013b) Chronic myelomonocytic leukemia: Casadevall, N., Lacombe, C., Romana, S.P., Des- R.K. & Walter, M.J. (2011) Recurrent mutations myelodysplastic or myeloproliferative? Best Prac- sen, P., Soulier, J., Viguie, F., Fontenay, M., in the U2AF1 splicing factor in myelodysplastic tice & Research. Clinical Haematology, 26, 387– Vainchenker, W. & Bernard, O.A. (2009) Muta- syndromes. Nature Genetics, 44, 53–57. 400. tion in TET2 in myeloid cancers. The New Eng- Greenberg, P.L., Stone, R.M., Al-Kali, A., Barta, Itzykson, R., Kosmider, O., Cluzeau, T., Mansat- land Journal of Medicine, 360, 2289–2301. S.K., Bejar, R., Bennett, J.M., Carraway, H., De De Mas, V., Dreyfus, F., Beyne-Rauzy, O., Ques- Duncavage, E.J., Uy, G.L., Petti, A.A., Miller, C.A., Castro, C.M., Deeg, H.J., DeZern, A.E., Fathi, nel, B., Vey, N., Gelsi-Boyer, V., Raynaud, S., Lee, Y.-S., Tandon, B., Gao, F., Fronick, C.C., A.T., Frankfurt, O., Gaensler, K., Garcia-Man- Preudhomme, C., Ades, L., Fenaux, P. & Fonte- O’Laughlin, M., Fulton, R.S., Wilson, R.K., ero, G., Griffiths, E.A., Head, D., Horsfall, R., nay, M.; Groupe Francophone des Myelodys- Jacoby, M.A., Cashen, A.F., Wartman, L.D., Johnson, R.A., Juckett, M., Klimek, V.M., Kom- plasies (GFM). (2011) Impact of TET2 Walter, M.J., Westervelt, P., Link, D.C., DiPer- rokji, R., Kujawski, L.A., Maness, L.J., O’Don- mutations on response rate to azacitidine in sio, J.F., Ley, T.J. & Welch, J.S. (2017) Muta- nell, M.R., Pollyea, D.A., Shami, P.J., Stein, B.L., myelodysplastic syndromes and low blast count tional landscape and response are conserved in Walker, A.R., Westervelt, P., Zeidan, A., Shead, acute myeloid leukemias. Leukemia, 25, 1147– peripheral blood of AML and MDS patients D.A. & Smith, C. (2017) Myelodysplastic syn- 1152. during decitabine therapy. Blood, 129, 1397– dromes, version 2.2017, NCCN clinical practice J€adersten, M., Saft, L., Smith, A., Kulasekararaj, A., 1401. guidelines in oncology. Journal of the National Pomplun, S., G€ ohring, G., Hedlund, A., Hast, Elena, C., Gallı, A., Such, E., Meggendorfer, M., Comprehensive Cancer Network: JNCCN, 15, 60– R., Schlegelberger, B., Porwit, A., Hellstr€ om- Germing, U., Rizzo, E., Cervera, J., Molteni, E., 87. Lindberg, E. & Mufti, G.J. (2011) TP53 muta- Fasan, A., Schuler, E., Ambaglio, I., Lopez-Pavia, Grossmann, V., Kohlmann, A., Eder, C., Haferlach, tions in low-risk myelodysplastic syndromes M., Zibellini, S., Kuendgen, A., Travaglino, E., C., Kern, W., Cross, N.C.P., Haferlach, T. & with del(5q) predict disease progression. Journal 620 ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd British Journal of Haematology, 2020, 188, 605–622 Guideline of Clinical Oncology: Official Journal of the Amer- Kulasekararaj, A.G., Smith, A.E., Mian, S.A., Stromburg, M., Brockmann, A., Tr€ umper, L., ican Society of Clinical Oncology, 29, 1971–1979. Mohamedali, A.M., Krishnamurthy, P., Lea, Sole, F. & Haase, D. (2018) Targeted deep Jaiswal, S., Fontanillas, P., Flannick, J., Manning, N.C., G€aken, J., Pennaneach, C., Ireland, R., sequencing of CD34+ cells from peripheral A., Grauman, P.V., Mar, B.G., Lindsley, R.C., Czepulkowski, B., Pomplun, S., Marsh, J.C. & blood can reproduce bone marrow molecular Mermel, C.H., Burtt, N., Chavez, A., Higgins, Mufti, G.J. (2013) TP53 mutations in myelodys- profile in myelodysplastic syndromes. American J.M., Moltchanov, V., Kuo, F.C., Kluk, M.J., plastic syndrome are strongly correlated with Journal of Hematology, 93, E152–E154. Henderson, B., Kinnunen, L., Koistinen, H.A., aberrations of chromosome 5, and correlate with Matthijs, G., Souche, E., Alders, M., Corveleyn, A., Ladenvall, C., Getz, G., Correa, A., Banahan, adverse prognosis. British Journal of Haematol- Eck, S., Feenstra, I., Race, V., Sistermans, E., B.F., Gabriel, S., Kathiresan, S., Stringham, ogy, 160, 660–672. Sturm, M., Weiss, M., Yntema, H., Bakker, E., H.M., McCarthy, M.I., Boehnke, M., Tuomile- Li, M.M., Datto, M., Duncavage, E.J., Kulkarni, S., Scheffer, H. & Bauer, P. (2016) Guidelines for hto, J., Haiman, C., Groop, L., Atzmon, G., Wil- Lindeman, N.I., Roy, S., Tsimberidou, A.M., diagnostic next-generation sequencing. European son, J.G., Neuberg, D., Altshuler, D. & Ebert, Vnencak-Jones, C.L., Wolff, D.J., Younes, A. & journal of human genetics: EJHG, 24, 1515. B.L. (2014) Age-related clonal hematopoiesis Nikiforova, M.N. (2017) Standards and guidelines McClure, R.F., Ewalt, M.D., Crow, J., Temple- associated with adverse outcomes. The New Eng- for the interpretation and reporting of sequence Smolkin, R.L., Pullambhatla, M., Sargent, R. & land Journal of Medicine, 371, 2488–2498. variants in cancer: a joint consensus recommenda- Kim, A.S. (2018) Clinical significance of DNA vari- Jaiswal, S., Natarajan, P., Silver, A.J., Gibson, C.J., tion of the Association for Molecular Pathology, ants in chronic myeloid neoplasms: a report of the Bick, A.G., Shvartz, E., McConkey, M., Gupta, American Society of Clinical Oncology, and Col- association for molecular pathology. The Journal of N., Gabriel, S., Ardissino, D., Baber, U., Meh- lege of American Pathologists. The Journal of molecular diagnostics: JMD, 20, 717–737. ran, R., Fuster, V., Danesh, J., Frossard, P., Sale- molecular diagnostics: JMD, 19, 4–23. Meggendorfer, M., Roller, A., Haferlach, T., Eder, heen, D., Melander, O., Sukhova, G.K., Makishima, H., Visconte, V., Sakaguchi, H., Jan- C., Dicker, F., Grossmann, V., Kohlmann, A., Neuberg, D., Libby, P., Kathiresan, S. & Ebert, kowska, A.M., Abu Kar, S., Jerez, A., Przy- Alpermann, T., Yoshida, K., Ogawa, S., Koeffler, B.L. (2017) Clonal hematopoiesis and risk of chodzen, B., Bupathi, M., Guinta, K., Afable, H.P., Kern, W., Haferlach, C. & Schnittger, S. atherosclerotic cardiovascular disease. The New M.G., Sekeres, M.A., Padgett, R.A., Tiu, R.V. & (2012) SRSF2 mutations in 275 cases with England Journal of Medicine, 377, 111–121. Maciejewski, J.P. (2012) Mutations in the spliceo- chronic myelomonocytic leukemia (CMML). Kennedy, J.A. & Ebert, B.L. (2017) Clinical impli- some machinery, a novel and ubiquitous pathway Blood, 120, 3080–3088. cations of genetic mutations in myelodysplastic in leukemogenesis. Blood, 119, 3203–3210. Meggendorfer, M., Haferlach, C., Kern, W. & syndrome. Journal of Clinical Oncology: Official Makishima, H., Yoshizato, T., Yoshida, K., Sekeres, Haferlach, T. (2017) Molecular analysis of Journal of the American Society of Clinical Oncol- M.A., Radivoyevitch, T., Suzuki, H., Przy- myelodysplastic syndrome with isolated deletion ogy, 35, 968–974. chodzen, B., Nagata, Y., Meggendorfer, M., of the long arm of chromosome 5 reveals a Kennedy, A.L. & Shimamura, A. (2019) Genetic Sanada, M., Okuno, Y., Hirsch, C., Kuzmanovic, specific spectrum of molecular mutations with predisposition to MDS: clinical features and clo- T., Sato, Y., Sato-Otsubo, A., LaFramboise, T., prognostic impact: a study on 123 patients and nal evolution. Blood, 133, 1071–1085. Hosono, N., Shiraishi, Y., Chiba, K., Haferlach, 27 genes. Haematologica, 102, 1502–1510. Kita-Sasai, Y., Horiike, S., Misawa, S., Kaneko, H., C., Kern, W., Tanaka, H., Shiozawa, Y., G omez- Ok, C.Y., Patel, K.P., Garcia-Manero, G., Routbort, Kobayashi, M., Nakao, M., Nakagawa, H., Fujii, Seguı, I., Husseinzadeh, H.D., Thota, S., Guinta, M.J., Fu, B., Tang, G., Goswami, M., Singh, R., H. & Taniwaki, M. (2001) International prog- K.M., Dienes, B., Nakamaki, T., Miyawaki, S., Kanagal-Shamanna, R., Pierce, S.A., Young, K.H., nostic scoring system and TP53 mutations are Saunthararajah, Y., Chiba, S., Miyano, S., Shih, Kantarjian, H.M., Medeiros, L.J., Luthra, R. & independent prognostic indicators for patients L.Y., Haferlach, T., Ogawa, S. & Maciejewski, Wang, S.A. (2015) Mutational profiling of ther- with myelodysplastic syndrome. British Journal J.P. (2017) Dynamics of clonal evolution in apy-related myelodysplastic syndromes and acute of Haematology, 115, 309–312. myelodysplastic syndromes. Nature Genetics, 49, myeloid leukemia by next generation sequencing, Kohlmann, A., Grossmann, V., Klein, H.-U., 204–212. a comparison with de novo diseases. Leukemia Schindela, S., Weiss, T., Kazak, B., Dicker, F., Malcovati, L., Hellstr€ om-Lindberg, E., Bowen, D., Research, 39, 348–354. Schnittger, S., Dugas, M., Kern, W., Haferlach, Ades, L., Cermak, J., Del Ca~ nizo, C., Della Papaemmanuil, E., Cazzola, M., Boultwood, J., C. & Haferlach, T. (2010) Next-generation Porta, M.G., Fenaux, P., Gattermann, N., Ger- Malcovati, L., Vyas, P., Bowen, D., Pellagatti, A., sequencing technology reveals a characteristic ming, U., Jansen, J.H., Mittelman, M., Mufti, Wainscoat, J.S., Hellstrom-Lindberg, E., Gamba- pattern of molecular mutations in 72.8% of G., Platzbecker, U., Sanz, G.F., Selleslag, D., corti-Passerini, C., Godfrey, A.L., Rapado, I., chronic myelomonocytic leukemia by detecting Skov-Holm, M., Stauder, R., Symeonidis, A., Cvejic, A., Rance, R., McGee, C., Ellis, P., frequent alterations in TET2, CBL, RAS, and van de Loosdrecht, A.A., de Witte, T. & Cazzola, Mudie, L.J., Stephens, P.J., McLaren, S., Massie, RUNX1. Journal of Clinical Oncology: Official M. (2013) Diagnosis and treatment of primary C.E., Tarpey, P.S., Varela, I., Nik-Zainal, S., Journal of the American Society of Clinical Oncol- myelodysplastic syndromes in adults: recom- Davies, H.R., Shlien, A., Jones, D., Raine, K., ogy, 28, 3858–3865. mendations from the European LeukemiaNet. Hinton, J., Butler, A.P., Teague, J.W., Baxter, Kon, A., Shih, L.-Y., Minamino, M., Sanada, M., Blood, 122, 2943–2964. E.J., Score, J., Galli, A., Della Porta, M.G., Tra- Shiraishi, Y., Nagata, Y., Yoshida, K., Okuno, Y., Malcovati, L., Papaemmanuil, E., Ambaglio, I., vaglino, E., Groves, M., Tauro, S., Munshi, Bando, M., Nakato, R., Ishikawa, S., Sato- Elena, C., Gallı, A., Della Porta, M.G., Trava- N.C., Anderson, K.C., El-Naggar, A., Fischer, A., Otsubo, A., Nagae, G., Nishimoto, A., Haferlach, glino, E., Pietra, D., Pascutto, C., Ubezio, M., Mustonen, V., Warren, A.J., Cross, N.C., Green, C., Nowak, D., Sato, Y., Alpermann, T., Naga- Bono, E., Da Via, M.C., Brisci, A., Bruno, F., A.R., Futreal, P.A., Stratton, M.R. & Campbell, saki, M., Shimamura, T., Tanaka, H., Chiba, K., Cremonesi, L., Ferrari, M., Boveri, E., Invernizzi, P.J. (2011) Somatic SF3B1 mutation in Yamamoto, R., Yamaguchi, T., Otsu, M., Obara, R., Campbell, P.J. & Cazzola, M. (2014) Driver myelodysplasia with ring sideroblasts. The New N., Sakata-Yanagimoto, M., Nakamaki, T., somatic mutations identify distinct disease enti- England Journal of Medicine, 365, 1384–1395. Ishiyama, K., Nolte, F., Hofmann, W.K., Miya- ties within myeloid neoplasms with myelodys- Papaemmanuil, E., Gerstung, M., Malcovati, L., waki, S., Chiba, S., Mori, H., Nakauchi, H., plasia. Blood, 124, 1513–1521. Tauro, S., Gundem, G., Van Loo, P., Yoon, C.J., Koeffler, H.P., Aburatani, H., Haferlach, T., Shi- Martin, R., Acha, P., Ganster, C., Palomo, L., Ellis, P., Wedge, D.C., Pellagatti, A., Shlien, A., rahige, K., Miyano, S. & Ogawa, S. (2013) Dierks, S., Fuster-Tormo, F., Mallo, M., Adema, Groves, M.J., Forbes, S.A., Raine, K., Hinton, J., Recurrent mutations in multiple components of V., Gomez-Marzo, P., De Haro, N., Solanes, N., Mudie, L.J., McLaren, S., Hardy, C., Latimer, C., the cohesin complex in myeloid neoplasms. Nat- Zamora, L., Xicoy, B., Shirneshan, K., Flach, J., Della Porta, M.G., O’Meara, S., Ambaglio, I., ure Genetics, 45, 1232–1237. Braulke, F., Schanz, J., Kominowski, A., Galli, A., Butler, A.P., Walldin, G., Teague, J.W., ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology 621 and John Wiley & Sons Ltd. British Journal of Haematology, 2020, 188, 605–622 Guideline Quek, L., Sternberg, A., Gambacorti-Passerini, M., Haferlach, T., Cross, N.C.P., Hochhaus, A., Thusberg, J., Olatubosun, A. & Vihinen, M. (2011) C., Cross, N.C., Green, A.R., Boultwood, J., Hofmann, W.-K. & Reiter, A. (2012) Activating Performance of mutation pathogenicity predic- Vyas, P., Hellstrom-Lindberg, E., Bowen, D., CBL mutations are associated with a distinct tion methods on missense variants. Human Cazzola, M., Stratton, M.R., Campbell, P.J. MDS/MPN phenotype. Annals of Hematology, Mutation, 32, 358–368. (2013) Clinical and biological implications of 91, 1713–1720. Traina, F., Visconte, V., Elson, P., Tabarroki, A., driver mutations in myelodysplastic syndromes. Shih, A.H., Abdel-Wahab, O., Patel, J.P. & Levine, Jankowska, A.M., Hasrouni, E., Sugimoto, Y., Blood, 122, 3616–3627; quiz 3699. R.L. (2012) The role of mutations in epigenetic Szpurka, H., Makishima, H., O’Keefe, C.L., Sek- Patnaik, M.M., Barraco, D., Lasho, T.L., Finke, regulators in myeloid malignancies. Nature eres, M.A., Advani, A.S., Kalaycio, M., Copelan, C.M., Hanson, C.A., Ketterling, R.P., Gangat, N. Reviews. Cancer, 12, 599–612. E.A., Saunthararajah, Y., Olalla Saad, S.T., & Tefferi, A. (2017) DNMT3A mutations are Smith, A.E., Kulasekararaj, A.G., Jiang, J., Mian, S., Maciejewski, J.P. & Tiu, R.V. (2014) Impact of associated with inferior overall and leukemia- Mohamedali, A., Gaken, J., Ireland, R., Czepulk- molecular mutations on treatment response to free survival in chronic myelomonocytic leuke- owski, B., Best, S. & Mufti, G.J. (2015) DNMT inhibitors in myelodysplasia and related mia. American Journal of Hematology, 92, 56–61. CSNK1A1 mutations and isolated del(5q) neoplasms. Leukemia, 28, 78–87. Pich, A., Riera, L., Sismondi, F., Godio, L., Davico abnormality in myelodysplastic syndrome: a ret- Vreeswijk, M.P.G., Kraan, J.N., van der Klift, Bonino, L., Marmont, F. & Francia di Celle, P. rospective mutational analysis. The Lancet. Hae- H.M., Vink, G.R., Cornelisse, C.J., Wijnen, J.T., (2009) JAK2V617F activating mutation is associ- matology, 2, e212–221. Bakker, E., van Asperen, C.J. & Devilee, P. ated with the myeloproliferative type of chronic Steensma, D.P., Bejar, R., Jaiswal, S., Lindsley, (2009) Intronic variants in BRCA1 and BRCA2 myelomonocytic leukaemia. Journal of Clinical R.C., Sekeres, M.A., Hasserjian, R.P. & Ebert, that affect RNA splicing can be reliably selected Pathology, 62, 798–801. B.L. (2015) Clonal hematopoiesis of indetermi- by splice-site prediction programs. Human Rack, K.A., van den Berg, E., Haferlach, C., Beverloo, nate potential and its distinction from Mutation, 30, 107–114. H.B., Costa, D., Espinet, B., Foot, N., Jeffries, S., myelodysplastic syndromes. Blood, 126, 9–16. Wall, M., Rayeroux, K.C., MacKinnon, R.N., Zor- Martin, K., O’Connor, S., Schoumans, J., Talley, P., Sukhai, M.A., Craddock, K.J., Thomas, M., Hansen, dan, A. & Campbell, L.J. (2012) ETV6 deletion Telford, N., Stioui, S., Zemanova, Z. & Hastings, A.R., Zhang, T., Siu, L., Bedard, P., Stockley, T.L. is a common additional abnormality in patients R.J. (2019) European recommendations and qual- & Kamel-Reid, S. (2016) A classification system with myelodysplastic syndromes or acute mye- ity assurance for cytogenomic analysis of haemato- for clinical relevance of somatic variants identi- loid leukemia and monosomy 7. Haematologica, logical neoplasms. Leukemia, 33, 1851–1867. fied in molecular profiling of cancer. Genetics in 97, 1933–1936. Ricci, C., Fermo, E., Corti, S., Molteni, M., Faric- Medicine: Official Journal of the American College Walter, M.J., Ding, L., Shen, D., Shao, J., Grillot, ciotti, A., Cortelezzi, A., Lambertenghi Deliliers, of Medical Genetics, 18, 128–136. M., McLellan, M., Fulton, R., Schmidt, H., Kal- G., Beran, M. & Onida, F. (2010) RAS muta- Takahashi, K., Pemmaraju, N., Strati, P., icki-Veizer, J., O’Laughlin, M., Kandoth, C., tions contribute to evolution of chronic Nogueras-Gonzalez, G., Ning, J., Bueso-Ramos, Baty, J., Westervelt, P., DiPersio, J.F., Mardis, myelomonocytic leukemia to the proliferative C., Luthra, R., Pierce, S., Cortes, J., Kantarjian, E.R., Wilson, R.K., Ley, T.J. & Graubert, T.A. variant. Clinical Cancer Research: An Official H. & Garcia-Manero, G. (2013) Clinical charac- (2011) Recurrent DNMT3A mutations in Journal of the American Association for Cancer teristics and outcomes of therapy-related patients with myelodysplastic syndromes. Leuke- Research, 16, 2246–2256. chronic myelomonocytic leukemia. Blood, 122, mia, 25, 1153–1158. Richards, S., Aziz, N., Bale, S., Bick, D., Das, S., 2807–2811; quiz 2920. Xie, M., Lu, C., Wang, J., McLellan, M.D., John- Gastier-Foster, J., Grody, W.W., Hegde, M., Tawana, K., Drazer, M.W. & Churpek, J.E. (2018) son, K.J., Wendl, M.C., McMichael, J.F., Sch- Lyon, E., Spector, E., Voelkerding, K., Rehm, Universal genetic testing for inherited suscepti- midt, H.K., Yellapantula, V., Miller, C.A., H.L.; ACMG Laboratory Quality Assurance bility in children and adults with myelodysplas- Ozenberger, B.A., Welch, J.S., Link, D.C., Wal- Committee. (2015) Standards and guidelines for tic syndrome and acute myeloid leukemia: are ter, M.J., Mardis, E.R., Dipersio, J.F., Chen, F., the interpretation of sequence variants: a joint we there yet? Leukemia, 32, 1482–1492. Wilson, R.K., Ley, T.J. & Ding, L. (2014) Age- consensus recommendation of the American Thol, F., Kade, S., Schlarmann, C., L€ offeld, P., related mutations associated with clonal College of Medical Genetics and Genomics and Morgan, M., Krauter, J., Wlodarski, M.W., hematopoietic expansion and malignancies. Nat- the Association for Molecular Pathology. Genet- K€olking, B., Wichmann, M., G€ orlich, K., ure Medicine, 20, 1472–1478. ics in Medicine: Official Journal of the American G€ohring, G., Bug, G., Ottmann, O., Niemeyer, Yoshida, K., Sanada, M., Shiraishi, Y., Nowak, D., College of Medical Genetics, 17, 405–424. C.M., Hofmann, W.-K., Schlegelberger, B., Gan- Nagata, Y., Yamamoto, R., Sato, Y., Sato-Otsubo, Richardson, H. (2002) Medical laboratories–require- ser, A. & Heuser, M. (2012) Frequency and A., Kon, A., Nagasaki, M., Chalkidis, G., Suzuki, ments for quality and competence: an ISO per- prognostic impact of mutations in SRSF2, Y., Shiosaka, M., Kawahata, R., Yamaguchi, T., spective. Vox Sanguinis, 83(Suppl. 1), 333–335. U2AF1, and ZRSR2 in patients with myelodys- Otsu, M., Obara, N., Sakata-Yanagimoto, M., Robinson, J.T., Thorvaldsd ottir, H., Winckler, W., plastic syndromes. Blood, 119, 3578–3584. Ishiyama, K., Mori, H., Nolte, F., Hofmann, Guttman, M., Lander, E.S., Getz, G. & Mesirov, Thota, S., Viny, A.D., Makishima, H., Spitzer, B., W.K., Miyawaki, S., Sugano, S., Haferlach, C., J.P. (2011) Integrative genomics viewer. Nature Radivoyevitch, T., Przychodzen, B., Sekeres, Koeffler, H.P., Shih, L.Y., Haferlach, T., Chiba, Biotechnology, 29, 24–6. M.A., Levine, R.L. & Maciejewski, J.P. (2014) S., Nakauchi, H., Miyano, S. & Ogawa, S. (2011) Schwaab, J., Ernst, T., Erben, P., Rinke, J., Sch- Genetic alterations of the cohesin complex genes Frequent pathway mutations of splicing machin- nittger, S., Str€ obel, P., Metzgeroth, G., Mossner, in myeloid malignancies. Blood, 124, 1790–1798. ery in myelodysplasia. Nature, 478, 64–69. 622 ª 2019 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd British Journal of Haematology, 2020, 188, 605–622