Special Report Guiding the global evolution of cytogenetic testing for hematologic malignancies Yassmine M. N. Akkari,1 Linda B. Baughn,2 Adrian M. Dubuc,3 Adam C. Smith,4 Mar Mallo,5 Paola Dal Cin,3 Maria Diez Campelo,6 Marta S. Gallego,7 Isabel Granada Font,8 Detlef T. Haase,9 Brigitte Schlegelberger,10 Irma Slavutsky,11 Cristina Mecucci,12 Ross L. Levine,13 Robert P. Hasserjian,14 Francesc Sol e,5 Brynn Levy,15 and Xinjie Xu2 1 Departments of Cytogenetics and Molecular Pathology, Legacy Health, Portland, OR; 2Division of Hematopathology, Department of Laboratory Medicine Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 and Pathology, Mayo Clinic, Rochester, MN; 3Department of Pathology, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA; 4Laboratory Medicine Program, University Health Network and Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada; 5MDS Group, Microarrays Unit, Josep Carreras Leukaemia Research Institute, Barcelona, Spain; 6Hematology Department University Hospital of Salamanca, IBSAL, Salamanca, Spain; 7Laboratory of Cytogenetics and Molecular Cytogenetics, Department of Clinical Pathology, Italian Hospital, Buenos Aires, Argentina; 8 Hematology Laboratory, Germans Trias i Pujol University Hospital–Catalan Institute of Oncology, Josep Carreras Leukemia Research Institute, Barcelona, Spain; 9Clinics of Hematology and Medical Oncology, University Medical Center G€ottingen, G€ ottingen, Germany; 10Department of Human Genetics, Hannover Medical School, Hannover, Germany; 11Laboratory Genetics of Lymphoid Malignancies, Institute of Experimental Medicine, Buenos Aires, Argentina; 12Labora- tory of Cytogenetics and Molecular Medicine, Hematology University of Perugia, Perugia, Italy; 13Department of Medicine, Human Oncology and Pathogenesis Program, Memorial Sloan-Kettering Cancer Center, New York, NY; 14Department of Pathology, Massachusetts General Hospital, Boston, MA; and 15College of Physicians and Surgeons, Columbia University Medical Center and the New York Presbyterian Hospital, New York, NY Cytogenetics has long represented a critical component consortium of laboratory geneticists, pathologists, and in the clinical evaluation of hematologic malignancies. oncologists, describe herein the advantages and limitations Chromosome banding studies provide a simultaneous of both conventional chromosome banding and novel snapshot of genome-wide copy number and structural sequencing technologies and share our considerations on variation, which have been shown to drive tumorigenesis, crucial next steps to implement these novel technologies in define diseases, and guide treatment. Technological the global clinical setting for a more accurate cytogenetic innovations in sequencing have ushered in our present-day evaluation, which may provide improved diagnosis and clinical genomics era. With recent publications highlighting treatment management. Considering the clinical, logistic, novel sequencing technologies as alternatives to technical, and financial implications, we provide points to conventional cytogenetic approaches, we, an international consider for the global evolution of cytogenetic testing. Introduction may not be obtained in culture to achieve a complete chromo- some banding study.22 Moreover, even when cells readily divide Genomic characterization has become an essential component and render ample metaphases, chromosome banding has lim- in the clinical management of hematologic malignancies, sup- ited resolution with challenges in detecting alterations ,10 Mb. porting diagnosis and prognostication as well as informing ther- Supplementary technologies, namely fluorescence in situ hybrid- apeutic decisions.1-7 Cytogenetics, the study of chromosome ization (FISH) and chromosomal microarray (CMA), were subse- number and structure, represents a central facet of diagnostic quently developed and overcame many of the inherent genomic analysis with a longstanding history. For the past 50 limitations of chromosome banding analysis.23 These platforms years, chromosome banding has permitted single-cell and do not necessarily require dividing cells and can detect abnor- genome-wide analysis of chromosomal alterations, generating a malities ,10 Mb, effectively expanding the resolution from large foundational knowledge base of chromosome abnormalities and chromosome bands down to gene-level imbalances. FISH is a their clinical associations.7-13 Presently, for most hematologic targeted assay that can be used to detect CNVs and SVs in tar- malignancies, chromosome banding is an essential component geted regions. While widely adopted, there exists great variabil- of the diagnostic workup that guides clinical care.1,2,14-16 Chro- ity in the clinical use of FISH testing, which is often predicated mosome banding has long remained the gold standard for cyto- on a priori knowledge of a specific gene or region of interest.24 genetic studies globally, primarily for its high success rate in Moreover, clinical laboratories are also largely reliant on the detecting critical copy number variation (CNV) and structural var- development of commercially available FISH probes.25 CMA is a iation (SV) known to initiate and drive disease progression.7,17-21 powerful whole-genome technology with the ability to detect CNV and copy-neutral loss-of-heterozygosity (CN-LOH) in the However, an inherent limitation of chromosome banding is the kilobase range using an unbiased approach.25-32 However, as dependency on cells dividing in vitro. In some hematologic CMA cannot detect balanced chromosomal rearrangements, malignancies, a sufficient number of analyzable metaphase cells which are hallmarks of many hematologic malignancies, its utility blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 2273 remains limited to diseases primarily driven by CNVs or unbal- Ongoing cancer research studies utilizing novel technologies anced SVs. CMA adoption worldwide has been hampered by continue to expose inherent limitations of current cytogenetic financial constraints and reimbursement practices. While several diagnostic testing. In hematologic malignancies, SVs involving evidence-based reviews have highlighted the clinical utility of enhancer hijackers can reposition promoter and/or enhancer ele- CMA for a diverse spectrum of hematologic malignancies, to ments of the genome (such as immunoglobulin enhancer date, CMA has not been universally incorporated into current sequences) and drive the overexpression of a nearby oncogene clinical guidelines.32-38 Given the rapid advancements of geno- without producing in-frame chimeric gene fusions.53-55 These mic technologies that have the potential to comprehensively types of abnormalities are relatively common in lymphoid malig- detect clinically significant cytogenetic alterations, the interna- nancies and have been shown to lead to false-negative56-58 or, tional diagnostic community must carefully evaluate the imple- in rare instances, false-positive FISH results.59 For example, the mentation of these next-generation cytogenetic methodologies MYC break-apart FISH probe demonstrates an approximately and their implications for clinical care by considering the follow- 4% false-negative rate in diffuse large B-cell lymphoma56 and a ing questions. 50% to 70% false-negative rate in PCNs.58,60 Novel molecular approaches have also identified new aberrations that play a role in cancer initiation and progression, some of which can be tar- Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 Are our current traditional cytogenetic geted using novel therapeutics.61 Understanding the limitations of current “gold standard” diagnostic tools and identifying tech- diagnostic assays sufficient, or could nologies that may be more effective in addressing these vulner- newer technologies improve clinical abilities will be crucial to the advancement of clinical care. management? Innovative molecular approaches designed to overcome present- For myeloid malignancies, chromosome banding and FISH can day limitations include mate pair sequencing (MPseq) and optical be used to detect the majority of the CNVs and SVs needed for genome mapping (OGM). MPseq is a variation of whole-genome appropriate clinical management.14,39-41 The high success rate sequencing (WGS) utilizing a specialized library preparation of at identifying acute myeloid leukemia (AML)-specific abnormali- long input DNA followed by circularization, fragmentation, and ties reflects the ability of myeloid cells to divide readily in cul- sequencing of smaller paired-end fragments.58,62 OGM electro- ture.42,43 Of 250 adult AML cases with concurrent chromosome phoreses high molecular weight DNA into nanochannels, linear- banding and FISH testing, a successful karyotype result (defined izes them for imaging, and creates a consensus genome map as having $20 sufficiently analyzable metaphases) was attained from processed images.63-65 Both MPseq and OGM can identify in 88% of cases, 10% of cases had between 1 and 19 meta- CNVs and SVs, including balanced rearrangements that escape phases, and only 2% of cases had no metaphases for analysis.44 detection by CMA. Recent studies have highlighted the advan- Supplementation by AML-directed FISH yielded a 98% concor- tages of OGM as an unbiased whole-genome technology in dance between chromosome and FISH analysis (in cases with an hematologic malignancies, demonstrating concordance with adequate chromosome study).44 Such evidence-based studies chromosome banding while revealing additional abnormalities provided the foundation for the clinical guidelines supported by missed by routine workups and refining complex karyotypes such the European LeukemiaNet (ELN) (www.leukemia-net.org/ as derivative and marker chromosomes.64,66-68 In addition, whole home), the National Comprehensive Cancer Network (NCCN) transcriptome sequencing (WTS) and targeted RNA-seq assays (www.nccn.org), and the Revised International Prognostic Scor- have been developed to enable the identification of expressed ing System (IPSS-R) (www.mds-foundation.org).5,20,21,44-47 gene fusions. These approaches have shown promising results in pediatric ALL,69 some lymphomas,70 multiple myeloma,71,72 The success of chromosome analysis is, however, not uniform T-cell lymphoblastic lymphoma,73 and AML.74 Generally, RNA- based assays detect transcribed genomic regions and highly across all hematologic malignancies.48 A successful chromosome expressed fusion products, which preclude their detection of SVs analysis would provide $20 metaphases of sufficient resolution involving enhancer hijackers that do not result in chimeric gene to identify chromosomal abnormalities representative of the fusions. In contrast, DNA-based WGS provides a comprehensive hematologic disease being evaluated. In contrast to myeloid and unbiased characterization of gene rearrangements. Impor- malignancies, the success rate of chromosome analysis in acute tantly, formalin-fixed, paraffin-embedded material, commonly lymphoblastic leukemia (ALL) is approximately 70% to 75%; fail- available for extramedullary hematologic malignancies such as ures are primarily attributed to substandard chromosome mor- lymphoma, generally only allows DNA-based assays such as phology rendering the cells unanalyzable, and/or poor in vitro WGS for fusion detection in these malignancies. Although tar- cell growth of the ALL clone with an inadequate number of ana- geted RNA-seq assays are commonly used in the context of lyzable metaphases.48 When chromosome banding studies are hematologic malignancies, WTS has not been widely adopted in successful, an ALL-specific clonal abnormality can be identified the clinical setting. in about 60% to 80% of ALL cases.22,48 Similar to ALL, the fre- quency of detecting an abnormal clone in chronic lymphocytic The technologies described above provide viable alternatives to leukemia (CLL) using a CpG oligodeoxynucleotide-stimulated traditional cytogenetic approaches (chromosome banding, FISH, culture is 50% to 70%, as reported in multilaboratory stud- and CMA) with improved resolution. However, all the aforemen- ies.49,50 In contrast, the frequency of identifying an abnormal tioned techniques examine the genome through a singular lens, plasma cell clone in the case of a plasma cell neoplasm (PCN) limited to cytogenetic aberrations. Today, diagnostic evaluation of such as multiple myeloma is only 10% to 20%51 since malignant hematologic disorders requires the evaluation of other types of plasma cells rarely undergo cell division, except in cases of genomic alterations, including single nucleotide variants (SNVs) advanced disease.52 (eg, TP53, RUNX1, IDH, NPM1, and FLT3-ITD).9,15,75-77 Detection 2274 blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 AKKARI et al Table 1. Advantages, limitations, and clinical utility of currently available genomic technologies in hematologic malignancies G-banding FISH CMA WGS Targeted sequencing panels RT-PCR MPSeq WTS OGM Analyte Chromosome DNA in DNA DNA DNA RNA RNA DNA RNA DNA in dividing cells interphase nuclei and metaphase Coverage Whole Targeted Whole Whole Targeted Targeted Targeted Whole Whole Whole Distinction of Yes Yes No No No No No No No No individual cell clones Analysis bias Yes Yes (if cultured) No No Yes Yes No No No No Turnaround 3-7 4 h to 2 d 3-7 7-14 7-14 7-14 4 h to 5 d 7-14 14-21 7-10 time (d) GUIDING THE EVOLUTION OF CYTOGENETIC TESTING Unmapped Yes No No No No No No No No Some Alu and region LINE detection elements Ability to Low Low High High High High High High High Low to medium multiplex Analytical 1*-3 out of 20 1-10 10-20 20-30 1-10 5-10 ~0.01 10 1-10 5-20 sensitivity metaphases (%) SVs Yes Yes No Yes (long-read No Gene fusion Limited Yes Yes Yes or short-read deep sequencing) CNVs Yes Yes Yes Yes Limited Limited Limited Yes Limited Yes SNVs No No No Yes Yes Yes No Limited Yes No Disease status Diagnosis, Diagnosis, Diagnosis, Diagnosis, Diagnosis, Diagnosis, Diagnosis, Diagnosis, Diagnosis, Diagnosis, disease disease relapse relapse disease disease disease relapse relapse relapse monitoring, monitoring, monitoring, monitoring, monitoring, relapse relapse relapse, MRD relapse, MRD MRD, relapse (if deep (if deep coverage) coverage) Well- High High High Low High High High Low Low Low established Cost 11 11 111 11111 111 111 11 1111 1111 111 blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 *Depending on the clinical situation, 1 metaphase with a recurring abnormality may be considered evidence for an abnormal clone. 2275 Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 of these alterations can be achieved using multiple molecular evolution of testing and clinical adoption of WGS vs those who methods, including reverse transcription polymerase chain reac- debated the real world and current feasibility of this transition. tion (RT-PCR), Sanger sequencing, or next-generation sequencing (NGS) approaches. Thus, for many hematologic malignancies, detection of clinically relevant genomic variants requires a battery Implementing new technologies in a of individual assays. Therefore, a single, comprehensive, and clinical diagnostic laboratory affordable assay that captures SNVs, CNVs, and SVs is highly desir- To justify the introduction of a new clinical genomic assay as a able. Please see Table 1 for the list of currently available genomic replacement for the current gold standard cytogenetic testing (ie, technologies, their advantages, and limitations. chromosome banding coupled with FISH), multiple factors should be carefully considered. These include the following intercon- Can a single assay provide comprehensive nected considerations: clinical, logistic, technical, and financial. genomic characterization? Clinical considerations The evolution of Sanger sequencing to NGS, coupled with the The utilization of new comprehensive molecular methodologies decreasing costs of DNA resequencing, ushered in an unprece- has the potential to improve clinical care by leading to a more Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 dented genomics era.78,79 In the current clinical diagnostics accurate diagnosis and enhancing treatment selection based on setting, NGS assays are frequently limited to targeted gene- higher resolution and more precise genomic data. In these specific panels that require regular updates as new clinically rel- instances, a genome-wide assessment may reveal uncommon evant genes are identified.80-82 However, there is considerable but known entities that are not routinely surveyed. Some exam- variability in content, analysis, and interpretation between differ- ple scenarios include, but are not limited to: ent genomics laboratories. 1. Hematologic malignancies with a broad spectrum of geno- Whole exome sequencing (WES) covers all exonic sequences and mic driver alterations of various classes that would require has the potential to eliminate much of the variance, at least relat- multiple assays to effectively evaluate, such as Ph-like ing to content. While WES costs have dramatically decreased B-ALL, PCN, CLL, and non-Hodgkin lymphomas. over the past 5 years, they nonetheless remain prohibitive in 2. Hematologic malignancies with well-documented enhancer many parts of the world. With each evolutionary advance, we see hijacking alterations associated with risks of false-negative a benefit but also recognize other deficiencies.83-86 Improved or false-positive findings by conventional diagnostic bioinformatics algorithms have enabled WES to detect most approaches (eg, IGH and/or MYC rearrangements). CNVs, albeit with variable performance, depending on the qual- 3. Cases in which chromosome studies are not successful or are ity and depth of the sequencing data and normalization parame- known to demonstrate an unacceptable failure rate (espe- ters. In addition, since WES excludes intronic and regulatory cially in follow-up studies of posttreatment specimens). genomic regions, it is unable to detect a large subset of SVs. This 4. Atypical clinical presentations warranting the identification shortfall is significant, as detection of some intronic SVs is essen- of genomic drivers for appropriate management (eg, mixed tial in the evaluation of certain hematologic malignancies. Thus, lineage leukemias). WES has the potential to rapidly adapt to new information stem- 5. Detection of cytogenetic abnormalities that require further ming from translational research, while it also suffers from the workup to completely characterize their origin and genomic inability to detect all pertinent genomic aberrations.84 makeup (eg, derivative and marker chromosomes). 6. Investigation of possible false-negative findings when a nor- WGS, particularly involving long-read or long-insert DNA frag- mal karyotype/FISH result is obtained, but clinical presenta- ments, is an attractive technology that can detect all 3 genomic tion and pathology support the evidence of a neoplasm. variants (SNVs, CNVs, and SVs) in an unbiased manner.87-89 Despite these advantages, WGS is currently not adopted for clinical use worldwide because of the complexity of WGS com- Technological advances have uncovered additional genomic pared with other technologies in relation to the processes, work- aberrations and complexities often unappreciated using conven- flows, and resources necessary for its implementation. At tional diagnostic approaches. However, interpretation of these present, WGS is primarily used in large academic centers, mostly novel findings with unclear clinical significance remains a chal- for translational research purposes. A recent publication by Dun- lenge and will likely require additional translational research ini- cavage et al proposed a paradigm shift in clinical laboratory tiatives to understand their clinical significance. In the interim, genomic testing by suggesting WGS as an alternative to cytoge- the question remains as to how to handle these new findings netic analysis in myeloid cancers.90 The authors performed WGS that are revealed when employing novel technologies. While in a cohort of 263 patients with myeloid malignancies and con- most laboratories are likely to only report abnormalities of firmed 40 recurrent translocations and 91 copy-number altera- known clinical significance, consideration should be given to tions previously identified by cytogenetic studies. In addition, possibly reporting novel abnormalities to potentially facilitate WGS identified additional genomic events (CNVs and/or SVs) in future translational research studies involving large patient 40 of 235 patients (17.0%). When used prospectively in 117 con- cohorts. Such studies would focus on correlating novel genomic secutive patients, WGS provided new genetic information in 29 abnormalities with patient outcomes and their response to ther- (24.8%) and changed the risk category in 19 (16.2%). This publi- apies. Indeed, this would require a clear description/disclaimer cation was met with both enthusiasm and skepticism,5,91-95 and of the unknown significance of the novel findings, which could subsequently sparked a discussion among clinical laboratorians potentially be reported under a special heading on the clinical globally, dividing those who were actively promoting an report. 2276 blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 AKKARI et al The value of this approach was recently demonstrated by Tyner In some circumstances, particularly when the tumor represents et al.96 In this study, the authors reported the integration of nearly 100% of the tested tissue, it is unclear if the abnormality whole-exome and transcriptome sequencing in the analysis of identified by NGS is germline or somatic. In such instances, simi- 672 patients with AML while examining recurrent mutations (and lar to other clinical testing scenarios, a nontumor sample can be combinations thereof) to determine drug sensitivity. This study requested (blood or skin) to evaluate the germline status. This summarized functional, genomic, and transcriptomic data from can be a desirable follow-up approach that minimizes the need this large, clinically characterized AML cohort to date and paved to perform side-by-side germline and somatic testing. Currently, the way for establishing the clinical utility of novel technologies as there are no guidelines to address all the above possibilities in myeloid disorders. Similar reports have been published for in oncology testing, each laboratory should establish its own other hematologic malignancies such as pediatric ALL and PCN. policies to address incidental findings, including the option to Nordlund et al used linked-read WGS to generate haplotype only analyze targeted regions. information and successfully detected CNVs and SVs in ALL patients, including cryptic and key abnormalities such as DUX4, We suggest that for early adopters of new technologies without ZNF384, and PAX5 gene rearrangements.97 Moreover, Hollein a full understanding of many of the novel variants that will be et al recently reported the use of WGS in combination with encountered, a tiered reporting strategy is necessary to clearly Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 RNA-seq in 211 diagnostic PCN samples and observed 92% delineate the well-established genomic alterations from other concordance of FISH and WGS results for CNVs and SVs evalua- variants of unknown significance. Such a tiered approach has tion.72 However, some abnormalities, such as MYC transloca- been recommended by numerous professional organizations in tions in PCN, are more reliably identified using WGS rather than the analysis of CNVs, SNVs, and CN-LOH in cancer.26,82,104-106 FISH approaches.60 Such translational studies afford the oppor- Moreover, as translational and clinical research studies continue tunity to correlate conventional diagnostic approaches with to deliver evidence-based data supporting the significance of novel genomic studies. In this instance, a comparative nomo- novel genomic findings, reanalysis of existing WGS data can be gram study could correlate the clinical significance of low-level achieved to update the patient’s diagnosis/prognosis/manage- abnormal clones detected by WGS with the number of abnor- ment if novel, clinically significant variants are identified, poten- mal metaphase cells detected by chromosome banding. This tially minimizing additional expenses associated with repeat would be critical to our future understanding of small clonal genomic testing. populations identified by WGS. As the clinical significance of most submicroscopic CNVs below the resolution of current cyto- Logistic considerations genetic profiling is not known, sufficiently powered studies are The impact of adopting new technologies such as WGS must be needed to delineate their clinical, prognostic, and therapeutic considered at a system level. On a global scale, the replacement relevance. Publicly available databases are also critically needed of conventional testing may be impractical at this time. Multiple to facilitate data sharing (both genomic and clinical information) factors must be considered and carefully navigated, including to enhance our understanding of these genomic findings within the availability of sequencing equipment, trained personnel, the clinical setting. This is by no means an easily accomplished infrastructure to support reagent transportation and distribution, task, as it will require the interpretation and integration of large impacts of turnaround time, complexity of analysis, development amounts of data for each patient while holding the promise of of an integrated workflow between the molecular and cytoge- improved clinical management. To this end, population refer- netics laboratories, computing and data storage infrastructure, ence databases such as the Genome Aggregation Database98 and Database of Genomic Variants,99 among others,100,101 have provided reference maps allowing improved interpretation of SNVs and CNVs, including some SVs. Recently, a population ref- 600 erence map of SVs has been made available,102 providing a ref- erence of SVs from nearly 15 000 genomes from a diverse Sequencing coverage depth population. This tool will be instrumental in the interpretation of incidental and benign SVs. 400 One commonly encountered clinical challenge in oncology test- ing is the identification of incidental findings of germline variants of clinical significance. To this end, the American College of Medical Genetics and Genomics has published guidance for the 200 reporting of secondary findings in the context of clinical exome and genome sequencing.103 The secondary findings mainte- nance working group has recommended a list of genes that should be evaluated in individuals undergoing clinical-grade sequencing. Based on these recommendations, these secondary 0 findings should also be reported when encountered in the anal- 0 5 10 15 20 ysis of SVs and CNVs using NGS. To simplify workflows, how- Limit of detection (LOD, %) ever, filters can be applied to allow a focused analysis on clinically significant regions related to the reason for testing. Figure 1. Sequencing coverage depth needed for LOD in the range of 1% Informed consent and genetic counseling should also be to 25% calculated based on confidence interval at 0.95 with threshold of recommended.103 minimum 3 unique sequencing reads according to the binomial distribution. GUIDING THE EVOLUTION OF CYTOGENETIC TESTING blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 2277 and cost (including reimbursement).107 Given that evaluation of Current published guidelines for clinical germline WGS specifically novel technologies may require a significant upfront investment note that laboratories are not required to validate all classes of var- in the forms of capital equipment, bioinformatics means, data iation that include all potential iterations of genomic lesions but storage capabilities, and human resources, the transition to any can instead adopt a phased approach to validations. However, disruptive technology must be well-planned in order to avoid dis- laboratories must establish operational familiarity with all quality ruption to patient care and confusion to the clinical stakeholders. control parameters and identify any factors that would adversely The gradual expansion of testing should enable harmonization of affect the detection of certain variants (eg, PCR amplification for laboratory processes, maintaining quality control and reporting library preparation and CNV detection). In certain clinical situa- guidelines, and facilitate the adoption of those technologies with tions, orthogonal confirmation of clinically actionable pathogenic the greatest improvements over present standards. calls may be warranted, although this additional effort would fur- ther increase the cost and turnaround time of clinical testing.105 Technical considerations The early adoption of WGS as an alternative to conventional cyto- Another challenge in the universal use of WGS is the inconsistency genetic testing has brought about questions regarding the need in the variant components of the bioinformatics pipelines used by to replace current cytogenetic approaches. Indeed, this each laboratory. These include alignment and variant calling in Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 “replacement” is much more likely to be an “evolution” with the addition to the transfer of data to new human genomic builds. In addition, the depth of sequencing coverage would optimally be gradual adoption of innovative technologies following the estab- adapted according to the reason for testing (diagnosis, response lishment of carefully considered disease lists and/or clinical scenar- to treatment, or minimal residual disease [MRD] assessment). In ios where a given technology shows the greatest promise. general, cancer NGS tests with the goal to guide targeted therapy Presently, many clinical situations optimally require the combinato- have a limit of detection (LOD) of around 5%. However, in the con- rial use of multiple technologies to appropriately balance time, text of MRD detection, cancer NGS tests may need to reliably cost, and clinical needs. Targeted assays may be supplemented if detect alterations down to #1%.111,112 The sensitivity of WGS is new technologies cannot return critical results in a timely fashion, typically lower than other testing methodologies such as targeted potentially delaying appropriate clinical management. In the con- RT-PCR. Thus, for MRD purposes, private and novel variants iden- text of AML, WGS will likely take a longer time to complete than tified by WGS at diagnosis could be used to customize an RT-PCR traditional approaches. Therefore, coupling WGS with more rapid, assay that would allow higher sensitivity. While Duncavage and targeted assays, such as FISH for PML::RARA or BCR::ABL1 fusion, colleagues suggested that WGS could be a viable alternative to may permit results to be returned in a timely fashion to determine cytogenetic analysis in myeloid cancers with a target coverage of initial therapy at diagnosis/presentation. Subsequently, WGS 60x for their entire cohort, this level of coverage is predicted to results can be delivered before the end of the initial induction ther- provide a LOD of 10% using 3 reads as the minimum number to apy for a more comprehensive genomic profiling that may provide call SNVs with 95% confidence interval based on binomial proba- novel but clinically important information for subsequent therapeu- bility distribution. This would be insufficient for either guiding tar- tic/risk stratification decisions. This approach requires built-in geted therapy or tracking MRD.112-114 In addition, a small subset redundancies in testing parameters until turnaround times can (4.3%) of their cohort had less than 25x genome coverage, corre- match present standards. sponding to a LOD of 23%. To reach a LOD of 5% and 1% for appropriate NGS tests to guide targeted therapy or to detect Despite the comprehensive nature of genome-wide assays such MRD, the sequence coverage needs to be at 124x and 625x, as WGS, there remain regions in the genome or critical aberra- respectively, which would increase sequencing cost significantly tions that are not amenable to WGS. For example, a fraction of (Figure 1). The LOD to detect CNV, on the other hand, varies the genome, such as subtelomeric and centromeric regions, is based on the size and genomic location of these alterations. For refractory to sequencing.108 Although some of these limitations CNVs that include junction sequences, split reads can support could be overcome with the use of ultralong read sequenc- breakpoint detection and improve the sensitivity compared with ing,109 this approach is not widely accessible. In addition, inter- CNV detected using read-depth alone.115 In the Duncavage et al nal tandem repeats (eg, FLT3-ITD) may escape detection by study, the following criteria were used to call CNVs .5Mb: (1) amplicon-based WGS and will likely require orthogonal Having at least 2 paired and 2 split reads supporting the break- approaches to uncover these types of genomic lesions.110 points, and (2) the ratio between coverage depth and normalized Table 2. Estimated cost for the adoption and validation of WGS in a clinical laboratory setting Item Estimated cost range ($) Comments Capital equipment 800 000-1 000 000 Ex: Illumina NovaSeq 6000 Bioinformatics pipeline 10 000-30 000 Ex: Illumina Dragon license Test validation 39 000-100 000 30-50 samples Staff salary 50 000-200 000 (varies per country) Technologist and research and development scientist (3-6 mo), bioinformatician (3-6 mo), laboratory geneticist and/or pathologist (1-3 mo) Total 899 000-1 330 000 2278 blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 AKKARI et al control being ,0.8 for deletion or .1.3 for duplication. This logistical barriers remain prohibitive for smaller laboratories, entails that the tumor burden is $20% to call a deletion .5Mb or especially those in low-income countries, to adopt WGS. $30% to call a duplication .5Mb. In the United States, reimbursement by private and public payers In general, LOD increases for smaller size CNVs, thus making it only occurs if a test has an associated current procedural termi- challenging to identify gene-level or exon-level CNVs ,5Mb with nology code (CPT; a medical code used to report medical, surgi- 20% to 30% tumor burden using 60x WGS. The LOD for cal, and diagnostic procedures and services to entities such as CN-LOH detection is still not well understood, and further studies physicians, health insurance companies, and accreditation organi- are needed. It is therefore critical that such issues be thoroughly zations) (www.cms.gov).119,120 The CPT codes for a given test are investigated so that laboratories can identify both advantages and linked to a specific technology (eg, chromosome banding/FISH/ limitations and ensure that alternative tests are available in the lab- RT-PCR/CMA/WES/WGS); however, a technology-associated oratory’s testing menu if, and when, a diagnostic result cannot be CPT code does not necessarily guarantee reimbursement. The achieved (eg, due to sequencing failures, gaps in coverage, com- newer the code, the more likely insurance companies are to deny plex repetitive sequences, appropriately tracking MRD, etc). payment, substantiating their actions by claiming the technology Therefore, if WGS is not successful, the laboratory can use conven- to be “experimental.” US Food and Drug Administration (FDA) Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 tional chromosome studies from samples that have been already approval of a specific technology-associated test essentially con- grown and stored as a fixed cell pellet. This illustrates the impor- verts an experimental test to one accepted for use in clinical diagnosis, thus facilitating reimbursement. However, considerable tance of maintaining some degree of redundancy, with compe- resources are needed to navigate the FDA approval process, tency in a more diversified testing portfolio to address the various making this pathway more likely to be initiated by commercial testing deficiencies defined above. Put simply, FISH and chromo- laboratories or large medical institutions. Reimbursement can be some banding may remain necessary in certain instances, particu- linked to FDA approval, particularly when a therapeutic drug is larly in instances of compromised WGS studies. All these associated with a specific diagnosis supported using an FDA- regulatory processes are time-consuming and will slow down the approved test (www.fda.gov).121 In such instances, access to speed of establishment of WGS for routine diagnostics. therapy may be tied to the specific testing dictated by pharma- ceutical companies. Therefore, even if a laboratory offers WGS, a Financial considerations specific targeted assay such as FISH may instead be required to The financial considerations for performing WGS as a clinical assure reimbursement for the disease-associated therapy. In test include both the cost to adopt and validate WGS and the countries with public health systems (eg, Canada, Germany, Italy, cost to routinely perform it in a clinical laboratory setting. The Spain, and the United Kingdom), government bodies work with major cost to adopt WGS is capital equipment, personnel cost scientific associations or physician organizations to determine for staff technologist/scientist with experience in NGS and bioin- which tests will be financially reimbursed and often have fixed formaticians to build and maintain the pipeline, reagents for test annual budgets that make adding expensive testing difficult.116 validation, data storage, database generation, and qualified lab- In many instances, centralized clinical laboratories then provide oratory geneticists or molecular pathologists for data interpreta- these approved tests. The same holds true for many Latin Ameri- tion and reporting. As shown in Table 2, the total cost is can countries, such as Argentina, that have a mix of public and estimated in the $0.9 to $1.5 million range in each laboratory. private health systems. Lagging reimbursement by public and pri- vate payers remains the primary obstacle to the widespread The total cost to run WGS as a clinical test with 60x coverage is adoption of novel technologies that offer more sensitive and estimated to be in the $1300 to $2500 range,90,116,117 with the cost comprehensive testing modalities. Unfortunately, this factor often breakdown to routinely perform WGS in a clinical laboratory setting discourages the adoption of new tests despite a body of litera- discussed in detail in Duncavage et al ($2 million for a laboratory ture that clearly demonstrates their clinical utility and validity. running 1000 WGS tests per year). This is significantly higher com- pared with chromosome banding, which remains in the range of Financial considerations remain key factors for clinical laboratories $150 to $750 (varies by country). Moreover, this cost is based on to consider adopting clinical WGS. The cost to adopt and validate the estimate of batching 10 to 20 samples per sequencing flow WGS, followed by routinely performing WGS while maintaining a cell, which may not be feasible for most laboratories, as batching reasonable turn-around-time, may not be currently applicable for will invariably interfere with maintaining an appropriate turn- most laboratories globally. In addition, without clear pathways for around-time for hematologic malignancy testing. In addition, reimbursement, creating a business plan to justify adopting the depending on the robustness of curated knowledge bases, the cost new technologies may be impractical at this time.116 for employing scientists and laboratory geneticists and/or patholo- gists to interpret and report cases will vary and remain a significant financial consideration. Optimistically, for future adoption of WGS, Toward real-time changes in clinical laboratories must batch patient samples with different clinical indi- guidelines cations, including hematologic malignancies, solid tumors, and con- Chromosome banding is an integral part of current clinical guide- genital conditions, which could help maintain a clinically lines and prognostic systems (NCCN, ELN, and IPSS-R). The value appropriate turn-around-time and decrease cost.118 of novel technologies has been highlighted in this review, and the frequent lag in their adoption constitutes an obstacle to improved Based on these cost considerations, currently, only large institu- diagnostics and patient outcomes. This is especially significant as tions in developed countries with substantial financial support access to the newest therapies through enrollment in clinical trials can afford the implementation of WGS. The financial and other relies on current guidelines. It is therefore important for our GUIDING THE EVOLUTION OF CYTOGENETIC TESTING blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 2279 laboratory and clinical communities to formulate a more dynamic single test may not replace the totality of tests that currently response to the evidence emerging from novel technologies and exist, the active exploration and adoption of novel approaches more rapidly and continuously update the guidelines. We argue in defined clinical settings may indeed allow for a continued and that clinical guidelines should adopt technology-agnostic standards controlled evolution resulting in the successful clinical implemen- so long as a laboratory appropriately validates a given technology. tation of WGS in the near future. Constraining diagnostic genomic laboratories to a particular testing methodology is counterproductive and decreases patient access to local laboratories that could potentially offer the same test by Acknowledgments another validated method in a more affordable manner. The authors thank Jill Kappers from the Mayo Clinic for providing administrative assistance in the preparation of this document. Alternative approaches to improving Supported by National Cancer Institute grant P01 CA108671 11 (R.L.L.). patient outcomes As improvement in patient outcomes remains the ultimate objec- tive of clinical genomic testing, it is important to consider alterna- Authorship Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 tive and innovative approaches that may reduce the necessity for Contribution: X.X., Y.M.N.A., L.B.B., A.M.D., and B.L. researched data for the article, provided substantial contribution to discussion and con- comprehensive genome-wide testing strategies. Clinical manage- tent, writing, and review/editing of the manuscript before submission; ment is often guided by the optimal therapy choice based on bio- M.M., C.M., R.L.L., R.P.H., and F.S. researched data for the article, pro- marker status: as more targeted therapies for specific biomarkers vided substantial contribution to discussion and content and review/edit- are developed, clinical management via personalized medicine ing of the manuscript before submission; and A.C.S., P.D.C., M.D.C., becomes paramount and may eliminate reliance on WGS alto- M.S.G., I.G.F., D.T.H., B.S., and I.S. reviewed and edited the manuscript before submission. gether. While we remain in the early days of single-cell sequenc- ing, a deeper understanding of clonal heterogeneity and a Conflict-of-interest disclosure: R.L.L. is on the supervisory board of Qia- change in therapeutic strategies based on these minor but poten- gen and is a scientific advisor to Imago, Mission Bio, Zentalis, Ajax, tially aggressive clones may become more widely used. In addi- Auron, Prelude, C4 Therapeutics, and Isoplexis; receives research sup- tion, the significance of novel therapies such as chimeric antigen port from and consulted for Celgene and Roche and has consulted for receptor T-cell or tumor mRNA vaccines may be agnostic to tumor Incyte, Janssen, Astellas, Morphosys, and Novartis; has received hono- biomarkers and become a “magic bullet” for some types of can- raria from Astra Zeneca, Roche, Lilly, and Amgen for invited lectures and cer therapy.122 Finally, it is unclear whether a combined omics Gilead for grant reviews. A.C.S. reports personal financial interests in Bionano Genomics. The remaining authors declare no competing finan- approach (genomic, transcriptomic, proteomic, methylomic) may cial interests. be a better and more comprehensive approach not only to under- stand but also to treat human disease. ORCID profiles: Y.M.N.A., 0000-0002-3072-7388; L.B.B., 0000-0001- 5229-4897; A.C.S., 0000-0001-9927-4914; M.M., 0000-0001-7741- 498X; P.D.C., 0000-0002-7426-5293; M.D.C., 0000-0002-1467-6779; Concluding remarks I.G.F., 0000-0002-4275-0104; B.S., 0000-0001-5256-1270; I.S., 0000- From a global health perspective, with the goal to provide equi- 0002-0611-2367; C.M., 0000-0002-1623-0148; R.P.H., 0000-0002- 7753-0697; F.S., 0000-0002-3251-2161; B.L., 0000-0002-4213-3480; table care for all patients worldwide, access to novel and sophis- X.X., 0000-0001-8373-1834. ticated genomic testing tools may not be immediately feasible across all clinical laboratories. Presently, there are significant impediments to the immediate adoption of WGS by clinical lab- Correspondence: Xinjie Xu, Division of Hematopathology, Division of oratories worldwide. While the clinical utility of WGS is undeni- Laboratory Genetics and Genomics, Department of Laboratory Medi- cine and Pathology, Mayo Clinic, 200 First Street SW, Rochester, MN able, chromosome banding and FISH with SNP array integration 55905; e-mail: xu.xinjie@mayo.edu; Brynn Levy, College of Physicians will likely continue to serve as the gold standard of cytogenetic and Surgeons, Columbia University Medical Center and the New York testing for hematologic malignancies. We remain optimistic that Presbyterian Hospital, 3959 Broadway CHC – Room 406b, New York, by establishing data-generating initiatives through translational NY 10032; e-mail: bl2185@cumc.columbia.edu; and Francesc Sol e, research, expediting the adoption of published guidelines (such MDS Group, Institut de Recerca Contra la Leuc emia Josep Carreras, as NCCN and ELN), validating novel technologies for clinical Edifici IJC, Campus ICO-Germans Trias i Pujol, Ctra de Can Ruti, Camı de les Escoles s/n, 08916 Badalona, Barcelona, Spain; e-mail: fsole@ utility, and ensuring appropriate turnaround times and reim- carrerasresearch.org. bursement policies, we can embrace the change and drive the evolution of genomic technologies. Looking ahead, and as our understanding of the genome and the molecular pathogenesis Footnote of disease continues to advance, we will undoubtedly find our- Submitted 3 October 2021; accepted 3 February 2022; prepublished selves looking into (epi)genomic, transcriptomic, and proteomic online on Blood First Edition 15 February 2022. DOI 10.1182/ tools for the full characterization of human diseases. While one blood.2021014309. REFERENCES of genetics. Blood. 2017;130(4): through cancer. N Engl J Med. 2019; 410-423. 381(22):2145-2156. 1. Taylor J, Xiao W, Abdel-Wahab O. Diagnosis and classification of 2. Nangalia J, Campbell PJ. Genome 3. Granada I, Palomo L, Ruiz-Xivill e N, Mallo M, hematologic malignancies on the basis sequencing during a patient’s journey Sol e F. Cytogenetics in the genomic era. Best 2280 blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 AKKARI et al Pract Res Clin Haematol. 2020;33(3): 19. Mitelman F, Heim S. Quantitative acute 30. O’Keefe C, McDevitt MA, Maciejewski JP. 101196. leukemia cytogenetics. Genes Copy neutral loss of heterozygosity: a Chromosomes Cancer. 1992;5(1):57-66. novel chromosomal lesion in myeloid 4. Ribeiro IP, Melo JB, Carreira IM. malignancies. Blood. 2010;115(14): Cytogenetics and cytogenomics evaluation 20. Haase D, Germing U, Schanz J, et al. New 2731-2739. in cancer. Int J Mol Sci. 2019;20(19):4711. insights into the prognostic impact of the karyotype in MDS and correlation with 31. Mohamedali A, G€ aken J, Twine NA, et al. 5. Greenberg PL, Tuechler H, Schanz J, et al. subtypes: evidence from a core dataset of Prevalence and prognostic significance of Revised international prognostic scoring 2124 patients. Blood. 2007;110(13): allelic imbalance by single-nucleotide poly- system for myelodysplastic syndromes. 4385-4395. morphism analysis in low-risk myelodysplas- Blood. 2012;120(12):2454-2465. tic syndromes. Blood. 2007;110(9): 21. Schanz J, T€ uchler H, Sol e F, et al. New 3365-3373. 6. Bernard E, Nannya Y, Hasserjian RP, et al. comprehensive cytogenetic scoring system Implications of TP53 allelic state for for primary myelodysplastic syndromes 32. Tiu RV, Gondek LP, O’Keefe CL, et al. genome stability, clinical presentation and (MDS) and oligoblastic acute myeloid Prognostic impact of SNP array karyotyping outcomes in myelodysplastic syndromes leukemia after MDS derived from an in myelodysplastic syndromes and related [published correction appears in Nat Med. international database merge. J Clin Oncol. myeloid malignancies. Blood. 2011;117(17): 2021;27(5):927]. Nat Med. 2020;26(10): 2012;30(8):820-829. 4552-4560. 1549-1556. 22. Moorman AV, Harrison CJ, Buck GA, et al; 33. Akkari YMN, Bruyere H, Hagelstrom RT, Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 7. Mitelman F. Recurrent chromosome Adult Leukaemia Working Party, Medical et al. Evidence-based review of genomic aberrations in cancer. Mutat Res. 2000; Research Council/National Cancer Research aberrations in B-lymphoblastic leukemia/ 462(2-3):247-253. Institute. Karyotype is an independent lymphoma: Report from the cancer geno- prognostic factor in adult acute mics consortium working group for lympho- 8. Mr ozek K, Bloomfield CD. Chromosome lymphoblastic leukemia (ALL): analysis of blastic leukemia. Cancer Genet. 2020;243: aberrations, gene mutations and cytogenetic data from patients treated on 52-72. expression changes, and prognosis in adult the Medical Research Council (MRC) acute myeloid leukemia. Hematology Am UKALLXII/Eastern Cooperative Oncology 34. Chun K, Wenger GD, Chaubey A, et al. Soc Hematol Educ Program. 2006;2006: Group (ECOG) 2993 trial. Blood. 2007; Assessing copy number aberrations and 169-177. 109(8):3189-3197. copy-neutral loss-of-heterozygosity across the genome as best practice: an evidence- 9. Ogawa S. Genetics of MDS. Blood. 2019; 23. Ronaghy A, Yang RK, Khoury JD, Kanagal- based review from the Cancer Genomics 133(10):1049-1059. Shamanna R. Clinical applications of Consortium (CGC) working group for chromosomal microarray testing in myeloid chronic lymphocytic leukemia. Cancer 10. Shallis RM, Ahmad R, Zeidan AM. The malignancies. Curr Hematol Malig Rep. Genet. 2018;228-229:236-250. genetic and molecular pathogenesis of 2020;15(3):194-202. myelodysplastic syndromes. Eur 35. Pugh TJ, Fink JM, Lu X, et al; Cancer J Haematol. 2018;101(3):260-271. 24. Wheeler FC, Kim AS, Mosse CA, Shaver Genomics Consortium Plasma Cell AC, Yenamandra A, Seegmiller AC. Limited Disorders Working Group. Assessing 11. Solomon E, Borrow J, Goddard AD. utility of fluorescence in situ hybridization genome-wide copy number aberrations Chromosome aberrations and cancer. for recurrent abnormalities in acute myeloid and copy-neutral loss-of-heterozygosity as Science. 1991;254(5035):1153-1160. leukemia at diagnosis and follow-up. Am J best practice: an evidence-based review 12. Yang JJ, Park TS, Wan TS. Recurrent Clin Pathol. 2018;149(5):418-424. from the Cancer Genomics Consortium cytogenetic abnormalities in acute myeloid working group for plasma cell disorders. 25. Rack KA, van den Berg E, Haferlach C, Cancer Genet. 2018;228-229:184-196. leukemia. Methods Mol Biol. 2017;1541: et al. European recommendations and 223-245. quality assurance for cytogenomic analysis 36. Kanagal-Shamanna R, Hodge JC, Tucker T, 13. Mitelman F. Catalogue of chromosome of haematological neoplasms. Leukemia. et al. Assessing copy number aberrations aberrations in cancer. Cytogenet Cell 2019;33(8):1851-1867. and copy neutral loss of heterozygosity Genet. 1983;36(1-2):1-515. across the genome as best practice: an 26. Mikhail FM, Biegel JA, Cooley LD, et al. evidence based review of clinical utility 14. Arber DA, Orazi A, Hasserjian R, et al. The Technical laboratory standards for from the cancer genomics consortium 2016 revision to the World Health interpretation and reporting of acquired (CGC) working group for myelodysplastic Organization classification of myeloid copy-number abnormalities and copy- syndrome, myelodysplastic/ neoplasms and acute leukemia Blood. neutral loss of heterozygosity in neoplastic myeloproliferative and myeloproliferative 2016;127(20):2391-2405. disorders: a joint consensus recommenda- neoplasms. Cancer Genet. 2018;228-229: tion from the American College of Medical 197-217. 15. de Haas V, Ismaila N, Advani A, et al. Initial Genetics and Genomics (ACMG) and the diagnostic work-up of acute leukemia: Cancer Genomics Consortium (CGC). 37. Afable MG II, Wlodarski M, Makishima H, ASCO clinical practice guideline endorse- Genet Med. 2019;21(9):1903-1916. et al. SNP array-based karyotyping: differ- ment of the College of American Patholo- ences and similarities between aplastic gists and American Society of Hematology 27. Simons A, Sikkema-Raddatz B, de Leeuw anemia and hypocellular myelodysplastic Guideline [published correction appears in N, Konrad NC, Hastings RJ, Schoumans J. syndromes. Blood. 2011;117(25): J Clin Oncol. 2019;37(3):239-253]. J Clin Genome-wide arrays in routine diagnostics 6876-6884. Oncol. 2019;37(7):612. of hematological malignancies. Hum Mutat. 2012;33(6):941-948. 38. Arenillas L, Mallo M, Ramos F, et al. Single 16. D€ ohner H, Estey E, Grimwade D, et al. nucleotide polymorphism array Diagnosis and management of AML in 28. Peterson JF, Aggarwal N, Smith CA, et al. karyotyping: a diagnostic and prognostic adults: 2017 ELN recommendations from Integration of microarray analysis into the tool in myelodysplastic syndromes with an international expert panel. Blood. 2017; clinical diagnosis of hematological unsuccessful conventional cytogenetic 129(4):424-447. malignancies: how much can we improve testing. Genes Chromosomes Cancer. cytogenetic testing? Oncotarget. 2015; 2013;52(12):1167-1177. 17. Heim S, Mitelman F. Cytogenetic analysis 6(22):18845-18862. in the diagnosis of acute leukemia. Cancer. 39. Adem a V, Hernandez JM, Abaigar M, et al. 1992;70(6 suppl):1701-1709. 29. Maciejewski JP, Tiu RV, O’Keefe C. Application of FISH 7q in MDS patients Application of array-based whole genome without monosomy 7 or 7q deletion by 18. Heim S, Mitelman F. Molecular screening scanning technologies as a cytogenetic conventional G-banding cytogenetics: does for new fusion genes in cancer. Nat Genet. tool in haematological malignancies. Br J -7/7q- detection by FISH have prognostic 2008;40(6):685-686. Haematol. 2009;146(5):479-488. value? Leuk Res. 2013;37(4):416-421. GUIDING THE EVOLUTION OF CYTOGENETIC TESTING blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 2281 40. Mallo M, Arenillas L, Espinet B, et al. newly diagnosed multiple myeloma. Blood hematological malignancy genomes by Fluorescence in situ hybridization improves Adv. 2020;4(10):2236-2244. optical genome mapping. Am J Hum the detection of 5q31 deletion in Genet. 2021;108(8):1423-1435. myelodysplastic syndromes without 52. Dewald GW, Kyle RA, Hicks GA, Greipp cytogenetic evidence of 5q-. PR. The clinical significance of cytogenetic 65. Barseghyan H, Tang W, Wang RT, et al. Haematologica. 2008;93(7):1001-1008. studies in 100 patients with multiple Next-generation mapping: a novel myeloma, plasma cell leukemia, or approach for detection of pathogenic 41. Haferlach T, Schmidts I. The power and amyloidosis. Blood. 1985;66(2):380-390. structural variants with a potential utility in potential of integrated diagnostics in acute clinical diagnosis. Genome Med. 2017; myeloid leukaemia. Br J Haematol. 2020; 53. Rovirosa L, Ramos-Morales A, Javierre BM. 9(1):90. 188(1):36-48. The genome in a three-dimensional con- text: deciphering the contribution of non- 66. Levy B, Baughn LB, Chartrand S, et al. A 42. Dexter TM, Allen TD, Lajtha LG. Conditions coding mutations at enhancers to blood national multicenter evaluation of the controlling the proliferation of cancer. Front Immunol. 2020;11:592087. clinical utility of optical genome mapping haemopoietic stem cells in vitro. J Cell for assessment of genomic aberrations in Physiol. 1977;91(3):335-344. 54. Montefiori LE, Bendig S, Gu Z, et al. acute myeloid leukemia. medRxiv. 2020: Enhancer hijacking drives oncogenic 2020.2011.2007.20227728. 43. Michaeli J, Lerer I, Rachmilewitz EA, Fibach BCL11B expression in lineage ambiguous E. Stimulation of proliferation of human stem cell leukemia. Cancer Discov. 2021; 67. Gupta A, Place M, Goldstein S, et al. myeloid leukemia cells in culture: 11(11):2846-2867. Single-molecule analysis reveals Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 applications for cytogenetic analysis. widespread structural variation in multiple 55. Rustad EH, Yellapantula VD, Glodzik D, myeloma. Proc Natl Acad Sci USA. 2015; Blood. 1986;68(3):790-793. et al. Revealing the impact of structural 112(25):7689-7694. 44. He R, Wiktor AE, Hanson CA, et al. variants in multiple myeloma. Blood Cancer Conventional karyotyping and fluorescence Discov. 2020;1(3):258-273. 68. Yang H, Garcia-Manero G, Rush D, et al. in situ hybridization: an effective utilization Application of optical genome mapping for 56. King RL, McPhail ED, Meyer RG, et al. comprehensive assessment of strategy in diagnostic adult acute myeloid False-negative rates for MYC fluorescence chromosomal structural variants for clinical leukemia [published correction appears in in situ hybridization probes in B-cell neo- evaluation of myelodysplastic syndromes. Am J Clin Pathol. 2015;144(1):175]. Am J plasms. Haematologica. 2019;104(6): medRxiv. 2021:2021.2001.2013.21249611. Clin Pathol. 2015;143(6):873-878. e248-e251. 45. Coleman JF, Theil KS, Tubbs RR, Cook JR. 69. Brown LM, Lonsdale A, Zhu A, et al. The 57. Peterson JF, Pitel BA, Smoley SA, et al. application of RNA sequencing for the Diagnostic yield of bone marrow and Elucidating a false-negative MYC break- diagnosis and genomic classification of peripheral blood FISH panel testing in apart fluorescence in situ hybridization pediatric acute lymphoblastic leukemia clinically suspected myelodysplastic probe study by next-generation sequenc- [published correction appears in Blood syndromes and/or acute myeloid leukemia: ing in a patient with high-grade B-cell lym- Adv. 2020;4(7):1217]. Blood Adv. 2020;4(5): a prospective analysis of 433 cases. Am J phoma with IGH/MYC and IGH/BCL2 930-942. Clin Pathol. 2011;135(6):915-920. rearrangements. Cold Spring Harb Mol 46. He R, Wiktor AE, Durnick DK, et al. Bone Case Stud. 2019;5(3):a004077. 70. Debackere K, Marcelis L, Demeyer S, et al. marrow conventional karyotyping and Fusion transcripts FYN-TRAF3IP2 and 58. Smadbeck J, Peterson JF, Pearce KE, et al. KHDRBS1-LCK hijack T cell receptor signal- fluorescence in situ hybridization: defining Mate pair sequencing outperforms ing in peripheral T-cell lymphoma, not oth- an effective utilization strategy for fluorescence in situ hybridization in the erwise specified. Nat Commun. 2021;12(1): evaluation of myelodysplastic syndromes. genomic characterization of multiple 3705. Am J Clin Pathol. 2016;146(1):86-94. myeloma. Blood Cancer J. 2019;9(12):103. 47. Baliakas P, Jeromin S, Iskas M, et al; ERIC, 71. Foltz SM, Gao Q, Yoon CJ, et al. Evolution 59. Lopes JL, Webley M, Pitel BA, et al. and structure of clinically relevant gene the European Research Initiative on CLL. Characterizing false-positive fluorescence in fusions in multiple myeloma. Nat Commun. Cytogenetic complexity in chronic situ hybridization results by mate-pair 2020;11(1):2666. lymphocytic leukemia: definitions, sequencing in a patient with chronic mye- associations, and clinical impact. Blood. loid leukemia and progression to myeloid 72. H€ollein A, Twardziok SO, Walter W, et al. 2019;133(11):1205-1216. blast crisis. Cancer Genet. 2020;243:48-51. The combination of WGS and RNA-Seq is superior to conventional diagnostic tests in 48. Mrozek K, Harper DP, Aplan PD. 60. Sharma N, Smadbeck JB, Abdallah N, et al. multiple myeloma: ready for prime time? Cytogenetics and molecular genetics of acute The prognostic role of MYC structural Cancer Genet. 2020;242:15-24. lymphoblastic leukemia. Hematol Oncol Clin variants identified by NGS and FISH in North Am. 2009;23(5):991-1010, v. multiple myeloma. Clin Cancer Res. 2021; 73. L opez-Nieva P, Fern andez-Navarro P, 27(19):5430-5439. Gra~na-Castro O, et al. Detection of novel 49. Heerema NA, Byrd JC, Dal Cin PS, et al; fusion-transcripts by RNA-Seq in T-cell lym- Chronic Lymphocytic Leukemia Research 61. Roberts KG, Li Y, Payne-Turner D, et al. phoblastic lymphoma. Sci Rep. 2019;9(1): Consortium. Stimulation of chronic Targetable kinase-activating lesions in 5179. lymphocytic leukemia cells with CpG Ph-like acute lymphoblastic leukemia. N oligodeoxynucleotide gives consistent Engl J Med. 2014;371(11):1005-1015. 74. Docking TR, Parker JDK, J€ adersten M, karyotypic results among laboratories: a et al. A clinical transcriptome approach to CLL Research Consortium (CRC) Study. 62. Aypar U, Smoley SA, Pitel BA, et al. Mate patient stratification and therapy selection Cancer Genet Cytogenet. 2010;203(2): pair sequencing improves detection of in acute myeloid leukemia. Nat Commun. 134-140. genomic abnormalities in acute myeloid 2021;12(1):2474. leukemia. Eur J Haematol. 2019;102(1): 50. Put N, Konings P, Rack K, et al; Belgian 87-96. 75. Papaemmanuil E, Gerstung M, Bullinger L, Cytogenetic Group for Hemato-Oncology et al. Genomic classification and prognosis (BCGHO). Improved detection of chromo- 63. Lestringant V, Duployez N, Penther D, in acute myeloid leukemia. N Engl J Med. somal abnormalities in chronic lymphocytic et al. Optical genome mapping, a 2016;374(23):2209-2221. leukemia by conventional cytogenetics promising alternative to gold standard using CpG oligonucleotide and interleukin- cytogenetic approaches in a series of acute 76. Haferlach T, Nagata Y, Grossmann V, et al. 2 stimulation: a Belgian multicentric study. lymphoblastic leukemias. Genes Landscape of genetic lesions in 944 Genes Chromosomes Cancer. 2009;48(10): Chromosomes Cancer. 2021;60(10): patients with myelodysplastic syndromes. 843-853. 657-667. Leukemia. 2014;28(2):241-247. 51. Mellors PW, Binder M, Ketterling RP, et al. 64. Neveling K, Mantere T, Vermeulen S, et al. 77. Papaemmanuil E, Gerstung M, Malcovati L, Metaphase cytogenetics and plasma cell Next-generation cytogenetics: et al; Chronic Myeloid Disorders Working proliferation index for risk stratification in comprehensive assessment of 52 Group of the International Cancer Genome 2282 blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 AKKARI et al Consortium. Clinical and biological 91. Nelson AC, Yohe SL. Cancer whole- disorders. Genes Chromosomes Cancer. implications of driver mutations in genome sequencing: the quest for compre- 2016;55(5):480-491. myelodysplastic syndromes. Blood. 2013; hensive genomic profiling in routine oncol- 122(22):3616-3627, quiz 3699. ogy care. J Mol Diagn. 2021;23(7):784-787. 105. Li MM, Datto M, Duncavage EJ, et al. Standards and guidelines for the 78. Goodwin S, McPherson JD, McCombie 92. Dolnik A, Schrezenmeier J, Bullinger L. interpretation and reporting of sequence WR. Coming of age: ten years of next- Genome sequencing in myeloid cancers. variants in cancer: a joint consensus generation sequencing technologies. Nat N Engl J Med. 2021;384(25):e106. recommendation of the Association for Rev Genet. 2016;17(6):333-351. Molecular Pathology, American Society of 93. Haferlach T, Hutter S, Meggendorfer M. Clinical Oncology, and College of 79. Wetterstrand KA. DNA sequencing costs: Genome sequencing in myeloid cancers. American Pathologists. J Mol Diagn. 2017; data from the NHGRI genome sequencing N Engl J Med. 2021;384(25):e106. 19(1):4-23. program. GSP; 2020. 94. Mack E, Kremer J, Neubauer A. Genome 106. Sirohi D, Schmidt RL, Aisner DL, et al. 80. Braggio E, Egan JB, Fonseca R, Stewart sequencing in myeloid cancers. N Engl J Multi-institutional evaluation of interrater AK. Lessons from next-generation sequenc- Med. 2021;384(25):e106. agreement of variant classification based ing analysis in hematological malignancies. on the 2017 Association for Molecular Blood Cancer J. 2013;3(7):e127. 95. Larkin KTM, Byrd JC. Whole-genome Pathology, American Society of Clinical sequencing for myeloid disease: one assay Oncology, and College of American 81. Bersanelli M, Travaglino E, Meggendorfer to stratify them all? Nat Rev Clin Oncol. Pathologists standards and guidelines for Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 M, et al. Classification and personalized 2021;18(9):543-544. the interpretation and reporting of prognostic assessment on the basis of sequence variants in cancer. J Mol Diagn. clinical and genomic features in 96. Tyner JW, Tognon CE, Bottomly D, et al. Functional genomic landscape of acute 2020;22(2):284-293. myelodysplastic syndromes. J Clin Oncol. 2021;39(11):1223-1233. myeloid leukaemia. Nature. 2018; 107. Mardis ER. The $1,000 genome, the 562(7728):526-531. $100,000 analysis? Genome Med. 2010; 82. Palomo L, Ib ~ ez M, Ab an aigar M, et al; 2(11):84. 97. Nordlund J, Marincevic-Zuniga Y, Cavelier Spanish Group of MDS (GESMD). Spanish L, et al. Refined detection and phasing of guidelines for the use of targeted deep 108. Vergult S, Van Binsbergen E, Sante T, et al. structural aberrations in pediatric acute sequencing in myelodysplastic syndromes Mate pair sequencing for the detection of lymphoblastic leukemia by linked-read and chronic myelomonocytic leukaemia. Br chromosomal aberrations in patients with whole-genome sequencing. Sci Rep. 2020; J Haematol. 2020;188(5):605-622. intellectual disability and congenital 10(1):2512. malformations. Eur J Hum Genet. 2014; 83. Bailey MH, Meyerson WU, Dursi LJ, et al; 22(5):652-659. 98. Karczewski KJ, Francioli LC, Tiao G, et al; PCAWG Consortium. Retrospective Genome Aggregation Database evaluation of whole exome and genome 109. Miga KH, Koren S, Rhie A, et al. Telomere- Consortium. The mutational constraint mutation calls in 746 cancer samples to-telomere assembly of a complete human spectrum quantified from variation in [published correction appears in Nat 141,456 humans [published correction X chromosome. Nature. 2020;585(7823): Commun. 2020.11(1):4748]. Nat Commun. appears in Nature. 2021;590(7846);E53]. 79-84. 2020;11(1):4748. Nature. 2020;581(7809):434-443. 110. He R, Devine DJ, Tu ZJ, et al. Hybridization 84. Rusch M, Nakitandwe J, Shurtleff S, et al. 99. MacDonald JR, Ziman R, Yuen RK, Feuk L, capture-based next generation sequencing Clinical cancer genomic profiling by three- Scherer SW. The Database of Genomic reliably detects FLT3 mutations and classi- platform sequencing of whole genome, Variants: a curated collection of structural fies FLT3-internal tandem duplication allelic whole exome and transcriptome. Nat Com- variation in the human genome. Nucleic ratio in acute myeloid leukemia: a compar- mun. 2018;9(1):3962. Acids Res. 2014;42(Database issue):D986- ative study to standard fragment analysis D992. [published correction appears in Mod 85. Trudsø LC, Andersen JD, Jacobsen SB, Pathol. 2020;33(3):514]. Mod Pathol. 2020; et al. A comparative study of single 100. Iafrate AJ, Feuk L, Rivera MN, et al. 33(3):334-343. nucleotide variant detection performance Detection of large-scale variation in the using three massively parallel sequencing human genome. Nat Genet. 2004;36(9): 111. Ben Lassoued A, Nivaggioni V, Gabert J. methods. PLoS One. 2020;15(9):e0239850. 949-951. Minimal residual disease testing in hematologic malignancies and solid cancer. 86. Zare F, Dow M, Monteleone N, Hosny A, 101. Sebat J, Lakshmi B, Troge J, et al. Large- Expert Rev Mol Diagn. 2014;14(6):699-712. Nabavi S. An evaluation of copy number scale copy number polymorphism in the variation detection tools for cancer using human genome. Science. 2004;305(5683): 112. Jennings LJ, Arcila ME, Corless C, et al. whole exome sequencing data. BMC 525-528. Guidelines for validation of next-generation Bioinformatics. 2017;18(1):286. sequencing-based oncology panels: a joint 102. Collins RL, Brand H, Karczewski KJ, et al; consensus recommendation of the Associa- 87. Kim JC, Zuzarte PC, Murphy T, et al. Genome Aggregation Database tion for Molecular Pathology and College Cryptic genomic lesions in adverse-risk Consortium. A structural variation reference of American Pathologists. J Mol Diagn. acute myeloid leukemia identified by inte- for medical and population genetics 2017;19(3):341-365. grated whole genome and transcriptome [published correction appears in Nature. sequencing. Leukemia. 2020;34(1):306-311. 2021;590(7846):E55]. Nature. 2020; 113. Petrackova A, Vasinek M, Sedlarikova L, 581(7809):444-451. et al. Standardization of sequencing 88. Roepman P, de Bruijn E, van Lieshout S, coverage depth in NGS: recommendation et al. Clinical validation of whole genome 103. Miller DT, Lee K, Chung WK, et al; ACMG for detection of clonal and subclonal sequencing for cancer diagnostics. J Mol Secondary Findings Working Group. mutations in cancer diagnostics. Front Diagn. 2021;23(7):816-833. ACMG SF v3.0 list for reporting of Oncol. 2019;9:851. secondary findings in clinical exome and 89. Logsdon GA, Vollger MR, Eichler EE. Long- genome sequencing: a policy statement of 114. Starks ER, Swanson L, Docking TR, et al. read human genome sequencing and its the American College of Medical Genetics Assessing limit of detection in clinical applications. Nat Rev Genet. 2020;21(10): and Genomics (ACMG) [published sequencing. J Mol Diagn. 2021;23(4): 597-614. correction appears in Genet Med. 455-466. 2021;23(8):1582-1584]. Genet Med. 2021; 90. Duncavage EJ, Schroeder MC, O’Laughlin 23(8):1381-1390. 115. Chen X, Schulz-Trieglaff O, Shaw R, et al. M, et al. Genome sequencing as an Manta: rapid detection of structural variants alternative to cytogenetic analysis in 104. Schoumans J, Suela J, Hastings R, et al. and indels for germline and cancer myeloid cancers. N Engl J Med. 2021; Guidelines for genomic array analysis in sequencing applications. Bioinformatics. 384(10):924-935. acquired haematological neoplastic 2016;32(8):1220-1222. GUIDING THE EVOLUTION OF CYTOGENETIC TESTING blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 2283 116. Gordon LG, White NM, Elliott TM, et al. genotyping in the cloud. BMC Med Geno- evidence and reimbursement. J Pers Med. Estimating the costs of genomic mics. 2015;8(1):64. 2014;4(2):163-175. sequencing in cancer control. BMC Health 119. Dotson P. CPT codes: what are they, why 122. Singh AK, McGuirk JP. CAR T cells: Serv Res. 2020;20(1):492. are they necessary, and how are they continuation in a revolution of 117. Schwarze K, Buchanan J, Taylor JC, developed? Adv Wound Care (New immunotherapy. Lancet Oncol. 2020;21(3): Rochelle). 2013;2(10):583-587. e168-e178. Wordsworth S. Are whole-exome and whole-genome sequencing approaches 120. Nusgart M. HCPCS coding: an integral part cost-effective? A systematic review of the of your reimbursement strategy. Adv © 2022 by The American Society of Hematology. Licensed literature. Genet Med. 2018;20(10): Wound Care (New Rochelle). 2013;2(10): under Creative Commons Attribution-NonCommercial- 1122-1130. 576-582. NoDerivatives 4.0 International (CC BY-NC-ND 4.0), 118. Souilmi Y, Lancaster AK, Jung JY, et al. 121. Cohen JP, Felix AE. Personalized permitting only noncommercial, nonderivative use with Scalable and cost-effective NGS medicine’s bottleneck: diagnostic test attribution. All other rights reserved. Downloaded from http://ashpublications.org/blood/article-pdf/139/15/2273/1891037/bloodbld2021014309.pdf by guest on 24 January 2023 2284 blood® 14 APRIL 2022 | VOLUME 139, NUMBER 15 AKKARI et al