High-Throughput Immunogenetics for Clinical and Research Applications in Immunohematology: Potential and Challenges This information is current as of September 29, 2021. Anton W. Langerak, Monika Brüggemann, Frédéric Davi, Nikos Darzentas, Jacques J. M. van Dongen, David Gonzalez, Gianni Cazzaniga, Véronique Giudicelli, Marie-Paule Lefranc, Mathieu Giraud, Elizabeth A. Macintyre, Michael Hummel, Christiane Pott, Patricia J. T. A. Groenen, Kostas Stamatopoulos and the Downloaded from EuroClonality-NGS Consortium J Immunol published online 17 April 2017 http://www.jimmunol.org/content/early/2017/04/16/jimmun

ol.1602050 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

by guest on September 29, 2021 • No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2017 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published April 17, 2017, doi:10.4049/jimmunol.1602050

High-Throughput Immunogenetics for Clinical and Research Applications in Immunohematology: Potential and Challenges € † ´ ´ ‡ x Anton W. Langerak,* Monika Bruggemann,{ Frederic Davi, Nikos‖ Darzentas, JacquesJ.M.vanDongen,*DavidGonzalez, Gianni Cazzaniga, Ve´ronique Giudicelli,# # †† ‡‡ Marie-Paule Lefranc, Mathieu Giraud,** Elizabethxx A. Macintyre, Michael{{ Hummel, Christiane Pott,† Patricia J. T. A. Groenen, Kostas Stamatopoulos, and the EuroClonality-NGS Consortium1 Analysis and interpretation of Ig and TCR rear- (IG/TR) diversity is the combined effect of molecular [mainly rangements in the conventional, low-throughput way V(D)J recombination] and cellular diversification processes Downloaded from have their limitations in terms of resolution, coverage, during maturation of B and T lymphocytes (1–5). and biases. With the advent of high-throughput, next- The result of Ag-independent B and T lymphocyte differ- generation sequencing (NGS) technologies, a deeper anal- entiation and diversification that occurs in bone marrow and ysis of Ig and/or TCR (IG/TR) gene rearrangements is thymus is a broadly diverse, polyclonal repertoire of Ag-specific receptors (naive or primary repertoire), whereas Ag-dependent

now within reach, which impacts on all main applications http://www.jimmunol.org/ of IG/TR immunogenetic analysis. To bridge the gener- maturation in the periphery further shapes the IG/TR repertoire ation gap from low- to high-throughput analysis, the through selection processes (immunocompetent or - EuroClonality-NGS Consortium has been formed, with experienced repertoire) (6, 7). IG/TR polyclonality is one end the main objectives to develop, standardize, and validate of a continuum of immune profiles (Fig. 1A). Upon Ag-specific triggering and during inflammation certain IG/TR specificities the entire workflow of IG/TR NGS assays for 1) clon- canpredominateandleadtooneormoresmallclones(i.e.,the ality assessment, 2) minimal residual disease detection, offspring of particular B or T lymphocytes) on top of the poly- and 3) repertoire analysis. This concerns the preanalytical clonal repertoire, thus reflecting a more oligoclonal immune

(sample preparation, target choice), analytical (amplifica- profile. At the other end of the continuum, significant outgrowth by guest on September 29, 2021 tion, NGS), and postanalytical (immunoinformatics) phases. of a single lymphocyte clone with particular Ag specificity would Here we critically discuss pitfalls and challenges of IG/TR lead to a monoclonal immune profile, the hallmark of lymphoid NGS methodology and its applications in hemato-oncology malignancies (Fig. 1A). and immunology. The Journal of Immunology, 2017, In view of the extensive diversification mechanisms, the 198: 000–000. probability that two independent B or clones carry exactly the same IG/TR gene rearrangement by chance alone is virtually pecific Ag recognition of the adaptive negligible. IG/TR gene rearrangements thus form unique ge- is mediated by a remarkably diverse repertoire of Ag netic markers that can be justifiably viewed as molecular sig- S receptors—Igs on B lymphocytes (plus Abs secreted natures (DNA fingerprints), which have been instrumental for by plasma cells) and TCRs on T lymphocytes—showing high understanding both normal and pathologic immune responses affinity for a particular Ag. Fundamental to Ig and/or TCR (8–22). In addition, the spectrum of IG/TR repertoire diversity

*Department of Immunology, Laboratory for Medical Immunology, Erasmus MC, ORCIDs: 0000-0002-2078-3220 (A.W.L.); 0000-0003-0580-5636 (D.G.); 0000-0003- University Medical Center, 3015 CN Rotterdam, the Netherlands; †Second Medical 2955-4528 (G.C.); 0000-0003-2741-8047 (M.G.); 0000-0003-0520-0493 (E.A.M.). Department,UniversityHospitalSchleswig-Holstein,24105Kiel,Germany; Received for publication December 15, 2016. Accepted for publication January 9, 2017. ‡De´partement d’He´matologie, Assistance Publique – Hoˆpitaux de Paris Hopital Pitie´- Salpeˆtrie`re and Universite´ Pierre et Marie Curie – Universite´ Paris IV, 75005 Paris, See related articles in this issue: IJspeert et al. (J. Immunol. 198, 4156; DOI: https://doi. x France; Molecular Medicine Program, Central European Institute of Technology, org/10.4049/jimmunol.1601921) and Boyer et al. (J. Immunol. 198, 4148; DOI: { Masaryk University, 625 00 Brno, Czech Republic; Centre for Research and https://doi.org/10.4049/jimmunol.1601924). ‖ Cell Biology, Queen’s University Belfast, Belfast BT9 7AE, United Kingdom; Centro This work was supported by EuroClonality. Ricerca Tettamanti, Clinica Pediatrica Universita` Milano-Bicocca, 20900 Monza, Italy; #UMR 9002 CNRS – Universite´ de Montpellier, 34396 Montpellier, France; **Centre Address correspondence and reprint requests to Dr. Anton W. Langerak, Laboratory for de Recherche en Informatique Signal et Automatique de Lille, CNRS, Universite´ de Medical Immunology, Department of Immunology, Erasmus MC, University Medical Lille, 59000 Lille, France; ††De´partement d’He´matologie, Assistance Publique – Hoˆpitaux Center, Nb-1248a, Wytemaweg 80, 3015 CN Rotterdam, the Netherlands. E-mail de Paris Necker-Enfants Malades and Paris Descartes, 75015 Paris, France; ‡‡Institut fur€ address: [email protected] xx Pathologie, Charite´ – Universita¨tsmedizin Berlin, D-10117 Berlin, Germany; Department Abbreviations used in this article: ALL, acute lymphoblastic leukemia; CLL, chronic of Pathology, Radboud University Nijmegen Medical Center, 6525 GA Nijmegen, {{ lymphocytic leukemia; FL, follicular ; IG/TR, Ig and/or TCR; MRD, min- the Netherlands; and Institute of Applied Biosciences, Center for Research and imal residual disease; NGS, next-generation sequencing; RepSeq, Repertoire Sequencing; Technology Hellas, GR-57001 Thessaloniki, Greece RQ-PCR, real-time quantitative PCR; SHM, somatic hypermutation. 1All listed authors are members of the EuroClonality-NGS Consortium and have written this article on behalf of the entire consortium. Copyright Ó 2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1602050 2 HIGH-THROUGHPUT IMMUNOGENETICS IN IMMUNOHEMATOLOGY

nological conditions, although the true diversity of the IG/TR repertoire is mostly only partly disclosed using low-throughput Sanger-based sequencing. Nevertheless, clonal repertoire analysis has proven clinically relevant, as exemplified by chronic lym- phocytic leukemia (CLL) where identifying IGHV gene muta- tional status is established as one of the most robust prognostic markers in CLL (standardized by the European Research Ini- tiative on CLL [termed ERIC]; ericll.org) (41, 42). Even though many of the low-throughput IG/TR assays have thus been optimized and standardized to a high level, inherently they may occasionally provide suboptimal and even misleading results. Depending on the diagnostic application, the causes of concern as well as the issues at stake are different. However, fundamental to all is the enormous potential di- versity of IG/TR gene rearrangements necessitating the use of multiplex PCR assays with multiple primers that—even in the most optimal situation—are always a compromise. Indeed,

PCR biases due to differential performance or misannealing Downloaded from of primers can lead to artificial asymmetries with regard to gene frequencies, resulting in a false impression of repertoire FIGURE 1. IG/TR repertoire diversity translating into clinical applications skewing or even clonality status. The use of consensus primers in low-throughput methodology and high-throughput immunogenetics. (A) in the amplification protocols, though practical, implies a less Continuum of IG/TR repertoire diversity ranging from true polyclonality (far than complete coverage and thus a less comprehensive view of left end) through oligoclonality and low level clonality (middle) to clear repertoire diversity. Moreover, consensus primers have a http://www.jimmunol.org/ B monoclonality (far right end). ( ) Schematic representation of how IG/TR tendency to miss clonal IG rearrangements in the presence repertoire diversity translates into low-throughput methodology-based de- tection of repertoire, MRD, and clonality testing (left to right). (C) Schematic of somatic hypermutation (SHM). Both have merged as relevant representation of how high-throughput methodology discloses the full IG/TR problems in studies of, especially, nonmalignant (i.e., oligo/ sequence information of the entire cell population, thus allowing much more polyclonal) repertoires, for example, in normal individuals information for repertoire analysis, MRD monitoring, and clonality assess- as well as in settings of vaccination, autoimmunity, or im- ment to be drawn. mune reconstitution after allogenic hematopoietic stem cell transplantation or drug-induced lymphocyte depletion. SHM also translates into immunogenetic profiles that are especially can also be the cause for the inability to detect the clonal by guest on September 29, 2021 useful for clinical/diagnostic purposes for pathophysiological population in B cell malignancies with high SHM load, conditions, that is, clonotypic repertoire analysis, minimal re- for example, follicular lymphoma (FL), , sidual disease (MRD) detection and monitoring, and clonality and others. assessment (Fig. 1B) (23–26). Furthermore, the IG/TR low-throughput approaches have their specific limitations in all diagnostic applications. Gene- Limitations in low-throughput and potential of high-throughput IG/ Scan analysis/spectratyping and heteroduplex analysis, widely TR immunogenetics used for evaluation of the clonality status (43–45), usually Multiplex PCR assays for the detection of clonally rearranged ensure diagnostic accuracy with an analytical sensitivity of IG/TR have been designed, standardized, and validated by maximally 5%, dependent upon the type of lymphoma and the BIOMED-2/EuroClonality Consortium (euroclonality.org) the context (e.g., specimen type and size, DNA quality and (27), which also established guidelines for interpretation of integrity). This limited dynamic range of spectratyping ren- clonality results (28). These BIOMED-2 multiplex assays are ders this approach suboptimal for determining low-level dis- now widely used to obtain corroborative evidence for the dif- semination or for monitoring purposes. RQ-PCR–based ferential diagnosis of reactive lesions versus malignant lympho- quantification of clone-specific IG/TR gene rearrangements can mas and for assessing clonal identity (25, 29–31), as well as for be highly informative for the detection of MRD; however, identification of appropriate PCR targets for MRD detection. sensitivity varies, depending on the PCR protocol and the 2 2 Highly sensitive (10 4 to 10 5) MRD analysis of follow-up relative size of the background of normal (polyclonal) B and samples upon treatment is achieved by using IG/TR rearrange- T lymphocytes. Also, oligoclonality at initial diagnosis and ments for design of patient-specific oligonucleotides in real-time clonal evolution of IG/TR gene rearrangements between di- quantitative (RQ-)PCR. Efforts by the EuroMRD Consortium agnosis and relapse, both relevant for acute lymphoblastic (euromrd.org) to standardize IG/TR-based RQ-PCR MRD leukemia (ALL) (46–48), or the occurrence of (ongoing) SHM diagnostics have resulted in improved technical guidelines and (for some mature B cell malignancies) (49), may give false- definitions of MRD positivity and MRD quantitative range to negative results. Traditional repertoire analysis, conducted by express MRD levels in a reproducible way across multiple centers Sanger sequencing of (subcloned) rearranged IG/TR PCR and therapeutic protocols (32). Such accurate MRD quantita- amplicons or by single cell strategies, is limited in depth by the tion is relevant for use as surrogate markers for clinical outcome strain on laboratory manpower and resources required, ham- and for actual monitoring in view of risk-based clinical decisions pering study of the dynamics of immune responses (e.g., in or group stratification (33–40). Finally, IG/TR repertoire at the vaccination) and clonal evolution (e.g., intraclonal diversifi- genomic or transcript level has been studied in many immu- cation of Ig genes in B cell malignancies). Moreover, using The Journal of Immunology 3 multiplex assays that do not encompass the entire V domain clonality approaches is not as deep as for MRD purposes, thus encoding region may lead to 1) ambiguities in IG/TR gene and allowing running more patient samples in parallel for the allele identification, and 2) incomplete appreciation of the true various clonality targets. Thus, NGS-based clonality assess- impact of SHM (41, 42). ment has the potential to replace the classical GeneScan ap- With the introduction of next-generation sequencing (NGS) proach. in immunogenetics, also collectively termed Repertoire Se- Capture-based detection of clonal IG/TR rearrangements and quencing (RepSeq) analysis (50–59), a deeper analysis of IG/ translocations. An alternative approach is to use a hybridiza- TR gene rearrangements is now within reach, which could tion capture or pull-down methodology to enrich for IG/TR have a profound impact on the three main applications of loci DNA using small probes or baits, potentially allowing for immunogenetic analysis (clonality assessment, MRD detec- less biased sequencing of all target regions. In addition, this tion and monitoring, and repertoire analysis; Table I). Based approach is suitable for parallel identification of chromosomal on the wealth of IG/TR sequences generated, much more translocations involving IG/TR gene loci, present in lym- information on the entire IG/TR repertoire diversity of the phoma and leukemia samples, in the same workflow (60, 61). cells can thus be disclosed (Fig. 1C). Finally, hybridization capture approaches allow the analysis of However, there are still many challenges toward routine all the IG/TR targets—including unproductive rearrange- clinical application that need to be overcome (Table II; see also ments—in a single assay, saving precious material, and also next paragraph for more details concerning the main appli- facilitates the use of a given target as a reference or baseline for cations). These are all subjects of study in the EuroClonality- normalization against the others, that is, comparing the level Downloaded from NGS Consortium (euroclonalityngs.org; coordinated by a of total rearranged B lymphocytes versus T lymphocytes, or steering group chaired by A. W. Langerak), which consists of clonal B lymphocytes out of total B lymphocytes. several EuroClonality laboratories experienced in the design of assays for detecting IG/TR rearrangements, supplemented Reappraisal of the meaning of clonality. The depth of resolution by laboratories with expertise in IG/TR gene-based MRD provided by NGS-based clonality assessment will almost certainly require a reappraisal of the term clonality. In par- studies (from the EuroMRD network) or IG/TR repertoire http://www.jimmunol.org/ studies and immunoinformatics (from the ERIC network). ticular, defining clonality simply based on the fact that a certain percentage of identical IG/TR rearrangement sequences can be Challenges in RepSeq data analysis for different applications found does not accurately address the clinical meaning and Clonality assessment. For NGS-based clonality assessment the implications of the term. On the other hand, the future following issues are at stake. amassment of a large body of NGS-based IG/TR sequence data obtained in the context of clonality assessment will certainly Assessing the clonal relationship between lesions. Although GeneScan analysis/spectratyping takes (identical) size of the reveal the true extent of repertoire skewing in healthy indi- viduals, in patients with reactive lesions or other immuno-

generated IG/TR amplicons as the basis for clonality assess- by guest on September 29, 2021 ment, NGS also provides the individual sequences to create so- logical conditions such as postinfection or vaccination, or in called clonotypes, a sequence-based compilation of identical oncology patients receiving immunotherapeutics. Therefore, a rearrangements. This allows a quantitative consideration of the major challenge in the quantitative NGS era will be to evaluate individual IG/TR rearrangements, and also helps to confirm and possibly revise the current definitions as to what constitutes the clonal relationship of the amplified rearrangements in clonality and what is the border, if any, to accurately distin- multiple lesions from different locations or in multiple lesions guish reactive from malignant lymphoproliferations in a over time (Fig. 2). NGS will also enable rapid identification clinicopathological context. One of the challenges is to define and evaluation of bone marrow involvement in lymphoma how and where borders should be drawn in the spectrum of patients. In addition, intraclonal diversity in mature B cell mono-, oligo-, and polyclonality in patients with lympho- malignancies that undergo continuous SHM processes can proliferations associated with viruses (e.g., HIV, EBV, or be evaluated via NGS. hepatitis C virus) and with lymphoproliferations suspicious for malignant lymphoma. Now for the first time, to our Complementarity of IG/TR targets to reduce the risk of missing knowledge, the question can be addressed whether clonal clones. NGS-based clonality assessment still relies on an initial multiplex PCR step, which will be hampered by the same limitations as all other PCR-based IG/TR studies, that is, polymorphisms and especially SHM that prevent primer Table I. Potential impact of IG/TR RepSeq analysis on different clinical annealing. Additionally, the diverse amplification efficiency applications leading to unequal amplification of different rearrangements in multiplex PCR potentially prevents precise quantification. Application Impact NGS-based clonality assessment thus also requires multiple Clonality testing Accurate evaluation of IG/TR clonal relationship in complementary IG/TR targets to ensure a high detection rate; multiple lesions (true relapse or dissemination versus for B cell malignancies a combination of IGH V-D-J and IGK new or secondary malignancy) Assessment of low-level dissemination of malignant clone V-J (and preferably also IGH D-J), and for T cell malignancies Assessment of intraclonal diversity of malignant clone a combination of TRB and TRG assays. Costs for NGS-based MRD monitoring More easy and unbiased identification of IG/TR targets clonality assays are affordable—also when compared with the Sensitive monitoring of malignant clone Evaluation of oligoclonal IG/TR heterogeneity at traditional approaches by GeneScan analysis. This can be diagnosis and clonal evolution of resistant clone accomplished by multiplexing of targets and samples for the Repertoire analysis Assessment of higher depth/coverage of IG/TR sequencing step following initial separate PCRs. In consid- rearrangements ering costs, it is worth noting that the required depth of NGS Assessment of higher number of IG/TR clonotypes 4 HIGH-THROUGHPUT IMMUNOGENETICS IN IMMUNOHEMATOLOGY

Table II. Challenges of IG/TR RepSeq analysis for different clinical Identification of the correct index clone. The standard way to applications identify a leukemia/lymphoma-associated index sequence is to perform IG/TR multiplex PCR followed by amplicon sequenc- Application Challenge ing and clonotype selection based on a frequency threshold Clonality testing Multiplexing and complementarity of IG/TR targets $5% of all analyzed sequences (Fig. 3A). This procedure is error Defining amount of starting material in the context of prone, because—depending on the clinical setting—IG/TR gene specimen type, DNA integrity, and estimated neoplastic cell load rearrangements of unrelated B and T lymphocyte clones can Defining value of clonal size, numerical cut-off values, account for a considerable fraction of amplified sequences and and limits of detection might be misinterpreted as leukemia/lymphoma-specific rear- Reappraisal/redefinition of the meaning of clonality rangements (53), in particular if the true malignant IG/TR re- Data analysis pipeline including visualization MRD monitoring Multiplexing and complementarity of IG/TR targets arrangement is missed by the applied primer set, for example, Amount and type of starting material due to harboring extensive SHM rates (66). Use of internal controls (e.g., spike-in) Definition of limits of detection (quantifiable, sensitivity) Correct MRD quantification. Published studies on IG/TR Correct for disproportional PCR amplification of amplicon sequencing frequently quantify the MRD level by rearrangements Data analysis pipeline including visualization counting the number of index sequences and dividing them Repertoire analysis Multiplexing and complete coverage of genes by the total number of sequenced amplicons. Given that Equal amplification and thus representation of genes the applied multiplex PCR assays only amplify rearranged IG/ Downloaded from Sequence information of entire V gene (IG loci) TR genes, although cells with the respective IG/TR gene Error correction prior to accurately defining mutations, polymorphisms in germline configuration are not targeted, this might lead Data analysis pipeline including visualization to considerable errors in MRD quantification, particularly in situations with a low number of polyclonal background B lymphocytes. For example, when IGH NGS is performed in a B cell malignancy after anti–B cell treatment resulting in a sizes and numerical cut-offs are of (any) biological or clinical http://www.jimmunol.org/ value. subtotal depletion of polyclonal B lymphocytes, preferential sequencing of IGH-rearranged B lymphocytes might lead to MRD detection. Despite recent developments offering proof of a considerable overestimation of MRD (Fig. 3B). Therefore, principle that NGS-based MRD assessment using IG/TR adequate internal controls have to be included, for which genes is workable in lymphoid leukemias and different approaches have been proposed: 1) different plas- and potentially even more sensitive than alternative options mids containing known IGH gene rearrangements (62); 2) (RQ-PCR, multicolor flow cytometry) (54, 55, 62–65), sev- synthetic control templates spiked at limiting dilution into eral issues remain to be addressed, which will be the topic of each sample to compute the average number of reads for each the sections below. sequenced spiked synthetic template (67); 3) spike-in of defined by guest on September 29, 2021

FIGURE 2. NGS-based clonality assessment. (A) IGK NGS analysis in a tonsil sample showing high reproducibility of the diversity in three different lab- oratories. Data presented as IGKV gene (x-axis) versus frequency (y-axis), highlighting different IGKJ genes in different colors. (B) Clonal identity in two lymph node biopsies (1 y time difference) as evidenced from identical clonotypes in the two consecutive samples. Data presented as CDR3 aa length (x-axis) versus frequency (y-axis), highlighting clonotype sequences in different colors. Visualization using ARResT/Interrogate (92). The Journal of Immunology 5

FIGURE 3. Clonotype frequencies and accurate MRD quantification. (A) IGH clonotype frequency in the peripheral blood of 12 cases of follicular lymphoma at diagnosis (left) and in normal buffy coat (right). A threshold of 5% is generally used to identify lymphoma/leukemia-related clonotypes at diagnosis. A Downloaded from significant number of clonotypes .2 and ,5% is seen in buffy coat. Dashed lines indicate the 5% threshold for index clone selection and 2 and 1% thresholds. (B) Correct MRD quantification is dependent on the background level of polyclonal B lymphocytes. MRD levels may greatly differ following chemotherapy versus B cell depletion therapy, yet might give rise to the same relative frequency of index sequences, thus necessitating the use of internal references for accurately calculating MRD levels. amounts of buffy coat (68). Again, in mature B cell malignancies on technical assay performance, the number of good quality like FL, where clonal heterogeneity is caused by ongoing SHM, reads, and the total number of cells analyzed. Given the http://www.jimmunol.org/ one has to follow the presence of evolved clonotypes to assess the complexity of NGS-based MRD assessment, successful par- correct MRD level during treatment or at relapse. ticipation in external quality control rounds should be a Detection of clonal heterogeneity and clonal evolution. In prerequisite for laboratories generating MRD data for clinical lymphoid malignancies different degrees and ways of IG/TR decision making. Notably, quality control rounds for RQ-PCR heterogeneity have been documented. In ALL ongoing gene and NGS-based MRD are being organized for experienced rearrangements lead to IG/TR oligoclonality (e.g., IGH D-J to laboratories by the EuroMRD Consortium twice yearly. IGH V-D-J changes, and VH substitutions of complete VH-

DH-JH rearrangements). Using an IGH NGS approach, IGH Repertoire analysis. NGS-based repertoire analysis requires op- by guest on September 29, 2021 oligoclonality was computationally identified in the vast ma- timization at the level of sequence methodology as well as at the jority of childhood B cell precursor–ALL cases by comparing level of data analysis. the IGH D-J stem of different complete IGH gene rear- Methodological challenges and possible solutions. Considering rangements with the IGH D-J stem of the index sequence; that Ag gene repertoires may vary considerably be- ∼10% of cases showed .1000 related sequences (53). tween different lymphocyte subpopulations (70–72), sorted Whether such potentially related sequences should also be cells should be used whenever possible, especially when ana- tracked in follow-up samples has not yet been prospectively lyzing complex repertoires of nonclonal lymphoid pop- analyzed. In contrast, in mature B cell malignancies, IGH ulations. In order to cover representative diversity and draw clonal heterogeneity is mainly the result of ongoing SHM, meaningful conclusions from experimental data, it is impor- leading to intraclonal diversity (69). This might lead to a tant to ensure adequate sampling, including biological as well decrease of amplification efficacy of the respective rearrange- as sequencing replicates. One microgram of DNA corre- ment in NGS and thereby to a low or even false-negative sponds to only 150,000 cells or up to 300,000 copies of a MRD result. This can at least partly be compensated by re- given IG/TR locus, which represents a minor proportion of lying on additional index sequences from other IG targets, if the total repertoire. Given that IG/TR transcript levels may present. vary considerably between cell populations (for instance more Validation, quality control, and standardized interpretation of than 100-fold difference in IG transcript levels between naive NGS-based MRD results. Technical guidelines, as developed B cells and plasma cells), mRNA use complicates the quan- for RQ-PCR based MRD analysis, are currently lacking for tification of clonal expansions and may lead to erroneous NGS-based MRD; moreover, data interpretation regarding the conclusions about clonal architecture. Nevertheless, starting definition of MRD positivity/negativity is very heterogeneous with mRNA allows the use of 59 RACE/template switching in the published literature. Considering the potential higher protocols that reduce amplification biases, but often at a cost sensitivity of NGS, minimal technical requirements have of shorter sequence length (70–74). Within the EuroClonality- therefore to be defined including the theoretical sensitivity for a NGS Consortium, gDNA was chosen as the template in mul- single sample analyzed for MRD. This is particularly relevant tiplex PCR with 59 IGH primers annealing to the leader with respect to the prognostic impact of certain MRD region to amplify the complete VH sequence. This is important thresholds; the sensitivity of NGS could indeed be higher than for 1) accurate identification of the rearranged germline V gene other methods, but is in fact a function of the DNA amount or and allele, and 2) robust determination of the SHM load in the the number of cells analyzed in a single sample. Furthermore, a IG repertoire. Optimization of primer sets was performed using clear definition of MRD positivity/negativity is needed based a collection of plasmids containing most functional V genes to 6 HIGH-THROUGHPUT IMMUNOGENETICS IN IMMUNOHEMATOLOGY ensure proper recognition of the targets and minimize ampli- of the SHM status in CLL, both approaches need to be compared. fication biases. Errors introduced during the amplification and Following a pilot study showing identical results in the vast sequencing phases complicate data interpretation, particularly in majority of CLL cases (93 out of 95, 98%), validation of the the case of IG gene repertoire analysis, where discriminating technique is currently ongoing in several EuroClonality-NGS between SHM versus error may prove challenging. Among the laboratories. Of note, satellite clones differing from the pre- error correction algorithms that have been proposed (75–77), dominant clone by nucleotide substitutions and more rarely by the most promising approach relies on molecular barcoding indels are often observed, which as mentioned earlier, could of each individual template molecule to improve data quality represent true SHM or artifacts (Fig. 4B). Furthermore, addi- and sequencing accuracy (78–80). Barcoding strategies have tional unrelated clonal VDJ sequences are detected at a lower also been developed on an mRNA template but suffer from frequency in a number of cases; whether these represent minor limited sequencing depth (79). Reliable repertoire analysis independent CLL clones or derive from other concomitant by amplification-based NGS strategies is dependent on com- lymphoproliferation(s) still remains elusive. parable amplification efficiency, which can be difficult to con- NGS profiling of complex repertoires may also prove trol in highly multiplexed strategies. Preferential amplification instrumental in dissecting normal and pathological immune can be detected by comparing repertoires during normal devel- responses (e.g., infection, immune reconstitution, vaccina- opment, as demonstrated for TRD rearrangements during thymic tion, allergy, or autoimmunity). Importantly, the prospective development (Fig. 4A). IG/TR loci that undergo deletion, such as large-scale accumulation of sequence data

IGK and TRD, also present particular challenges for repertoire in various pathological contexts may help unveil immuno- Downloaded from analysis. genetic signatures that are distinctive in certain entities, and Finally, Ag receptors are heterodimers and sequence in- provide valuable insight regarding pathogenic mechanisms of formation of both chains is necessary for thorough repertoire diseases and effect of immunomodulating therapies, immune analysis. Although, in lymphoid tumors, the chain pair of the surveillance defects, or even normal immune system con- clonotypic Ag receptor can be determined fairly easily from the stitution. dominant clonotypes, it is virtually impossible to do so in Our recent NGS study of the T cell repertoire in CLL http://www.jimmunol.org/ polyclonal populations with a vastly heterogeneous sequence documented oligoclonal expansions, with T cell clones being composition. NGS technologies are evolving to investigate persistent and further expanding over time and other T cell paired chain repertoires; methods to circumvent this problem clones being shared by different patients, hence appearing to include 1) two-dimensional barcoding system tagging H chain be disease specific. Altogether, these findings revealed an as yet and L chain V genes of the BcR Ig within individual cells (81), unappreciated extent of TCR skewing, with implications for and 2) partitioning single cells in emulsion-based droplets future interventions into CLL microenvironment interactions followed by a linkage RT-PCR producing a composite BcR Ig and interdependencies (85).

H chain and L chain V region product (51, 82–84). by guest on September 29, 2021 NGS-based repertoire data analysis in malignant and Computational RepSeq analysis nonmalignant lymphocytes. To test whether NGS could re- In all cases, molecular analysis of IG/TR genes eventually place the conventional Sanger sequencing for accurate assessment concerns the in silico analysis of the nucleotide sequences of

FIGURE 4. NGS-based repertoire analysis. (A) TRD repertoire during human thymocyte development. Different types of TRD rearrangements are present at different frequencies during human thymic development. (B) Additional clonotypic sequences as detected by NGS in different CLL samples. Each of the two dominant clonotypic sequences (a productive IGHV3-11 rearrangement and a nonproductive IGHV3-47 rearrangement) is surrounded by minor satellite se- quences that differ from the main one by unique point mutations. Whether they are true SHM or PCR/sequence artifacts is currently unknown. Visualization using Vidjil software (91). The Journal of Immunology 7 their rearrangements. Although NGS enables us to explore elements with a collection of dedicated computational re- immune repertoires and responses in their immense variability sources and tools in the form of IMGT (imgt.org) (87–90), and complexity, it also naturally produces vast amounts of Vidjil (vidjil.org) (91), and most recently ARResT/Interrogate complex data that render these steps highly nontrivial. (bat.infspire.org/arrest/) (92). Raw NGS data processing is a critical but often neglected Specific challenges for MRD detection. A true computational first step. It can involve statistical demultiplexing to ensure challenge for clonotype assessment in the MRD setting is the minimal digital contamination (i.e., misassignment of reads to pronounced ongoing SHM of the IG locus in FL leading to samples due to noisy barcodes) while maximizing read re- clonal evolution and heterogeneity during treatment course and covery and thus depth (86), prefiltering of capture data to at relapse. Here, the acceptance criteria for the numbers of remove non-IG/TR sequences and thus aid computational mismatches in relation to the diagnostic clonotype have to efficiency, paired-end and multilane joining to maximize respect a continuous mutation process of FL cells, and the clonal quality and information content, and primer annotation and relationship of heterogeneous sequences has therefore to be trimming. This last component has many important func- carefully defined to assess a correct MRD value over time. tions: assay development, assessment of run quality and Overall, high reproducibility and a standardized NGS-based amplicon (and thus junctional) completeness, removal of the quantification are particularly important for the comparabil- artificial primer sequences before immunogenetic annotation, ity of NGS-based MRD in the setting of prospective multicenter amplification bias correction when combined with a control trials, as is the goal of EuroClonality-NGS. sample (e.g., highlighting misbehaving primers), and com- Downloaded from putational efficiency. Specific challenges for repertoire analysis. In addition to raw data Processed reads are then immunogenetically annotated (e.g., processing, V(D)J gene assignment, and clonotype identifi- involved V and J genes and alleles, and exact CDR3 sequence). cation, Ag receptor gene repertoire studies require additional Standardization is important here, and although the use of sequence analysis extending from diversity profiles to clonal different underlying algorithms is arguably inevitable, it is architecture, CDR3 length and characteristics, clonal dy- important to at least safeguard the consistent use of germline http://www.jimmunol.org/ namics (if temporal samples are analyzed), and clonotype sequences, rules for CDR3 identification and translation comparisons between different lymphoid populations/ (including C104 and W/F 118 anchor positions), and no- individuals/disease entities. The complexity of the data menclature of V, D, and J genes and alleles. At this stage, and generated would argue for concomitant analysis via inde- especially for clinical applications, it is also desirable to identify pendent existing pipelines (87–92) or newly published ones incomplete and special/uncommon rearrangements. The (93, 94). eventual classification of reads into all these rearrangement types is in fact necessary when the experimental design,

(i.e., when the amplification of specific rearrangements takes Conclusions by guest on September 29, 2021 place in separate tubes) requires the normalization of abun- RepSeq analysis has quickly found its way into hematology and dances. It also allows classifying reads with no signs of any immunology research. Implementation of this high-throughput rearrangement as unknown and excluding them from the total technology into routine clinical applications requires stan- sample read count used as the basis for relative abundances. dardization, validation, and application-specific challenges that This supervised abundance calculation is critical in clinical should be covered in a network of laboratories with specialists applications when thresholds are used for diagnosis or prog- that bring immunobiological knowledge, technical experience nosis or therapy decisions. on NGS methodology, and immunoinformatics expertise. Only Basic immunogenetic annotation can be used to construct then will RepSeq analysis be fully exploited for its high potential clonotypes (73). The exact definition of a clonotype has been in diagnostic and translational research, with the full benefit diverse in the literature and has depended on the underlying for patients. The EuroClonality-NGS Consortium was formed question, experiment, and data. In general, a clonotype can be to bridge the immunogenetics generation gap from low- considered as a distinct rearrangement event, in which case it throughput IG/TR analysis to high-throughput RepSeq anal- can and has been used to report repertoires in an expression- ysis. Main objectives of the EuroClonality-NGS Consortium are independent manner. to develop, standardize, and validate IG/TR NGS assays for the Eventually, mining these inherently complex data for in- different clinical applications in a platform-independent way. formation requires additional functionalities, such as filtering Even though several such assays have already been published, and visualization, within an interactive graphical environment there still is a need for optimization in assay development that provides flexibility and enables expert user input. How- with the aim to ensure better coverage of all the genes and also ever, this interactivity needs to be bidirectional and include to evaluate other types of rearrangements (partial IGH D–J feedback and guidance to the user toward unambiguous and rearrangements, IGK locus rearrangements including those consistent conclusions. involving the k-deleting element, partial TRB D–J rearrange- Taken together and properly standardized and validated, ments, IG/TR translocations, etc.). these key elements (i.e., comprehensive and consistent an- One of the most important aspects in the implementation of notation of sequences; true representation of the repertoire; RepSeq analysis in both research and routine diagnostic practice meaningful clonotype definition and quantitation; end-user- concerns standardization of the entire NGS workflow, which friendly but unambiguous visualization) can form a consis- pertains not only to the analytical phase, but also to the pre- tent computational platform usable by both researchers and analytical (e.g., sample preparation and target choice) and the clinicians. Within EuroClonality-NGS complementary postanalytical phases (e.g., immunoinformatics pipeline, data immunoinformatics expertise is available that covers these key visualization, and interpretation). Another very important aspect 8 HIGH-THROUGHPUT IMMUNOGENETICS IN IMMUNOHEMATOLOGY of implementing RepSeq analysis for different applications re- 17. Nadel, B., A. Tang, G. Lugo, V. Love, G. Escuro, and A. J. Feeney. 1998. De- creased frequency of rearrangement due to the synergistic effect of nucleotide lates to validation of the technology against standard method- changes in the heptamer and nonamer of the recombination signal sequence of the ologies (GeneScan analysis, Sanger sequencing, RQ-PCR, and V kappa gene A2b, which is associated with increased susceptibility of Navajos to multiparameter flow cytometry) via large-scale, multilaboratory Haemophilus influenzae type b disease. J. Immunol. 161: 6068–6073. 18. Naylor, K., G. Li, A. N. Vallejo, W. W. Lee, K. Koetz, E. Bryl, J. Witkowski, testing of clinical samples in the context of clinical trials. J. Fulbright, C. M. Weyand, and J. J. Goronzy. 2005. The influence of age on T cell Eventually, this should provide robust tools and methodology, generation and TCR diversity. J. Immunol. 174: 7446–7452. 19. Suzuki, N., T. Harada, S. Mihara, and T. Sakane. 1996. Characterization of a which will allow exploiting the full potential of this powerful new germline Vk gene encoding cationic anti-DNA antibody and role of receptor editing technology in diagnostic patient care. An additional confounding for development of the autoantibody in patients with systemic lupus erythematosus. J. Clin. Invest. 98: 1843–1850. factor, at least for diagnostic applications of RepSeq analysis, is 20. Weller, S., M. C. Braun, B. K. Tan, A. Rosenwald, C. Cordier, M. E. Conley, the lack of expertise and guidelines for implementation, evalu- A. Plebani, D. S. Kumararatne, D. Bonnet, O. Tournilhac, et al. 2004. Human ation, and clinical translation of the results. Both validation blood IgM “memory” B cells are circulating splenic marginal zone B cells harboring a prediversified immunoglobulin repertoire. Blood 104: 3647–3654. studies and guideline development are among the key activities of 21. Xu, J. L., and M. M. Davis. 2000. Diversity in the CDR3 region of V(H) is suf- the EuroClonality-NGS Consortium. ficient for most antibody specificities. Immunity 13: 37–45. 22. Zemlin, M., K. Bauer, M. Hummel, S. Pfeiffer, S. Devers, C. Zemlin, H. Stein, and H. T. Versmold. 2001. The diversity of rearranged immunoglobulin heavy chain variable region genes in peripheral blood B cells of preterm infants is restricted by Acknowledgments short third complementarity-determining regions but not by limited gene segment We thank all members of the EuroClonality-NGS Consortium for their input in usage. Blood 97: 1511–1513. the discussions, Jos Rijntjes and Florian Thonier for preparing figures, Marieke 23. Damle, R. N., T. Wasil, F. Fais, F. Ghiotto, A. Valetto, S. L. Allen, A. Buchbinder,

D. Budman, K. Dittmar, J. Kolitz, et al. 1999. Ig V gene mutation status and CD38 Downloaded from Bitter for artwork and general support in management of the EuroClonality- expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood NGS Consortium, and Bibi van Bodegom for secretarial assistance. 94: 1840–1847. 24. Hamblin, T. J., Z. Davis, A. Gardiner, D. G. Oscier, and F. K. Stevenson. 1999. Unmutated Ig V(H) genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood 94: 1848–1854. Disclosures 25. van Krieken, J. H., A. W. Langerak, E. A. Macintyre, M. Kneba, E. Hodges, The EuroClonality network receives royalty income from InVivoScribe Tech- R. G. Sanz, G. J. Morgan, A. Parreira, T. J. Molina, J. Cabec¸adas, et al. 2007. Im- nologies. proved reliability of lymphoma diagnostics via PCR-based clonality testing: report of the BIOMED-2 Concerted Action BHM4-CT98-3936. Leukemia 21: 201–206. http://www.jimmunol.org/ 26. Willemse, M. J., T. Seriu, K. Hettinger, E. d’Aniello, W. C. Hop, E. R. Panzer- Grumayer,€ A. Biondi, M. Schrappe, W. A. Kamps, G. Masera, et al. 2002. De- References tection of minimal residual disease identifies differences in treatment response 1. Tonegawa, S. 1983. Somatic generation of antibody diversity. Nature 302: 575–581. between T-ALL and precursor B-ALL. Blood 99: 4386–4393. 2. Davis, M. M., and P. J. Bjorkman. 1988. T-cell antigen receptor genes and T-cell 27. van Dongen, J. J., A. W. Langerak, M. Bruggemann,€ P. A. Evans, M. Hummel, recognition. Nature 334: 395–402. F. L. Lavender, E. Delabesse, F. Davi, E. Schuuring, R. Garcı´a-Sanz, et al. 2003. 3. Schlissel, M. S. 2003. Regulating antigen-receptor gene assembly. Nat. Rev. Design and standardization of PCR primers and protocols for detection of clonal Immunol. 3: 890–899. immunoglobulin and T-cell receptor gene recombinations in suspect lymphopro- 4. Lefranc, M.-P., and G. Lefranc. 2001. The T Cell Receptor Factsbook. Academic liferations: report of the BIOMED-2 Concerted Action BMH4-CT98-3936. Leu- Press, San Diego, London. kemia 17: 2257–2317. € 5. Lefranc, M.-P., and G. Lefranc. 2001. The Immunoglobulin Factsbook. Academic 28. Langerak, A. W., P. J. Groenen, M. Bruggemann, K. Beldjord, C. Bellan, L. Bonello, by guest on September 29, 2021 Press, San Diego, London. E. Boone, G. I. Carter, M. Catherwood, F. Davi, et al. 2012. EuroClonality/ 6. Monroe, J. G., and K. Dorshkind. 2007. Fate decisions regulating bone marrow and BIOMED-2 guidelines for interpretation and reporting of Ig/TCR clonality testing peripheral B lymphocyte development. Adv. Immunol. 95: 1–50. in suspected lymphoproliferations. Leukemia 26: 2159–2171. 7. von Boehmer, H., and F. Melchers. 2010. Checkpoints in lymphocyte development 29. Appenzeller, S., C. Gilissen, J. Rijntjes, B. B. Tops, A. Kastner-van Raaij, and autoimmune disease. Nat. Immunol. 11: 14–20. K. M. Hebeda, L. Nissen, B. E. Dutilh, J. H. van Krieken, and P. J. Groenen. 2015. 8. Cappione, A. J., A. E. Pugh-Bernard, J. H. Anolik, and I. Sanz. 2004. Lupus IgG Immunoglobulin rearrangement analysis from multiple lesions in the same patient VH4.34 antibodies bind to a 220-kDa glycoform of CD45/B220 on the surface of using next-generation sequencing. Histopathology 67: 843–858. human B lymphocytes. J. Immunol. 172: 4298–4307. 30. Geurts-Giele, W. R., I. L. Wolvers-Tettero, W. N. Dinjens, K. H. Lam, and 9. Chang, Q., J. Abadi, P. Alpert, and L. Pirofski. 2000. A pneumococcal capsular A. W. Langerak. 2013. Successive B-cell lymphomas mostly reflect recurrences polysaccharide vaccine induces a repertoire shift with increased VH3 expression in rather than unrelated primary lymphomas. Am. J. Clin. Pathol. 140: 114–126. peripheral B cells from human immunodeficiency virus (HIV)-uninfected but not 31. Langerak, A. W., T. J. Molina, F. L. Lavender, D. Pearson, T. Flohr, C. Sambade, HIV-infected persons. J. Infect. Dis. 181: 1313–1321. E. Schuuring, T. Al Saati, J. J. van Dongen, and J. H. van Krieken. 2007. Poly- 10. Fais, F., F. Ghiotto, S. Hashimoto, B. Sellars, A. Valetto, S. L. Allen, P. Schulman, merase chain reaction-based clonality testing in tissue samples with reactive lym- V. P. Vinciguerra, K. Rai, L. Z. Rassenti, et al. 1998. Chronic lymphocytic leukemia phoproliferations: usefulness and pitfalls. A report of the BIOMED-2 Concerted B cells express restricted sets of mutated and unmutated antigen receptors. J. Clin. Action BMH4-CT98-3936. Leukemia 21: 222–229. Invest. 102: 1515–1525. 32. van der Velden, V. H. J., G. Cazzaniga, A. Schrauder, J. Hancock, P. Bader, 11. Hadzidimitriou, A., A. Agathangelidis, N. Darzentas, F. Murray, M. H. Delfau- E. R. Panzer-Grumayer, T. Flohr, R. Sutton, H. Cave, H. O. Madsen, et al. 2007. Larue, L. B. Pedersen, A. N. Lopez, A. Dagklis, P. Rombout, K. Beldjord, et al. Analysis of minimal residual disease by Ig/TCR gene rearrangements: guidelines for 2011. Is there a role for antigen selection in mantle cell lymphoma? Immunogenetic interpretation of real-time quantitative PCR data. Leukemia 21: 604–611. support from a series of 807 cases. Blood 118: 3088–3095. 33. Bruggemann,€ M., T. Raff, T. Flohr, N. Go¨kbuget, M. Nakao, J. Droese, 12. Ho¨hn, H., C. Neukirch, K. Freitag, A. Necker, W. Hitzler, B. Seliger, and S. Luschen,€ C. Pott, M. Ritgen, U. Scheuring, et al. 2006. Clinical significance of M. J. Maeurer. 2002. Longitudinal analysis of the T-cell receptor (TCR)-VA and -VB repertoire in CD81 T cells from individuals immunized with recombinant minimal residual disease quantification in adult patients with standard-risk acute hepatitis B surface antigen. Clin. Exp. Immunol. 129: 309–317. lymphoblastic leukemia. Blood 107: 1116–1123. 13. Khan, N., N. Shariff, M. Cobbold, R. Bruton, J. A. Ainsworth, A. J. Sinclair, 34. Hillmen, P., D. R. Cohen, K. Cocks, A. Pettitt, H. A. Sayala, A. C. Rawstron, L. Nayak, and P. A. Moss. 2002. Cytomegalovirus seropositivity drives the CD8 D. B. Kennedy, C. Fegan, D. W. Milligan, J. Radford, et al. 2011. A randomized T cell repertoire toward greater clonality in healthy elderly individuals. J. Immunol. phase II trial of fludarabine, cyclophosphamide and mitoxantrone (FCM) with or 169: 1984–1992. without rituximab in previously treated chronic lymphocytic leukaemia. Br. J. 14. Klein, U., K. Rajewsky, and R. Kuppers.€ 1998. Human immunoglobulin (Ig)M1 Haematol. 152: 570–578. IgD1 peripheral blood B cells expressing the CD27 cell surface antigen carry so- 35. Kwok, M., A. C. Rawstron, A. Varghese, P. A. Evans, S. J. O’Connor, C. Doughty, matically mutated variable region genes: CD27 as a general marker for somatically D. J. Newton, P. Moreton, and P. Hillmen. 2016. Minimal residual disease is an mutated (memory) B cells. J. Exp. Med. 188: 1679–1689. independent predictor for 10-year progression-free and overall survival in CLL. 15. Matsutani, T., T. Yoshioka, Y. Tsuruta, S. Iwagami, T. Toyosaki-Maeda, Blood 128: 2770–2773. T. Horiuchi, A. B. Miura, A. Watanabe, G. Takada, R. Suzuki, and M. Hirokawa. 36. Ladetto, M., C. Lobetti-Bodoni, B. Mantoan, M. Ceccarelli, C. Boccomini, 2000. Restricted usage of T-cell receptor alpha-chain variable region (TCRAV) and E. Genuardi, A. Chiappella, L. Baldini, G. Rossi, A. Pulsoni, et al. 2013. Per- T-cell receptor beta-chain variable region (TCRBV) repertoires after human allo- sistence of minimal residual disease in bone marrow predicts outcome in fol- geneic haematopoietic transplantation. Br. J. Haematol. 109: 759–769. licular lymphomas treated with a rituximab-intensive program. Blood 122: 16. Miura, Y., C. J. Thoburn, E. C. Bright, M. Sommer, S. Lefell, M. Ueda, 3759–3766. S. Nakao, and A. D. Hess. 2001. Characterization of the T-cell repertoire in 37. Moreton, P., B. Kennedy, G. Lucas, M. Leach, S. M. Rassam, A. Haynes, J. Tighe, autologous graft-versus-host disease (GVHD): evidence for the involvement of D. Oscier, C. Fegan, A. Rawstron, and P. Hillmen. 2005. Eradication of minimal antigen-driven T-cell response in the development of autologous GVHD. Blood residual disease in B-cell chronic lymphocytic leukemia after alemtuzumab therapy 98: 868–876. is associated with prolonged survival. J. Clin. Oncol. 23: 2971–2979. The Journal of Immunology 9

38. Pott, C., E. Hoster, M. H. Delfau-Larue, K. Beldjord, S. Bo¨ttcher, V. Asnafi, throughput sequencing detects minimal residual disease in acute T lymphoblastic A. Plonquet, R. Siebert, E. Callet-Bauchu, N. Andersen, et al. 2010. Molecular leukemia. Sci. Transl. Med. 4: 134ra63. remission is an independent predictor of clinical outcome in patients with mantle 59. Wu, Y. C., D. Kipling, H. S. Leong, V. Martin, A. A. Ademokun, and D. K. Dunn- cell lymphoma after combined immunochemotherapy: a European MCL intergroup Walters. 2010. High-throughput immunoglobulin repertoire analysis distinguishes study. Blood 115: 3215–3223. between human IgM memory and switched memory B-cell populations. Blood 116: 39. Puig, N., M. E. Sarasquete, A. Balanzategui, J. Martı´nez, B. Paiva, H. Garcı´a, 1070–1078. S. Fumero, C. Jime´nez, M. Alcoceba, M. C. Chillo´n, et al. 2014. Critical evaluation 60.Walker,B.A.,C.P.Wardell,D.C.Johnson,M.F.Kaiser,D.B.Begum, of ASO RQ-PCR for minimal residual disease evaluation in multiple myeloma. A N. B. Dahir, F. M. Ross, F. E. Davies, D. Gonzalez, and G. J. Morgan. 2013. comparative analysis with flow cytometry. Leukemia 28: 391–397. Characterization of IGH locus breakpoints in multiple myeloma indicates a 40. Rawstron, A. C., W. M. Gregory, R. M. de Tute, F. E. Davies, S. E. Bell, subset of translocations appear to occur in pregerminal center B cells. Blood 121: M. T. Drayson, G. Cook, G. H. Jackson, G. J. Morgan, J. A. Child, and R. G. Owen. 3413–3419. 2015. Minimal residual disease in myeloma by flow cytometry: independent prediction 61. Wren, D., B. A. Walker, M. Bruggemann,€ M. A. Catherwood, C. Pott, of survival benefit per log reduction. Blood 125: 1932–1935. K. Stamatopoulos, A. W. Langerak, D. Gonzalez, et al. 2017. Comprehensive 41. Ghia, P., K. Stamatopoulos, C. Belessi, C. Moreno, S. Stilgenbauer, F. Stevenson, translocation and clonality detection in lymphoproliferative disorders by next gen- F. Davi, and R. Rosenquist, European Research Initiative on CLL. 2007. ERIC eration sequencing. Haematologica 102: e57–e60. recommendations on IGHV gene mutational status analysis in chronic lymphocytic 62.Faham,M.,J.Zheng,M.Moorhead,V.E.Carlton,P.Stow,E.Coustan- leukemia. Leukemia 21: 1–3. Smith, C. H. Pui, and D. Campana. 2012. Deep-sequencing approach for 42. Langerak, A. W., F. Davi, P. Ghia, A. Hadzidimitriou, F. Murray, K. N. Potter, minimal residual disease detectioninacutelymphoblasticleukemia.Blood 120: R. Rosenquist, K. Stamatopoulos, and C. Belessi, European Research Initiative on 5173–5180. CLL (ERIC). 2011. Immunoglobulin sequence analysis and prognostication in 63. Ladetto, M., M. Bruggemann,€ L. Monitillo, S. Ferrero, F. Pepin, D. Drandi, CLL: guidelines from the ERIC review board for reliable interpretation of prob- D. Barbero, A. Palumbo, R. Passera, M. Boccadoro, et al. 2014. Next-generation lematic cases. Leukemia 25: 979–984. sequencing and real-time quantitative PCR for minimal residual disease detection in 43. Bottaro, M., E. Berti, A. Biondi, N. Migone, and L. Crosti. 1994. Heteroduplex B-cell disorders. Leukemia 28: 1299–1307. analysis of T-cell receptor gamma gene rearrangements for diagnosis and monitoring 64. Logan, A. C., N. Vashi, M. Faham, V. Carlton, K. Kong, I. Bun˜o, J. Zheng, of cutaneous T-cell lymphomas. Blood 83: 3271–3278. M. Moorhead, M. Klinger, B. Zhang, et al. 2014. Immunoglobulin and T cell Downloaded from ń 44. Langerak, A. W., T. Szczepa ski, M. van der Burg, I. L. Wolvers-Tettero, and receptor gene high-throughput sequencing quantifies minimal residual disease in J. J. van Dongen. 1997. Heteroduplex PCR analysis of rearranged T cell receptor acute lymphoblastic leukemia and predicts post-transplantation relapse and survival. genes for clonality assessment in suspect T cell proliferations. Leukemia 11: 2192– Biol. Blood Marrow Transplant. 20: 1307–1313. 2199. 65. Rawstron, A. C., C. Fazi, A. Agathangelidis, N. Villamor, R. Letestu, J. Nomdedeu, 45. Linke, B., I. Bolz, A. Fayyazi, M. von Hofen, C. Pott, J. Bertram, W. Hiddemann, C. Palacio, O. Stehlikova, K. A. Kreuzer, S. Liptrot, et al. 2016. A complementary and M. Kneba. 1997. Automated high resolution PCR fragment analysis for role of multiparameter flow cytometry and high-throughput sequencing for minimal identification of clonally rearranged immunoglobulin heavy chain genes. Leukemia residual disease detection in chronic lymphocytic leukemia: an European Research 11: 1055–1062. Initiative on CLL study. Leukemia 30: 929–936. 46. Bruggemann, M., A. Schrauder, T. Raff, H. Pfeifer, M. Dworzak, O. G. Ottmann, 66. Martinez-Lopez, J., J. J. Lahuerta, F. Pepin, M. Gonza´lez, S. Barrio, R. Ayala, http://www.jimmunol.org/ V. Asnafi, A. Baruchel, R. Bassan, Y. Benoit, et al. 2010. Standardized MRD N. Puig, M. A. Montalban, B. Paiva, L. Weng, et al. 2014. Prognostic value of deep quantification in European ALL trials: proceedings of the second international sequencing method for minimal residual disease detection in multiple myeloma. symposium on MRD assessment in Kiel, Germany, 18-20 September 2008. Leu- Blood 123: 3073–3079. kemia 24: 521–535. 67. Wu, D., R. O. Emerson, A. Sherwood, M. L. Loh, A. Angiolillo, B. Howie, J. Vogt, 47. de Haas, V., O. J. Verhagen, A. E. von dem Borne, W. Kroes, H. van den Berg, and M. Rieder, I. Kirsch, C. Carlson, et al. 2014. Detection of minimal residual disease C. E. van der Schoot. 2001. Quantification of minimal residual disease in children in B lymphoblastic leukemia by high-throughput sequencing of IGH. Clin. Cancer with oligoclonal B-precursor acute lymphoblastic leukemia indicates that the clones Res. 20: 4540–4548. that grow out during relapse already have the slowest rate of reduction during in- 68.Kotrova,M.,K.Muzikova,E.Mejstrikova,M.Novakova,V.Bakardjieva- duction therapy. Leukemia 15: 134–140. Mihaylova, K. Fiser, J. Stuchly, M. Giraud, M. Salson, C. Pott, et al. 2015. The 48. Germano, G., L. del Giudice, S. Palatron, E. Giarin, G. Cazzaniga, A. Biondi, and predictive strength of next-generation sequencing MRD detection for relapse G. Basso. 2003. Clonality profile in relapsed precursor-B-ALL children by Gen-

compared with current methods in childhood ALL. Blood 126: 1045–1047. by guest on September 29, 2021 eScan and sequencing analyses. Consequences on minimal residual disease moni- 69. Stamatopoulos, K., C. Kosmas, C. Belessi, N. Stavroyianni, P. Kyriazopoulos, and toring. Leukemia 17: 1573–1582. T. Papadaki. 2000. Molecular insights into the immunopathogenesis of follicular 49. Sutton, L. A., E. Kostareli, A. Hadzidimitriou, N. Darzentas, A. Tsaftaris, A. Anagnostopoulos, R. Rosenquist, and K. Stamatopoulos. 2009. Extensive lymphoma. Immunol. Today 21: 298–305. intraclonal diversification in a subgroup of chronic lymphocytic leukemia patients 70. Calis, J. J., and B. R. Rosenberg. 2014. Characterizing immune repertoires by with stereotyped IGHV4-34 receptors: implications for ongoing interactions with high throughput sequencing: strategies and applications. Trends Immunol. 35: antigen. Blood 114: 4460–4468. 581–590. 50. Boyd, S. D., E. L. Marshall, J. D. Merker, J. M. Maniar, L. N. Zhang, B. Sahaf, 71. Georgiou, G., G. C. Ippolito, J. Beausang, C. E. Busse, H. Wardemann, and C. D. Jones, B. B. Simen, B. Hanczaruk, K. D. Nguyen, et al. 2009. Measurement S. R. Quake. 2014. The promise and challenge of high-throughput sequencing of and clinical monitoring of human lymphocyte clonality by massively parallel VDJ the antibody repertoire. Nat. Biotechnol. 32: 158–168. pyrosequencing. Sci. Transl. Med. 1: 12ra23. 72. Six, A., M. E. Mariotti-Ferrandiz, W. Chaara, S. Magadan, H. P. Pham, 51. DeKosky, B. J., G. C. Ippolito, R. P. Deschner, J. J. Lavinder, Y. Wine, M. P. Lefranc, T. Mora, V. Thomas-Vaslin, A. M. Walczak, and P. Boudinot. B. M. Rawlings, N. Varadarajan, C. Giesecke, T. Do¨rner, S. F. Andrews, et al. 2013. The past, present, and future of immune repertoire biology - the rise of next- 2013. High-throughput sequencing of the paired human immunoglobulin heavy generation repertoire analysis. Front. Immunol. 4: 413. and light chain repertoire. Nat. Biotechnol. 31: 166–169. 73. Li, S., M. P. Lefranc, J. J. Miles, E. Alamyar, V. Giudicelli, P. Duroux, J. D. Freeman, 52. Freeman, J. D., R. L. Warren, J. R. Webb, B. H. Nelson, and R. A. Holt. 2009. V. D. Corbin, J. P. Scheerlinck, M. A. Frohman, et al. 2013. IMGT/HighV QUEST Profiling the T-cell receptor beta-chain repertoire by massively parallel sequencing. paradigm for T cell receptor IMGT clonotype diversity and next generation repertoire Genome Res. 19: 1817–1824. immunoprofiling. Nat. Commun. 4: 2333. 53. Gawad, C., F. Pepin, V. E. Carlton, M. Klinger, A. C. Logan, D. B. Miklos, 74. Mamedov, I. Z., O. V. Britanova, I. V. Zvyagin, M. A. Turchaninova, D. A. Bolotin, M. Faham, G. Dahl, and N. Lacayo. 2012. Massive evolution of the immuno- E. V. Putintseva, Y. B. Lebedev, and D. M. Chudakov. 2013. Preparing unbiased T- globulin heavy chain locus in children with B precursor acute lymphoblastic leu- cell receptor and antibody cDNA libraries for the deep next generation sequencing kemia. Blood 120: 4407–4417. profiling. Front. Immunol. 4: 456. 54. Logan, A. C., H. Gao, C. Wang, B. Sahaf, C. D. Jones, E. L. Marshall, I. Bun˜o, 75.Estorninho,M.,V.B.Gibson,D.Kronenberg-Versteeg,Y.F.Liu,C.Ni,K.Cerosaletti, R. Armstrong, A. Z. Fire, K. I. Weinberg, et al. 2011. High-throughput VDJ se- and M. Peakman. 2013. A novel approach to tracking antigen-experienced CD4 T cells quencing for quantification of minimal residual disease in chronic lymphocytic into functional compartments via tandem deep and shallow TCR clonotyping. J. leukemia and immune reconstitution assessment. Proc. Natl. Acad. Sci. USA 108: Immunol. 191: 5430–5440. 21194–21199. 76. Greiff, V., U. Menzel, U. Haessler, S. C. Cook, S. Friedensohn, T. A. Khan, 55. Logan, A. C., B. Zhang, B. Narasimhan, V. Carlton, J. Zheng, M. Moorhead, M. Pogson, I. Hellmann, and S. T. Reddy. 2014. Quantitative assessment of the M. R. Krampf, C. D. Jones, A. N. Waqar, M. Faham, et al. 2013. Minimal residual robustness of next-generation sequencing of antibody variable gene repertoires from disease quantification using consensus primers and high-throughput IGH se- immunized mice. BMC Immunol. 15: 40. quencing predicts post-transplant relapse in chronic lymphocytic leukemia. Leuke- 77. Nguyen, P., J. Ma, D. Pei, C. Obert, C. Cheng, and T. L. Geiger. 2011. Identi- mia 27: 1659–1665. fication of errors introduced during high throughput sequencing of the T cell re- 56. Robins, H. S., S. K. Srivastava, P. V. Campregher, C. J. Turtle, J. Andriesen, ceptor repertoire. BMC Genomics 12: 106. S. R. Riddell, C. S. Carlson, and E. H. Warren. 2010. Overlap and effective size of 78. Khan, T. A., S. Friedensohn, A. R. Gorter de Vries, J. Straszewski, H. J. Ruscheweyh, the human CD81 T cell receptor repertoire. Sci. Transl. Med. 2: 47ra64. and S. T. Reddy. 2016. Accurate and predictive antibody repertoire profiling by 57.Wang,C.,C.M.Sanders,Q.Yang,H.W.Schroeder,Jr.,E.Wang, molecular amplification fingerprinting. Sci. Adv. 2: e1501371. F.Babrzadeh,B.Gharizadeh,R.M.Myers,J.R.Hudson,Jr.,R.W.Davis,and 79. Shugay, M., O. V. Britanova, E. M. Merzlyak, M. A. Turchaninova, J. Han. 2010. High throughput sequencing reveals a complex pattern of dynamic I. Z. Mamedov, T. R. Tuganbaev, D. A. Bolotin, D. B. Staroverov, interrelationships among human T cell subsets. Proc.Natl.Acad.Sci.USA107: E. V. Putintseva, K. Plevova, et al. 2014. Towards error-free profiling of immune 1518–1523. repertoires. Nat. Methods 11: 653–655. 58. Wu, D., A. Sherwood, J. R. Fromm, S. S. Winter, K. P. Dunsmore, M. L. Loh, 80. Turchaninova, M. A., A. Davydov, O. V. Britanova, M. Shugay, V. Bikos, H. A. Greisman, D. E. Sabath, B. L. Wood, and H. Robins. 2012. High- E. S. Egorov, V. I. Kirgizova, E. M. Merzlyak, D. B. Staroverov, D. A. Bolotin, et al. 10 HIGH-THROUGHPUT IMMUNOGENETICS IN IMMUNOHEMATOLOGY

2016. High-quality full-length immunoglobulin profiling with unique molecular 88. Aouinti, S., V. Giudicelli, P. Duroux, D. Malouche, S. Kossida, and M. P. Lefranc. barcoding. Nat. Protoc. 11: 1599–1616. 2016. IMGT/StatClonotype for pairwise evaluation and visualization of NGS IG 81. Busse, C. E., I. Czogiel, P. Braun, P. F. Arndt, and H. Wardemann. 2014. Single- and TR IMGT clonotype (AA) diversity or expression from IMGT/HighV- cell based high-throughput sequencing of full-length immunoglobulin heavy and QUEST. Front. Immunol. 7: 339. light chain genes. Eur. J. Immunol. 44: 597–603. 89. Lefranc, M. P. 2014. Immunoglobulin and T cell receptor genes: IMGT(Ò) and the 82. McDaniel, J. R., B. J. DeKosky, H. Tanno, A. D. Ellington, and G. Georgiou. birth and rise of immunoinformatics. Front. Immunol. 5: 22. 2016. Ultra-high-throughput sequencing of the immune receptor repertoire from 90. Lefranc, M. P., V. Giudicelli, P. Duroux, J. Jabado-Michaloud, G. Folch, millions of lymphocytes. Nat. Protoc. 11: 429–442. S. Aouinti, E. Carillon, H. Duvergey, A. Houles, T. Paysan-Lafosse, et al. 2015. 83. Agathangelidis, A., N. Darzentas, A. Hadzidimitriou, X. Brochet, F. Murray, X. J. Yan, IMGTÒ, the international ImMunoGeneTics information systemÒ 25 years on. Z. Davis, E. J. van Gastel-Mol, C. Tresoldi, C. C. Chu, et al. 2012. Stereotyped B-cell Nucleic Acids Res. 43: D413–D422. receptors in one-third of chronic lymphocytic leukemia: a molecular classification with 91. Giraud, M., M. Salson, M. Duez, C. Villenet, S. Quief, A. Caillault, N. Grardel, implications for targeted therapies. Blood 119: 4467–4475. C. Roumier, C. Preudhomme, and M. Figeac. 2014. Fast multiclonal clusterization 84. Hadzidimitriou, A., N. Darzentas, F. Murray, T. Smilevska, E. Arvaniti, C. Tresoldi, of V(D)J recombinations from high-throughput sequencing. BMC Genomics 15: A. Tsaftaris, N. Laoutaris, A. Anagnostopoulos, F. Davi, et al. 2009. Evidence for the 409. significant role of immunoglobulin light chains in antigen recognition and selection in 92.Bystry,V.,T.Reigl,A.Krejci,M.Demko,B.Hanakova,A.Grioni, chronic lymphocytic leukemia. Blood 113: 403–411. 85.Vardi,A.,A.Agathangelidis,E.Stalika,M.Karypidou,A.Siorenta,A.Anagnostopoulos, H. Knecht, M. Schlitt, P. Dreger, L. Sellner, et al. 2017. ARResT/ R. Rosenquist, A. Hadzidimitriou, P. Ghia, L. A. Sutton, and K. Stamatopoulos. 2016. Interrogate: an interactive immunoprofiler for IG/TR NGS data. Bio- Antigen selection shapes the T-cell repertoire in chronic lymphocytic leukemia. Clin. informatics 33: 435–437. Cancer Res. 22: 167–174. 93. IJspeert, H., P. A. van Schouwenburg, D. van Zessen, I. Pico-Knijnenburg, 86. Renaud, G., U. Stenzel, T. Maricic, V. Wiebe, and J. Kelso. 2015. deML: robust A. P. Stubbs, and M. van der Burg. 2017. Antigen Receptor Galaxy: a user- demultiplexing of Illumina sequences using a likelihood-based approach. Bio- friendly, web-based tool for analysis and visualization of T and B cell re- informatics 31: 770–772. ceptor repertoire data. J. Immunol. 198: XXX–XXX. 87. Alamyar, E., P. Duroux, M. P. Lefranc, and V. Giudicelli. 2012. IMGT(Ò) tools 94. Boyer, F., H. Boutouil, I. Dalloul, Z. Dalloul, J. Cook-Moreau, J.-C. Aldigier, for the nucleotide analysis of immunoglobulin (IG) and T cell receptor (TR) V-(D)- C. Carrion, B. Herve, E. Scaon, M. Cogne´, and S. Pe´ron. 2017. CSReport: a new J repertoires, polymorphisms, and IG mutations: IMGT/V-QUEST and IMGT/ computational tool designed for automatic analysis of class switch recombination Downloaded from HighV-QUEST for NGS. Methods Mol. Biol. 882: 569–604. junctions sequenced by high-throughput sequencing. J. Immunol. 198: XXX–XXX. http://www.jimmunol.org/ by guest on September 29, 2021