TRAF3 loss-of-function reveals the noncanonical NF-κB pathway as a therapeutic target in diffuse large B cell lymphoma.

Here, we report recurrent focal deletions of the chr14q32.31-32 locus, including TRAF3, a negative regulator of NF-κB signaling, in de novo diffuse large B cell lymphoma (DLBCL) (24/324 cases). Integrative analysis revealed an association between TRAF3 copy number loss with accumulation of NIK, the central noncanonical (NC) NF-κB kinase, and increased NC NF-κB pathway activity. Accordingly, TRAF3 genetic ablation in isogenic DLBCL model systems caused upregulation of NIK and enhanced NC NF-κB downstream signaling. Knockdown or pharmacological inhibition of NIK in TRAF3-deficient cells differentially impaired their proliferation and survival, suggesting an acquired onco-addiction to NC NF-κB. TRAF3 ablation also led to exacerbated secretion of the immunosuppressive cytokine IL-10. Coculturing of TRAF3-deficient DLBCL cells with CD8+ T cells impaired the induction of Granzyme B and interferon (IFN) γ, which were restored following neutralization of IL-10. Our findings corroborate a direct relationship between TRAF3 genetic alterations and NC NF-κB activation, and highlight NIK as a potential therapeutic target in a defined subset of DLBCL.

Loss of CREBBP and KMT2D cooperate to accelerate lymphomagenesis and shape the lymphoma immune microenvironment.

Despite regulating overlapping gene enhancers and pathways, CREBBP and KMT2D mutations recurrently co-occur in germinal center (GC) B cell-derived lymphomas, suggesting potential oncogenic cooperation. Herein, we report that combined haploinsufficiency of Crebbp and Kmt2d induces a more severe mouse lymphoma phenotype (vs either allele alone) and unexpectedly confers an immune evasive microenvironment manifesting as CD8+ T-cell exhaustion and reduced infiltration. This is linked to profound repression of immune synapse genes that mediate crosstalk with T-cells, resulting in aberrant GC B cell fate decisions. From the epigenetic perspective, we observe interaction and mutually dependent binding and function of CREBBP and KMT2D on chromatin. Their combined deficiency preferentially impairs activation of immune synapse-responsive super-enhancers, pointing to a particular dependency for both co-activators at these specialized regulatory elements. Together, our data provide an example where chromatin modifier mutations cooperatively shape and induce an immune-evasive microenvironment to facilitate lymphomagenesis.

Evaluating Outcome Prediction via Baseline, End-of-Treatment, and Delta Radiomics on PET-CT Images of Primary Mediastinal Large B-Cell Lymphoma.

OBJECTIVES: Accurate outcome prediction is important for making informed clinical decisions in cancer treatment. In this study, we assessed the feasibility of using changes in radiomic features over time (Delta radiomics: absolute and relative) following chemotherapy, to predict relapse/progression and time to progression (TTP) of primary mediastinal large B-cell lymphoma (PMBCL) patients. MATERIAL AND METHODS: Given the lack of standard staging PET scans until 2011, only 31 out of 103 PMBCL patients in our retrospective study had both pre-treatment and end-of-treatment (EoT) scans. Consequently, our radiomics analysis focused on these 31 patients who underwent [18F]FDG PET-CT scans before and after R-CHOP chemotherapy. Expert manual lesion segmentation was conducted on their scans for delta radiomics analysis, along with an additional 19 EoT scans, totaling 50 segmented scans for single time point analysis. Radiomics features (on PET and CT), along with maximum and mean standardized uptake values (SUVmax and SUVmean), total metabolic tumor volume (TMTV), tumor dissemination (Dmax), total lesion glycolysis (TLG), and the area under the curve of cumulative standardized uptake value-volume histogram (AUC-CSH) were calculated. We additionally applied longitudinal analysis using radial mean intensity (RIM) changes. For prediction of relapse/progression, we utilized the individual coefficient approximation for risk estimation (ICARE) and machine learning (ML) techniques (K-Nearest Neighbor (KNN), Linear Discriminant Analysis (LDA), and Random Forest (RF)) including sequential feature selection (SFS) following correlation analysis for feature selection. For TTP, ICARE and CoxNet approaches were utilized. In all models, we used nested cross-validation (CV) (with 10 outer folds and 5 repetitions, along with 5 inner folds and 20 repetitions) after balancing the dataset using Synthetic Minority Oversampling TEchnique (SMOTE). RESULTS: To predict relapse/progression using Delta radiomics between the baseline (staging) and EoT scans, the best performances in terms of accuracy and F1 score (F1 score is the harmonic mean of precision and recall, where precision is the ratio of true positives to the sum of true positives and false positives, and recall is the ratio of true positives to the sum of true positives and false negatives) were achieved with ICARE (accuracy = 0.81 ± 0.15, F1 = 0.77 ± 0.18), RF (accuracy = 0.89 ± 0.04, F1 = 0.87 ± 0.04), and LDA (accuracy = 0.89 ± 0.03, F1 = 0.89 ± 0.03), that are higher compared to the predictive power achieved by using only EoT radiomics features. For the second category of our analysis, TTP prediction, the best performer was CoxNet (LASSO feature selection) with c-index = 0.67 ± 0.06 when using baseline + Delta features (inclusion of both baseline and Delta features). The TTP results via Delta radiomics were comparable to the use of radiomics features extracted from EoT scans for TTP analysis (c-index = 0.68 ± 0.09) using CoxNet (with SFS). The performance of Deauville Score (DS) for TTP was c-index = 0.66 ± 0.09 for n = 50 and 0.67 ± 03 for n = 31 cases when using EoT scans with no significant differences compared to the radiomics signature from either EoT scans or baseline + Delta features (p-value> 0.05). CONCLUSION: This work demonstrates the potential of Delta radiomics and the importance of using EoT scans to predict progression and TTP from PMBCL [18F]FDG PET-CT scans.

ARID1A orchestrates SWI/SNF-mediated sequential binding of transcription factors with ARID1A loss driving pre-memory B cell fate and lymphomagenesis.

ARID1A, a subunit of the canonical BAF nucleosome remodeling complex, is commonly mutated in lymphomas. We show that ARID1A orchestrates B cell fate during the germinal center (GC) response, facilitating cooperative and sequential binding of PU.1 and NF-kB at crucial genes for cytokine and CD40 signaling. The absence of ARID1A tilts GC cell fate toward immature IgM+CD80-PD-L2- memory B cells, known for their potential to re-enter new GCs. When combined with BCL2 oncogene, ARID1A haploinsufficiency hastens the progression of aggressive follicular lymphomas (FLs) in mice. Patients with FL with ARID1A-inactivating mutations preferentially display an immature memory B cell-like state with increased transformation risk to aggressive disease. These observations offer mechanistic understanding into the emergence of both indolent and aggressive ARID1A-mutant lymphomas through the formation of immature memory-like clonal precursors. Lastly, we demonstrate that ARID1A mutation induces synthetic lethality to SMARCA2/4 inhibition, paving the way for potential precision therapy for high-risk patients.

TMTV-Net: fully automated total metabolic tumor volume segmentation in lymphoma PET/CT images - a multi-center generalizability analysis.

PURPOSE: Total metabolic tumor volume (TMTV) segmentation has significant value enabling quantitative imaging biomarkers for lymphoma management. In this work, we tackle the challenging task of automated tumor delineation in lymphoma from PET/CT scans using a cascaded approach. METHODS: Our study included 1418 2-[18F]FDG PET/CT scans from four different centers. The dataset was divided into 900 scans for development/validation/testing phases and 518 for multi-center external testing. The former consisted of 450 lymphoma, lung cancer, and melanoma scans, along with 450 negative scans, while the latter consisted of lymphoma patients from different centers with diffuse large B cell, primary mediastinal large B cell, and classic Hodgkin lymphoma cases. Our approach involves resampling PET/CT images into different voxel sizes in the first step, followed by training multi-resolution 3D U-Nets on each resampled dataset using a fivefold cross-validation scheme. The models trained on different data splits were ensemble. After applying soft voting to the predicted masks, in the second step, we input the probability-averaged predictions, along with the input imaging data, into another 3D U-Net. Models were trained with semi-supervised loss. We additionally considered the effectiveness of using test time augmentation (TTA) to improve the segmentation performance after training. In addition to quantitative analysis including Dice score (DSC) and TMTV comparisons, the qualitative evaluation was also conducted by nuclear medicine physicians. RESULTS: Our cascaded soft-voting guided approach resulted in performance with an average DSC of 0.68 ± 0.12 for the internal test data from developmental dataset, and an average DSC of 0.66 ± 0.18 on the multi-site external data (n = 518), significantly outperforming (p < 0.001) state-of-the-art (SOTA) approaches including nnU-Net and SWIN UNETR. While TTA yielded enhanced performance gains for some of the comparator methods, its impact on our cascaded approach was found to be negligible (DSC: 0.66 ± 0.16). Our approach reliably quantified TMTV, with a correlation of 0.89 with the ground truth (p < 0.001). Furthermore, in terms of visual assessment, concordance between quantitative evaluations and clinician feedback was observed in the majority of cases. The average relative error (ARE) and the absolute error (AE) in TMTV prediction on external multi-centric dataset were ARE = 0.43 ± 0.54 and AE = 157.32 ± 378.12 (mL) for all the external test data (n = 518), and ARE = 0.30 ± 0.22 and AE = 82.05 ± 99.78 (mL) when the 10% outliers (n = 53) were excluded. CONCLUSION: TMTV-Net demonstrates strong performance and generalizability in TMTV segmentation across multi-site external datasets, encompassing various lymphoma subtypes. A negligible reduction of 2% in overall performance during testing on external data highlights robust model generalizability across different centers and cancer types, likely attributable to its training with resampled inputs. Our model is publicly available, allowing easy multi-site evaluation and generalizability analysis on datasets from different institutions.

Integrative genomic analysis identifies unique immune environments associated with immunotherapy response in diffuse large B cell lymphoma.

Most diffuse large B-cell lymphoma (DLBCL) patients treated with bispecific antibodies (BsAb) or chimeric antigen receptor (CAR) T cells fail to achieve durable treatment responses, underscoring the need for a deeper understanding of mechanisms that regulate the immune environment and response to treatment. Here, an integrative, multi-omic approach was employed to characterize DLBCL immune environments, which effectively segregated DLBCLs into four quadrants - termed DLBCL-immune quadrants (IQ) - defined by cell-of-origin and immune-related gene set expression scores. Recurrent genomic alterations were enriched in each IQ, suggesting that lymphoma cell-intrinsic alterations contribute to orchestrating unique DLBCL immune environments. In relapsed/refractory DLBCL patients, DLBCL-IQ assignment correlated significantly with clinical benefit with the CD20 x CD3 BsAb, mosunetuzumab, but not with CD19-directed CAR T cells. DLBCL-IQ provides a new framework to conceptualize the DLBCL immune landscape and uncovers the differential impact of the endogenous immune environment on outcomes to BsAb and CAR T cell treatment.

Chronic Lymphocytic Leukemia IGHV Somatic Hypermutation Detection by Targeted Capture Next-Generation Sequencing.

BACKGROUND: Somatic hypermutation (SHM) status of the immunoglobulin heavy variable (IGHV) gene plays a crucial role in determining the prognosis and treatment of patients with chronic lymphocytic leukemia (CLL). A common approach for determining SHM status is multiplex polymerase chain reaction and Sanger sequencing of the immunoglobin heavy locus; however, this technique is low throughput, is vulnerable to failure, and does not allow multiplexing with other diagnostic assays. METHODS: Here we designed and validated a DNA targeted capture approach to detect immunoglobulin heavy variable somatic hypermutation (IGHV SHM) status as a submodule of a larger next-generation sequencing (NGS) panel that also includes probes for ATM, BIRC3, CHD2, KLHL6, MYD88, NOTCH1, NOTCH2, POT1, SF3B1, TP53, and XPO1. The assay takes as input FASTQ files and outputs a report containing IGHV SHM status and V allele usage following European Research Initiative on CLL guidelines. RESULTS: We validated the approach on 35 CLL patient samples, 34 of which were characterized using Sanger sequencing. The NGS panel identified the IGHV SHM status of 34 of 35 CLL patients. We showed 100% sensitivity and specificity among the 33 CLL samples with both NGS and Sanger sequencing calls. Furthermore, we demonstrated that this panel can be combined with additional targeted capture panels to detect prognostically important CLL single nucleotide variants, insertions/deletions, and copy number variants (TP53 copy number loss). CONCLUSIONS: A targeted capture approach to IGHV SHM detection can be integrated into broader sequencing panels, allowing broad CLL prognostication in a single molecular assay.

Multisite clinical cross-validation and variant interpretation of a next generation sequencing panel for lymphoid cancer prognostication.

AIMS: Genomic sequencing of lymphomas is under-represented in routine clinical testing despite having prognostic and predictive value. Clinical implementation is challenging due to a lack of consensus on reportable targets and a paucity of reference samples. We organised a cross-validation study of a lymphoma-tailored next-generation sequencing panel between two College of American Pathologists (CAP)-accredited clinical laboratories to mitigate these challenges. METHODS: A consensus for the genomic targets was discussed between the two institutes based on recurrence in diffuse large B-cell lymphoma, follicular lymphoma, mantle cell lymphoma, chronic lymphocytic leukaemia and T-cell lymphomas. Using the same genomic targets, each laboratory ordered libraries independently and a cross-validation study was designed to exchange samples (8 cell lines and 22 clinical samples) and their FASTQ files. RESULTS: The sensitivity of the panel when comparing different library preparation and bioinformatic workflows was between 97% and 99% and specificity was 100% when a 5% limit of detection cut-off was applied. To evaluate how the current standards for variant classification of tumours apply to lymphomas, the Association for Molecular Pathology/American Society of Clinical Oncology/CAP and OncoKB classification systems were applied to the panel. The majority of variants were assigned a possibly actionable class or likely pathogenic due to more limited evidence in the literature. CONCLUSIONS: The cross-validation study highlights the benefits of sample and data exchange for clinical validation and provided a framework for reporting the findings in lymphoid malignancies.

CD5 Gene Signature Identifies Diffuse Large B-Cell Lymphomas Sensitive to Bruton's Tyrosine Kinase Inhibition.

PURPOSE: A genetic classifier termed LymphGen accurately identifies diffuse large B-cell lymphoma (DLBCL) subtypes vulnerable to Bruton's tyrosine kinase inhibitors (BTKis), but is challenging to implement in the clinic and fails to capture all DLBCLs that benefit from BTKi-based therapy. Here, we developed a novel CD5 gene expression signature as a biomarker of response to BTKi-based therapy in DLBCL. METHODS: CD5 immunohistochemistry (IHC) was performed on 404 DLBCLs to identify CD5 IHC+ and CD5 IHC- cases, which were subsequently characterized at the molecular level through mutational and transcriptional analyses. A 60-gene CD5 gene expression signature (CD5sig) was constructed using genes differentially expressed between CD5 IHC+ and CD5 IHC- non-germinal center B-cell-like (non-GCB DLBCL) DLBCLs. This CD5sig was applied to external DLBCL data sets, including pretreatment biopsies from patients enrolled in the PHOENIX study (n = 584) to define the extent to which the CD5sig could identify non-GCB DLBCLs that benefited from the addition of ibrutinib to rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP). RESULTS: CD5 expression was observed in 12% of non-GCB DLBCLs. CD5+ DLBCLs displayed transcriptional features of B-cell receptor (BCR) activation and were enriched for BCR-activating mutations known to correlate with BTKi sensitivity. However, most CD5+ DLBCLs lacked canonical BCR-activating mutations or were LymphGen-unclassifiable (LymphGen-Other). The CD5sig recapitulated these findings in multiple independent data sets, indicating its utility in identifying DLBCLs with genetic and nongenetic bases for BCR dependence. Supporting this notion, CD5sig+ DLBCLs derived a selective survival advantage from the addition of ibrutinib to R-CHOP in the PHOENIX study, independent of LymphGen classification. CONCLUSION: CD5sig is a useful biomarker to identify DLBCLs vulnerable to BTKi-based therapies and complements current biomarker approaches by identifying DLBCLs with genetic and nongenetic bases for BTKi sensitivity.

Distinct Hodgkin lymphoma subtypes defined by noninvasive genomic profiling.

The scarcity of malignant Hodgkin and Reed-Sternberg cells hampers tissue-based comprehensive genomic profiling of classic Hodgkin lymphoma (cHL). By contrast, liquid biopsies show promise for molecular profiling of cHL due to relatively high circulating tumour DNA (ctDNA) levels1-4. Here we show that the plasma representation of mutations exceeds the bulk tumour representation in most cases, making cHL particularly amenable to noninvasive profiling. Leveraging single-cell transcriptional profiles of cHL tumours, we demonstrate Hodgkin and Reed-Sternberg ctDNA shedding to be shaped by DNASE1L3, whose increased tumour microenvironment-derived expression drives high ctDNA concentrations. Using this insight, we comprehensively profile 366 patients, revealing two distinct cHL genomic subtypes with characteristic clinical and prognostic correlates, as well as distinct transcriptional and immunological profiles. Furthermore, we identify a novel class of truncating IL4R mutations that are dependent on IL-13 signalling and therapeutically targetable with IL-4Rα-blocking antibodies. Finally, using PhasED-seq5, we demonstrate the clinical value of pretreatment and on-treatment ctDNA levels for longitudinally refining cHL risk prediction and for detection of radiographically occult minimal residual disease. Collectively, these results support the utility of noninvasive strategies for genotyping and dynamic monitoring of cHL, as well as capturing molecularly distinct subtypes with diagnostic, prognostic and therapeutic potential.

Spatially Resolved Tumor Microenvironment Predicts Treatment Outcomes in Relapsed/Refractory Hodgkin Lymphoma.

PURPOSE: About a third of patients with relapsed or refractory classic Hodgkin lymphoma (r/r CHL) succumb to their disease after high-dose chemotherapy followed by autologous stem-cell transplantation (HDC/ASCT). Here, we aimed to describe spatially resolved tumor microenvironment (TME) ecosystems to establish novel biomarkers associated with treatment failure in r/r CHL. PATIENTS AND METHODS: We performed imaging mass cytometry (IMC) on 71 paired primary diagnostic and relapse biopsies using a marker panel specific to CHL biology. For each cell type in the TME, we calculated a spatial score measuring the distance of nearest neighbor cells to the malignant Hodgkin Reed Sternberg cells within the close interaction range. Spatial scores were used as features in prognostic model development for post-ASCT outcomes. RESULTS: Highly multiplexed IMC data revealed shared TME patterns in paired diagnostic and early r/r CHL samples, whereas TME patterns were more divergent in pairs of diagnostic and late relapse samples. Integrated analysis of IMC and single-cell RNA sequencing data identified unique architecture defined by CXCR5+ Hodgkin and Reed Sternberg (HRS) cells and their strong spatial relationship with CXCL13+ macrophages in the TME. We developed a prognostic assay (RHL4S) using four spatially resolved parameters, CXCR5+ HRS cells, PD1+CD4+ T cells, CD68+ tumor-associated macrophages, and CXCR5+ B cells, which effectively separated patients into high-risk versus low-risk groups with significantly different post-ASCT outcomes. The RHL4S assay was validated in an independent r/r CHL cohort using a multicolor immunofluorescence assay. CONCLUSION: We identified the interaction of CXCR5+ HRS cells with ligand-expressing CXCL13+ macrophages as a prominent crosstalk axis in relapsed CHL. Harnessing this TME biology, we developed a novel prognostic model applicable to r/r CHL biopsies, RHL4S, opening new avenues for spatial biomarker development.

ESQmodel: biologically informed evaluation of 2-D cell segmentation quality in multiplexed tissue images.

MOTIVATION: Single cell segmentation is critical in the processing of spatial omics data to accurately perform cell type identification and analyze spatial expression patterns. Segmentation methods often rely on semi-supervised annotation or labeled training data which are highly dependent on user expertise. To ensure the quality of segmentation, current evaluation strategies quantify accuracy by assessing cellular masks or through iterative inspection by pathologists. While these strategies each address either the statistical or biological aspects of segmentation, there lacks a unified approach to evaluating segmentation accuracy. RESULTS: In this article, we present ESQmodel, a Bayesian probabilistic method to evaluate single cell segmentation using expression data. By using the extracted cellular data from segmentation and a prior belief of cellular composition as input, ESQmodel computes per cell entropy to assess segmentation quality by how consistent cellular expression profiles match with cell type expectations. AVAILABILITY AND IMPLEMENTATION: Source code is available on Github at: https://github.com/Roth-Lab/ESQmodel.

The pathobiology of select adolescent young adult lymphomas.

Lymphoid cancers are among the most frequent cancers diagnosed in adolescents and young adults (AYA), ranging from approximately 30%-35% of cancer diagnoses in adolescent patients (age 10-19) to approximately 10% in patients aged 30-39 years. Moreover, the specific distribution of lymphoid cancer types varies by age with substantial shifts in the subtype distributions between pediatric, AYA, adult, and older adult patients. Currently, biology studies specific to AYA lymphomas are rare and therefore insight into age-related pathogenesis is incomplete. This review focuses on the paradigmatic epidemiology and pathogenesis of select lymphomas, occurring in the AYA patient population. With the example of posttransplant lymphoproliferative disorders, nodular lymphocyte-predominant Hodgkin lymphoma, follicular lymphoma (incl. pediatric-type follicular lymphoma), and mediastinal lymphomas (incl. classic Hodgkin lymphoma, primary mediastinal large B cell lymphoma and mediastinal gray zone lymphoma), we here illustrate the current state-of-the-art in lymphoma classification, recent molecular insights including genomics, and translational opportunities. To improve outcome and quality of life, international collaboration in consortia dedicated to AYA lymphoma is needed to overcome challenges related to siloed biospecimens and data collections as well as to develop studies designed specifically for this unique population.

Relapse Timing Is Associated With Distinct Evolutionary Dynamics in Diffuse Large B-Cell Lymphoma.

PURPOSE: Diffuse large B-cell lymphoma (DLBCL) is cured in more than 60% of patients, but outcomes remain poor for patients experiencing disease progression or relapse (refractory or relapsed DLBCL [rrDLBCL]), particularly if these events occur early. Although previous studies examining cohorts of rrDLBCL have identified features that are enriched at relapse, few have directly compared serial biopsies to uncover biological and evolutionary dynamics driving rrDLBCL. Here, we sought to confirm the relationship between relapse timing and outcomes after second-line (immuno)chemotherapy and determine the evolutionary dynamics that underpin that relationship. PATIENTS AND METHODS: Outcomes were examined in a population-based cohort of 221 patients with DLBCL who experienced progression/relapse after frontline treatment and were treated with second-line (immuno)chemotherapy with an intention-to-treat with autologous stem-cell transplantation (ASCT). Serial DLBCL biopsies from a partially overlapping cohort of 129 patients underwent molecular characterization, including whole-genome or whole-exome sequencing in 73 patients. RESULTS: Outcomes to second-line therapy and ASCT are superior for late relapse (>2 years postdiagnosis) versus primary refractory (<9 months) or early relapse (9-24 months). Diagnostic and relapse biopsies were mostly concordant for cell-of-origin classification and genetics-based subgroup. Despite this concordance, the number of mutations exclusive to each biopsy increased with time since diagnosis, and late relapses shared few mutations with their diagnostic counterpart, demonstrating a branching evolution pattern. In patients with highly divergent tumors, many of the same genes acquired new mutations independently in each tumor, suggesting that the earliest mutations in a shared precursor cell constrain tumor evolution toward the same genetics-based subgroups at both diagnosis and relapse. CONCLUSION: These results suggest that late relapses commonly represent genetically distinct and chemotherapy-naïve disease and have implications for optimal patient management.

Genetic subdivisions of follicular lymphoma defined by distinct coding and noncoding mutation patterns.

Follicular lymphoma (FL) accounts for ∼20% of all new lymphoma cases. Increases in cytological grade are a feature of the clinical progression of this malignancy, and eventual histologic transformation (HT) to the aggressive diffuse large B-cell lymphoma (DLBCL) occurs in up to 15% of patients. Clinical or genetic features to predict the risk and timing of HT have not been described comprehensively. In this study, we analyzed whole-genome sequencing data from 423 patients to compare the protein coding and noncoding mutation landscapes of untransformed FL, transformed FL, and de novo DLBCL. This revealed 2 genetically distinct subgroups of FL, which we have named DLBCL-like (dFL) and constrained FL (cFL). Each subgroup has distinguishing mutational patterns, aberrant somatic hypermutation rates, and biological and clinical characteristics. We implemented a machine learning-derived classification approach to stratify patients with FL into cFL and dFL subgroups based on their genomic features. Using separate validation cohorts, we demonstrate that cFL status, whether assigned with this full classifier or a single-gene approximation, is associated with a reduced rate of HT. This implies distinct biological features of cFL that constrain its evolution, and we highlight the potential for this classification to predict HT from genetic features present at diagnosis.