Distinct immune and genomic signatures predict resistance to ibrutinib therapy in Waldenström macroglobulinemia

Distinct immune and genomic signatures predict resistance to ibrutinib therapy in Waldenström macroglobulinemia | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Distinct immune and genomic signatures predict resistance to ibrutinib therapy in Waldenström macroglobulinemia Tina Bagratuni, Christos Vlachos, Maria Sakkou, Ioannis Kollias, and 15 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8594385/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Despite recent advances in treatment strategies of Waldenström macroglobulinemia (WM) patients, the disease remains incurable. While several genomic features predict poor outcomes with ibrutinib, the contribution of immune dysfunction to tumor behavior remains unclear. Single-cell RNA sequencing on longitudinal bone marrows from 37 Waldenström macroglobulinemia patients before and after ibrutinib therapy profiled 348,000 cells and identified distinct immune phenotypes associated with ibrutinib progression. This was primarily characterized by the accumulation of T-effector cells, Tregs and pro-inflammatory M1-like monocytes, alongside a depletion of naïve T-cells. The tumor architecture in progressing patients, exhibited a unique transcriptomic signature, also validated in an external cohort of 47 WM patients, driven by genes including LTB, NFKBIA, DUSP2 , NR4A1 leading to shorter progression-free survival. Mutational profiling using whole-genome sequencing identified mutational signature SBS1/SBS5 being significantly associated with poorer outcome. Finally, we demonstrate that integrating tumor with immune cell compartments can significantly improve ibrutinib response prediction scoring. Biological sciences/Cancer/Haematological cancer/Myeloma Biological sciences/Cancer/Tumour heterogeneity Waldenström macroglobulinemia ibrutinib single cell transcriptome bone marrow microenvironment Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Introduction Waldenström’s macroglobulinemia (WM) is a rare B-cell lymphoproliferative disorder characterized by the abnormal growth and clonal expansion of small mature B lymphocytes and plasma cells in the bone marrow and lymphoid tissues, 1,2 which secrete IgM monoclonal immunoglobulin. About a decade ago, the BTK inhibitor, ibrutinib, became the first approved agent for the treatment of symptomatic WM patients 3 , and despite the unpreceded high response rates in terms of IgM reduction and clinical improvement, about one third of treated patients fail to respond 4 , 5 , 6 or show delayed responses with limited consensus regarding the substantial heterogeneity in the response type, mechanisms of resistance, and causes of treatment failure 7 – 14 . One of the well-known predictive factors of ibrutinib treatment outcome, is the presence of MYD88 and CXCR4 mutations 15 – 17 , with patients with MYD88 WT genotype have very low probability of major response, while patients harboring both MYD88 L265P and CXCR4 mutations exhibit inferior response rates 3 , 4 , 18 . Additionally, almost half of WM patients who progress on ibrutinib, display BTK mutations at the binding site of ibrutinib or its downstream mediator PLCg2 5,6 while differential gene expression of several genes as well as acquired mutations in chromosomes 6q and 8p has also been linked to ibrutinib resistance 19 , 20 . These observations suggest that ibrutinib’s therapeutic effects may extend beyond direct tumor burden reduction, highlighting the need to elucidate the broader genomic and immune mechanisms. In this study, we investigate the complex interplay between tumour-acquired genetic features and the composition of the microenvironment in the promotion of ibrutinib resistance in WM patients by integrating single-cell RNA sequencing (scRNAseq) and whole-genome sequencing (WGS), to examine bone marrow tumour and immune compartments from 37 symptomatic WM patients treated with ibrutinib, alongside with a validation cohort of 47 WM patients (Fig. 1 ). Using this approach, we were able to define biological networks behind the variable clinical responses seen in patients treated with ibrutinib and the impact of BTK inhibition on the crosstalk between WM clones and the surrounding microenvironment. Results Study population The study included 37 consecutive consenting patients with previously untreated WM who received ibrutinib therapy, as per the standard clinical practice. After a median follow-up of 1.7 years (range 0.58–4.08 years), 7 (19%) patients achieved very good partial response (VGPR); 17 (45%) achieved partial response (PR), 4 (11%) patients achieved minimum response (MR) whereas 9 (25%) patients had stable or progressive disease (SD and PD) after 6 months of ibrutinib therapy. Patients achieving PR or better after 6 months of therapy were defined as the responder group (RG) while patients with less than PR were defined as the non-responder group (NRG), so that 24 patients were included in the RG and 13 patients in the NRG. In the NRG cohort, 4 out of 13 patients (30%) ultimately achieved a PR to ibrutinib with a median time to response of 3.1 years (range 1.5–5.2 years) while the remaining 9 patients did not improve their response. The key clinical and genetic features of the RG and the NRG are summarized in Suppl. Tables 1 and 2 and Suppl. Figure 1. The landscape of bone marrow mononuclear cells in WM at single cell resolution We performed scRNAseq on 37 paired bone marrow samples at diagnosis (T1) and at 6 months (T2) after start of ibrutinib treatment (74 total). In addition, scRNAseq analysis was performed on samples from 5 patients at 12 months post start of ibrutinib (T3) and 1 patient at relapse 24-month post therapy (T4) as well as on two healthy donors. Upon preprocessing, alignment and quality control steps including removal of the low-quality reads and cells, a total of 348,000 high-quality bone marrow mononuclear cells (BMMCs) were analyzed. We also integrated publicly available normal bone marrow single cell data set, annotated as reference (~ 300,000 cells), to enhance our analysis for comparison among healthy individuals and patients. We generated a BM cells atlas, and by using Uniform Manifold Approximation and Projection (UMAP) approach, we were able to resolve 13 distinct cell clusters which consisted 32% of T-cells, 26% of B-cells, 12% of Monocytes ,6% of NK-cells and 1% of plasma cells (Fig. 2 A, B, Suppl Fig. 2A). Ibrutinib treatment in the entire patient cohort led to a significant reduction in the B cell ( p < 0.01) and plasma cell ( p = 0.01) populations, while classical CD14 + monocytes ( p < 0.01), NK/T cells ( p < 0.01) and classical dendritic cells ( p < 0.01) were significantly increased (Fig. 2 C). This finding demonstrates that, beyond its direct effect on the malignant cells, the treatment also exerts profound effects on the immune compartment. Focusing further on the two response groups, we observed that at T1, the B-cell compartment was markedly enriched in the NRG, whereas progenitor and T cells were more enriched in the RG. By T2, B-cells significantly decreased in both groups ( p < 0.01), while monocytes were significantly increased in the RG only ( p < 0.01, Fig. 2 D and E, Suppl Fig. 2B). Furthermore, although the number of clonal plasma cells at diagnosis was similar between both groups, a significant reduction at T2 was observed only in the RG ( p = 0.04, Suppl Fig. 2B). The contribution of immune subpopulations to the ibrutinib resistance Focusing on the myeloid compartment, we identified four myeloid subpopulations including CD14⁺ classical and CD16⁺ non-classical monocytes, conventional dendritic cells (cDCs, CD1C⁺), and a smaller fraction of monocyte activated platelets (MAP) cells (PPBP) (Fig. 3 A) which were further subdivided into seven transcriptionally distinct clusters (C0–C6, Fig. 3 B). Classical CD14 + monocytes were characterized by robust expression of M1-like pro-inflammatory mediators ( S100A8 , IL1B , CXCL2 ) and signaling molecules ( IRAK2 ) and exhibited enriched interferon (IFN)-stimulated gene ( ISG15 ) expression, indicative of the induction of a highly inflammatory transcriptional program (Fig. 3 D). Assessing the relative abundance of monocytic states at T1 in both groups compared to the reference, we observed that CD16 + /FCGR3A (C2) and the inflammatory IFN activated monocyte state (C4), were massively increased in all WM patients (Suppl. Figure 3A). Although these populations were homogenously present within the monocyte compartment at T1 in both response groups, at T2 a significant induction of the CD14 + monocytes was observed in both groups (p < 0.01) accompanied with a decrease of the CD16 + subpopulation (p < 0.01, Fig. 3 C). An increase was observed at T2 in cluster C0 ( IRAK2 and CXCL2 ) in the NRG group compared to RG ( p = 0.16 vs p = 0.49, log2Fc = 0.8 vs 0.27,Fig. 3 E) which was further evidenced by the increased inflammatory pathway activity score, produced by the expression of 29 genes including CXCL8 , IL6 , IL4 , TGFB2 observed in the NRG (Suppl. Figure 3B). Conversely, significant expansion of IFN + monocytes (C4) was observed at T2 only in the RG (p < 0.001), potentially indicating an enhanced activation of the innate immune response that may be linked to ibrutinib efficacy (Fig. 3 F). Additionally, our analysis identified a neutrophil population with similar expression patterns between RG and NRG at T1 (Suppl. Figure 3C) however, at T2, a cluster characterized by genes such as MMP8 and MMP9 was more enriched in the RG group ( p = 0.3, Fig. 3 G, Suppl. Figure 3D). Overall, gene set enrichment analysis (GSEA) revealed that at T2 pathways including TNF and NF-κb signaling, inflammation and IFNγ response were enriched driven by the upregulation of genes including CCL3 , CCL4 , CXCL8 , NFKBIZ and NFKBIA (Fig. 3 H). Focusing on the T cell compartment, we resolved 9 distinct subpopulations including both naive and memory subsets of CD4 + and CD8 + , regulatory T (Tregs), MAIT, and gamma-delta (γδ) T-cells (Fig. 4 A). GZMB CD8 + , memory CD4 + T, and Tregs were markedly increased in patients compared to the reference (Suppl. Figure 4A). At T1, an enrichment of GZMB⁺ CD8⁺ cells ( p = 0.07) was seen in the NRG combined with a significantly reduced expression of naive CD4⁺ T-cell frequencies ( p < 0.01) compared to RG. Trajectory analysis also revealed an accumulation of GZMK⁺, GZMB⁺ effector, and γδ cells in NRG in the terminal region of the pseudotime trajectory, suggesting a shift toward terminal effector states (Fig. 4 B). At T2, we observed an expansion of GZMK⁺ CD8⁺ cells in both groups, with more pronounced effects in the NRG ( p < 0.01), highlighting the limited cytotoxicity observed in this group. Concurrently, CD4 + naive cells were significantly reduced only in the RG possibly indicating that ibrutinib treatment leads to a global shift in the T cell compartment towards a more effector-like and functionally engaged immune response (Fig. 4 C). Furthermore, as opposed to NRG, the RG exhibited a marked suppression of Tregs after therapy ( p = 0.02), possibly reflecting the more favorable outcome observed in this group leading to higher cytotoxic T-cell responses. AUCell scoring based on a curated set of T-cell exhaustion-associated genes including CTLA4 , PDCD1 , LAG3 , TIGIT , PD1 , HAVCR2 demonstrated that although at T1, NRG displayed a slightly lower score compared to RG, after treatment a significant reduction of this signature was observed in the RG only ( p = 0.01), suggesting that the NRG maintained the terminally differentiated and dysfunctional T-cell profile even after treatment (Suppl. Figure 4B).Functional analysis of the IFNγ response revealed that activity score of this pathway was markedly enriched in NRG, further highlighting the higher inflammatory state of these patients (Fig. 4 D). Finally, sub-clustering of the NK-cell compartment resolved five clusters; a progenitor-like CD56 bright SELL + associated with lower toxic functions, an activated FOS + CD56 bright , a CD160 memory-like regulatory CD56 bright , a transitional/ early cytotoxic CD56 dim FGFBP2 + and an NKT-like (Fig. 4 E). At T1, almost all subpopulations were evenly distributed among the two response groups with a minimal enrichment of the SELL + cluster in the NRG and CD160 + in the RG. Ibrutinib treatment resulted in a marked enrichment of regulatory CD56 bright subsets, particularly CD160 + ( p = 0.02 in NRG and p = 0.06 in RG) and FOS⁺ NKs ( p = 0.14 in NRG), suggesting a shift toward a less cytotoxic, less activated NK-cell phenotype especially in the NRG (Fig. 4 F). Furthermore, functional analysis showed a marked suppression of the NK cytotoxicity pathway during treatment, observed mainly in the NRG, indicating that successful response to ibrutinib is potentially linked to continued NK cell cytotoxic activity (p = 0.05, Fig. 4 G). Furthermore, the ratio of CD56 bright /CD56 dim NK cells increased between T1 and T2 in both groups, with more prominent effects in NRG (0.025% (0–0.4) versus 0.31% (0–2), p = 0.001, Fig. 4 H). The transcriptomic features of clonal B compartment and the impact of distinct B-cell molecular subtypes on ibrutinib response Focusing on the transcriptomic features of clonal B-cells (cBc) and their response to ibrutinib, we initially observed that most of the polyclonal cells of the patient samples clustered together while cBc created individual clusters each of which originated from a single patient. Furthermore, while MYD88 MUT patients tended to cluster more closely together, neither CXCR4 mutation status nor response to ibrutinib revealed consistent transcriptional patterns across different patient groups (Fig. 5 A, Suppl. Figure 5A). As expected, memory B cells constituted the majority of the cBc in both groups while a small proportion of pro-B cells was observed only in the RG (Suppl. 5B). Correlation of patients BM infiltration undergoing ibrutinib treatment with cBc, we observed that at both T1 and T2, the frequency of cBc was significantly higher in the NRG ( p = 0.02, Suppl. Figure 5C) while ibrutinib led to a significant reduction of this compartment in both groups ( p < 0.01, Fig. 5 B). The presence of MYD88 and CXCR4 mutations had an impact on the response to ibrutinib, with a significant reduction of cBc observed in MYD88 MUT and CXCR4 WT patients (Suppl. Figure 5D). Importantly, a significant upregulation of CXCR4 expression was observed in the NRG at T2 ( p = 0.01) and a significant reduction in MYD88 expression in the RG ( p = 0.01, Fig. 5 B). Differential gene expression (DEG) analysis among response groups, showed increased expression of MAST4 , XIST , PTK2 , BLK and MAP4K4 genes at T1 and higher expression levels of genes such as PRDM4 , PCDH9 , IRF4 , and MAPK8 at T2 (Suppl. 5E-F, Fig. 5 C). Interestingly, a subset of NRG patients who eventually responded to ibrutinib after 12 months of treatment (4/13 patients, NRtoR) exhibited a transcriptomic profile at T2 (key genes XBP1 , CD9 , MZB1 , NFKBIA , CXCR4 and JCHAIN) that highly overlapped genes observed in pretreated RG patients while patients with sustained resistance (NR to NR) showed no such convergence, suggesting that delayed responders may acquire RG-like transcriptional features over time. (Fig. 5 D). Functional analysis indicated that IL6 activity at T2 was significantly reduced in the RG only ( p = 0.03) possibly suggesting that this activity remains prolonged in the NRG (Suppl. Suppl. 5G). Furthermore, CNV profiling revealed chromosomal amplifications of chromosomes 12 and 18 observed exclusively in the NRG ( p = 0.03 for chr12, p = 0.11 for chr18), while deletion of chromosome 6q was equally distributed in both groups ( p = 0.7) (Fig. 5 E). We next sought to determine whether our cohort of patients exhibited distinct B cell subtypes with memory B-cell (MBC) like or plasma-cell (PC) like features 21 – 23 . Within our cohort, 12 out of 24 patients in the RG (50%) and 8 out of 13 in the NRG (61%) exhibited MBC-like features, while the remaining patients in both groups were classified as PC-like (Fig. 5 F, G, Suppl. Figure 6A); however no significant differences in subtype distribution among the response groups. Interestingly our results showed that all MYD88 WT patients (n = 6) displayed the MBC-like subtype which were also classified as NRG. At diagnosis, the NRG displayed a significantly higher number of cBc in this MBC subtype compared to the RG ( p = 0.02) while PC subtype was similar between both groups. At T2, both groups displayed a marked reduction in the MBC-like signature ( p = 0.06 and p = 0.01, respectively) although the effects were more pronounced in the NRG, possibly indicating that transcriptional changes in the MBC-like signature act independently of ibrutinib response (Fig. 5 H). Conversely, the PC-like signature was notably reduced in the RG ( p = 0.11) while remained unchanged in the NRG. Overall, no effects in PFS were observed between the groups ( p = 0.77, Fig. 5 I). Identification of gene expression signatures associated with ibrutinib resistance To further address the significant heterogeneity observed between the two response groups, we applied Bayesian non-negative matrix factorization on the cBc, utilizing 2,000 highly variable genes based on the MYD88 and CXCR4 mutation status. Four gene expression signatures (GEX1-GEX4) were identified with key marker genes including LTB and MARCKS in GEX1, DUSP2 and TNFAIP3 in GEX2, CD9 and CHST15 in GEX3, and MS4A1 and NOTCH2 in GEX4, each of which affected different signaling pathways (Fig. 6 A, Suppl. Figure 7A-B). GEX3 ( MYD88 MUT , CXCR4 WT , n = 14) was predominantly found in the RG, GEX4 ( MYD88 WT , CXCR4 WT , n = 6) was primarily found in the NRG while GEX2 ( MYD88 WT , CXCR4 MUT ) belonged to the NRG (one patient). Focusing on the GEX1 ( MYD88 MUT , CXCR4 MUT , n = 14), we noticed that this was found in 6 NRG and 8 RG patients. At diagnosis, genes within the GEX1 signature overexpressed in RG included EZR and DUSP1 (MAPK pathway), SSR4 and MZB1 (ER processing), HIFX (hypoxia response) and MSI2 (WNT signaling) while genes overexpressed in NRG included AFF3 and BACH2 (BCR signaling), HLA-DQB1 and HLA-DRB5 (antigen presenting) and LTB (NF-κB/TNF signaling). Upon treatment, genes such DUSP2 ( MAPK pathway), RHOH (T cell receptor signaling), NFKBIA ( NF-κB signaling), NR4A1 (immune regulation) and JCHAIN were enriched in NRG (Fig. 6 B). These findings suggest that in non-responding patients with CXCR4 mutations, resistance to ibrutinib is influenced by genes that primarily affect BCR signaling and immune regulation. To further validate the above signatures and evaluate their clinical significance, we analyzed bulk RNA-seq data from an external cohort of 47 patients 24 before the initiation of ibrutinib, from the Dana Farber Cancer Institute (DFCI, Suppl. Table 3). Within the external cohort, 12 patients had disease progression to ibrutinib (25%). In line with our results from the initial cohort, 83% of patients (20/24) with MYD88 MUT / CXCR4 WT genotype were associated with the GEX3 signature while 82% of patients (19/23) with MYD88 MUT / CXCR4 MUT were associated with the GEX1 signature. Interestingly, about 50% of patients (12/23) within the GEX1 signature also shared features of the GEX2 signature and 42% (5/12) of the progressed patients displayed the combined GEX1/GEX2 signature (Suppl. Figure 7C). The remaining 58% of the progressed patients (7/12) displayed features of the GEX3 signature meaning that almost 25% of patients with CXCR4 WT genotype progressed to ibrutinib therapy. To gain further insights into the transcriptional evolution and resistance mechanisms of residual cBc, we conducted unsupervised clustering of all B cells from each patient individually in the initial cohort. Within the NRG, we observe an increase of cBc after treatment in the RG. Comparing the dynamic shifts of the cBc at T1 and T2 we found that the dominant clone at T1 remained the most prevalent after treatment in all patients, regardless of response group. In 2 of the 37 patients, we identified the presence of two separate clones: one with different IgHV rearrangements (clonotype 1 and 2) classified as RG and another with independent kappa and lambda light chain clones, classified as NRG. Focusing on the patient with the 2 clonotypes, we observed two patterns of transcriptional signatures: one resembling the GEX3 signature (major clone, clonotype 1) and one to the GEX2-like (minor clone, clonotype 2). Interestingly, at T2, clonotype 1 decreased by nearly 25%, while clonotype 2-GEX2-like clone showed a simultaneous increase, possibly indicating the gradual expansion of a resistant clone gradually increasing over time. This clone was mainly characterized by the abundant expression of LTB , NFKBIA , DUSP2, NR4A1 and NR4A2 genes (Fig. 6 C). In fact, the LTB and NFKBIA genes appear to be highly expressed in patients whose cBc increase after treatment, suggesting their potential involvement in mechanisms of ibrutinib resistance. Finally, to determine whether transcriptomic features at diagnosis could predict long-term outcomes, we conducted a PFS analysis. The combination of high GEX1/GEX2 expression score resulted in shorter PFS ( p = 0.066) while significant shorter PFS was observed in patients with low GEX3 ( p = 0.0035) and high GEX4 expression score ( p = 0.0046, Fig. 6 D). The results from the external cohort validated the impact of the different expression patterns in PFS, with the combination of high GEX1/ GEX2 expression score resulting in a shorter PFS ( p = 0.018) while low GEX3 had an almost significant impact on PFS ( p = 0.05, Suppl. Figure 7D). Specifically, high expression of LTB was significantly associated with shorter PFS ( p = 0.043). Furthermore, we developed a classifier using an Elastic Net model for each response group, where genes such as JUN , PRDM5 , HS2ST1 , BCL7A , LTB and others had the highest importance scores for the model (Suppl. Table 4) with significant impact in PFS ( p < 0.0001, log-rank test, Fig. 6 E) which was also observed in the validation cohort (PFS, p = 0.038, Fig. 6 F). The interactome landscape underlying ibrutinib responsiveness and the role of immune microenvironment in predicting outcome To investigate how the TME influences the functional behavior of cBc in patients with varying responses to ibrutinib, we analyzed cell-cell interactions between the cBC population and the TME. Our findings revealed that cBc from all patients at diagnosis were able to send signals to specific cell compartments including naïve, memory and cytotoxic T cells, NK cells, MAITs and pDCs. Of particular interest was the CD74:CD44 (macrophage inhibitor factor, MIF) cell interaction involved in many processes (ERK1/ERK2, TLR4, p53 apoptosis) 25 – 28 showing higher cell interaction in the RG at diagnosis compared to the NRG (Fig. 7 A, Suppl Fig. 8); however, upon treatment this interaction was reduced in RG and upregulated in the NRG, possibly suggesting a higher dependency of this interaction in the RG which is disrupted upon ibrutinib therapy resulting in a better restoration of immune balance in this group of patients. We also observe a broad spectrum of HLA-related genes:CD44 ligand-receptor interaction between pDCs and cBc at T1 in both groups, however at T2 this interaction is disrupted and enhanced between other cell types including classical and non-classical monocytes, naïve and memory T cells. These data support the presence B cell proliferative elements driven by different cellular interactions at diagnosis, and their disruption upon treatment with marked effects in the RG. Finally, we assessed the ability to predict ibrutinib response by integrating cBc and immune cell compartments in a multivariable framework by employing a receiver operative characteristic (ROC) analysis providing an AUC score which stratifies patients in high- and low-risk groups. Incorporation of cBc alone yielded an AUC value of 0.72 and was associated with shorter PFS ( p = 0.078, Fig. 7 B-C). Lower T-cell proportion also correlated with significantly shorter PFS ( p = 0.022, Fig. 7 D). Importantly, incorporating the most predictive T-cell subsets (naïve CD4 and CD8 cells and GZMB) enhanced prediction accuracy (AUC = 0.87) where high B cell levels combined with low T-cell proportions were associated with shorter PFS ( p = 0.022, Fig. 7 D). On the other hand, patients with low levels of cBc and high levels of T-cells had the best outcome. Finally, the incorporation of specific monocyte compartments increased the AUC score to 0.944 (Fig. 7 B). This data further highlights the importance of the immune microenvironment in partially determining the depth of ibrutinib response and the significant advantage of these integrative models. Impact of the mutational profile of the B cell clone in ibrutinib resistance To determine the mutational profile of two response groups, WGS was performed in 10 RG and 8 NRG patients at both timepoints. This analysis revealed a similar mutational burden in the NRG and RG which was not affected by ibrutinib therapy (Fig. 8 A,B). Focusing on genes recurrently mutated in WM and other lymphomas, 29–31 we found that the most frequently mutated genes in RG included MYD88 (90%), CXCR4 (20%), CAST (25%, ERK1/2 signalling), TP53 (10%) and CD79B (10%) while in the NRG included MYD88 (62%), CXCR4 (38%), IGLL5 (38%), KMT2D (38%) (Fig. 8 C, Suppl. Figure 9A ). In line with the scRNA-seq results, CNV profiling revealed similar differences between the RG and NRG (Fig. 8 D). Clonal evolution analysis showed a branching evolution in 3 of 17 patients (17%), with 2 cases in the NRG and 1 in the RG who ultimately progressed after 12 months of therapy (RtoNR) (Fig. 8 E, Suppl. Figure 9B). Most of the acquired alterations at T2 included genes such as IL10R , RUNX1 , FAT3 , IRF4 , RBM6 , RHOT1 some of which have been previously identified by our group 32 . Finally, consistent with our prior findings 31 , 32 , mutational signature analysis identified four main single base substitution (SBS) signatures: clock-like aging signatures (SBS1 and SBS5), germinal-centre-associated polymerase eta (POLH) somatic hypermutation (SBS9), and SBS8 (Fig. 8 F). However in contrast to previous studies 31 , 32 , we also found the presence of the APOBEC mutational activity (SBS13) in about 65% of patients (11/16) BUT with a very low contribution (3%), which also appeared to be more enriched in males compared to females ( p = 0.16). Furthermore, significantly higher levels of SBS1 and SBS5 signatures were observed in the NRG group at both time points compared to the RG group ( p = 0.02),which also had an impact on the PFS ( p = 0.00059), a pattern which has also been observed in a subset of high-risk multiple myeloma patients 33 . Discussion Despite the marked clinical benefit of ibrutinib therapy in WM, the underlying mechanisms that contribute to the heterogeneity of patient responses remain poorly understood. In this study, we performed an in-depth analysis of both the tumor and immune microenvironment transcriptome in WM patients before and after ibrutinib treatment. Our findings reveal a distinct tumor-intrinsic transcriptional signature associated with treatment response, which is also linked to changes in the TME, suggesting that this interplay may significantly influence therapeutic efficacy and patient outcomes (Fig. 9). We demonstrated highly enriched terminally differentiated GZMB CD8 + T-cells and SELL-expressing NK cells in the NRG, indicating an expansion of these subpopulations with increasing disease severity possibly affecting the interaction of WM cells with the immune system and the optimum efficacy of ibrutinib therapy. In fact, previous studies have shown that WM patients with high tumor infiltration exhibit higher levels of exhausted CD8 + T-cells compared to healthy donors 34 . In the myeloid compartment, we observed the enrichment of the IFN + monocytes in the NRG, possibly promoting a tumor pro-survival microenvironment with a BTK-independent ecosystem which could contribute to WM disease progression 35 . On the other hand, responders display a more functionally competent T-cell repertoire through the expression of naïve CD4 + and CD8 + T-cells. Therefore the overall combined phenotype of expanded T-cell effector population expressing low levels of cytotoxicity-related genes along with the depletion of naïve T cell states which has been shown to reduce the TCR clonality 36 , is possibly associated with poorer outcomes and this has been described in several other malignancies 37 – 39 . Furthermore, these effects, possibly contribute to a higher inflammatory microenvironment through the enrichment of IFN + monocytes which could play a role in the treatment failure in non-responding patients. Following treatment the significant reduction of the Tregs only observed in the responding patients, possibly suggests that the ibrutinib’s efficacy is potentially linked to the restoration of T-cell fitness 46 . Conversely, while both response groups exhibited elevated T-cell exhaustion scores compared to the reference before the start of therapy, the responder’s group showed a greater reduction upon treatment, potentially reflecting a more robust recovery of T-cell function compared to non-responders. Furthermore, a more pronounced expansion of the pro-inflammatory M1-like monocytic population was observed in NRG after treatment, suggesting a shift towards a more pro-inflammatory state and an IRAK-NFκB driven signaling that could sustain tumor support BTK-independent manner. Recent studies have focused on inflammatory WM (iWM) which is present in one-third of patients 42 – 44 and notably in the context of BTK inhibitor therapy, the inflammatory syndrome decreased during the hematological response in BTKi-treated iWM 45 . When characterizing tumor-intrinsic mechanisms among the two response groups we observed significant intra-tumor heterogeneity, which partly explains the differential responses seen with ibrutinib therapy. We initially demonstrated that resistance to ibrutinib by the disease persistence, was associated with a higher cBc burden at diagnosis in the non-responders primarily driven by elevated expression of genes regulating BCR signaling ( IKZF3 46,47 , SYK 48 , 49 ), chromatin modification ( XIST 50 , HDAC9 51 ), and antigen presentation ( HLA-DQA1 52 ). Ibrutinib treatment led to distinct transcriptional evolution between the two response groups, with only the RG showing enrichment in apoptotic and negative regulation of the p38/MAPK pathways, while NRG retained a transcriptional program resembling the pre-treatment RG states. Classification of patients into the previously described 21 – 23 MBC-like and PC-like molecular subtypes demonstrated that, although all patients could be assigned to one of these subtypes, response to ibrutinib was independent of the subtype classification. Overall, ibrutinib treatment resulted in a markedly greater reduction of cBc in the MBC-like group in both groups while reduction of cBc in the PC-like subtype was only seen in the RG, possibly suggesting that a more resistant clone could be linked with PC-like features. To further uncover the tumor-intrinsic mechanisms of ibrutinib resistance, we defined gene expression signatures of cBc classified by MYD88 and CXCR4 mutation status. GEX1 ( MYD88 MUT , CXCR4 MUT ) was similarly distributed among the RG and the NRG patients, where genes overexpressed in RG were linked to JNK/MAPK, ER processing, and WNT signaling, while in NRG were involved in BCR, antigen presentation, and TNF/NF-κB signaling. Importantly, most of these findings were also validated with the use of bulk RNA-seq data from an external cohort of 47 patients, which highlighted the significance of genes mainly involved in GEX1-GEX3 expression signatures and their potential prognostic value in predicting ibrutinib response. Furthermore, clonal transcriptional evolution in each patient highlighted resistance mechanisms closely linked to the GEX2 signature, notably through LTB, NFKBIA, DUSP2, NR4A1 , and NR4A2 genes mainly involved in TNFα/NFκB signaling pathways. Most of these genes have been previously associated with other lymphomas and multiple myeloma (MM) 53 – 58 . To further understand how the transcriptome functions in the context of the clinical outcome, we identified several genes, including LTB , that had a significant impact in PFS. High GEX1/GEX2 scores were associated with shorter PFS, while low GEX3 significantly correlated with shorter PFS both in the training and the validation cohort. Additionally, the Elastic Net classifier identified key genes like IRF4 , NR4A1 , MYC , BCL7A , and LTB that contributed most to ibrutinib resistance prediction with significant impact on PFS which was also further validated in the external cohort of patients confirming their essential effect on PFS. Further mutational analysis showed that NRG displayed more mutations in genes such as CXCR4 , KMT2D , IGLL5 ,and ARID1A , also previously described by our group 32 . Additionally, we observed for the first time the presence of the APOBEC signature in WM patients (although with a relatively low contribution) while the age-related mutational signatures SBS1/SBS5 were significantly enriched in the NRG group, which may reflect an underlying mechanism of resistance consistent with the patients’ clinical profile, a finding also described in several cancers 33 , 59 . Cell- cell communication identified MIF (CD74:CD44) as the major regulator among B- and T-cells in RG and HLA-related genes among B-cells and T/NK-cells in NRG. MIF transcription is regulated by the BCR activity and activates pathways including ERK1/2, MAPK and NFκB however after successful BTK inhibition the normal BCR-driven regulatory network that controls MIF production is disrupted. Hence, although MIF is disrupted in RG after therapy, is further induced in NRG possibly indicating that the cells in this group of patients are activated by alternative pathways. The induction of MIF in NRG after therapy could also suggest that ERK1/2, MAPK and NFκB are reactivated which will eventually lead to successful BTK inhibition. Finally, we observed that by combining TME immune signatures with the proportion of clonal B-cells could better predict the ibrutinib efficacy outcomes, providing an accurate prediction of responders and non-responders. Overall, this study offers important insights into the mechanisms driving resistance to BTKi based therapies in WM, such as ibrutinib, and highlights the value of integrated multi-omic strategies in guiding optimal treatment strategies informed by the underlying molecular profiles. Methods Patients and sample processing Bone marrow aspirates were collected from 37 newly diagnosed symptomatic WM patients at 3 time points, (i) diagnosis, (ii) 6 months and (iii) 12 months upon ibrutinib therapy initiation. Two healthy donors were also included. Prior to any study procedure, all participants of this study provided written informed consent. The research was approved by the Institutional Review Board (IRB) and Scientific committee of “Alexandra” Hospital (nos. 56463 and 27). All research was conducted in accordance with the Declaration of Helsinki. For the single cell analysis, frozen bone marrow mononuclear cells (BMMCs) were thawed in phosphate buffered saline (PBS) with 0.01% bovine serum albumin (BSA). For whole genome sequencing (WGS), lymphoplasmacytic cells from BM aspirates were collected by CD19 + Auto-MACS micro-bead selection (Miltenyi-Biotech) as previously described 32 , 60 . 10x single cell RNA sequencing Single-cell 5’ RNA and V(D)J) libraries were generated using the Chromium Next GEM Single Cell 5' Reagent Kits v2 (Dual Index) (10X Genomics) following the manufacturer’s protocol. Briefly, BMMCs were loaded onto the Chromium Next GEM Chip K followed by gel beads in emulsion (GEM) generation on the Chromium X. GEM-RT reactions were performed followed by breakage of the GEMs and cDNA cleanup using Dynabeads MyOne Silane beads. For the 5’ RNA, cDNA amplification was performed with 13 amplification cycles, followed by reaction cleanup using SPRIselect Reagent Kit (Beckman Coulter). Total cDNA quantification was then performed using the Qubit 1X dsDNA HS Assay Kit (Thermo Fisher Scientific). Subsequently, enzymatic fragmentation, library construction, and additional SPRIselect bead clean-ups were performed to generate the final indexed libraries. Sample index PCR cycles were determined based on the total cDNA yield and ranged between 14 and 16 cycles. Library quality was assessed using the Agilent High Sensitivity DNA kit and libraries were quantified using the Qubit 1X dsDNA HS Assay kit (Thermo Fisher Scientific). V(D)Js amplification from the total cDNA was performed in two PCR cycles using 2 sequential, B cell oligos mixes. V(D)J libraries were constructed, further by enzymatic fragmentation of the V(D)J enriched cDNA and library indexing. Similar to the RNAseq libraries quantitation and quality were assessed with both Qubit 1X dsDNA HS Assay kit (Thermo Fisher Scientific) and Agilent High Sensitivity DNA kit. Some samples as indicated were analysed with the Chromium Next GEM Single Cell 3ʹ Reagent Kits v3.1 (Dual Index) (CG000315 Rev F). The libraries were generated following the manufacturer’s protocol. Single cell Bone marrow suspensions were loaded onto the Chromium Next GEM Chip G followed by gel beads in emulsion generation on the Chromium X. GEM-RT reactions were performed and 3’ RNA cDNA amplification was performed with 12 amplification cycles, followed by reaction clean-up using SPRIselect Reagent Kit (Beckman Coulter). Subsequently, enzymatic fragmentation, library construction, and additional SPRIselect bead clean-ups were performed to generate the final indexed libraries. Sample index PCR cycles were determined based on the total cDNA yield and ranged between 14 and 16 cycles. Library quality was assessed using the Agilent High Sensitivity DNA kit and libraries were quantified using the Qubit 1X dsDNA HS Assay kit (Thermo Fisher Scientific). Libraries were pooled and sequenced with the Illumina, NovaSeq 6000, platform, PE, Dual Index, (26 bp read 1, 90 bp read 2, 10 cycles i5 index and 10 cycles i7 index). As recommended, we aimed at obtaining, minimum 5,000 read pairs per cell for V(D)J Dual Index library and a minimum of 20,000 read pairs per cell for 5' Gene Expression Dual Index library. Single-cell RNA sequencing data analysis CellRanger (v.7.2.0) 61 was used to align scRNA reads to the reference and create count matrices. Analysis was performed using the R (v4.3.1) and Python (3.12) programming languages. Seurat (v5.0.0) 62 was used for the processing of scRNA data. Ambient RNA correction was performed using SoupX(v1.6.2) 63 . scDblFinder(v1.16.0) 64 was used to identify and remove doublets. Additional filtering was performed removing cells with 8000 genes, mitochondrial RNA > 10% and ribosomal RNA > 40%. Integration of single cells was performed with Harmony (v.0.1.0) 65 resulting in clusters corresponding to distinct cell types. To further remove low quality cells that might have passed the initial quality control, additional subclustering of the cells corresponding to each of the major cell types (T-cells, NK-cell, Monocytes, B-cells, plasma cells), was performed. Sub-clusters expressing non-specific marker genes for each cell type were removed. Annotation of the clusters was performed using canonical marker genes and validated using CellTypist 66 . To enable systematic comparisons across patients, we integrated the data from all individuals and all time-points by applying Harmony and Louvain clustering (with resolution 0.4). Overall, we annotated 127,432 tumor and immune cells at T1 (B-cells, n = 44,328 (35%); plasma cells, n = 2,570 (2%); T cells, n = 39,412 (31%); NK, n = 8,693 (7%); monocytes, n = 13,007(10%); progenitors, n = 16,137 (12%) and 20,9341 tumor and immune cells at T2 (B-cells, n = 46,840 (22%); plasma cells, n = 1,986 (1%); T cells, n = 72,651 (35%); NK, n = 16,452 (8%); monocytes, n = 28,644 (13%); progenitors, n = 29,808 (14%). Sub-clustering and cell sub-type proportion analysis Each major cell type of the tumor microenvironment (T-cells, NK-cells, Monocytes,Neutrophils) was subjected to sub-clustering to identify cell-subtypes and estimate changes in the proportions of the subtypes across the study groups. Integration of T-cells was performed using Harmony 65 and Seurat’s CCA integration was applied on NK-cells and monocytes. Annotation of the sub-clusters was performed using canonical marker genes. To evaluate the differences in the proportions of cell subtypes across groups, we normalized the number of cells from each subtype for every individual by the total number of cells of the corresponding type. Samples with < 50 cells were excluded for this part of the analysis. Statistically significant differences were estimated using the Wilcoxon singed-rank test (un-paired when comparing responding groups of the same timepoint and paired when comparing across timepoints). Integration and label transfer between the patient data and the single-cell bone marrow reference 67 , was performed using Symphony(v.0.1.1) 68 . Samples of the reference data with < 50 cells for a major cell type were excluded from the analysis. Identification of malignant B cells To identify malignant (clonal) B-cells, data from each sample were analyzed separately. For 25 patients additional scBCR sequencing information was performed allowing us to assign single cells to clonotypes with unique V(D)J combinations. Clusters where multiple clonotypes were found to be clustering together, were annotated as healthy B-cells. Clusters that were enriched with a unique clonotype were annotated as malignant. To validate the clonotype-based annotation and identify malignant cells in the remaining 12 samples for which scBCR information was not available we applied the K/L ratio approach, as previously described 69 . scBCR-seq and K/L ratio measurement identified 6% healthy and 94% cBc at T1 and 4% healthy an 96% cBc at T2. Inference of copy number variants from single-cell RNA seq data To identify CNVs from the single-cell data, the healthy and malignant B-cells from every sample (across the different timepoints) were processed separately. CNV inference was performed using Numbat (v1.3.2-1) 70 . To avoid potential artifacts, different custom references were used for the 3’ and 5’ sequenced data. The 3’ reference included 2136 B cells from a healthy donor previously described by our group 69 , and the 5’ reference included 1589 B cells obtained from the two healthy donors sequenced for the current study. For the CNV visualization across groups the GenVisR(v.1.34.0) package was used 71 . To statistically assess the difference of the CNVs occurrence across groups, we used Fisher's exact test. Identification of PC-like/MBC-like molecular subtypes For this we used sciRED, a matrix factorization approach, to identify gene expression programs to determine sources of variation in our cohort of patients at diagnosis. PCA analysis on the set of genes associated with the most important factor from sciRed using as biological covariate the response group, revealed a clear distinction between samples that exhibited distinct MBC-like or PC-like features. The separation of the samples across the first principal component showed that the main source of variation across all patients originated from the B-cell subtype they belong to, indicating that the subtype of origin might be an important factor for the progression of the disease and the response to therapy. We validated these findings by determining a plasma cell scoring approach, using established gene markers associated with these subtypes. Briefly, two scores were estimated for each cell, one determined by the expression of PC-like features such as XBP1 , IRF4 , MZB1 , JCHAIN , CD138 , CD9 , DUSP22 , and the second by the expression of MBC-like features such as CD74 , BACH2 , CD20 , SPI1 , HLA-DRA , RACK1. We next divided the two scores to determine a single score between 0 and 1 for each cell, with higher scores being indicative of a PC-like subtype. Gene expression signature (GEX) analysis Gene expression signature (GEX) analysis was performed using the Python package sciRed 72 , with library size and assay type (3’/5’) used as technical covariates. As input, we used the raw counts of the top 2000 variable genes. Only malignant B-cells were considered for this analysis. sciRed was applied using as biological covariate the response group and the MYD88/CXCR4 mutation status. The top 400 genes with the highest positive (200 genes) and negative (200 genes) association with the most important factors were considered for further analysis. GLMnet risk scoring We used the glmnet package (family=”cox”) 73 to estimate a risk score for each patient, based on single-cell gene expression data from malignant B cells at diagnosis, retaining only the top 1,000 variable genes. Briefly, glmnet fits an elastic net regression model that combines L1 (lasso) and L2 (ridge) coefficients, enabling us to identify genes that can help distinguish high-risk from low-risk patients. We used an alpha parameter of 0.03 maintaining 64 genes with coefficient > 0. The final score provides a combined estimate of the contribution of every selected variable (gene) that can be used as a quantitative measurement for the patient’s risk of disease progression. Validation of gene expression signatures using an external cohort To validate our findings from the gene expression signature profile analysis we used bulk-RNA sequencing data of CD19 + cells from 47 WM patients. RNA sequencing reads were aligned using STAR 74 and transcript counts from Salmon 75 were aggregated into gene counts using tximport. Downstream analysis was performed using DeSeq2 76 , and the vst() function was used to log-transform the data. Survival analysis Kaplan-Meier analysis was performed using the survival (v3.7) R package. To estimate survival curves for individual genes, we initially calculated the average gene expression per sample for the gene of interest, followed by the average gene expression across the whole cohort at diagnosis. Samples were labeled as '1' when gene expression exceeded the average cohort expression and as '0' when gene expression was below the average cohort expression. Progression-free survival (PFS) was defined as the event of interest. To estimate survival curves for the gene expression signatures, we initially run the R package AUCell (v.1.24.0) to estimate a single score per cell using the expression of genes highly correlated to the factor of interest. We then estimated the median per sample to obtain a single score per patient. Finally, samples were labeled as “1” or “0” according to the sample score (larger or lower from the defined cutoff). Cell-cell communication analysis CellChat 77 was used to identify significant interactions between cell types of the tumor microenvironment. CellChat was performed four times to identify interactions across the following cohorts: T1-RG, T1-NRG, T2-RG, and T2-NRG. The function mergeCellChat() was then used to combine pairs of datasets, for downstream analysis and comparison across cohorts. Whole genome sequencing (WGS) analysis Burrows-Wheeler Aligner 78 (BWA v0.7.17) was utilized to map the paired-end reads to the human reference genome (hg38). SAMtools 79 was used for sorting the BAM files, and Picard tools ( https://github.com/broadinstitute/picard ) was utilized to mark duplicate reads. For the matched tumor - germline samples somatic SNV and INDEL detection was performed with MuTect2 80 ,Strelka2 81 and Varscan2 82 . Only variants identified by at least two variant callers were processed for further analysis. Somatic CNV detection was performed using Battenberg 83 . The VCF files that included the somatic SNVs and INDELs of each sample were annotated using VEP and converted into MAF files using vcf2maf (v1.6.21) ( https://github.com/mskcc/vcf2maf ). Maftools 84 was used was downstream analysis and visualization of variants in oncoplots. For the signature analysis, we used the fitting algorithm mmsig 85 (1000 iterations, 0.01 cosine similarity threshold) to confirm the presence and estimate the contribution of 6 SBS signatures: SBS1, SBS5, SBS8, SBS9, SBS2, SBS13, SBS84. Clonal evolution of longitudinal samples We used Retcher 86 to reconstruct the clonal evolution of patients treated with ibrutinib using sequencing information from both timepoints (T1 and T2). Retcher, employs a probabilistic model to reconstruct the tumor phylogeny based on somatic SNVs and CNVs identified across sequential timepoints, forming clusters of variants with similar cellular prevalence. Analysis of the resulting clusters allowed us to identify samples with branching and linear evolution. Declarations Funding: This study has been supported by the IWMF-LLS Research Roadmap Initiative. Author Contribution T.B. designed study, analyzed data, wrote paper C.V. analyzed data, wrote paper M.S. performed experiments, analyzed data, wrote paper I.K. performed experiments N.M-K. performed experiments C.L. performed experiments K.T. performed experiments A.P. performed experiments K.C. performed experiments A.P. performed experiments E.K. performed experiments F.T. provided clinical data M.G. wrote paper E.T. wrote paper Z.H. provided validation cohort, wrote paper S.T wrote paper M.D. wrote paper E.K. wrote paper References Owen RG, Treon SP, Al-Katib A et al (2003) Clinicopathological definition of Waldenstrom’s macroglobulinemia: consensus panel recommendations from the Second International Workshop on Waldenstrom’s Macroglobulinemia. Semin Oncol . ;30(2):110–115. 10.1053/sonc.2003.50082 Campo E, Swerdlow SH, Harris NL, Pileri S, Stein H, Jaffe ES (2011) The 2008 WHO classification of lymphoid neoplasms and beyond: evolving concepts and practical applications. Blood 117(19):5019–5032. 10.1182/blood-2011-01-293050 Treon SP, Tripsas CK, Meid K et al (2015) Ibrutinib in previously treated Waldenström’s macroglobulinemia. N Engl J Med 372(15):1430–1440. 10.1056/NEJMoa1501548 Treon SP, Xu L, Hunter Z (2015) MYD88 Mutations and Response to Ibrutinib in Waldenström’s Macroglobulinemia. N Engl J Med 373(6):584–586. 10.1056/NEJMc1506192 Xu L, Tsakmaklis N, Yang G et al (2017) Acquired mutations associated with ibrutinib resistance in Waldenström macroglobulinemia. Blood 129(18):2519–2525. 10.1182/blood-2017-01-761726 Chen JG, Liu X, Munshi M et al (2018) BTK(Cys481Ser) drives ibrutinib resistance via ERK1/2 and protects BTK(wild-type) MYD88-mutated cells by a paracrine mechanism. Blood 131(18):2047–2059. 10.1182/blood-2017-10-811752 Paiva B, Corchete LA, Vidriales MB et al (2015) The cellular origin and malignant transformation of Waldenström macroglobulinemia. Blood 125(15):2370–2380. 10.1182/blood-2014-09-602565 Treon SP, Hunter ZR, Aggarwal A et al (2006) Characterization of familial Waldenstrom’s macroglobulinemia. Ann Oncol 17(3):488–494. 10.1093/annonc/mdj111 Poulain S, Braggio E, Roumier C et al (2011) High-throughput genomic analysis in Waldenström’s macroglobulinemia. Clin Lymphoma Myeloma Leuk 11(1):106–108. 10.3816/CLML.2011.n.021 Braggio E, Philipsborn C, Novak A, Hodge L, Ansell S, Fonseca R (2012) Molecular pathogenesis of Waldenstrom’s macroglobulinemia. Haematologica 97(9):1281–1290. 10.3324/haematol.2012.068478 Paiva B, Montes MC, García-Sanz R et al (2014) Multiparameter flow cytometry for the identification of the Waldenström’s clone in IgM-MGUS and Waldenström’s Macroglobulinemia: new criteria for differential diagnosis and risk stratification. Leukemia 28(1):166–173. 10.1038/leu.2013.124 San Miguel JF, Vidriales MB, Ocio E et al (2003) Immunophenotypic analysis of Waldenstrom’s macroglobulinemia. Semin Oncol 30(2):187–195. 10.1053/sonc.2003.50074 Morice WG, Chen D, Kurtin PJ, Hanson CA, McPhail ED (2009) Novel immunophenotypic features of marrow lymphoplasmacytic lymphoma and correlation with Waldenström’s macroglobulinemia. Mod Pathol 22(6):807–816. 10.1038/modpathol.2009.34 Remstein ED, Hanson CA, Kyle RA, Hodnefield JM, Kurtin PJ (2003) Despite apparent morphologic and immunophenotypic heterogeneity, Waldenstrom’s macroglobulinemia is consistently composed of cells along a morphologic continuum of small lymphocytes, plasmacytoid lymphocytes, and plasma cells. Semin Oncol 30(2):182–186. 10.1053/sonc.2003.50073 Cao Y, Hunter ZR, Liu X et al (2015) CXCR4 WHIM-like frameshift and nonsense mutations promote ibrutinib resistance but do not supplant MYD88(L265P) -directed survival signalling in Waldenström macroglobulinaemia cells. Br J Haematol 168(5):701–707. 10.1111/bjh.13200 Cao Y, Hunter ZR, Liu X et al (2015) The WHIM-like CXCR4(S338X) somatic mutation activates AKT and ERK, and promotes resistance to ibrutinib and other agents used in the treatment of Waldenstrom’s Macroglobulinemia. Leukemia 29(1):169–176. 10.1038/leu.2014.187 Castillo JJ, Xu L, Gustine JN et al (2019) CXCR4 mutation subtypes impact response and survival outcomes in patients with Waldenström macroglobulinaemia treated with ibrutinib. Br J Haematol 187(3):356–363. 10.1111/bjh.16088 Treon SP, Meid K, Gustine J et al (2021) Long-Term Follow-Up of Ibrutinib Monotherapy in Symptomatic, Previously Treated Patients With Waldenström Macroglobulinemia. J Clin Oncol 39(6):565–575. 10.1200/JCO.20.00555 Jiménez C, Chan GG, Xu L et al (2020) Genomic evolution of ibrutinib-resistant clones in Waldenström macroglobulinaemia. Br J Haematol 189(6):1165–1170. 10.1111/bjh.16463 Richardson K, Castillo JJ, Sarosiek SR et al (2024) Identification of robust predictors for ibrutinib response by multiomics in MYD88-mutated Waldenström macroglobulinemia. Blood Adv 8(9):2133–2137. 10.1182/bloodadvances.2023012111 Roos-Weil D, Giacopelli B, Armand M et al (2020) Identification of 2 DNA methylation subtypes of Waldenström macroglobulinemia with plasma and memory B-cell features. Blood 136(5):585–595. 10.1182/blood.2020005081 Hunter Z, Guerrera M, Tsakmaklis N et al (2025) The Evolution and Subtypes of Waldenstrom Macroglobulinemia: Findings from a Multi-Omic Analysis of 249 Treatment Naive MYD88L265P Mutated Patients. 10.21203/rs.3.rs-6473628/v1 Gagler DC, Ghamlouch H, Zhang D et al (2025) A multiomic analysis of Waldenström macroglobulinemia defines distinct disease subtypes. Blood . Published online 7 May :blood.2024028164. 10.1182/blood.2024028164 Richardson K, Castillo JJ, Sarosiek SR et al (2024) Identification of robust predictors for ibrutinib response by multiomics in MYD88-mutated Waldenström macroglobulinemia. Blood Adv 8(9):2133–2137. 10.1182/bloodadvances.2023012111 Mitchell RA, Liao H, Chesney J et al (2002) Macrophage migration inhibitory factor (MIF) sustains macrophage proinflammatory function by inhibiting p53: regulatory role in the innate immune response. Proc Natl Acad Sci U S A 99(1):345–350. 10.1073/pnas.012511599 Hudson JD, Shoaibi MA, Maestro R, Carnero A, Hannon GJ, Beach DH (1999) A proinflammatory cytokine inhibits p53 tumor suppressor activity. J Exp Med 190(10):1375–1382. 10.1084/jem.190.10.1375 Nguyen MT, Lue H, Kleemann R et al (2003) The cytokine macrophage migration inhibitory factor reduces pro-oxidative stress-induced apoptosis. J Immunol 170(6):3337–3347. 10.4049/jimmunol.170.6.3337 Roger T, David J, Glauser MP, Calandra T (2001) MIF regulates innate immune responses through modulation of Toll-like receptor 4. Nature 414(6866):920–924. 10.1038/414920a Hunter ZR, Xu L, Tsakmaklis N et al (2018) Insights into the genomic landscape of MYD88 wild-type Waldenström macroglobulinemia. Blood Adv 2(21):2937–2946. 10.1182/bloodadvances.2018022962 Hunter ZR, Xu L, Yang G et al (2014) The genomic landscape of Waldenstrom macroglobulinemia is characterized by highly recurring MYD88 and WHIM-like CXCR4 mutations, and small somatic deletions associated with B-cell lymphomagenesis. Blood 123(11):1637–1646. 10.1182/blood-2013-09-525808 Maclachlan KH, Bagratuni T, Kastritis E et al (2023) Waldenström macroglobulinemia whole genome reveals prolonged germinal center activity and late copy number aberrations. Blood Adv 7(6):971–981. 10.1182/bloodadvances.2022008876 Bagratuni TD, Theologi O, Vlachos CD et al Genomic landscape of IgM-MGUS and patients with stable or progressive asymptomatic Waldenström macroglobulinemia. Blood . Published online 25 September 2025:blood.2025030177. 10.1182/blood.2025030177 Samur MK, Aktas Samur A, Fulciniti M et al (2020) Genome-Wide Somatic Alterations in Multiple Myeloma Reveal a Superior Outcome Group. J Clin Oncol 38(27):3107–3118. 10.1200/JCO.20.00461 Sun H, Fang T, Wang T et al (2022) Single-cell profiles reveal tumor cell heterogeneity and immunosuppressive microenvironment in Waldenström macroglobulinemia. J Transl Med 20(1):576. 10.1186/s12967-022-03798-6 Sklavenitis-Pistofidis R, Konishi Y, Heilpern-Mallory D et al (2025) Single-cell RNA sequencing defines distinct disease subtypes and reveals hypo-responsiveness to interferon in asymptomatic Waldenstrom’s Macroglobulinemia. Nat Commun 16(1):1480. 10.1038/s41467-025-56323-w Aran A, Garrigós L, Curigliano G, Cortés J, Martí M (2022) Evaluation of the TCR Repertoire as a Predictive and Prognostic Biomarker in Cancer: Diversity or Clonality? Cancers (Basel) 14(7). 10.3390/cancers14071771 Pilcher WC, Yao L, Gonzalez-Kozlova E et al A single-cell atlas characterizes dysregulation of the bone marrow immune microenvironment associated with outcomes in multiple myeloma. Preprint posted online 17 May 2024. 10.1101/2024.05.15.593193 Lian J, Yue Y, Yu W, Zhang Y (2025) Correction: Immunosenescence: a key player in cancer development. J Hematol Oncol 18(1):25. 10.1186/s13045-025-01682-6 Barilà G, Pavan L, Vedovato S et al (2021) Treatment Induced Cytotoxic T-Cell Modulation in Multiple Myeloma Patients. Front Oncol 11:682658. 10.3389/fonc.2021.682658 Sacco A, Desantis V, Celay J et al (2023) Targeting the immune microenvironment in Waldenström macroglobulinemia via halting the CD40/CD40-ligand axis. Blood 141(21):2615–2628. 10.1182/blood.2022019240 Sklavenitis-Pistofidis R, Konishi Y, Heilpern-Mallory D et al (2025) Single-cell RNA sequencing defines distinct disease subtypes and reveals hypo-responsiveness to interferon in asymptomatic Waldenstrom’s Macroglobulinemia. Nat Commun 16(1):1480. 10.1038/s41467-025-56323-w Elessa D, Debureaux PE, Villesuzanne C et al (2022) Inflammatory Waldenström’s macroglobulinaemia: A French monocentric retrospective study of 67 patients. Br J Haematol 197(6):728–735. 10.1111/bjh.18157 Forgeard N, Baron M, Caron J et al (2022) Inflammation in Waldenström macroglobulinemia is associated with 6q deletion and need for treatment initiation. Haematologica 107(11):2720–2724. 10.3324/haematol.2022.281053 Debureaux PE, Forgeard N, Elessa D et al (2024) Inflammation is predictive of outcome in Waldenström macroglobulinemia treated by Bruton tyrosine kinase inhibitors: a multicentric real-life study. Haematologica 109(1):325–330. 10.3324/haematol.2023.283141 Debureaux PE, Forgeard N, Elessa D et al (2024) Inflammation is predictive of outcome in Waldenström macroglobulinemia treated by Bruton tyrosine kinase inhibitors: a multicentric real-life study. Haematologica 109(1):325–330. 10.3324/haematol.2023.283141 Bertrand E, Jouy N, Manier S et al (2017) Role of IRF4 in resistance to immunomodulatory (IMid) compounds(®) in Waldenström’s macroglobulinemia. Oncotarget 8(68):112917–112927. 10.18632/oncotarget.22872 Maura F, Boyle EM, Coffey D et al (2023) Genomic and immune signatures predict clinical outcome in newly diagnosed multiple myeloma treated with immunotherapy regimens. Nat Cancer 4(12):1660–1674. 10.1038/s43018-023-00657-1 Argyropoulos KV, Vogel R, Ziegler C et al (2016) Clonal B cells in Waldenström’s macroglobulinemia exhibit functional features of chronic active B-cell receptor signaling. Leukemia 30(5):1116–1125. 10.1038/leu.2016.8 Kuiatse I, Baladandayuthapani V, Lin HY et al (2015) Targeting the Spleen Tyrosine Kinase with Fostamatinib as a Strategy against Waldenström Macroglobulinemia. Clin Cancer Res 21(11):2538–2545. 10.1158/1078-0432.CCR-14-1462 Agrelo R, Souabni A, Novatchkova M et al (2009) SATB1 defines the developmental context for gene silencing by Xist in lymphoma and embryonic cells. Dev Cell 16(4):507–516. 10.1016/j.devcel.2009.03.006 Sun JY, Xu L, Tseng H et al (2011) Histone deacetylase inhibitors demonstrate significant preclinical activity as single agents, and in combination with bortezomib in Waldenström’s macroglobulinemia. Clin Lymphoma Myeloma Leuk 11(1):152–156. 10.3816/CLML.2011.n.036 Roccaro AM, Sacco A, Jia X et al (2010) microRNA-dependent modulation of histone acetylation in Waldenstrom macroglobulinemia. Blood 116(9):1506–1514. 10.1182/blood-2010-01-265686 Hunter ZR, Xu L, Tsakmaklis N et al (2018) Insights into the genomic landscape of MYD88 wild-type Waldenström macroglobulinemia. Blood Adv 2(21):2937–2946. 10.1182/bloodadvances.2018022962 Hartmann S, Schuhmacher B, Rausch T et al (2016) Highly recurrent mutations of SGK1, DUSP2 and JUNB in nodular lymphocyte predominant Hodgkin lymphoma. Leukemia 30(4):844–853. 10.1038/leu.2015.328 Fechter K, Feichtinger J, Prochazka K et al (2018) Cytoplasmic location of NR4A1 in aggressive lymphomas is associated with a favourable cancer specific survival. Sci Rep 8(1):14528. 10.1038/s41598-018-32972-4 Derenzini E, Mazzara S, Melle F et al (2021) A three-gene signature based on MYC, BCL-2 and NFKBIA improves risk stratification in diffuse large B-cell lymphoma. Haematologica 106(9):2405–2416. 10.3324/haematol.2019.236455 Nagy B, Ferrer A, Larramendy ML et al (2003) Lymphotoxin beta expression is high in chronic lymphocytic leukemia but low in small lymphocytic lymphoma: a quantitative real-time reverse transcriptase polymerase chain reaction analysis. Haematologica 88(6):654–658 Bolli N, Biancon G, Moarii M et al (2018) Analysis of the genomic landscape of multiple myeloma highlights novel prognostic markers and disease subgroups. Leukemia 32(12):2604–2616. 10.1038/s41375-018-0037-9 Lopez-Gomez J, Cerrato-Izaguirre D, Quezada-Maldonado EM et al (2025) Molecular Characterization of Oral Squamous Cell Carcinoma in Mexican Patients: A Genomic and Epidemiological Overview. Cancers 17(20). 10.3390/cancers17203282 Bagratuni T, Markou A, Patseas D et al (2022) Determination of MYD88L265P mutation fraction in IgM monoclonal gammopathies. Blood Adv 6(1):189–199. 10.1182/bloodadvances.2021005354 Zheng GXY, Terry JM, Belgrader P et al (2017) Massively parallel digital transcriptional profiling of single cells. Nat Commun 8:14049. 10.1038/ncomms14049 Satija R, Farrell JA, Gennert D, Schier AF, Regev A (2015) Spatial reconstruction of single-cell gene expression data. Nat Biotechnol 33(5):495–502. 10.1038/nbt.3192 Young MD, Behjati S (2020) SoupX removes ambient RNA contamination from droplet-based single-cell RNA sequencing data. Gigascience 9(12). 10.1093/gigascience/giaa151 Germain PL, Lun A, Garcia Meixide C, Macnair W, Robinson MD (2021) Doublet identification in single-cell sequencing data using scDblFinder. F1000Res 10:979. 10.12688/f1000research.73600.2 Korsunsky I, Millard N, Fan J et al (2019) Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods 16(12):1289–1296. 10.1038/s41592-019-0619-0 Domínguez Conde C, Xu C, Jarvis LB et al (2022) Cross-tissue immune cell analysis reveals tissue-specific features in humans. Science 376(6594):eabl5197. 10.1126/science.abl5197 Lee NYS, Li M, Ang KS, Chen J (2023) Establishing a human bone marrow single cell reference atlas to study ageing and diseases. Front Immunol 14:1127879. 10.3389/fimmu.2023.1127879 Kang JB, Nathan A, Weinand K et al (2021) Efficient and precise single-cell reference atlas mapping with Symphony. Nat Commun 12(1):5890. 10.1038/s41467-021-25957-x Bagratuni T, Aktypi F, Theologi O et al (2024) Single-cell analysis of MYD88(L265P) and MYD88(WT) Waldenström macroglobulinemia patients. Hemasphere 8(2):e27. 10.1002/hem3.27 Gao T, Soldatov R, Sarkar H et al (2023) Haplotype-aware analysis of somatic copy number variations from single-cell transcriptomes. Nat Biotechnol 41(3):417–426. 10.1038/s41587-022-01468-y Skidmore ZL, Wagner AH, Lesurf R et al (2016) GenVisR: Genomic Visualizations in R. Bioinformatics 32(19):3012–3014. 10.1093/bioinformatics/btw325 Pouyabahar D, Andrews T, Bader GD (2025) Interpretable single-cell factor decomposition using sciRED. Nat Commun 16(1):1878. 10.1038/s41467-025-57157-2 Friedman J, Hastie T, Tibshirani R (2010) Regularization Paths for Generalized Linear Models via Coordinate Descent. J Stat Softw 33(1):1–22 Dobin A, Davis CA, Schlesinger F et al (2013) STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29(1):15–21. 10.1093/bioinformatics/bts635 Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C (2017) Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods 14(4):417–419. 10.1038/nmeth.4197 Love MI, Huber W, Anders S (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15(12):550. 10.1186/s13059-014-0550-8 Jin S, Guerrero-Juarez CF, Zhang L et al (2021) Inference and analysis of cell-cell communication using CellChat. Nat Commun 12(1):1088. 10.1038/s41467-021-21246-9 Li H Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv: Genomics . Published online 2013. https://api.semanticscholar.org/CorpusID:14669139 Danecek P, Bonfield JK, Liddle J et al (2021) Twelve years of SAMtools and BCFtools. Gigascience 10(2). 10.1093/gigascience/giab008 Benjamin D, Sato T, Cibulskis K, Getz G, Stewart C, Lichtenstein L Calling Somatic SNVs and Indels with Mutect2. bioRxiv . Published online 2019. 10.1101/861054 Kim S, Scheffler K, Halpern AL et al (2018) Strelka2: fast and accurate calling of germline and somatic variants. Nat Methods 15(8):591–594. 10.1038/s41592-018-0051-x Koboldt DC, Zhang Q, Larson DE et al (2012) VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Res 22(3):568–576. 10.1101/gr.129684.111 Nik-Zainal S, Van Loo P, Wedge DC et al (2012) The life history of 21 breast cancers. Cell 149(5):994–1007. 10.1016/j.cell.2012.04.023 Mayakonda A, Lin DC, Assenov Y, Plass C, Koeffler HP (2018) Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res 28(11):1747–1756. 10.1101/gr.239244.118 Rustad EH, Nadeu F, Angelopoulos N et al (2021) mmsig: a fitting approach to accurately identify somatic mutational signatures in hematological malignancies. Commun Biol 4(1):424. 10.1038/s42003-021-01938-0 Wang Z, Fang Y, Wang R, Kong L, Liang S, Tao S (2024) Reconstructing tumor clonal heterogeneity and evolutionary relationships based on tumor DNA sequencing data. Brief Bioinform 25(6). 10.1093/bib/bbae516 Additional Declarations There is NO Competing Interest. Supplementary Files Supplementaltables.docx Supplemental tables SupplementalFi.legends.docx Supplemental Figure legends SupplFigure1FS.png Supplemental Figure 1 Supplfigure6FS.png Supplemental Figure 6 Supplfigure8FS.png Supplemental Figure 8 Supplfigure9FS.png Supplemental Figure 9 Supplfigure7FS.png Supplemental Figure 7 Supplfigure4FS.png Supplemental Figure 4 Supplfigure5FS.png Supplemental Figure 5 Supplfigure3FS.png Supplemental Figure 3 Supplfigure2FS.png Supplemental Figure 2 Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8594385","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":578911639,"identity":"48364012-c4cb-4480-bcf6-998fc14d5479","order_by":0,"name":"Tina Bagratuni","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYDACCRiDmfkAAwMbETp4IFoMgFrYEkjVwsBjQJwWe+nmx68rav4kbmfn+brhQxlDYj/7AebXFfhskTlmZnnmmEHizmbebTdnnGNInNmTwGZ5Bq/DEswMG9gMEjcc5t12m7eNIXHDgQQ2wwa8WtK/GTb8A2nheXb7L0jL+QeEtOQYP2xsA2thu80I0nIjgfkhXi13zpQxNvYZG284zGZ2s+echPHMGQ/bGPFpYZ/dvvljwzc52Q3nDz+78aPMRrafP/nwR3xagIBNAokDYjO2SeBSCwXMHwiLjIJRMApGwYgGAHbjUk9kHdphAAAAAElFTkSuQmCC","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":true,"prefix":"","firstName":"Tina","middleName":"","lastName":"Bagratuni","suffix":""},{"id":578911640,"identity":"79892866-91f4-49e5-a4c3-f5f5b75ea33b","order_by":1,"name":"Christos Vlachos","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Christos","middleName":"","lastName":"Vlachos","suffix":""},{"id":578911641,"identity":"bd2bfd3f-b63c-47e1-ae02-e510ef5be2cb","order_by":2,"name":"Maria Sakkou","email":"","orcid":"","institution":"BSRC \"Alexander Fleming\"","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"","lastName":"Sakkou","suffix":""},{"id":578911642,"identity":"67e6f535-3c3a-4219-acd8-79755b4a084a","order_by":3,"name":"Ioannis Kollias","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Ioannis","middleName":"","lastName":"Kollias","suffix":""},{"id":578911643,"identity":"0b0c2d8b-879b-4510-9908-497d1b57b37a","order_by":4,"name":"Ourania Theologi","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Ourania","middleName":"","lastName":"Theologi","suffix":""},{"id":578911644,"identity":"3e89d04e-921e-4948-b19d-36a33c2b15c9","order_by":5,"name":"Nefeli Mavrianou Koutsoukou","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Nefeli","middleName":"Mavrianou","lastName":"Koutsoukou","suffix":""},{"id":578911645,"identity":"4e82c358-8211-4d9a-9ae5-7578ed7db844","order_by":6,"name":"Christine-Ivy Liacos","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Christine-Ivy","middleName":"","lastName":"Liacos","suffix":""},{"id":578911646,"identity":"e8326c4b-3a34-4b13-a5b3-917632c17c47","order_by":7,"name":"Konstantina Taouxi","email":"","orcid":"https://orcid.org/0009-0007-0954-3908","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Konstantina","middleName":"","lastName":"Taouxi","suffix":""},{"id":578911647,"identity":"e5f7a7d2-15e7-4265-a795-00df6ff3bdf7","order_by":8,"name":"Alexandra Papadimou","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Alexandra","middleName":"","lastName":"Papadimou","suffix":""},{"id":578911648,"identity":"f26f4ef7-a66f-491f-a811-68a3ed947cae","order_by":9,"name":"Katerina Chrisostomidou","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Katerina","middleName":"","lastName":"Chrisostomidou","suffix":""},{"id":578911649,"identity":"f49e3288-845e-4d1f-925e-0d6ca514955e","order_by":10,"name":"Agisilaos Papadopoulos","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Agisilaos","middleName":"","lastName":"Papadopoulos","suffix":""},{"id":578911650,"identity":"584de35b-e24c-4ac2-b025-5e04b0855eb3","order_by":11,"name":"Eleni Kalafati","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Eleni","middleName":"","lastName":"Kalafati","suffix":""},{"id":578911651,"identity":"f011535c-aa48-4db7-a555-1ed135c80e4f","order_by":12,"name":"Foteini Theodorakakou","email":"","orcid":"","institution":"National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Foteini","middleName":"","lastName":"Theodorakakou","suffix":""},{"id":578911652,"identity":"a184c1cf-90ca-40ce-b666-6f4513912048","order_by":13,"name":"Maria Gavriatopoulou","email":"","orcid":"https://orcid.org/0000-0002-6244-1229","institution":"National and Kapodistrian University of Athens, School of Medicine, Athens, Greece","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"","lastName":"Gavriatopoulou","suffix":""},{"id":578911653,"identity":"af10d94c-b072-4280-9492-c74088f08e6d","order_by":14,"name":"Evangelos Terpos","email":"","orcid":"https://orcid.org/0000-0001-5133-1422","institution":"National and Kapodistrian University of Athens, School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Evangelos","middleName":"","lastName":"Terpos","suffix":""},{"id":578911654,"identity":"825caedb-d1e0-4431-99a2-ba9c7c8e6d6a","order_by":15,"name":"Zachary Hunter","email":"","orcid":"https://orcid.org/0000-0002-1689-1691","institution":"Dana-Farber Cancer Institute","correspondingAuthor":false,"prefix":"","firstName":"Zachary","middleName":"","lastName":"Hunter","suffix":""},{"id":578911655,"identity":"f2d93a83-ccb5-4526-b334-1652c2b8f946","order_by":16,"name":"Steven TREON","email":"","orcid":"https://orcid.org/0000-0001-6393-6154","institution":"Dana Farber Cancer Institute","correspondingAuthor":false,"prefix":"","firstName":"Steven","middleName":"","lastName":"TREON","suffix":""},{"id":578911656,"identity":"dc194258-f168-4cca-ac89-d6aa336bcfc9","order_by":17,"name":"Meletios Dimopoulos","email":"","orcid":"https://orcid.org/0000-0001-8990-3254","institution":"School of Medicine, National and Kapodistrian University of Athens","correspondingAuthor":false,"prefix":"","firstName":"Meletios","middleName":"","lastName":"Dimopoulos","suffix":""},{"id":578911657,"identity":"ae0d46a7-2758-4df4-8616-7360c4fbe7ba","order_by":18,"name":"Efstathios Kastritis","email":"","orcid":"https://orcid.org/0000-0001-8191-5832","institution":"National and Kapodistrian University of Athens, School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Efstathios","middleName":"","lastName":"Kastritis","suffix":""}],"badges":[],"createdAt":"2026-01-13 16:52:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8594385/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8594385/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100964486,"identity":"5e43c199-bbd4-4f1a-a1da-3689b2abea9b","added_by":"auto","created_at":"2026-01-23 09:07:47","extension":"png","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":836854,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/ec995856201f9959b481b838.png"},{"id":101203316,"identity":"392f8c68-9ee5-4a86-a447-d7a462785ed2","added_by":"auto","created_at":"2026-01-27 09:39:22","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6829374,"visible":true,"origin":"","legend":"","description":"","filename":"IbrutinibresistantWMmanuscriptfinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/d42203a8d3665bdee6da1ba9.docx"},{"id":101203685,"identity":"bab828c8-0cf8-4045-aea9-6e15eaff9f91","added_by":"auto","created_at":"2026-01-27 09:40:25","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6910562,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/4c710784fb712cd4b7b4e1db.png"},{"id":101202618,"identity":"0af21602-cdb4-4a63-82f0-bfddc594794a","added_by":"auto","created_at":"2026-01-27 09:36:46","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2813647,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/acffb06efec8ee69eebe9754.png"},{"id":100964473,"identity":"b806d43c-e039-4425-b683-059f47fe672a","added_by":"auto","created_at":"2026-01-23 09:07:46","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6705872,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/ba7ed2fb381cf76de65c93c0.png"},{"id":100964476,"identity":"61d65a31-a9e8-4e6d-aa89-24ef57c335de","added_by":"auto","created_at":"2026-01-23 09:07:46","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1079407,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/8bf2d54b338ac7751a6e4726.png"},{"id":101296617,"identity":"8442675e-e537-4dde-9fec-7fdd11c0aa1c","added_by":"auto","created_at":"2026-01-28 09:17:16","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1863399,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/7674ccc600e4f59d8f192160.png"},{"id":100964480,"identity":"83c3a83f-3a1d-4e9e-acf2-b5b5cbec58bb","added_by":"auto","created_at":"2026-01-23 09:07:46","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1515676,"visible":true,"origin":"","legend":"","description":"","filename":"Figure7FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/3c2cd3b10078fa33da37c1a2.png"},{"id":100964488,"identity":"7136b826-9de5-4112-b2dd-093bf2ec2eec","added_by":"auto","created_at":"2026-01-23 09:07:47","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1551402,"visible":true,"origin":"","legend":"","description":"","filename":"Figure8FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/db60c6b5ed9bfcd042ce6f47.png"},{"id":100964487,"identity":"6f7cdee3-a59d-43a3-b21f-0c4783c5f654","added_by":"auto","created_at":"2026-01-23 09:07:47","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1465176,"visible":true,"origin":"","legend":"","description":"","filename":"Figure9FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/3aace10c85daae1a62bacb79.png"},{"id":100964491,"identity":"ae894947-3958-4b8a-9cfe-a6c3fcdececa","added_by":"auto","created_at":"2026-01-23 09:07:47","extension":"json","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18354,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS26003578.json","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/ced75fe2cee248c829b8aced.json"},{"id":101751159,"identity":"6eb25dce-9647-4fc7-8083-e207ca9b3d62","added_by":"auto","created_at":"2026-02-03 10:17:02","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":336095,"visible":true,"origin":"","legend":"","description":"","filename":"SupplFigure1FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/f0c833850f6a07eccb054533.png"},{"id":100964510,"identity":"85d99b32-b23a-419a-8a9c-61fb30d8ed2e","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1030428,"visible":true,"origin":"","legend":"","description":"","filename":"Supplfigure2FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/e3b83f5bed912e73fb97058d.png"},{"id":101203763,"identity":"47e10897-2d87-431f-8b0d-363338d98c2e","added_by":"auto","created_at":"2026-01-27 09:40:37","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1353807,"visible":true,"origin":"","legend":"","description":"","filename":"Supplfigure3FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/6c0973eaef8e1be219a2b629.png"},{"id":100964500,"identity":"38648c3b-d581-4c1a-826a-b655d7452eb1","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2867552,"visible":true,"origin":"","legend":"","description":"","filename":"Supplfigure4FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/2ea763eaf99a6b1d9c34d080.png"},{"id":100964504,"identity":"95378430-6c40-4c70-9e30-a92a2804332f","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2645493,"visible":true,"origin":"","legend":"","description":"","filename":"Supplfigure5FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/0840835f1bb58b34e9082872.png"},{"id":101397897,"identity":"65a994d5-7438-46ba-9122-dd9b3993f912","added_by":"auto","created_at":"2026-01-29 09:37:58","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129560,"visible":true,"origin":"","legend":"","description":"","filename":"Supplfigure6FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/89b85d5e0e9ac68ff3a1069a.png"},{"id":101202612,"identity":"13c01872-6aa9-49ce-a5d1-eec6eedff8f5","added_by":"auto","created_at":"2026-01-27 09:36:43","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1395046,"visible":true,"origin":"","legend":"","description":"","filename":"Supplfigure7FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/e41ded4ba7ff822c6750147c.png"},{"id":101203375,"identity":"cf812d3c-6246-4c5d-8b99-09131aa88220","added_by":"auto","created_at":"2026-01-27 09:39:30","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":600827,"visible":true,"origin":"","legend":"","description":"","filename":"Supplfigure8FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/53c5ca13d614f29255162111.png"},{"id":100964506,"identity":"87051ba3-9a7b-4c0d-869c-ee7f9007fe60","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":571593,"visible":true,"origin":"","legend":"","description":"","filename":"Supplfigure9FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/b45630d069f1d66efd25f9a7.png"},{"id":101203516,"identity":"18b8ac91-c6eb-4838-beb9-3e9c763169d2","added_by":"auto","created_at":"2026-01-27 09:39:58","extension":"docx","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18657,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementalFi.legends.docx","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/b35eac4c77122c8b788102c0.docx"},{"id":101202691,"identity":"dce5a1f7-868c-4e1d-9770-1c061a9eb065","added_by":"auto","created_at":"2026-01-27 09:37:10","extension":"docx","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19985,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaltables.docx","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/9f15124c43366163be091c06.docx"},{"id":100964509,"identity":"d3975534-1d8e-4edf-b6ab-3554cc626abe","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"xml","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":221013,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS260035780enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/073490a1758599155fcbe4a1.xml"},{"id":100964489,"identity":"c7deb42f-0995-4eb3-975d-e1860b14eecd","added_by":"auto","created_at":"2026-01-23 09:07:47","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":836854,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/4f6347cdcc429a547e4b2756.png"},{"id":100964503,"identity":"9db36e9b-f9ab-4ff9-8196-835f647b834a","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6910562,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/bf309f0408dc6111d0111416.png"},{"id":101202664,"identity":"dec8c3f1-02c0-430b-93a3-1578c646ab72","added_by":"auto","created_at":"2026-01-27 09:36:58","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2813647,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/4e49127680af1790a48866d1.png"},{"id":101296665,"identity":"46314611-8261-499e-9137-4d84f44bc1d1","added_by":"auto","created_at":"2026-01-28 09:18:23","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6705872,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/1bb2dbc2fd14cbed254f8468.png"},{"id":101202909,"identity":"ae1de598-7704-4a32-8533-0c718d7a63d9","added_by":"auto","created_at":"2026-01-27 09:38:07","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1079407,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/3e0af376918028ec5e6d37bb.png"},{"id":101202806,"identity":"32a46e1a-3f07-4f7b-b25d-288f5d2be542","added_by":"auto","created_at":"2026-01-27 09:37:49","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1863399,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/6f374ac9622bcff006585fa3.png"},{"id":101202700,"identity":"98be3e43-79bc-475e-b2b1-72cbde05c8cb","added_by":"auto","created_at":"2026-01-27 09:37:17","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1515676,"visible":true,"origin":"","legend":"","description":"","filename":"Figure7FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/3f2016e2b1e1423bfe5dfd6b.png"},{"id":100964511,"identity":"13b03279-fadf-4545-9e3d-0c3384bdc68e","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1551402,"visible":true,"origin":"","legend":"","description":"","filename":"Figure8FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/cccfd674955a6692f4f8545f.png"},{"id":101202697,"identity":"f77c3fda-884e-4d44-b75e-cd709c29a6dd","added_by":"auto","created_at":"2026-01-27 09:37:16","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1465176,"visible":true,"origin":"","legend":"","description":"","filename":"Figure9FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/930c9d666c0e61186a3bf85b.png"},{"id":100964499,"identity":"b72f3aa9-c413-446d-9f01-79fe61d0fa01","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":162831,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/01a4f454729061c177f26968.png"},{"id":101397680,"identity":"05a35461-d005-4073-a1e1-cd37cc493d69","added_by":"auto","created_at":"2026-01-29 09:35:05","extension":"png","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3026231,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/9b893d2fc9b5da0e4ad4937c.png"},{"id":100964505,"identity":"b45ba2eb-0e27-4ba6-bb3d-81f724e6c58f","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":873741,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/4e45ad3f8542fa8c03d6021d.png"},{"id":101296666,"identity":"ded3f457-e10a-4f75-babe-a63a8350692b","added_by":"auto","created_at":"2026-01-28 09:18:23","extension":"png","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2372021,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/2cc01ce4091f6a01ea04300e.png"},{"id":101202652,"identity":"5b63187d-f960-46b0-bda6-2fb614e13230","added_by":"auto","created_at":"2026-01-27 09:36:54","extension":"png","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":383197,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/d51d8be724367cedc19e2f54.png"},{"id":100964513,"identity":"b902f0e2-3042-4704-8d3a-a31ca17c9906","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":682578,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure6FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/68cd525bf259f96886a9518a.png"},{"id":100964501,"identity":"e8fcdfb5-91a8-448c-9a16-d111a6e08012","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":38,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":603749,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure7FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/688cb6b1818fb67fa3d89537.png"},{"id":101751281,"identity":"94582cc5-be41-4634-ae6f-3ddbf32d4f5c","added_by":"auto","created_at":"2026-02-03 10:18:54","extension":"png","order_by":39,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":871293,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure8FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/1ffd5730aa87d8c980b282b0.png"},{"id":101296644,"identity":"95d8db23-1e46-4405-be89-52cc26982768","added_by":"auto","created_at":"2026-01-28 09:17:57","extension":"png","order_by":40,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":262849,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure9FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/56e712f9a0b8de1ddf410ff1.png"},{"id":100964517,"identity":"99cdd799-edb8-40a2-b3dc-33d757ff8e61","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"xml","order_by":41,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":217486,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS260035780structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/aeeeaf0a07eebc904d1da1af.xml"},{"id":101297247,"identity":"bf0b21c1-5352-43d2-8f51-5f188d17ce32","added_by":"auto","created_at":"2026-01-28 09:26:10","extension":"html","order_by":42,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":244881,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/cd86ee29e6eaa7be21275486.html"},{"id":101296637,"identity":"136653f5-f767-4703-8671-9dd394230660","added_by":"auto","created_at":"2026-01-28 09:17:36","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":836854,"visible":true,"origin":"","legend":"\u003cp\u003eSchematic overview of the study design, patient cohort, sample processing, and analysis workflow. Created with Biorender.com\u003c/p\u003e","description":"","filename":"Figure1FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/88f353710fbeebb4d545349e.png"},{"id":101202570,"identity":"47e72a1c-880d-4165-a0c7-e09f9460fa91","added_by":"auto","created_at":"2026-01-27 09:36:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":6910562,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSingle- cell immune and B cell atlas of all WM patients. A. Uniform manifold approximation and projection.\u003c/strong\u003e \u003cstrong\u003eA. \u003c/strong\u003eUniform manifold approximation and projection (UMAP) embedding of 348,00 bone marrow mononuclear cells (BMMCs) collected from 37 WM patients. \u003cstrong\u003eB.\u003c/strong\u003e Dot plot displaying the average scaled expression of selected marker genes used for precise cluster annotation. Expression is visualized on a red-blue color scale, with the size of each dot corresponding to the percent expression. Dot plots are split by lineage into different immune cell types. \u003cstrong\u003eC.\u003c/strong\u003ePairwise comparison of overall subpopulations of samples with both timepoints available. P-values were estimated using Wilcoxon-paired test\u003cstrong\u003e. D.\u003c/strong\u003eLandscape of B cell and immune environment according to timepoint and response group. \u003cstrong\u003eE.\u003c/strong\u003e A stacked bar chart, displaying the average per-patient cell type composition according to timepoint and response group. Clusters are colored by their lineage.\u003c/p\u003e","description":"","filename":"Figure2FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/d231b2f3e6d88d9e18a1509d.png"},{"id":101202698,"identity":"10235483-40d6-4504-9d1f-2b3c55b6d1e5","added_by":"auto","created_at":"2026-01-27 09:37:16","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":2813647,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnalysis of monocyte subpopulations. A. \u003c/strong\u003eUMAP colored by monocyte annotation (left) and by monocyte clusters (right). \u003cstrong\u003eB.\u003c/strong\u003e Box plot showing changes in monocytes according to response and timepoint. \u003cstrong\u003eC. \u0026nbsp;\u003c/strong\u003eFeature plot highlighting the main characteristic of each cluster.\u003cstrong\u003e D. \u003c/strong\u003eBox plot representing the proportions of each monocytic cluster (0, 2 and 4) according to response and timepoint\u003cstrong\u003e. E. \u003c/strong\u003eVolcano plot displaying the differentially expressed genes (DEGs) between timepoint (T1) and timepoint 2 (T2) in the entire cohort.\u003cstrong\u003e \u0026nbsp;F. \u003c/strong\u003eBar plot displaying gene set enrichment analysis results for the DEGs shown in\u003cstrong\u003e E.\u003c/strong\u003e Box plot representing the proportions of each neutrophilic cluster (0, 1, 2 and 3) according to response and timepoint\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure3FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/016f503d11a460b55fe40627.png"},{"id":101203309,"identity":"54a7bc70-a221-4528-89c5-6ce313421cf2","added_by":"auto","created_at":"2026-01-27 09:39:22","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":6705872,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnalysis of T and NK subpopulations. A. \u003c/strong\u003eUMAP colored by T cell subpopulations and dot plot displaying the average scaled expression of selected T cell marker genes used for precise cluster annotation. Expression is visualized on a red-blue color scale, with the size of each dot corresponding to the percent expression. Dot plots are split by lineage into different immune cell types. \u003cstrong\u003eB.\u003c/strong\u003e T cell lineage pseudotime trajectory analysis and enrichment analysis of pseudotime-dependent genes in all T cell states. C. Box plot showing changes in specific T cells subpopulations according to response and timepoint. \u003cstrong\u003eD.\u003c/strong\u003e Box plot showing changes of activity score hallmark IFNG response pathway according to response. \u003cstrong\u003eE.\u003c/strong\u003e UMAP colored by NK cell subpopulations and dot plot displaying the average scaled expression of selected NK cell marker genes used for precise cluster annotation. \u003cstrong\u003eF.\u003c/strong\u003e Box plot showing changes in specific NK cells subpopulations according to response and timepoint. \u003cstrong\u003eG.\u003c/strong\u003e Box plot showing changes of activity score KEGG NK cytotoxicity pathway according to response. \u003cstrong\u003eH.\u003c/strong\u003eUMAP colored by NK CD56 bright and dim subpopulations and bar plots of NK bright/dim ratio according to response and timepoint.\u003c/p\u003e","description":"","filename":"Figure4FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/2a925afe3c699286d56d5c67.png"},{"id":101202669,"identity":"bd83ab32-3b6f-411d-b6d9-6f34f2ee8f10","added_by":"auto","created_at":"2026-01-27 09:37:00","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1079407,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnalysis of malignant B cell compartment A. \u003c/strong\u003eUMAP visualizations of B cells colored by healthy and malignant B cell state (top left), \u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eL265P \u003c/em\u003e\u003c/sup\u003emutation status (bottom left), \u003cem\u003eCXCR4 \u003c/em\u003emutation status (top right) and response status (bottom right). B. Bar plots showing malignant B cell proportion (top left), \u003cem\u003eMYD88 \u003c/em\u003eexpression (top right) \u003cem\u003eCXCR4\u003c/em\u003e expression (bottom left) according to response and timepoint. Colored dots show mutation status. \u003cstrong\u003eC.\u003c/strong\u003e Bar plot displaying gene set enrichment analysis results for the DEGs according to response and timepoint. \u003cstrong\u003eD.\u003c/strong\u003e Circle plot showing gene expression overlap between the RG at T1 and NRG at T2. \u003cstrong\u003eE.\u003c/strong\u003e Copy number variation (CNV) distribution of all chromosomes among the RG and the NRG. Red color displays the gains while the blue the deletions. \u003cstrong\u003eF. \u003c/strong\u003ePrincipal component analysis showing the plasma cell (PC)-like and memory B cell (MBC)-like molecular subtyping across the entire cohort of patients. \u003cstrong\u003eG.\u003c/strong\u003e Heatmap analysis showing the key marker genes used for the distribution of PC-like and MBC-like features. H. Box plots showing changes of cBc in PC-like and MBC-like patients according to response and timepoint\u003cstrong\u003e. I.\u003c/strong\u003e Volcano plot of DEGs among PC-like and MBC-like patients J. Progression-free survival (PFS) of PC-like and MBC-like patients.\u003c/p\u003e","description":"","filename":"Figure5FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/c432d79c3550a4f9079a0a39.png"},{"id":101203682,"identity":"c42d4c77-89c2-41b2-a63f-78e9dbf3c66e","added_by":"auto","created_at":"2026-01-27 09:40:25","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1863399,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMolecular gene expression signatures associated with ibrutinib response. A. \u003c/strong\u003eHeatmap analysis of four gene expression signatures (GEX1-GEX4) according to \u003cem\u003eMYD88\u003c/em\u003e and \u003cem\u003eCXCR4\u003c/em\u003emutation status. \u003cstrong\u003eB. \u003c/strong\u003eVolcano plot of DEGs between RG and NRG at T1 (top) and T2 (bottom). \u003cstrong\u003eC. \u003c/strong\u003eClonal evolution of 4 patients from T1 to T2 based on clonotype proportion and violin plots of key genes expressed in each malignant/clonotype compartment. \u003cstrong\u003eD.\u003c/strong\u003e PFS according to GEX1-GEX4 expression signatures (top and middle left), \u003cem\u003eLTB \u003c/em\u003eexpression (middle right), elasticnet high and low score in the entire cohort of patients (bottom left) and the external cohort (bottom right).\u003c/p\u003e","description":"","filename":"Figure6FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/053fdac7df7032cd55a99ceb.png"},{"id":100964494,"identity":"cc23e127-6003-4ad1-8c28-d36f69dd1a36","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1515676,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMutational landscape associated with ibrutinib response. A. \u003c/strong\u003eProportion of variants determined by single nucleotide polymorphisms (SNPs), insertion and deletion across the RG and the NRG patients. \u003cstrong\u003eB.\u003c/strong\u003e Proportion of variants according to response and timepoint. Top mutated genes at diagnosis in \u003cstrong\u003eC.\u003c/strong\u003e RG patients and \u003cstrong\u003eD.\u003c/strong\u003e NRG patients. \u003cstrong\u003eE.\u003c/strong\u003e CNV distribution of all chromosomes among the RG and the NRG. Red color displays the gains while the blue the deletions. \u003cstrong\u003eF.\u003c/strong\u003e Clonal evolution determined by the timing of gene mutation patterns (clonal-subclonal). \u003cstrong\u003eG.\u003c/strong\u003e Single base substitution (SBS) signature distribution in patients at both time points. \u003cstrong\u003eH.\u003c/strong\u003e Proportion of SBS1/SBS5 signatures according to response and their effect on PFS.\u003c/p\u003e","description":"","filename":"Figure7FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/d14129b42d4dbc615fb90c41.png"},{"id":101202634,"identity":"5fc17a3f-82a5-4955-b0c5-0d9919d27810","added_by":"auto","created_at":"2026-01-27 09:36:50","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1551402,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eInteractome analysis and prediction analysis using immune signatures. A. \u003c/strong\u003eChord diagram indicating the cell-cell communication network in our dataset. Chords are colored by the ‘sender’ cell type (population expressing the ligand) and point towards the ‘receiver’ cell type (population expressing the receptor). Sender was assigned to the clonal B cells. The blue colors indicate the responder group while the red colors the non-responder group. \u003cstrong\u003eB.\u003c/strong\u003e Receiver operating characteristic (ROC) curves for response prediction models based on tumor and immune cell atlas alone and in combination as shown in colored specific group of models. \u003cstrong\u003eC.\u003c/strong\u003e PFS using the composition of B cell % alone D. PFS using the composition of T cell % alone E. PFS using the composition of both B cell % and T cell %.\u003c/p\u003e","description":"","filename":"Figure8FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/5029e430bf07af8f5e5db0d8.png"},{"id":100964496,"identity":"f9fbe541-bc3a-465c-8b15-fbde106a5078","added_by":"auto","created_at":"2026-01-23 09:07:48","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":1465176,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure9FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/efaa9213bf648c54df33cf08.png"},{"id":100964469,"identity":"9aec4f27-1276-44d7-90ea-7d68777ca2a3","added_by":"auto","created_at":"2026-01-23 09:07:46","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":19985,"visible":true,"origin":"","legend":"Supplemental tables","description":"","filename":"Supplementaltables.docx","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/0f66d449c1e264b91151faf8.docx"},{"id":101202579,"identity":"4da9c4ac-b6a8-47ab-883a-a8001c3b4cb6","added_by":"auto","created_at":"2026-01-27 09:36:31","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":18657,"visible":true,"origin":"","legend":"Supplemental Figure legends","description":"","filename":"SupplementalFi.legends.docx","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/7dac6e2c18fb53a1a23b539c.docx"},{"id":101202730,"identity":"53497169-4e2d-4131-8889-5a52c923ae4f","added_by":"auto","created_at":"2026-01-27 09:37:21","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":336095,"visible":true,"origin":"","legend":"Supplemental Figure 1","description":"","filename":"SupplFigure1FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/cca0069c86df1ab73b8c2025.png"},{"id":101202727,"identity":"638aa5b5-93d7-4cfa-8a91-d121b5a544f5","added_by":"auto","created_at":"2026-01-27 09:37:21","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":129560,"visible":true,"origin":"","legend":"Supplemental Figure 6","description":"","filename":"Supplfigure6FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/5a1e895ea0cca0a4739d0fae.png"},{"id":101203444,"identity":"55a5081b-87f8-4e30-9d9e-cc2cd651f49b","added_by":"auto","created_at":"2026-01-27 09:39:42","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":600827,"visible":true,"origin":"","legend":"Supplemental Figure 8","description":"","filename":"Supplfigure8FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/5d75422cf73b154923358777.png"},{"id":100964483,"identity":"0ab79782-fd78-4f9f-99c3-6c0a4a0ece28","added_by":"auto","created_at":"2026-01-23 09:07:46","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":571593,"visible":true,"origin":"","legend":"Supplemental Figure 9","description":"","filename":"Supplfigure9FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/27841c24fa12546f7b4a6036.png"},{"id":101203020,"identity":"fbde1898-7998-44d4-92f8-3d8bc2d268ac","added_by":"auto","created_at":"2026-01-27 09:38:33","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":1395046,"visible":true,"origin":"","legend":"Supplemental Figure 7","description":"","filename":"Supplfigure7FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/da36bce3a572b8cb1b5871d2.png"},{"id":101202679,"identity":"f0394819-cac8-4325-828b-f7d18482d7f0","added_by":"auto","created_at":"2026-01-27 09:37:05","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":2867552,"visible":true,"origin":"","legend":"Supplemental Figure 4","description":"","filename":"Supplfigure4FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/28d262f61d4dbcd3b3598813.png"},{"id":100964493,"identity":"867cf530-0f52-441e-93d9-1ab7cc781b43","added_by":"auto","created_at":"2026-01-23 09:07:47","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":2645493,"visible":true,"origin":"","legend":"Supplemental Figure 5","description":"","filename":"Supplfigure5FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/ad06a93f49ba3a1ad8bd5efc.png"},{"id":101203318,"identity":"0a3a7a63-0c3e-4b78-84f8-fb78d554b802","added_by":"auto","created_at":"2026-01-27 09:39:22","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":1353807,"visible":true,"origin":"","legend":"Supplemental Figure 3","description":"","filename":"Supplfigure3FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/74e1684089b87f25ff11551f.png"},{"id":101202648,"identity":"e2646dff-4809-4151-b66c-3a213c8236a7","added_by":"auto","created_at":"2026-01-27 09:36:54","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":1030428,"visible":true,"origin":"","legend":"Supplemental Figure 2","description":"","filename":"Supplfigure2FS.png","url":"https://assets-eu.researchsquare.com/files/rs-8594385/v1/1eab17040d1e28703e1368ff.png"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Distinct immune and genomic signatures predict resistance to ibrutinib therapy in Waldenström macroglobulinemia","fulltext":[{"header":"Introduction","content":"\u003cp\u003eWaldenstr\u0026ouml;m\u0026rsquo;s macroglobulinemia (WM) is a rare B-cell lymphoproliferative disorder characterized by the abnormal growth and clonal expansion of small mature B lymphocytes and plasma cells in the bone\u003c/p\u003e \u003cp\u003emarrow and lymphoid tissues,\u003csup\u003e1,2\u003c/sup\u003ewhich secrete IgM monoclonal immunoglobulin. About a decade ago, the BTK inhibitor, ibrutinib, became the first approved agent for the treatment of symptomatic WM patients\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e, and despite the unpreceded high response rates in terms of IgM reduction and clinical improvement, about one third of treated patients fail to respond\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e or show delayed responses with limited consensus regarding the substantial heterogeneity in the response type, mechanisms of resistance, and causes of treatment failure\u003csup\u003e\u003cspan additionalcitationids=\"CR8 CR9 CR10 CR11 CR12 CR13\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. One of the well-known predictive factors of ibrutinib treatment outcome, is the presence of \u003cem\u003eMYD88\u003c/em\u003e and \u003cem\u003eCXCR4\u003c/em\u003e mutations\u003csup\u003e\u003cspan additionalcitationids=\"CR16\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, with patients with \u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e genotype have very low probability of major response, while patients harboring both \u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eL265P\u003c/em\u003e\u003c/sup\u003e and \u003cem\u003eCXCR4\u003c/em\u003e mutations exhibit inferior response rates\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Additionally, almost half of WM patients who progress on ibrutinib, display \u003cem\u003eBTK\u003c/em\u003e mutations at the binding site of ibrutinib or its downstream mediator \u003cem\u003ePLCg2\u003c/em\u003e\u003csup\u003e5,6\u003c/sup\u003e while differential gene expression of several genes as well as acquired mutations in chromosomes 6q and 8p has also been linked to ibrutinib resistance\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThese observations suggest that ibrutinib\u0026rsquo;s therapeutic effects may extend beyond direct tumor burden reduction, highlighting the need to elucidate the broader genomic and immune mechanisms. In this study, we investigate the complex interplay between tumour-acquired genetic features and the composition of the microenvironment in the promotion of ibrutinib resistance in WM patients by integrating single-cell RNA sequencing (scRNAseq) and whole-genome sequencing (WGS), to examine bone marrow tumour and immune compartments from 37 symptomatic WM patients treated with ibrutinib, alongside with a validation cohort of 47 WM patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Using this approach, we were able to define biological networks behind the variable clinical responses seen in patients treated with ibrutinib and the impact of BTK inhibition on the crosstalk between WM clones and the surrounding microenvironment.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy population\u003c/h2\u003e \u003cp\u003eThe study included 37 consecutive consenting patients with previously untreated WM who received ibrutinib therapy, as per the standard clinical practice. After a median follow-up of 1.7 years (range 0.58\u0026ndash;4.08 years), 7 (19%) patients achieved very good partial response (VGPR); 17 (45%) achieved partial response (PR), 4 (11%) patients achieved minimum response (MR) whereas 9 (25%) patients had stable or progressive disease (SD and PD) after 6 months of ibrutinib therapy. Patients achieving PR or better after 6 months of therapy were defined as the responder group (RG) while patients with less than PR were defined as the non-responder group (NRG), so that 24 patients were included in the RG and 13 patients in the NRG. In the NRG cohort, 4 out of 13 patients (30%) ultimately achieved a PR to ibrutinib with a median time to response of 3.1 years (range 1.5\u0026ndash;5.2 years) while the remaining 9 patients did not improve their response. The key clinical and genetic features of the RG and the NRG are summarized in Suppl. Tables\u0026nbsp;1 and 2 and Suppl. Figure\u0026nbsp;1.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eThe landscape of bone marrow mononuclear cells in WM at single cell resolution\u003c/h3\u003e\n\u003cp\u003eWe performed scRNAseq on 37 paired bone marrow samples at diagnosis (T1) and at 6 months (T2) after start of ibrutinib treatment (74 total). In addition, scRNAseq analysis was performed on samples from 5 patients at 12 months post start of ibrutinib (T3) and 1 patient at relapse 24-month post therapy (T4) as well as on two healthy donors. Upon preprocessing, alignment and quality control steps including removal of the low-quality reads and cells, a total of 348,000 high-quality bone marrow mononuclear cells (BMMCs) were analyzed. We also integrated publicly available normal bone marrow single cell data set, annotated as reference (~\u0026thinsp;300,000 cells), to enhance our analysis for comparison among healthy individuals and patients. We generated a BM cells atlas, and by using Uniform Manifold Approximation and Projection (UMAP) approach, we were able to resolve 13 distinct cell clusters which consisted 32% of T-cells, 26% of B-cells, 12% of Monocytes ,6% of NK-cells and 1% of plasma cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, B, Suppl Fig.\u0026nbsp;2A). Ibrutinib treatment in the entire patient cohort led to a significant reduction in the B cell (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01) and plasma cell (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.01) populations, while classical CD14\u003csup\u003e+\u003c/sup\u003e monocytes (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01), NK/T cells (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01) and classical dendritic cells (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01) were significantly increased (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). This finding demonstrates that, beyond its direct effect on the malignant cells, the treatment also exerts profound effects on the immune compartment. Focusing further on the two response groups, we observed that at T1, the B-cell compartment was markedly enriched in the NRG, whereas progenitor and T cells were more enriched in the RG. By T2, B-cells significantly decreased in both groups (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01), while monocytes were significantly increased in the RG only (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD and E, Suppl Fig.\u0026nbsp;2B). Furthermore, although the number of clonal plasma cells at diagnosis was similar between both groups, a significant reduction at T2 was observed only in the RG (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.04, Suppl Fig.\u0026nbsp;2B).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eThe contribution of immune subpopulations to the ibrutinib resistance\u003c/h3\u003e\n\u003cp\u003eFocusing on the myeloid compartment, we identified four myeloid subpopulations including CD14⁺ classical and CD16⁺ non-classical monocytes, conventional dendritic cells (cDCs, CD1C⁺), and a smaller fraction of monocyte activated platelets (MAP) cells (PPBP) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA) which were further subdivided into seven transcriptionally distinct clusters (C0\u0026ndash;C6, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Classical CD14\u003csup\u003e+\u003c/sup\u003e monocytes were characterized by robust expression of M1-like pro-inflammatory mediators (\u003cem\u003eS100A8\u003c/em\u003e, \u003cem\u003eIL1B\u003c/em\u003e, \u003cem\u003eCXCL2\u003c/em\u003e) and signaling molecules (\u003cem\u003eIRAK2\u003c/em\u003e) and exhibited enriched interferon (IFN)-stimulated gene (\u003cem\u003eISG15\u003c/em\u003e) expression, indicative of the induction of a highly inflammatory transcriptional program (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). Assessing the relative abundance of monocytic states at T1 in both groups compared to the reference, we observed that CD16\u003csup\u003e+\u003c/sup\u003e/FCGR3A (C2) and the inflammatory IFN activated monocyte state (C4), were massively increased in all WM patients (Suppl. Figure\u0026nbsp;3A). Although these populations were homogenously present within the monocyte compartment at T1 in both response groups, at T2 a significant induction of the CD14\u003csup\u003e+\u003c/sup\u003e monocytes was observed in both groups (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01) accompanied with a decrease of the CD16\u003csup\u003e+\u003c/sup\u003e subpopulation (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). An increase was observed at T2 in cluster C0 (\u003cem\u003eIRAK2\u003c/em\u003e and \u003cem\u003eCXCL2\u003c/em\u003e) in the NRG group compared to RG (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.16 vs \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.49, log2Fc\u0026thinsp;=\u0026thinsp;0.8 vs 0.27,Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE) which was further evidenced by the increased inflammatory pathway activity score, produced by the expression of 29 genes including \u003cem\u003eCXCL8\u003c/em\u003e, \u003cem\u003eIL6\u003c/em\u003e, \u003cem\u003eIL4\u003c/em\u003e, \u003cem\u003eTGFB2\u003c/em\u003e observed in the NRG (Suppl. Figure\u0026nbsp;3B). Conversely, significant expansion of IFN\u003csup\u003e+\u003c/sup\u003e monocytes (C4) was observed at T2 only in the RG (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), potentially indicating an enhanced activation of the innate immune response that may be linked to ibrutinib efficacy (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). Additionally, our analysis identified a neutrophil population with similar expression patterns between RG and NRG at T1 (Suppl. Figure\u0026nbsp;3C) however, at T2, a cluster characterized by genes such as \u003cem\u003eMMP8\u003c/em\u003e and \u003cem\u003eMMP9\u003c/em\u003e was more enriched in the RG group (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.3, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG, Suppl. Figure\u0026nbsp;3D). Overall, gene set enrichment analysis (GSEA) revealed that at T2 pathways including TNF and NF-κb signaling, inflammation and IFNγ response were enriched driven by the upregulation of genes including \u003cem\u003eCCL3\u003c/em\u003e, \u003cem\u003eCCL4\u003c/em\u003e, \u003cem\u003eCXCL8\u003c/em\u003e, \u003cem\u003eNFKBIZ\u003c/em\u003e and \u003cem\u003eNFKBIA\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFocusing on the T cell compartment, we resolved 9 distinct subpopulations including both naive and memory subsets of CD4\u003csup\u003e+\u003c/sup\u003e and CD8\u003csup\u003e+\u003c/sup\u003e, regulatory T (Tregs), MAIT, and gamma-delta (γδ) T-cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). GZMB CD8\u003csup\u003e+\u003c/sup\u003e, memory CD4\u003csup\u003e+\u003c/sup\u003e T, and Tregs were markedly increased in patients compared to the reference (Suppl. Figure\u0026nbsp;4A). At T1, an enrichment of GZMB⁺ CD8⁺ cells (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.07) was seen in the NRG combined with a significantly reduced expression of naive CD4⁺ T-cell frequencies (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01) compared to RG. Trajectory analysis also revealed an accumulation of GZMK⁺, GZMB⁺ effector, and γδ cells in NRG in the terminal region of the pseudotime trajectory, suggesting a shift toward terminal effector states (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). At T2, we observed an expansion of GZMK⁺ CD8⁺ cells in both groups, with more pronounced effects in the NRG (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01), highlighting the limited cytotoxicity observed in this group. Concurrently, CD4\u003csup\u003e+\u003c/sup\u003e naive cells were significantly reduced only in the RG possibly indicating that ibrutinib treatment leads to a global shift in the T cell compartment towards a more effector-like and functionally engaged immune response (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). Furthermore, as opposed to NRG, the RG exhibited a marked suppression of Tregs after therapy (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02), possibly reflecting the more favorable outcome observed in this group leading to higher cytotoxic T-cell responses. AUCell scoring based on a curated set of T-cell exhaustion-associated genes including \u003cem\u003eCTLA4\u003c/em\u003e, \u003cem\u003ePDCD1\u003c/em\u003e, \u003cem\u003eLAG3\u003c/em\u003e, \u003cem\u003eTIGIT\u003c/em\u003e, \u003cem\u003ePD1\u003c/em\u003e, \u003cem\u003eHAVCR2\u003c/em\u003e demonstrated that although at T1, NRG displayed a slightly lower score compared to RG, after treatment a significant reduction of this signature was observed in the RG only (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.01), suggesting that the NRG maintained the terminally differentiated and dysfunctional T-cell profile even after treatment (Suppl. Figure\u0026nbsp;4B).Functional analysis of the IFNγ response revealed that activity score of this pathway was markedly enriched in NRG, further highlighting the higher inflammatory state of these patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFinally, sub-clustering of the NK-cell compartment resolved five clusters; a progenitor-like CD56\u003csub\u003ebright\u003c/sub\u003e \u003cem\u003eSELL\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e associated with lower toxic functions, an activated FOS\u003csup\u003e+\u003c/sup\u003e CD56\u003csub\u003ebright\u003c/sub\u003e, a CD160 memory-like regulatory CD56\u003csub\u003ebright\u003c/sub\u003e, a transitional/ early cytotoxic CD56\u003csub\u003edim\u003c/sub\u003e \u003cem\u003eFGFBP2\u003c/em\u003e\u003csup\u003e+\u003c/sup\u003e and an NKT-like (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE). At T1, almost all subpopulations were evenly distributed among the two response groups with a minimal enrichment of the \u003cem\u003eSELL\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e cluster in the NRG and \u003cem\u003eCD160\u003c/em\u003e\u003csup\u003e\u003cem\u003e+\u003c/em\u003e\u003c/sup\u003e in the RG. Ibrutinib treatment resulted in a marked enrichment of regulatory CD56\u003csub\u003ebright\u003c/sub\u003e subsets, particularly CD160\u003csup\u003e+\u003c/sup\u003e (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02 in NRG and \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.06 in RG) and \u003cem\u003eFOS⁺\u003c/em\u003e NKs (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.14 in NRG), suggesting a shift toward a less cytotoxic, less activated NK-cell phenotype especially in the NRG (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF). Furthermore, functional analysis showed a marked suppression of the NK cytotoxicity pathway during treatment, observed mainly in the NRG, indicating that successful response to ibrutinib is potentially linked to continued NK cell cytotoxic activity (p\u0026thinsp;=\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG). Furthermore, the ratio of CD56\u003csub\u003ebright\u003c/sub\u003e/CD56\u003csub\u003edim\u003c/sub\u003e NK cells increased between T1 and T2 in both groups, with more prominent effects in NRG (0.025% (0\u0026ndash;0.4) versus 0.31% (0\u0026ndash;2), \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH).\u003c/p\u003e \u003cp\u003e \u003cb\u003eThe transcriptomic features of clonal B compartment and the impact of distinct B-cell molecular subtypes on ibrutinib response\u003c/b\u003e \u003c/p\u003e \u003cp\u003eFocusing on the transcriptomic features of clonal B-cells (cBc) and their response to ibrutinib, we initially observed that most of the polyclonal cells of the patient samples clustered together while cBc created individual clusters each of which originated from a single patient. Furthermore, while \u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e patients tended to cluster more closely together, neither \u003cem\u003eCXCR4\u003c/em\u003e mutation status nor response to ibrutinib revealed consistent transcriptional patterns across different patient groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, Suppl. Figure\u0026nbsp;5A). As expected, memory B cells constituted the majority of the cBc in both groups while a small proportion of pro-B cells was observed only in the RG (Suppl. 5B). Correlation of patients BM infiltration undergoing ibrutinib treatment with cBc, we observed that at both T1 and T2, the frequency of cBc was significantly higher in the NRG (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02, Suppl. Figure\u0026nbsp;5C) while ibrutinib led to a significant reduction of this compartment in both groups (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). The presence of \u003cem\u003eMYD88\u003c/em\u003e and \u003cem\u003eCXCR4\u003c/em\u003e mutations had an impact on the response to ibrutinib, with a significant reduction of cBc observed in \u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e and \u003cem\u003eCXCR4\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e patients (Suppl. Figure\u0026nbsp;5D). Importantly, a significant upregulation of \u003cem\u003eCXCR4\u003c/em\u003e expression was observed in the NRG at T2 (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.01) and a significant reduction in \u003cem\u003eMYD88\u003c/em\u003e expression in the RG (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.01, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). Differential gene expression (DEG) analysis among response groups, showed increased expression of \u003cem\u003eMAST4\u003c/em\u003e, \u003cem\u003eXIST\u003c/em\u003e, \u003cem\u003ePTK2\u003c/em\u003e, \u003cem\u003eBLK\u003c/em\u003e and \u003cem\u003eMAP4K4\u003c/em\u003e genes at T1 and higher expression levels of genes such as \u003cem\u003ePRDM4\u003c/em\u003e, \u003cem\u003ePCDH9\u003c/em\u003e, \u003cem\u003eIRF4\u003c/em\u003e, and \u003cem\u003eMAPK8\u003c/em\u003e at T2 (Suppl. 5E-F, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). Interestingly, a subset of NRG patients who eventually responded to ibrutinib after 12 months of treatment (4/13 patients, NRtoR) exhibited a transcriptomic profile at T2 (key genes \u003cem\u003eXBP1\u003c/em\u003e, \u003cem\u003eCD9\u003c/em\u003e, \u003cem\u003eMZB1\u003c/em\u003e, \u003cem\u003eNFKBIA\u003c/em\u003e, \u003cem\u003eCXCR4\u003c/em\u003e and \u003cem\u003eJCHAIN)\u003c/em\u003e that highly overlapped genes observed in pretreated RG patients while patients with sustained resistance (NR to NR) showed no such convergence, suggesting that delayed responders may acquire RG-like transcriptional features over time. (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). Functional analysis indicated that IL6 activity at T2 was significantly reduced in the RG only (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.03) possibly suggesting that this activity remains prolonged in the NRG (Suppl. Suppl. 5G). Furthermore, CNV profiling revealed chromosomal amplifications of chromosomes 12 and 18 observed exclusively in the NRG (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.03 for chr12, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.11 for chr18), while deletion of chromosome 6q was equally distributed in both groups (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.7) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe next sought to determine whether our cohort of patients exhibited distinct B cell subtypes with memory B-cell (MBC) like or plasma-cell (PC) like features\u003csup\u003e\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Within our cohort, 12 out of 24 patients in the RG (50%) and 8 out of 13 in the NRG (61%) exhibited MBC-like features, while the remaining patients in both groups were classified as PC-like (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF, G, Suppl. Figure\u0026nbsp;6A); however no significant differences in subtype distribution among the response groups. Interestingly our results showed that all \u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e patients (n\u0026thinsp;=\u0026thinsp;6) displayed the MBC-like subtype which were also classified as NRG. At diagnosis, the NRG displayed a significantly higher number of cBc in this MBC subtype compared to the RG (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02) while PC subtype was similar between both groups. At T2, both groups displayed a marked reduction in the MBC-like signature (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.06 and p\u0026thinsp;=\u0026thinsp;0.01, respectively) although the effects were more pronounced in the NRG, possibly indicating that transcriptional changes in the MBC-like signature act independently of ibrutinib response (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eH). Conversely, the PC-like signature was notably reduced in the RG (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.11) while remained unchanged in the NRG. Overall, no effects in PFS were observed between the groups (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.77, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eI).\u003c/p\u003e\n\u003ch3\u003eIdentification of gene expression signatures associated with ibrutinib resistance\u003c/h3\u003e\n\u003cp\u003eTo further address the significant heterogeneity observed between the two response groups, we applied Bayesian non-negative matrix factorization on the cBc, utilizing 2,000 highly variable genes based on the \u003cem\u003eMYD88\u003c/em\u003e and \u003cem\u003eCXCR4\u003c/em\u003e mutation status. Four gene expression signatures (GEX1-GEX4) were identified with key marker genes including \u003cem\u003eLTB\u003c/em\u003e and \u003cem\u003eMARCKS\u003c/em\u003e in GEX1, \u003cem\u003eDUSP2\u003c/em\u003e and \u003cem\u003eTNFAIP3\u003c/em\u003e in GEX2, \u003cem\u003eCD9\u003c/em\u003e and \u003cem\u003eCHST15\u003c/em\u003e in GEX3, and \u003cem\u003eMS4A1\u003c/em\u003e and \u003cem\u003eNOTCH2\u003c/em\u003e in GEX4, each of which affected different signaling pathways (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA, Suppl. Figure\u0026nbsp;7A-B). GEX3 (\u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e, \u003cem\u003eCXCR4\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e, n\u0026thinsp;=\u0026thinsp;14) was predominantly found in the RG, GEX4 (\u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e, \u003cem\u003eCXCR4\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e, n\u0026thinsp;=\u0026thinsp;6) was primarily found in the NRG while GEX2 (\u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e, \u003cem\u003eCXCR4\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e) belonged to the NRG (one patient). Focusing on the GEX1 (\u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e, \u003cem\u003eCXCR4\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e, n\u0026thinsp;=\u0026thinsp;14), we noticed that this was found in 6 NRG and 8 RG patients. At diagnosis, genes within the GEX1 signature overexpressed in RG included \u003cem\u003eEZR\u003c/em\u003e and \u003cem\u003eDUSP1\u003c/em\u003e (MAPK pathway), \u003cem\u003eSSR4\u003c/em\u003e and \u003cem\u003eMZB1\u003c/em\u003e (ER processing), \u003cem\u003eHIFX\u003c/em\u003e (hypoxia response) and \u003cem\u003eMSI2\u003c/em\u003e (WNT signaling) while genes overexpressed in NRG included \u003cem\u003eAFF3\u003c/em\u003e and \u003cem\u003eBACH2\u003c/em\u003e (BCR signaling), \u003cem\u003eHLA-DQB1\u003c/em\u003e and \u003cem\u003eHLA-DRB5\u003c/em\u003e (antigen presenting) and \u003cem\u003eLTB\u003c/em\u003e (NF-κB/TNF signaling). Upon treatment, genes such \u003cem\u003eDUSP2 (\u003c/em\u003eMAPK pathway), \u003cem\u003eRHOH\u003c/em\u003e (T cell receptor signaling), \u003cem\u003eNFKBIA (\u003c/em\u003eNF-κB signaling), \u003cem\u003eNR4A1\u003c/em\u003e (immune regulation) and \u003cem\u003eJCHAIN\u003c/em\u003e were enriched in NRG (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). These findings suggest that in non-responding patients with \u003cem\u003eCXCR4\u003c/em\u003e mutations, resistance to ibrutinib is influenced by genes that primarily affect BCR signaling and immune regulation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo further validate the above signatures and evaluate their clinical significance, we analyzed bulk RNA-seq data from an external cohort of 47 patients\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e before the initiation of ibrutinib, from the Dana Farber Cancer Institute (DFCI, Suppl. Table\u0026nbsp;3). Within the external cohort, 12 patients had disease progression to ibrutinib (25%). In line with our results from the initial cohort, 83% of patients (20/24) with \u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e/ \u003cem\u003eCXCR4\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e genotype were associated with the GEX3 signature while 82% of patients (19/23) with \u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e/ \u003cem\u003eCXCR4\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e were associated with the GEX1 signature. Interestingly, about 50% of patients (12/23) within the GEX1 signature also shared features of the GEX2 signature and 42% (5/12) of the progressed patients displayed the combined GEX1/GEX2 signature (Suppl. Figure\u0026nbsp;7C). The remaining 58% of the progressed patients (7/12) displayed features of the GEX3 signature meaning that almost 25% of patients with \u003cem\u003eCXCR4\u003c/em\u003e\u003csup\u003e\u003cem\u003eWT\u003c/em\u003e\u003c/sup\u003e genotype progressed to ibrutinib therapy.\u003c/p\u003e \u003cp\u003eTo gain further insights into the transcriptional evolution and resistance mechanisms of residual cBc, we conducted unsupervised clustering of all B cells from each patient individually in the initial cohort. Within the NRG, we observe an increase of cBc after treatment in the RG. Comparing the dynamic shifts of the cBc at T1 and T2 we found that the dominant clone at T1 remained the most prevalent after treatment in all patients, regardless of response group. In 2 of the 37 patients, we identified the presence of two separate clones: one with different IgHV rearrangements (clonotype 1 and 2) classified as RG and another with independent kappa and lambda light chain clones, classified as NRG. Focusing on the patient with the 2 clonotypes, we observed two patterns of transcriptional signatures: one resembling the GEX3 signature (major clone, clonotype 1) and one to the GEX2-like (minor clone, clonotype 2). Interestingly, at T2, clonotype 1 decreased by nearly 25%, while clonotype 2-GEX2-like clone showed a simultaneous increase, possibly indicating the gradual expansion of a resistant clone gradually increasing over time. This clone was mainly characterized by the abundant expression of \u003cem\u003eLTB\u003c/em\u003e, \u003cem\u003eNFKBIA\u003c/em\u003e, DUSP2, NR4A1 and \u003cem\u003eNR4A2\u003c/em\u003e genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). In fact, the \u003cem\u003eLTB\u003c/em\u003e and \u003cem\u003eNFKBIA\u003c/em\u003e genes appear to be highly expressed in patients whose cBc increase after treatment, suggesting their potential involvement in mechanisms of ibrutinib resistance.\u003c/p\u003e \u003cp\u003eFinally, to determine whether transcriptomic features at diagnosis could predict long-term outcomes, we conducted a PFS analysis. The combination of high GEX1/GEX2 expression score resulted in shorter PFS (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.066) while significant shorter PFS was observed in patients with low GEX3 (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0035) and high GEX4 expression score (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0046, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). The results from the external cohort validated the impact of the different expression patterns in PFS, with the combination of high GEX1/ GEX2 expression score resulting in a shorter PFS (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.018) while low GEX3 had an almost significant impact on PFS (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.05, Suppl. Figure\u0026nbsp;7D). Specifically, high expression of \u003cem\u003eLTB\u003c/em\u003e was significantly associated with shorter PFS (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.043). Furthermore, we developed a classifier using an Elastic Net model for each response group, where genes such as \u003cem\u003eJUN\u003c/em\u003e, \u003cem\u003ePRDM5\u003c/em\u003e, \u003cem\u003eHS2ST1\u003c/em\u003e, \u003cem\u003eBCL7A\u003c/em\u003e, \u003cem\u003eLTB\u003c/em\u003e and others had the highest importance scores for the model (Suppl. Table\u0026nbsp;4) with significant impact in PFS (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001, log-rank test, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE) which was also observed in the validation cohort (PFS, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.038, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF).\u003c/p\u003e\n\u003ch3\u003eThe interactome landscape underlying ibrutinib responsiveness and the role of immune microenvironment in predicting outcome\u003c/h3\u003e\n\u003cp\u003eTo investigate how the TME influences the functional behavior of cBc in patients with varying responses to ibrutinib, we analyzed cell-cell interactions between the cBC population and the TME. Our findings revealed that cBc from all patients at diagnosis were able to send signals to specific cell compartments including na\u0026iuml;ve, memory and cytotoxic T cells, NK cells, MAITs and pDCs. Of particular interest was the CD74:CD44 (macrophage inhibitor factor, MIF) cell interaction involved in many processes (ERK1/ERK2, TLR4, p53 apoptosis) \u003csup\u003e\u003cspan additionalcitationids=\"CR26 CR27\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e showing higher cell interaction in the RG at diagnosis compared to the NRG (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA, Suppl Fig.\u0026nbsp;8); however, upon treatment this interaction was reduced in RG and upregulated in the NRG, possibly suggesting a higher dependency of this interaction in the RG which is disrupted upon ibrutinib therapy resulting in a better restoration of immune balance in this group of patients. We also observe a broad spectrum of HLA-related genes:CD44 ligand-receptor interaction between pDCs and cBc at T1 in both groups, however at T2 this interaction is disrupted and enhanced between other cell types including classical and non-classical monocytes, na\u0026iuml;ve and memory T cells. These data support the presence B cell proliferative elements driven by different cellular interactions at diagnosis, and their disruption upon treatment with marked effects in the RG.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFinally, we assessed the ability to predict ibrutinib response by integrating cBc and immune cell compartments in a multivariable framework by employing a receiver operative characteristic (ROC) analysis providing an AUC score which stratifies patients in high- and low-risk groups. Incorporation of cBc alone yielded an AUC value of 0.72 and was associated with shorter PFS (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.078, Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB-C). Lower T-cell proportion also correlated with significantly shorter PFS (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.022, Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD). Importantly, incorporating the most predictive T-cell subsets (na\u0026iuml;ve CD4 and CD8 cells and GZMB) enhanced prediction accuracy (AUC\u0026thinsp;=\u0026thinsp;0.87) where high B cell levels combined with low T-cell proportions were associated with shorter PFS (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.022, Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD). On the other hand, patients with low levels of cBc and high levels of T-cells had the best outcome. Finally, the incorporation of specific monocyte compartments increased the AUC score to 0.944 (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB). This data further highlights the importance of the immune microenvironment in partially determining the depth of ibrutinib response and the significant advantage of these integrative models.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eImpact of the mutational profile of the B cell clone in ibrutinib resistance\u003c/h2\u003e \u003cp\u003eTo determine the mutational profile of two response groups, WGS was performed in 10 RG and 8 NRG patients at both timepoints. This analysis revealed a similar mutational burden in the NRG and RG which was not affected by ibrutinib therapy (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA,B). Focusing on genes recurrently mutated in WM and other lymphomas, \u003csup\u003e29\u0026ndash;31\u003c/sup\u003e we found that the most frequently mutated genes in RG included \u003cem\u003eMYD88\u003c/em\u003e (90%), \u003cem\u003eCXCR4\u003c/em\u003e (20%), \u003cem\u003eCAST\u003c/em\u003e (25%, ERK1/2 signalling), \u003cem\u003eTP53\u003c/em\u003e (10%) and \u003cem\u003eCD79B\u003c/em\u003e (10%) while in the NRG included \u003cem\u003eMYD88\u003c/em\u003e (62%), \u003cem\u003eCXCR4\u003c/em\u003e (38%), \u003cem\u003eIGLL5\u003c/em\u003e (38%), \u003cem\u003eKMT2D\u003c/em\u003e (38%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eC, Suppl. Figure\u0026nbsp;9A ). In line with the scRNA-seq results, CNV profiling revealed similar differences between the RG and NRG (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eD). Clonal evolution analysis showed a branching evolution in 3 of 17 patients (17%), with 2 cases in the NRG and 1 in the RG who ultimately progressed after 12 months of therapy (RtoNR) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eE, Suppl. Figure\u0026nbsp;9B). Most of the acquired alterations at T2 included genes such as \u003cem\u003eIL10R\u003c/em\u003e, \u003cem\u003eRUNX1\u003c/em\u003e, \u003cem\u003eFAT3\u003c/em\u003e, \u003cem\u003eIRF4\u003c/em\u003e, \u003cem\u003eRBM6\u003c/em\u003e, \u003cem\u003eRHOT1\u003c/em\u003e some of which have been previously identified by our group\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFinally, consistent with our prior findings\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e, mutational signature analysis identified four main single base substitution (SBS) signatures: clock-like aging signatures (SBS1 and SBS5), germinal-centre-associated polymerase eta (POLH) somatic hypermutation (SBS9), and SBS8 (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eF). However in contrast to previous studies\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e, we also found the presence of the APOBEC mutational activity (SBS13) in about 65% of patients (11/16) BUT with a very low contribution (3%), which also appeared to be more enriched in males compared to females (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.16). Furthermore, significantly higher levels of SBS1 and SBS5 signatures were observed in the NRG group at both time points compared to the RG group (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02),which also had an impact on the PFS (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.00059), a pattern which has also been observed in a subset of high-risk multiple myeloma patients\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eDespite the marked clinical benefit of ibrutinib therapy in WM, the underlying mechanisms that contribute to the heterogeneity of patient responses remain poorly understood. In this study, we performed an in-depth analysis of both the tumor and immune microenvironment transcriptome in WM patients before and after ibrutinib treatment. Our findings reveal a distinct tumor-intrinsic transcriptional signature associated with treatment response, which is also linked to changes in the TME, suggesting that this interplay may significantly influence therapeutic efficacy and patient outcomes (Fig.\u0026nbsp;9).\u003c/p\u003e \u003cp\u003eWe demonstrated highly enriched terminally differentiated GZMB CD8\u003csup\u003e+\u003c/sup\u003e T-cells and SELL-expressing NK cells in the NRG, indicating an expansion of these subpopulations with increasing disease severity possibly affecting the interaction of WM cells with the immune system and the optimum efficacy of ibrutinib therapy. In fact, previous studies have shown that WM patients with high tumor infiltration exhibit higher levels of exhausted CD8\u003csup\u003e+\u003c/sup\u003e T-cells compared to healthy donors\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. In the myeloid compartment, we observed the enrichment of the IFN\u003csup\u003e+\u003c/sup\u003e monocytes in the NRG, possibly promoting a tumor pro-survival microenvironment with a BTK-independent ecosystem which could contribute to WM disease progression\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. On the other hand, responders display a more functionally competent T-cell repertoire through the expression of na\u0026iuml;ve CD4\u003csup\u003e+\u003c/sup\u003e and CD8\u003csup\u003e+\u003c/sup\u003e T-cells. Therefore the overall combined phenotype of expanded T-cell effector population expressing low levels of cytotoxicity-related genes along with the depletion of na\u0026iuml;ve T cell states which has been shown to reduce the TCR clonality\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e, is possibly associated with poorer outcomes and this has been described in several other malignancies\u003csup\u003e\u003cspan additionalcitationids=\"CR38\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Furthermore, these effects, possibly contribute to a higher inflammatory microenvironment through the enrichment of IFN\u003csup\u003e+\u003c/sup\u003e monocytes which could play a role in the treatment failure in non-responding patients. Following treatment the significant reduction of the Tregs only observed in the responding patients, possibly suggests that the ibrutinib\u0026rsquo;s efficacy is potentially linked to the restoration of T-cell fitness\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. Conversely, while both response groups exhibited elevated T-cell exhaustion scores compared to the reference before the start of therapy, the responder\u0026rsquo;s group showed a greater reduction upon treatment, potentially reflecting a more robust recovery of T-cell function compared to non-responders. Furthermore, a more pronounced expansion of the pro-inflammatory M1-like monocytic population was observed in NRG after treatment, suggesting a shift towards a more pro-inflammatory state and an IRAK-NFκB driven signaling that could sustain tumor support BTK-independent manner. Recent studies have focused on inflammatory WM (iWM) which is present in one-third of patients\u003csup\u003e\u003cspan additionalcitationids=\"CR43\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e and notably in the context of BTK inhibitor therapy, the inflammatory syndrome decreased during the hematological response in BTKi-treated iWM\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWhen characterizing tumor-intrinsic mechanisms among the two response groups we observed significant intra-tumor heterogeneity, which partly explains the differential responses seen with ibrutinib therapy. We initially demonstrated that resistance to ibrutinib by the disease persistence, was associated with a higher cBc burden at diagnosis in the non-responders primarily driven by elevated expression of genes regulating BCR signaling (\u003cem\u003eIKZF3\u003c/em\u003e\u003csup\u003e46,47\u003c/sup\u003e, \u003cem\u003eSYK\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e,\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e), chromatin modification (\u003cem\u003eXIST\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e, \u003cem\u003eHDAC9\u003c/em\u003e\u003csup\u003e51\u003c/sup\u003e), and antigen presentation (\u003cem\u003eHLA-DQA1\u003c/em\u003e\u003csup\u003e52\u003c/sup\u003e). Ibrutinib treatment led to distinct transcriptional evolution between the two response groups, with only the RG showing enrichment in apoptotic and negative regulation of the p38/MAPK pathways, while NRG retained a transcriptional program resembling the pre-treatment RG states. Classification of patients into the previously described\u003csup\u003e\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e MBC-like and PC-like molecular subtypes demonstrated that, although all patients could be assigned to one of these subtypes, response to ibrutinib was independent of the subtype classification. Overall, ibrutinib treatment resulted in a markedly greater reduction of cBc in the MBC-like group in both groups while reduction of cBc in the PC-like subtype was only seen in the RG, possibly suggesting that a more resistant clone could be linked with PC-like features.\u003c/p\u003e \u003cp\u003eTo further uncover the tumor-intrinsic mechanisms of ibrutinib resistance, we defined gene expression signatures of cBc classified by \u003cem\u003eMYD88\u003c/em\u003e and \u003cem\u003eCXCR4\u003c/em\u003e mutation status. GEX1 (\u003cem\u003eMYD88\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e, \u003cem\u003eCXCR4\u003c/em\u003e\u003csup\u003e\u003cem\u003eMUT\u003c/em\u003e\u003c/sup\u003e) was similarly distributed among the RG and the NRG patients, where genes overexpressed in RG were linked to JNK/MAPK, ER processing, and WNT signaling, while in NRG were involved in BCR, antigen presentation, and TNF/NF-κB signaling. Importantly, most of these findings were also validated with the use of bulk RNA-seq data from an external cohort of 47 patients, which highlighted the significance of genes mainly involved in GEX1-GEX3 expression signatures and their potential prognostic value in predicting ibrutinib response. Furthermore, clonal transcriptional evolution in each patient highlighted resistance mechanisms closely linked to the GEX2 signature, notably through \u003cem\u003eLTB, NFKBIA, DUSP2, NR4A1\u003c/em\u003e, and \u003cem\u003eNR4A2\u003c/em\u003e genes mainly involved in TNFα/NFκB signaling pathways. Most of these genes have been previously associated with other lymphomas and multiple myeloma (MM)\u003csup\u003e\u003cspan additionalcitationids=\"CR54 CR55 CR56 CR57\" citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e. To further understand how the transcriptome functions in the context of the clinical outcome, we identified several genes, including \u003cem\u003eLTB\u003c/em\u003e, that had a significant impact in PFS. High GEX1/GEX2 scores were associated with shorter PFS, while low GEX3 significantly correlated with shorter PFS both in the training and the validation cohort. Additionally, the Elastic Net classifier identified key genes like \u003cem\u003eIRF4\u003c/em\u003e, \u003cem\u003eNR4A1\u003c/em\u003e, \u003cem\u003eMYC\u003c/em\u003e, \u003cem\u003eBCL7A\u003c/em\u003e, and \u003cem\u003eLTB\u003c/em\u003e that contributed most to ibrutinib resistance prediction with significant impact on PFS which was also further validated in the external cohort of patients confirming their essential effect on PFS. Further mutational analysis showed that NRG displayed more mutations in genes such as \u003cem\u003eCXCR4\u003c/em\u003e, \u003cem\u003eKMT2D\u003c/em\u003e, \u003cem\u003eIGLL5\u003c/em\u003e,and \u003cem\u003eARID1A\u003c/em\u003e, also previously described by our group\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. Additionally, we observed for the first time the presence of the APOBEC signature in WM patients (although with a relatively low contribution) while the age-related mutational signatures SBS1/SBS5 were significantly enriched in the NRG group, which may reflect an underlying mechanism of resistance consistent with the patients\u0026rsquo; clinical profile, a finding also described in several cancers\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCell- cell communication identified MIF (CD74:CD44) as the major regulator among B- and T-cells in RG and HLA-related genes among B-cells and T/NK-cells in NRG. MIF transcription is regulated by the BCR activity and activates pathways including ERK1/2, MAPK and NFκB however after successful BTK inhibition the normal BCR-driven regulatory network that controls MIF production is disrupted. Hence, although MIF is disrupted in RG after therapy, is further induced in NRG possibly indicating that the cells in this group of patients are activated by alternative pathways. The induction of MIF in NRG after therapy could also suggest that ERK1/2, MAPK and NFκB are reactivated which will eventually lead to successful BTK inhibition.\u003c/p\u003e \u003cp\u003eFinally, we observed that by combining TME immune signatures with the proportion of clonal B-cells could better predict the ibrutinib efficacy outcomes, providing an accurate prediction of responders and non-responders.\u003c/p\u003e \u003cp\u003eOverall, this study offers important insights into the mechanisms driving resistance to BTKi based therapies in WM, such as ibrutinib, and highlights the value of integrated multi-omic strategies in guiding optimal treatment strategies informed by the underlying molecular profiles.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePatients and sample processing\u003c/h2\u003e \u003cp\u003eBone marrow aspirates were collected from 37 newly diagnosed symptomatic WM patients at 3 time points, (i) diagnosis, (ii) 6 months and (iii) 12 months upon ibrutinib therapy initiation. Two healthy donors were also included. Prior to any study procedure, all participants of this study provided written informed consent. The research was approved by the Institutional Review Board (IRB) and Scientific committee of \u0026ldquo;Alexandra\u0026rdquo; Hospital (nos. 56463 and 27). All research was conducted in accordance with the Declaration of Helsinki. For the single cell analysis, frozen bone marrow mononuclear cells (BMMCs) were thawed in phosphate buffered saline (PBS) with 0.01% bovine serum albumin (BSA). For whole genome sequencing (WGS), lymphoplasmacytic cells from BM aspirates were collected by CD19\u003csup\u003e+\u003c/sup\u003e Auto-MACS micro-bead selection (Miltenyi-Biotech) as previously described\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003cb\u003e10x single cell RNA sequencing\u003c/b\u003e \u003c/p\u003e \u003cp\u003eSingle-cell 5\u0026rsquo; RNA and V(D)J) libraries were generated using the Chromium Next GEM Single Cell 5' Reagent Kits v2 (Dual Index) (10X Genomics) following the manufacturer\u0026rsquo;s protocol. Briefly, BMMCs were loaded onto the Chromium Next GEM Chip K followed by gel beads in emulsion (GEM) generation on the Chromium X. GEM-RT reactions were performed followed by breakage of the GEMs and cDNA cleanup using Dynabeads MyOne Silane beads. For the 5\u0026rsquo; RNA, cDNA amplification was performed with 13 amplification cycles, followed by reaction cleanup using SPRIselect Reagent Kit (Beckman Coulter). Total cDNA quantification was then performed using the Qubit 1X dsDNA HS Assay Kit (Thermo Fisher Scientific). Subsequently, enzymatic fragmentation, library construction, and additional SPRIselect bead clean-ups were performed to generate the final indexed libraries. Sample index PCR cycles were determined based on the total cDNA yield and ranged between 14 and 16 cycles. Library quality was assessed using the Agilent High Sensitivity DNA kit and libraries were quantified using the Qubit 1X dsDNA HS Assay kit (Thermo Fisher Scientific). V(D)Js amplification from the total cDNA was performed in two PCR cycles using 2 sequential, B cell oligos mixes. V(D)J libraries were constructed, further by enzymatic fragmentation of the V(D)J enriched cDNA and library indexing. Similar to the RNAseq libraries quantitation and quality were assessed with both Qubit 1X dsDNA HS Assay kit (Thermo Fisher Scientific) and Agilent High Sensitivity DNA kit. Some samples as indicated were analysed with the Chromium Next GEM Single Cell 3ʹ Reagent Kits v3.1 (Dual Index) (CG000315 Rev F). The libraries were generated following the manufacturer\u0026rsquo;s protocol. Single cell Bone marrow suspensions were loaded onto the Chromium Next GEM Chip G followed by gel beads in emulsion generation on the Chromium X. GEM-RT reactions were performed and 3\u0026rsquo; RNA cDNA amplification was performed with 12 amplification cycles, followed by reaction clean-up using SPRIselect Reagent Kit (Beckman Coulter). Subsequently, enzymatic fragmentation, library construction, and additional SPRIselect bead clean-ups were performed to generate the final indexed libraries. Sample index PCR cycles were determined based on the total cDNA yield and ranged between 14 and 16 cycles. Library quality was assessed using the Agilent High Sensitivity DNA kit and libraries were quantified using the Qubit 1X dsDNA HS Assay kit (Thermo Fisher Scientific).\u003c/p\u003e \u003cp\u003eLibraries were pooled and sequenced with the Illumina, NovaSeq 6000, platform, PE, Dual Index, (26 bp read 1, 90 bp read 2, 10 cycles i5 index and 10 cycles i7 index). As recommended, we aimed at obtaining, minimum 5,000 read pairs per cell for V(D)J Dual Index library and a minimum of 20,000 read pairs per cell for 5' Gene Expression Dual Index library.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eSingle-cell RNA sequencing data analysis\u003c/h2\u003e \u003cp\u003eCellRanger (v.7.2.0)\u003csup\u003e61\u003c/sup\u003e was used to align scRNA reads to the reference and create count matrices. Analysis was performed using the R (v4.3.1) and Python (3.12) programming languages. Seurat (v5.0.0)\u003csup\u003e62\u003c/sup\u003e was used for the processing of scRNA data. Ambient RNA correction was performed using SoupX(v1.6.2)\u003csup\u003e63\u003c/sup\u003e. scDblFinder(v1.16.0)\u003csup\u003e64\u003c/sup\u003e was used to identify and remove doublets. Additional filtering was performed removing cells with \u0026lt;\u0026thinsp;300 or \u0026gt;\u0026thinsp;8000 genes, mitochondrial RNA\u0026thinsp;\u0026gt;\u0026thinsp;10% and ribosomal RNA\u0026thinsp;\u0026gt;\u0026thinsp;40%. Integration of single cells was performed with Harmony (v.0.1.0)\u003csup\u003e65\u003c/sup\u003e resulting in clusters corresponding to distinct cell types. To further remove low quality cells that might have passed the initial quality control, additional subclustering of the cells corresponding to each of the major cell types (T-cells, NK-cell, Monocytes, B-cells, plasma cells), was performed. Sub-clusters expressing non-specific marker genes for each cell type were removed. Annotation of the clusters was performed using canonical marker genes and validated using CellTypist\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. To enable systematic comparisons across patients, we integrated the data from all individuals and all time-points by applying Harmony and Louvain clustering (with resolution 0.4). Overall, we annotated 127,432 tumor and immune cells at T1 (B-cells, n\u0026thinsp;=\u0026thinsp;44,328 (35%); plasma cells, n\u0026thinsp;=\u0026thinsp;2,570 (2%); T cells, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39,412 (31%); NK, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;8,693 (7%); monocytes, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;13,007(10%); progenitors, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;16,137 (12%) and 20,9341 tumor and immune cells at T2 (B-cells, n\u0026thinsp;=\u0026thinsp;46,840 (22%); plasma cells, n\u0026thinsp;=\u0026thinsp;1,986 (1%); T cells, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;72,651 (35%); NK, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;16,452 (8%); monocytes, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;28,644 (13%); progenitors, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;29,808 (14%).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eSub-clustering and cell sub-type proportion analysis\u003c/h2\u003e \u003cp\u003eEach major cell type of the tumor microenvironment (T-cells, NK-cells, Monocytes,Neutrophils) was subjected to sub-clustering to identify cell-subtypes and estimate changes in the proportions of the subtypes across the study groups. Integration of T-cells was performed using Harmony\u003csup\u003e\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e and Seurat\u0026rsquo;s CCA integration was applied on NK-cells and monocytes. Annotation of the sub-clusters was performed using canonical marker genes. To evaluate the differences in the proportions of cell subtypes across groups, we normalized the number of cells from each subtype for every individual by the total number of cells of the corresponding type. Samples with \u0026lt;\u0026thinsp;50 cells were excluded for this part of the analysis. Statistically significant differences were estimated using the Wilcoxon singed-rank test (un-paired when comparing responding groups of the same timepoint and paired when comparing across timepoints). Integration and label transfer between the patient data and the single-cell bone marrow reference\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e, was performed using Symphony(v.0.1.1)\u003csup\u003e68\u003c/sup\u003e. Samples of the reference data with \u0026lt;\u0026thinsp;50 cells for a major cell type were excluded from the analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of malignant B cells\u003c/h2\u003e \u003cp\u003eTo identify malignant (clonal) B-cells, data from each sample were analyzed separately. For 25 patients additional scBCR sequencing information was performed allowing us to assign single cells to clonotypes with unique V(D)J combinations. Clusters where multiple clonotypes were found to be clustering together, were annotated as healthy B-cells. Clusters that were enriched with a unique clonotype were annotated as malignant. To validate the clonotype-based annotation and identify malignant cells in the remaining 12 samples for which scBCR information was not available we applied the K/L ratio approach, as previously described\u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e. scBCR-seq and K/L ratio measurement identified 6% healthy and 94% cBc at T1 and 4% healthy an 96% cBc at T2.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eInference of copy number variants from single-cell RNA seq data\u003c/h2\u003e \u003cp\u003eTo identify CNVs from the single-cell data, the healthy and malignant B-cells from every sample (across the different timepoints) were processed separately. CNV inference was performed using Numbat (v1.3.2-1)\u003csup\u003e70\u003c/sup\u003e. To avoid potential artifacts, different custom references were used for the 3\u0026rsquo; and 5\u0026rsquo; sequenced data. The 3\u0026rsquo; reference included 2136 B cells from a healthy donor previously described by our group\u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e, and the 5\u0026rsquo; reference included 1589 B cells obtained from the two healthy donors sequenced for the current study. For the CNV visualization across groups the GenVisR(v.1.34.0) package was used\u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e. To statistically assess the difference of the CNVs occurrence across groups, we used Fisher's exact test.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of PC-like/MBC-like molecular subtypes\u003c/h2\u003e \u003cp\u003eFor this we used sciRED, a matrix factorization approach, to identify gene expression programs to determine sources of variation in our cohort of patients at diagnosis. PCA analysis on the set of genes associated with the most important factor from sciRed using as biological covariate the response group, revealed a clear distinction between samples that exhibited distinct MBC-like or PC-like features. The separation of the samples across the first principal component showed that the main source of variation across all patients originated from the B-cell subtype they belong to, indicating that the subtype of origin might be an important factor for the progression of the disease and the response to therapy. We validated these findings by determining a plasma cell scoring approach, using established gene markers associated with these subtypes. Briefly, two scores were estimated for each cell, one determined by the expression of PC-like features such as \u003cem\u003eXBP1\u003c/em\u003e, \u003cem\u003eIRF4\u003c/em\u003e, \u003cem\u003eMZB1\u003c/em\u003e, \u003cem\u003eJCHAIN\u003c/em\u003e, \u003cem\u003eCD138\u003c/em\u003e, \u003cem\u003eCD9\u003c/em\u003e, \u003cem\u003eDUSP22\u003c/em\u003e, and the second by the expression of MBC-like features such as \u003cem\u003eCD74\u003c/em\u003e, \u003cem\u003eBACH2\u003c/em\u003e, \u003cem\u003eCD20\u003c/em\u003e, \u003cem\u003eSPI1\u003c/em\u003e, \u003cem\u003eHLA-DRA\u003c/em\u003e, \u003cem\u003eRACK1.\u003c/em\u003eWe next divided the two scores to determine a single score between 0 and 1 for each cell, with higher scores being indicative of a PC-like subtype.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eGene expression signature (GEX) analysis\u003c/h2\u003e \u003cp\u003eGene expression signature (GEX) analysis was performed using the Python package sciRed\u003csup\u003e\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e, with library size and assay type (3\u0026rsquo;/5\u0026rsquo;) used as technical covariates. As input, we used the raw counts of the top 2000 variable genes. Only malignant B-cells were considered for this analysis. sciRed was applied using as biological covariate the response group and the \u003cem\u003eMYD88/CXCR4\u003c/em\u003e mutation status. The top 400 genes with the highest positive (200 genes) and negative (200 genes) association with the most important factors were considered for further analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eGLMnet risk scoring\u003c/h2\u003e \u003cp\u003eWe used the glmnet package (family=\u0026rdquo;cox\u0026rdquo;)\u003csup\u003e73\u003c/sup\u003e to estimate a risk score for each patient, based on single-cell gene expression data from malignant B cells at diagnosis, retaining only the top 1,000 variable genes. Briefly, glmnet fits an elastic net regression model that combines L1 (lasso) and L2 (ridge) coefficients, enabling us to identify genes that can help distinguish high-risk from low-risk patients. We used an alpha parameter of 0.03 maintaining 64 genes with coefficient\u0026thinsp;\u0026gt;\u0026thinsp;0. The final score provides a combined estimate of the contribution of every selected variable (gene) that can be used as a quantitative measurement for the patient\u0026rsquo;s risk of disease progression.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eValidation of gene expression signatures using an external cohort\u003c/h2\u003e \u003cp\u003eTo validate our findings from the gene expression signature profile analysis we used bulk-RNA sequencing data of CD19\u003csup\u003e+\u003c/sup\u003e cells from 47 WM patients. RNA sequencing reads were aligned using STAR\u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e and transcript counts from Salmon\u003csup\u003e\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e were aggregated into gene counts using tximport. Downstream analysis was performed using DeSeq2\u003csup\u003e76\u003c/sup\u003e, and the vst() function was used to log-transform the data.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eSurvival analysis\u003c/h2\u003e \u003cp\u003eKaplan-Meier analysis was performed using the survival (v3.7) R package. To estimate survival curves for individual genes, we initially calculated the average gene expression per sample for the gene of interest, followed by the average gene expression across the whole cohort at diagnosis. Samples were labeled as '1' when gene expression exceeded the average cohort expression and as '0' when gene expression was below the average cohort expression. Progression-free survival (PFS) was defined as the event of interest.\u003c/p\u003e \u003cp\u003eTo estimate survival curves for the gene expression signatures, we initially run the R package AUCell (v.1.24.0) to estimate a single score per cell using the expression of genes highly correlated to the factor of interest. We then estimated the median per sample to obtain a single score per patient. Finally, samples were labeled as \u0026ldquo;1\u0026rdquo; or \u0026ldquo;0\u0026rdquo; according to the sample score (larger or lower from the defined cutoff).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eCell-cell communication analysis\u003c/h2\u003e \u003cp\u003eCellChat\u003csup\u003e\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e was used to identify significant interactions between cell types of the tumor microenvironment. CellChat was performed four times to identify interactions across the following cohorts: T1-RG, T1-NRG, T2-RG, and T2-NRG. The function mergeCellChat() was then used to combine pairs of datasets, for downstream analysis and comparison across cohorts.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eWhole genome sequencing (WGS) analysis\u003c/h2\u003e \u003cp\u003eBurrows-Wheeler Aligner\u003csup\u003e\u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e\u003c/sup\u003e (BWA v0.7.17) was utilized to map the paired-end reads to the human reference genome (hg38). SAMtools\u003csup\u003e\u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e\u003c/sup\u003e was used for sorting the BAM files, and Picard tools (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/broadinstitute/picard\u003c/span\u003e\u003cspan address=\"https://github.com/broadinstitute/picard\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e)\u003c/span\u003e was utilized to mark duplicate reads. For the matched tumor - germline samples somatic SNV and INDEL detection was performed with MuTect2\u003csup\u003e80\u003c/sup\u003e,Strelka2\u003csup\u003e81\u003c/sup\u003e and Varscan2\u003csup\u003e82\u003c/sup\u003e. Only variants identified by at least two variant callers were processed for further analysis. Somatic CNV detection was performed using Battenberg\u003csup\u003e\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e\u003c/sup\u003e. The VCF files that included the somatic SNVs and INDELs of each sample were annotated using VEP and converted into MAF files using vcf2maf (v1.6.21) (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/mskcc/vcf2maf\u003c/span\u003e\u003cspan address=\"https://github.com/mskcc/vcf2maf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e).\u003c/span\u003e Maftools\u003csup\u003e\u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e\u003c/sup\u003e was used was downstream analysis and visualization of variants in oncoplots. For the signature analysis, we used the fitting algorithm mmsig\u003csup\u003e\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e\u003c/sup\u003e (1000 iterations, 0.01 cosine similarity threshold) to confirm the presence and estimate the contribution of 6 SBS signatures: SBS1, SBS5, SBS8, SBS9, SBS2, SBS13, SBS84.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eClonal evolution of longitudinal samples\u003c/h2\u003e \u003cp\u003eWe used Retcher\u003csup\u003e\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e\u003c/sup\u003e to reconstruct the clonal evolution of patients treated with ibrutinib using sequencing information from both timepoints (T1 and T2). Retcher, employs a probabilistic model to reconstruct the tumor phylogeny based on somatic SNVs and CNVs identified across sequential timepoints, forming clusters of variants with similar cellular prevalence. Analysis of the resulting clusters allowed us to identify samples with branching and linear evolution.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding:\u003c/h2\u003e \u003cp\u003eThis study has been supported by the IWMF-LLS Research Roadmap Initiative.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eAuthor Contribution\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eT.B. designed study, analyzed data, wrote paper\u003c/p\u003e\n\u003cp\u003eC.V. analyzed data, wrote paper\u003c/p\u003e\n\u003cp\u003eM.S. performed experiments, analyzed data, wrote paper\u003c/p\u003e\n\u003cp\u003eI.K. performed experiments\u003c/p\u003e\n\u003cp\u003eN.M-K. performed experiments\u003c/p\u003e\n\u003cp\u003eC.L. performed experiments\u003c/p\u003e\n\u003cp\u003eK.T. performed experiments\u003c/p\u003e\n\u003cp\u003eA.P. performed experiments\u003c/p\u003e\n\u003cp\u003eK.C. performed experiments\u003c/p\u003e\n\u003cp\u003eA.P. performed experiments\u003c/p\u003e\n\u003cp\u003eE.K. performed experiments\u003c/p\u003e\n\u003cp\u003eF.T. provided clinical data\u003c/p\u003e\n\u003cp\u003eM.G. wrote paper\u003c/p\u003e\n\u003cp\u003eE.T. wrote paper\u003c/p\u003e\n\u003cp\u003eZ.H. provided validation cohort, wrote paper\u003c/p\u003e\n\u003cp\u003eS.T wrote paper\u003c/p\u003e\n\u003cp\u003eM.D. wrote paper\u003c/p\u003e\n\u003cp\u003eE.K. wrote paper\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eOwen RG, Treon SP, Al-Katib A et al (2003) Clinicopathological definition of Waldenstrom\u0026rsquo;s macroglobulinemia: consensus panel recommendations from the Second International Workshop on Waldenstrom\u0026rsquo;s Macroglobulinemia. \u003cem\u003eSemin Oncol\u003c/em\u003e. ;30(2):110\u0026ndash;115. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/sonc.2003.50082\u003c/span\u003e\u003cspan address=\"10.1053/sonc.2003.50082\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCampo E, Swerdlow SH, Harris NL, Pileri S, Stein H, Jaffe ES (2011) The 2008 WHO classification of lymphoid neoplasms and beyond: evolving concepts and practical applications. Blood 117(19):5019\u0026ndash;5032. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood-2011-01-293050\u003c/span\u003e\u003cspan address=\"10.1182/blood-2011-01-293050\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTreon SP, Tripsas CK, Meid K et al (2015) Ibrutinib in previously treated Waldenstr\u0026ouml;m\u0026rsquo;s macroglobulinemia. N Engl J Med 372(15):1430\u0026ndash;1440. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1056/NEJMoa1501548\u003c/span\u003e\u003cspan address=\"10.1056/NEJMoa1501548\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTreon SP, Xu L, Hunter Z (2015) MYD88 Mutations and Response to Ibrutinib in Waldenstr\u0026ouml;m\u0026rsquo;s Macroglobulinemia. N Engl J Med 373(6):584\u0026ndash;586. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1056/NEJMc1506192\u003c/span\u003e\u003cspan address=\"10.1056/NEJMc1506192\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu L, Tsakmaklis N, Yang G et al (2017) Acquired mutations associated with ibrutinib resistance in Waldenstr\u0026ouml;m macroglobulinemia. Blood 129(18):2519\u0026ndash;2525. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood-2017-01-761726\u003c/span\u003e\u003cspan address=\"10.1182/blood-2017-01-761726\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen JG, Liu X, Munshi M et al (2018) BTK(Cys481Ser) drives ibrutinib resistance via ERK1/2 and protects BTK(wild-type) MYD88-mutated cells by a paracrine mechanism. Blood 131(18):2047\u0026ndash;2059. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood-2017-10-811752\u003c/span\u003e\u003cspan address=\"10.1182/blood-2017-10-811752\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaiva B, Corchete LA, Vidriales MB et al (2015) The cellular origin and malignant transformation of Waldenstr\u0026ouml;m macroglobulinemia. Blood 125(15):2370\u0026ndash;2380. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood-2014-09-602565\u003c/span\u003e\u003cspan address=\"10.1182/blood-2014-09-602565\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTreon SP, Hunter ZR, Aggarwal A et al (2006) Characterization of familial Waldenstrom\u0026rsquo;s macroglobulinemia. Ann Oncol 17(3):488\u0026ndash;494. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/annonc/mdj111\u003c/span\u003e\u003cspan address=\"10.1093/annonc/mdj111\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoulain S, Braggio E, Roumier C et al (2011) High-throughput genomic analysis in Waldenstr\u0026ouml;m\u0026rsquo;s macroglobulinemia. Clin Lymphoma Myeloma Leuk 11(1):106\u0026ndash;108. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3816/CLML.2011.n.021\u003c/span\u003e\u003cspan address=\"10.3816/CLML.2011.n.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBraggio E, Philipsborn C, Novak A, Hodge L, Ansell S, Fonseca R (2012) Molecular pathogenesis of Waldenstrom\u0026rsquo;s macroglobulinemia. Haematologica 97(9):1281\u0026ndash;1290. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3324/haematol.2012.068478\u003c/span\u003e\u003cspan address=\"10.3324/haematol.2012.068478\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaiva B, Montes MC, Garc\u0026iacute;a-Sanz R et al (2014) Multiparameter flow cytometry for the identification of the Waldenstr\u0026ouml;m\u0026rsquo;s clone in IgM-MGUS and Waldenstr\u0026ouml;m\u0026rsquo;s Macroglobulinemia: new criteria for differential diagnosis and risk stratification. Leukemia 28(1):166\u0026ndash;173. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/leu.2013.124\u003c/span\u003e\u003cspan address=\"10.1038/leu.2013.124\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSan Miguel JF, Vidriales MB, Ocio E et al (2003) Immunophenotypic analysis of Waldenstrom\u0026rsquo;s macroglobulinemia. Semin Oncol 30(2):187\u0026ndash;195. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/sonc.2003.50074\u003c/span\u003e\u003cspan address=\"10.1053/sonc.2003.50074\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorice WG, Chen D, Kurtin PJ, Hanson CA, McPhail ED (2009) Novel immunophenotypic features of marrow lymphoplasmacytic lymphoma and correlation with Waldenstr\u0026ouml;m\u0026rsquo;s macroglobulinemia. Mod Pathol 22(6):807\u0026ndash;816. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/modpathol.2009.34\u003c/span\u003e\u003cspan address=\"10.1038/modpathol.2009.34\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRemstein ED, Hanson CA, Kyle RA, Hodnefield JM, Kurtin PJ (2003) Despite apparent morphologic and immunophenotypic heterogeneity, Waldenstrom\u0026rsquo;s macroglobulinemia is consistently composed of cells along a morphologic continuum of small lymphocytes, plasmacytoid lymphocytes, and plasma cells. Semin Oncol 30(2):182\u0026ndash;186. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/sonc.2003.50073\u003c/span\u003e\u003cspan address=\"10.1053/sonc.2003.50073\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao Y, Hunter ZR, Liu X et al (2015) CXCR4 WHIM-like frameshift and nonsense mutations promote ibrutinib resistance but do not supplant MYD88(L265P) -directed survival signalling in Waldenstr\u0026ouml;m macroglobulinaemia cells. Br J Haematol 168(5):701\u0026ndash;707. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/bjh.13200\u003c/span\u003e\u003cspan address=\"10.1111/bjh.13200\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao Y, Hunter ZR, Liu X et al (2015) The WHIM-like CXCR4(S338X) somatic mutation activates AKT and ERK, and promotes resistance to ibrutinib and other agents used in the treatment of Waldenstrom\u0026rsquo;s Macroglobulinemia. Leukemia 29(1):169\u0026ndash;176. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/leu.2014.187\u003c/span\u003e\u003cspan address=\"10.1038/leu.2014.187\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCastillo JJ, Xu L, Gustine JN et al (2019) CXCR4 mutation subtypes impact response and survival outcomes in patients with Waldenstr\u0026ouml;m macroglobulinaemia treated with ibrutinib. Br J Haematol 187(3):356\u0026ndash;363. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/bjh.16088\u003c/span\u003e\u003cspan address=\"10.1111/bjh.16088\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTreon SP, Meid K, Gustine J et al (2021) Long-Term Follow-Up of Ibrutinib Monotherapy in Symptomatic, Previously Treated Patients With Waldenstr\u0026ouml;m Macroglobulinemia. J Clin Oncol 39(6):565\u0026ndash;575. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1200/JCO.20.00555\u003c/span\u003e\u003cspan address=\"10.1200/JCO.20.00555\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJim\u0026eacute;nez C, Chan GG, Xu L et al (2020) Genomic evolution of ibrutinib-resistant clones in Waldenstr\u0026ouml;m macroglobulinaemia. Br J Haematol 189(6):1165\u0026ndash;1170. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/bjh.16463\u003c/span\u003e\u003cspan address=\"10.1111/bjh.16463\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichardson K, Castillo JJ, Sarosiek SR et al (2024) Identification of robust predictors for ibrutinib response by multiomics in MYD88-mutated Waldenstr\u0026ouml;m macroglobulinemia. Blood Adv 8(9):2133\u0026ndash;2137. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/bloodadvances.2023012111\u003c/span\u003e\u003cspan address=\"10.1182/bloodadvances.2023012111\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoos-Weil D, Giacopelli B, Armand M et al (2020) Identification of 2 DNA methylation subtypes of Waldenstr\u0026ouml;m macroglobulinemia with plasma and memory B-cell features. Blood 136(5):585\u0026ndash;595. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood.2020005081\u003c/span\u003e\u003cspan address=\"10.1182/blood.2020005081\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHunter Z, Guerrera M, Tsakmaklis N et al (2025) The Evolution and Subtypes of Waldenstrom Macroglobulinemia: Findings from a Multi-Omic Analysis of 249 Treatment Naive MYD88L265P Mutated Patients. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.21203/rs.3.rs-6473628/v1\u003c/span\u003e\u003cspan address=\"10.21203/rs.3.rs-6473628/v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGagler DC, Ghamlouch H, Zhang D et al (2025) A multiomic analysis of Waldenstr\u0026ouml;m macroglobulinemia defines distinct disease subtypes. \u003cem\u003eBlood\u003c/em\u003e. Published online 7 May :blood.2024028164. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood.2024028164\u003c/span\u003e\u003cspan address=\"10.1182/blood.2024028164\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichardson K, Castillo JJ, Sarosiek SR et al (2024) Identification of robust predictors for ibrutinib response by multiomics in MYD88-mutated Waldenstr\u0026ouml;m macroglobulinemia. Blood Adv 8(9):2133\u0026ndash;2137. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/bloodadvances.2023012111\u003c/span\u003e\u003cspan address=\"10.1182/bloodadvances.2023012111\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMitchell RA, Liao H, Chesney J et al (2002) Macrophage migration inhibitory factor (MIF) sustains macrophage proinflammatory function by inhibiting p53: regulatory role in the innate immune response. Proc Natl Acad Sci U S A 99(1):345\u0026ndash;350. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1073/pnas.012511599\u003c/span\u003e\u003cspan address=\"10.1073/pnas.012511599\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHudson JD, Shoaibi MA, Maestro R, Carnero A, Hannon GJ, Beach DH (1999) A proinflammatory cytokine inhibits p53 tumor suppressor activity. J Exp Med 190(10):1375\u0026ndash;1382. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1084/jem.190.10.1375\u003c/span\u003e\u003cspan address=\"10.1084/jem.190.10.1375\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNguyen MT, Lue H, Kleemann R et al (2003) The cytokine macrophage migration inhibitory factor reduces pro-oxidative stress-induced apoptosis. J Immunol 170(6):3337\u0026ndash;3347. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4049/jimmunol.170.6.3337\u003c/span\u003e\u003cspan address=\"10.4049/jimmunol.170.6.3337\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoger T, David J, Glauser MP, Calandra T (2001) MIF regulates innate immune responses through modulation of Toll-like receptor 4. Nature 414(6866):920\u0026ndash;924. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/414920a\u003c/span\u003e\u003cspan address=\"10.1038/414920a\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHunter ZR, Xu L, Tsakmaklis N et al (2018) Insights into the genomic landscape of MYD88 wild-type Waldenstr\u0026ouml;m macroglobulinemia. Blood Adv 2(21):2937\u0026ndash;2946. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/bloodadvances.2018022962\u003c/span\u003e\u003cspan address=\"10.1182/bloodadvances.2018022962\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHunter ZR, Xu L, Yang G et al (2014) The genomic landscape of Waldenstrom macroglobulinemia is characterized by highly recurring MYD88 and WHIM-like CXCR4 mutations, and small somatic deletions associated with B-cell lymphomagenesis. Blood 123(11):1637\u0026ndash;1646. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood-2013-09-525808\u003c/span\u003e\u003cspan address=\"10.1182/blood-2013-09-525808\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaclachlan KH, Bagratuni T, Kastritis E et al (2023) Waldenstr\u0026ouml;m macroglobulinemia whole genome reveals prolonged germinal center activity and late copy number aberrations. Blood Adv 7(6):971\u0026ndash;981. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/bloodadvances.2022008876\u003c/span\u003e\u003cspan address=\"10.1182/bloodadvances.2022008876\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBagratuni TD, Theologi O, Vlachos CD et al Genomic landscape of IgM-MGUS and patients with stable or progressive asymptomatic Waldenstr\u0026ouml;m macroglobulinemia. \u003cem\u003eBlood\u003c/em\u003e. Published online 25 September 2025:blood.2025030177. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood.2025030177\u003c/span\u003e\u003cspan address=\"10.1182/blood.2025030177\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSamur MK, Aktas Samur A, Fulciniti M et al (2020) Genome-Wide Somatic Alterations in Multiple Myeloma Reveal a Superior Outcome Group. J Clin Oncol 38(27):3107\u0026ndash;3118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1200/JCO.20.00461\u003c/span\u003e\u003cspan address=\"10.1200/JCO.20.00461\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun H, Fang T, Wang T et al (2022) Single-cell profiles reveal tumor cell heterogeneity and immunosuppressive microenvironment in Waldenstr\u0026ouml;m macroglobulinemia. J Transl Med 20(1):576. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12967-022-03798-6\u003c/span\u003e\u003cspan address=\"10.1186/s12967-022-03798-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSklavenitis-Pistofidis R, Konishi Y, Heilpern-Mallory D et al (2025) Single-cell RNA sequencing defines distinct disease subtypes and reveals hypo-responsiveness to interferon in asymptomatic Waldenstrom\u0026rsquo;s Macroglobulinemia. Nat Commun 16(1):1480. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-025-56323-w\u003c/span\u003e\u003cspan address=\"10.1038/s41467-025-56323-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAran A, Garrig\u0026oacute;s L, Curigliano G, Cort\u0026eacute;s J, Mart\u0026iacute; M (2022) Evaluation of the TCR Repertoire as a Predictive and Prognostic Biomarker in Cancer: Diversity or Clonality? Cancers (Basel) 14(7). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/cancers14071771\u003c/span\u003e\u003cspan address=\"10.3390/cancers14071771\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePilcher WC, Yao L, Gonzalez-Kozlova E et al A single-cell atlas characterizes dysregulation of the bone marrow immune microenvironment associated with outcomes in multiple myeloma. Preprint posted online 17 May 2024. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1101/2024.05.15.593193\u003c/span\u003e\u003cspan address=\"10.1101/2024.05.15.593193\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLian J, Yue Y, Yu W, Zhang Y (2025) Correction: Immunosenescence: a key player in cancer development. J Hematol Oncol 18(1):25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13045-025-01682-6\u003c/span\u003e\u003cspan address=\"10.1186/s13045-025-01682-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaril\u0026agrave; G, Pavan L, Vedovato S et al (2021) Treatment Induced Cytotoxic T-Cell Modulation in Multiple Myeloma Patients. Front Oncol 11:682658. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fonc.2021.682658\u003c/span\u003e\u003cspan address=\"10.3389/fonc.2021.682658\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSacco A, Desantis V, Celay J et al (2023) Targeting the immune microenvironment in Waldenstr\u0026ouml;m macroglobulinemia via halting the CD40/CD40-ligand axis. Blood 141(21):2615\u0026ndash;2628. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood.2022019240\u003c/span\u003e\u003cspan address=\"10.1182/blood.2022019240\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSklavenitis-Pistofidis R, Konishi Y, Heilpern-Mallory D et al (2025) Single-cell RNA sequencing defines distinct disease subtypes and reveals hypo-responsiveness to interferon in asymptomatic Waldenstrom\u0026rsquo;s Macroglobulinemia. Nat Commun 16(1):1480. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-025-56323-w\u003c/span\u003e\u003cspan address=\"10.1038/s41467-025-56323-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElessa D, Debureaux PE, Villesuzanne C et al (2022) Inflammatory Waldenstr\u0026ouml;m\u0026rsquo;s macroglobulinaemia: A French monocentric retrospective study of 67 patients. Br J Haematol 197(6):728\u0026ndash;735. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/bjh.18157\u003c/span\u003e\u003cspan address=\"10.1111/bjh.18157\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eForgeard N, Baron M, Caron J et al (2022) Inflammation in Waldenstr\u0026ouml;m macroglobulinemia is associated with 6q deletion and need for treatment initiation. Haematologica 107(11):2720\u0026ndash;2724. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3324/haematol.2022.281053\u003c/span\u003e\u003cspan address=\"10.3324/haematol.2022.281053\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDebureaux PE, Forgeard N, Elessa D et al (2024) Inflammation is predictive of outcome in Waldenstr\u0026ouml;m macroglobulinemia treated by Bruton tyrosine kinase inhibitors: a multicentric real-life study. Haematologica 109(1):325\u0026ndash;330. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3324/haematol.2023.283141\u003c/span\u003e\u003cspan address=\"10.3324/haematol.2023.283141\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDebureaux PE, Forgeard N, Elessa D et al (2024) Inflammation is predictive of outcome in Waldenstr\u0026ouml;m macroglobulinemia treated by Bruton tyrosine kinase inhibitors: a multicentric real-life study. Haematologica 109(1):325\u0026ndash;330. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3324/haematol.2023.283141\u003c/span\u003e\u003cspan address=\"10.3324/haematol.2023.283141\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBertrand E, Jouy N, Manier S et al (2017) Role of IRF4 in resistance to immunomodulatory (IMid) compounds(\u0026reg;) in Waldenstr\u0026ouml;m\u0026rsquo;s macroglobulinemia. Oncotarget 8(68):112917\u0026ndash;112927. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.18632/oncotarget.22872\u003c/span\u003e\u003cspan address=\"10.18632/oncotarget.22872\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaura F, Boyle EM, Coffey D et al (2023) Genomic and immune signatures predict clinical outcome in newly diagnosed multiple myeloma treated with immunotherapy regimens. Nat Cancer 4(12):1660\u0026ndash;1674. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s43018-023-00657-1\u003c/span\u003e\u003cspan address=\"10.1038/s43018-023-00657-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArgyropoulos KV, Vogel R, Ziegler C et al (2016) Clonal B cells in Waldenstr\u0026ouml;m\u0026rsquo;s macroglobulinemia exhibit functional features of chronic active B-cell receptor signaling. Leukemia 30(5):1116\u0026ndash;1125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/leu.2016.8\u003c/span\u003e\u003cspan address=\"10.1038/leu.2016.8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuiatse I, Baladandayuthapani V, Lin HY et al (2015) Targeting the Spleen Tyrosine Kinase with Fostamatinib as a Strategy against Waldenstr\u0026ouml;m Macroglobulinemia. Clin Cancer Res 21(11):2538\u0026ndash;2545. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1158/1078-0432.CCR-14-1462\u003c/span\u003e\u003cspan address=\"10.1158/1078-0432.CCR-14-1462\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAgrelo R, Souabni A, Novatchkova M et al (2009) SATB1 defines the developmental context for gene silencing by Xist in lymphoma and embryonic cells. Dev Cell 16(4):507\u0026ndash;516. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.devcel.2009.03.006\u003c/span\u003e\u003cspan address=\"10.1016/j.devcel.2009.03.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun JY, Xu L, Tseng H et al (2011) Histone deacetylase inhibitors demonstrate significant preclinical activity as single agents, and in combination with bortezomib in Waldenstr\u0026ouml;m\u0026rsquo;s macroglobulinemia. Clin Lymphoma Myeloma Leuk 11(1):152\u0026ndash;156. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3816/CLML.2011.n.036\u003c/span\u003e\u003cspan address=\"10.3816/CLML.2011.n.036\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoccaro AM, Sacco A, Jia X et al (2010) microRNA-dependent modulation of histone acetylation in Waldenstrom macroglobulinemia. Blood 116(9):1506\u0026ndash;1514. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/blood-2010-01-265686\u003c/span\u003e\u003cspan address=\"10.1182/blood-2010-01-265686\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHunter ZR, Xu L, Tsakmaklis N et al (2018) Insights into the genomic landscape of MYD88 wild-type Waldenstr\u0026ouml;m macroglobulinemia. Blood Adv 2(21):2937\u0026ndash;2946. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/bloodadvances.2018022962\u003c/span\u003e\u003cspan address=\"10.1182/bloodadvances.2018022962\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHartmann S, Schuhmacher B, Rausch T et al (2016) Highly recurrent mutations of SGK1, DUSP2 and JUNB in nodular lymphocyte predominant Hodgkin lymphoma. Leukemia 30(4):844\u0026ndash;853. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/leu.2015.328\u003c/span\u003e\u003cspan address=\"10.1038/leu.2015.328\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFechter K, Feichtinger J, Prochazka K et al (2018) Cytoplasmic location of NR4A1 in aggressive lymphomas is associated with a favourable cancer specific survival. Sci Rep 8(1):14528. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-018-32972-4\u003c/span\u003e\u003cspan address=\"10.1038/s41598-018-32972-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDerenzini E, Mazzara S, Melle F et al (2021) A three-gene signature based on MYC, BCL-2 and NFKBIA improves risk stratification in diffuse large B-cell lymphoma. Haematologica 106(9):2405\u0026ndash;2416. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3324/haematol.2019.236455\u003c/span\u003e\u003cspan address=\"10.3324/haematol.2019.236455\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNagy B, Ferrer A, Larramendy ML et al (2003) Lymphotoxin beta expression is high in chronic lymphocytic leukemia but low in small lymphocytic lymphoma: a quantitative real-time reverse transcriptase polymerase chain reaction analysis. Haematologica 88(6):654\u0026ndash;658\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBolli N, Biancon G, Moarii M et al (2018) Analysis of the genomic landscape of multiple myeloma highlights novel prognostic markers and disease subgroups. Leukemia 32(12):2604\u0026ndash;2616. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41375-018-0037-9\u003c/span\u003e\u003cspan address=\"10.1038/s41375-018-0037-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLopez-Gomez J, Cerrato-Izaguirre D, Quezada-Maldonado EM et al (2025) Molecular Characterization of Oral Squamous Cell Carcinoma in Mexican Patients: A Genomic and Epidemiological Overview. Cancers 17(20). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/cancers17203282\u003c/span\u003e\u003cspan address=\"10.3390/cancers17203282\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBagratuni T, Markou A, Patseas D et al (2022) Determination of MYD88L265P mutation fraction in IgM monoclonal gammopathies. Blood Adv 6(1):189\u0026ndash;199. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/bloodadvances.2021005354\u003c/span\u003e\u003cspan address=\"10.1182/bloodadvances.2021005354\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng GXY, Terry JM, Belgrader P et al (2017) Massively parallel digital transcriptional profiling of single cells. Nat Commun 8:14049. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ncomms14049\u003c/span\u003e\u003cspan address=\"10.1038/ncomms14049\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSatija R, Farrell JA, Gennert D, Schier AF, Regev A (2015) Spatial reconstruction of single-cell gene expression data. Nat Biotechnol 33(5):495\u0026ndash;502. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nbt.3192\u003c/span\u003e\u003cspan address=\"10.1038/nbt.3192\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYoung MD, Behjati S (2020) SoupX removes ambient RNA contamination from droplet-based single-cell RNA sequencing data. Gigascience 9(12). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/gigascience/giaa151\u003c/span\u003e\u003cspan address=\"10.1093/gigascience/giaa151\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGermain PL, Lun A, Garcia Meixide C, Macnair W, Robinson MD (2021) Doublet identification in single-cell sequencing data using scDblFinder. F1000Res 10:979. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.12688/f1000research.73600.2\u003c/span\u003e\u003cspan address=\"10.12688/f1000research.73600.2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKorsunsky I, Millard N, Fan J et al (2019) Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods 16(12):1289\u0026ndash;1296. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41592-019-0619-0\u003c/span\u003e\u003cspan address=\"10.1038/s41592-019-0619-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDom\u0026iacute;nguez Conde C, Xu C, Jarvis LB et al (2022) Cross-tissue immune cell analysis reveals tissue-specific features in humans. Science 376(6594):eabl5197. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.abl5197\u003c/span\u003e\u003cspan address=\"10.1126/science.abl5197\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee NYS, Li M, Ang KS, Chen J (2023) Establishing a human bone marrow single cell reference atlas to study ageing and diseases. Front Immunol 14:1127879. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2023.1127879\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2023.1127879\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKang JB, Nathan A, Weinand K et al (2021) Efficient and precise single-cell reference atlas mapping with Symphony. Nat Commun 12(1):5890. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-021-25957-x\u003c/span\u003e\u003cspan address=\"10.1038/s41467-021-25957-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBagratuni T, Aktypi F, Theologi O et al (2024) Single-cell analysis of MYD88(L265P) and MYD88(WT) Waldenstr\u0026ouml;m macroglobulinemia patients. Hemasphere 8(2):e27. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/hem3.27\u003c/span\u003e\u003cspan address=\"10.1002/hem3.27\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao T, Soldatov R, Sarkar H et al (2023) Haplotype-aware analysis of somatic copy number variations from single-cell transcriptomes. Nat Biotechnol 41(3):417\u0026ndash;426. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41587-022-01468-y\u003c/span\u003e\u003cspan address=\"10.1038/s41587-022-01468-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSkidmore ZL, Wagner AH, Lesurf R et al (2016) GenVisR: Genomic Visualizations in R. Bioinformatics 32(19):3012\u0026ndash;3014. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bioinformatics/btw325\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/btw325\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePouyabahar D, Andrews T, Bader GD (2025) Interpretable single-cell factor decomposition using sciRED. Nat Commun 16(1):1878. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-025-57157-2\u003c/span\u003e\u003cspan address=\"10.1038/s41467-025-57157-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFriedman J, Hastie T, Tibshirani R (2010) Regularization Paths for Generalized Linear Models via Coordinate Descent. J Stat Softw 33(1):1\u0026ndash;22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDobin A, Davis CA, Schlesinger F et al (2013) STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29(1):15\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bioinformatics/bts635\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/bts635\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatro R, Duggal G, Love MI, Irizarry RA, Kingsford C (2017) Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods 14(4):417\u0026ndash;419. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nmeth.4197\u003c/span\u003e\u003cspan address=\"10.1038/nmeth.4197\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLove MI, Huber W, Anders S (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15(12):550. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13059-014-0550-8\u003c/span\u003e\u003cspan address=\"10.1186/s13059-014-0550-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJin S, Guerrero-Juarez CF, Zhang L et al (2021) Inference and analysis of cell-cell communication using CellChat. Nat Commun 12(1):1088. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-021-21246-9\u003c/span\u003e\u003cspan address=\"10.1038/s41467-021-21246-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi H Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. \u003cem\u003earXiv: Genomics\u003c/em\u003e. Published online 2013. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://api.semanticscholar.org/CorpusID:14669139\u003c/span\u003e\u003cspan address=\"https://api.semanticscholar.org/CorpusID:14669139\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDanecek P, Bonfield JK, Liddle J et al (2021) Twelve years of SAMtools and BCFtools. Gigascience 10(2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/gigascience/giab008\u003c/span\u003e\u003cspan address=\"10.1093/gigascience/giab008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenjamin D, Sato T, Cibulskis K, Getz G, Stewart C, Lichtenstein L Calling Somatic SNVs and Indels with Mutect2. \u003cem\u003ebioRxiv\u003c/em\u003e. Published online 2019. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1101/861054\u003c/span\u003e\u003cspan address=\"10.1101/861054\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim S, Scheffler K, Halpern AL et al (2018) Strelka2: fast and accurate calling of germline and somatic variants. Nat Methods 15(8):591\u0026ndash;594. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41592-018-0051-x\u003c/span\u003e\u003cspan address=\"10.1038/s41592-018-0051-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoboldt DC, Zhang Q, Larson DE et al (2012) VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Res 22(3):568\u0026ndash;576. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1101/gr.129684.111\u003c/span\u003e\u003cspan address=\"10.1101/gr.129684.111\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNik-Zainal S, Van Loo P, Wedge DC et al (2012) The life history of 21 breast cancers. Cell 149(5):994\u0026ndash;1007. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2012.04.023\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2012.04.023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMayakonda A, Lin DC, Assenov Y, Plass C, Koeffler HP (2018) Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res 28(11):1747\u0026ndash;1756. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1101/gr.239244.118\u003c/span\u003e\u003cspan address=\"10.1101/gr.239244.118\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRustad EH, Nadeu F, Angelopoulos N et al (2021) mmsig: a fitting approach to accurately identify somatic mutational signatures in hematological malignancies. Commun Biol 4(1):424. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s42003-021-01938-0\u003c/span\u003e\u003cspan address=\"10.1038/s42003-021-01938-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Z, Fang Y, Wang R, Kong L, Liang S, Tao S (2024) Reconstructing tumor clonal heterogeneity and evolutionary relationships based on tumor DNA sequencing data. Brief Bioinform 25(6). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bib/bbae516\u003c/span\u003e\u003cspan address=\"10.1093/bib/bbae516\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":false,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Waldenström macroglobulinemia, ibrutinib, single cell, transcriptome, bone marrow microenvironment","lastPublishedDoi":"10.21203/rs.3.rs-8594385/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8594385/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDespite recent advances in treatment strategies of Waldenstr\u0026ouml;m macroglobulinemia (WM) patients, the disease remains incurable. While several genomic features predict poor outcomes with ibrutinib, the contribution of immune dysfunction to tumor behavior remains unclear. Single-cell RNA sequencing on longitudinal bone marrows from 37 Waldenstr\u0026ouml;m macroglobulinemia patients before and after ibrutinib therapy profiled 348,000 cells and identified distinct immune phenotypes associated with ibrutinib progression. This was primarily characterized by the accumulation of T-effector cells, Tregs and pro-inflammatory M1-like monocytes, alongside a depletion of na\u0026iuml;ve T-cells. The tumor architecture in progressing patients, exhibited a unique transcriptomic signature, also validated in an external cohort of 47 WM patients, driven by genes including \u003cem\u003eLTB, NFKBIA, DUSP2\u003c/em\u003e, \u003cem\u003eNR4A1\u003c/em\u003e leading to shorter progression-free survival. Mutational profiling using whole-genome sequencing identified mutational signature SBS1/SBS5 being significantly associated with poorer outcome. Finally, we demonstrate that integrating tumor with immune cell compartments can significantly improve ibrutinib response prediction scoring.\u003c/p\u003e","manuscriptTitle":"Distinct immune and genomic signatures predict resistance to ibrutinib therapy in Waldenström macroglobulinemia","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-23 09:07:41","doi":"10.21203/rs.3.rs-8594385/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"17ef2ee6-23bb-4074-8a8f-ed4dc4ec824f","owner":[],"postedDate":"January 23rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":61597198,"name":"Biological sciences/Cancer/Haematological cancer/Myeloma"},{"id":61597199,"name":"Biological sciences/Cancer/Tumour heterogeneity"}],"tags":[],"updatedAt":"2026-04-06T10:50:35+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-23 09:07:41","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8594385","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8594385","identity":"rs-8594385","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}