Cytogenetic abnormalities (CAs) are known to be the preponderant prognostic factor in multiple myeloma. Our team has recently developed a prognostic score based on 6 CAs, with which del(1p32) appears to be the second worst abnormality after del(17p). This study aimed to confirm the adverse effect of 1p32 deletion in patients with newly diagnosed multiple myeloma (NDMM). Among 2551 patients with newly diagnosed multiple myeloma, 11% were harboring del(1p32). Their overall survival (OS) was significantly inferior compared with patients without del(1p32) (median OS: 49 months vs 124 months). Likewise, progression-free survival was significantly shorter. More importantly, biallelic del(1p32) conferred a dramatically poorer prognosis than a monoallelic del(1p32) (median OS: 25 months vs 60 months). As expected, the OS of patients with del(1p32) significantly decreased when this abnormality was associated with other high-risk CAs [del(17p), t(4;14), or gain(1q)]. In the multivariate analysis, del(1p32) appeared as a negative prognostic factor; after adjustment for age and treatment, the risk of progression was 1.3 times higher among patients harboring del(1p32), and the risk of death was 1.9 times higher. At the dawn of risk-adapted treatment strategies, we have confirmed the adverse effect of del(1p32) in multiple myeloma and the relevance of its assessment at diagnosis.

Multiple myeloma (MM) is a B cell malignancy involving terminally differentiated plasma cells that is often initially driven by translocations of the IGHgene at 14q32 (tIGH, observed in ~35% of patients). It is believed that the main cause of tIGH is activation-induced cytidine deaminase (AID), which is a protein that mediates antibody class switching. This is based on data from MM cell lines, where tIGH is located most often in breakpoints in the switch regions. However, it can also be in the VDJ regions, which implicates dysregulated VDJ recombination. To fully understand the mechanisms behind tIGH, we must first fully catalog the spectrum of these translocations and assess their association with other genomic features. Here, we analyzed deep whole-genome sequencing from 1257 newly diagnosed patients enrolled in the IFM or DFCI trials as well as the MMRF cohort. In addition, targeted sequencing data were collected from 4078 patients in the IFM network. tIGH breakpoints were classified according to their overlap with the switch, 3' regulatory (3'RR) or VDJ regions. If a breakpoint did not overlap with any of these regions, it was considered "intergenic". Overall, we identified tIGH in 38% of patients, and among all tIGH breakpoints, switch region translocations were the most frequent (59.5%) followed by 24.8% in the VDJ regions, 12% in intergenic regions and 3.7% in the 3'RR. We next looked at how common translocations in MM were related to various breakpoint locations. t(4;14) occurred almost exclusively in switch regions (95.7%, 95% CI 92.7-97.7%, p value < 2.2e-16). For other translocations (t(6;14), t(11;14), t(14;16), t(14;20)), tIGH tended to be in the switch regions (max 62%) but they were also present in VDJ regions (range 12.6 to 30.9%). On the other hand, for MYC-targeting translocations (t[8;14]), the most common breakpoints were in intergenic (37.8%, 95%CI 28.1-48.4%, p value = 9.5e-11) regions and the 3'RR. To investigate the mechanisms that translocate IGHto intergenic regions, we first searched for homology and found that intergenic breakpoint hotspots have a high sequence similarity between each other and IgHG1/2/3/4 (cosine distance > 0.80) unlike IgHA and IgHM. Because AID is involved in somatic hypermutation, we next investigated the mutational load of tIGH and found a significant increase in mutations (Kruskal-Wallis p value = 0.03) from VDJ to switch to intergenic patients. However, the intergenic group had significantly lower AID-related mutations on both an absolute (t test p value = 0.007) and normalized (t test p value = 0.002) scale. In addition, the intergenic group had increased genomic instability, as assessed by CNA segmentation and the percentage of the genome that was either amplified or deleted (Wilcoxon p value = 0.048). Furthermore, genes involved in genome instability and class switch-associated genes, like BLM and TP53BP1, were only mutated in the intergenic group. There were no KRAS mutations in the intergenic group (p value = 0.011), whereas BRAFmutations were much more common (p value=0.008, OR=17.3). Overall, our study indicates that multiple factors contribute to tIGH and these factors correlate to MM subgroups. tIGH in patients with t(4;14) is clearly associated with faulty class switching, likely caused by AID, whereas patients with other translocations appear to have multiple potential mechanisms. Interestingly, even within the MM subgroups, those patients with tIGH in an intergenic region have increased hypodiploidy frequency, implicating a mechanism that also plays a role in genomic stability.

Multiple myeloma (MM) is a heterogeneous disease, and cytogenetic abnormalities such as t(4;14) and del17p are well-established independent high-risk features in newly diagnosed patients. In two clinical trials (N=1422), the IFM2009 and Determination, patients with del17p or t(4;14) had a hazard ratio of 1.9 (95% CI 1.5-2.4) and 1.6 (95% CI 1.2-2.1), respectively. However, the same study also showed that 25% of the remaining patients had a similar short PFS without any identified known high-risk features. To understand the genomic etiology of the high-risk in these patients, we generated WGS data from 640 newly diagnosed patients (GAMER dataset) in 3 categories (low risk [PFS > 36 months] without known risk features (LR), high-risk [PFS < 18 months] without known risk features (unknownHR, uHR) and high-risk with known risk features) (HR). Our primary focus was to find factors that contribute to high-risk without known risk features. As a validation set, we used WGS samples from newly diagnosed patients who enrolled in the DFCI 2009 study. Overall, we observed that t(14;16) was the only primary translocation that differs between uHR (7%) and the LR group (1%) (p value < 0.01). We found more frequent driver mutations in TP53(7%), ATM(6%), and EGR1(4%), in the uHR group (FDR < 0.05). In uHR group, we found that the cancer cell fraction carrying these mutations was significantly lower compared to double hits events observed in HR (p value = 6.4e-05), suggesting their presence in high-risk subclones. Chromosome 1q gain vs amplification were associated with similar outcomes in these patients (p value = 0.81), but both significantly differed from chr1q wild type (p value = 1.8e-07). High-risk patients with and without known risk features had a significant enrichment of mutations in the Genome Integrity pathways (24%) compared to the LR group. However, the enrichment in MAPK signaling pathway mutations (60%) was significantly different between the two high-risk groups. Features that contribute to genome integrity, such as structural variants (insertion, deletion, translocation and inversion, p value<0.001), mutational load (p=0.00025), and genomic scar scores (p=0.0007), were all significantly lower in the low-risk group compared to the uHR. The number of copy number alterations that have not been described before, such as del16, del8p and gain 9 and 19, were also significantly different between low-risk and high-risk patients without known risk features. We further searched for genomic factors that contribute to the classification of uHR using multiple statistical modeling and machine learning tools. All models were run using a 10-fold cross-validation for choosing variables to put into the final model. Using all variables in the final model, we created a classification model using the variables' significance and the model's accuracy. Out of 27 variables, we found that 7 (mutational load, structural variants, del8p, t(14;16), Genome Integrity Pathway, MAPK pathway and ISS > 2) significantly contributed to the classification model. On training and validation data, this model successfully detected the risk group (p value = 3.5e-09 and 0.0027, respectively). Our dataset gives the ability to identify genomic risk markers that are not currently considered in newly diagnosed MM and allows us to extend our understanding beyond traditional risk. The absence of unique and mutually exclusive markers between uHR and LR suggests the additive risk model in MM patients without known high risk markers. Additional risk features might be hidden in the transcriptome and regulation of the transcriptome. We are currently building an integrative model from the same patients with WGS, RNA-seq and methylation data. This will allow us to fine-tune our risk model and help us develop simplified and complex models on MM risk beyond known cytogenetic risk features.

Background: Multiple myeloma (MM) is the second hematological malignancy in the Western countries. Currently, the Revised International Staging System (R-ISS) is widely used to assess patients’ prognosis. Cytogenetic abnormalities (CA) affecting the chromosome 1, gain 1q and del(1p32), were not included in the new criteria of HR CA despite their relatively high frequencies (respectively 35% and 11%), due to insufficient data in the study. However, we have recently confirmed the significant impact on del(1p32) on myeloma‘s prognosis, being the second most pejorative abnormality, just after del(17p). Aims: The aim of this study is to update our data about the prognostic impact of del(1p32) on a large cohort of NDMM patients. Methods: Clinical data were obtained from 2551 NDMM patients, followed up for ≥ 36 months or having died or progressed within 36 months post treatment. Informed consent was obtained for all included patients. 1258 patients were treated by an intensive therapy. Follow-up duration was estimated using reverse Kaplan-Meier method. Overall survival (OS) and progression free survival (PFS) curves were estimated using the Kaplan-Meier method and were compared using the log-rank test. Tests were two-sided, and P < .05 was considered significant. All analyses were performed using R version 4.1.1. Results: We observed 12.4% of patients displaying del(1p32), which was the expected proportion. Median follow-up was 67.4 months. The OS of patients harboring del(1p32) was twice as short as the OS of patients without del(1p32) (60.4 and 123.9 months, respectively, P < 0.0001) (Figure 1). Likewise, PFS was significantly shorter in del(1p32) patients (18.1 and 29.1 months, respectively, P < 0.0001). These poorer outcomes were observed even in patients treated with an intensive therapy (del(1p32) vs. no del(1p32); PFS 26.5 vs. 37.0 months, P < 0.0001; OS 72.0 vs. 127.4 months, P < 0.0001). We observed higher frequencies of del(17p) and gain 1q in del(1p32) patients (21.3% and 53.2% respectively), compared to general MM population. To check if the poor survival is not only due to these higher levels of association, we have decided to focus on patients harboring del(1p32) without the main high-risk (HR) CA. HR CA were defined by the presence of del(17p), t(4;14) and/or gain 1q. Patients without HR CA had a lower PFS and OS when they carried del(1p32) (del(1p32) vs. no del(1p32); PFS 25.6 vs. 34.8 months, P = 0.0004; OS 83.0 vs. 136.1 months, P = 0.0002). It is now widely admitted that cumulating HR CA worsen the prognosis. Thus, we have assessed the effect of additional CA on the prognosis of del(1p32) patients. Additional CA were CA defined as HR CA above. As expected, the overall survival of del(1p32) patients significantly decreases when this abnormality was associated with other CA (OS: del(1p32) alone 83.0 months, del(1p32) with 1 HR CA 45.8 months, del(1p32) with 2 HR CA or more 36.5 months, P < 0.0001). Image:Summary/Conclusion: Here we have confirmed the pejorative impact of del(1p32) in multiple myeloma on the largest cohort of NDMM patients ever evaluated to our knowledge (316 del(1p32) patients). Our results demonstrate the importance of the detection of del(1p32) at diagnosis because of its huge impact on the prognosis, especially in the era of risk-adapted treatment strategy. The multivariate analysis is in progress.

Primary plasma cell leukemia (pPCL) is an aggressive form of multiple myeloma (MM) that has not benefited from recent therapeutic advances in the field. Because it is very rare and heterogeneous, it remains poorly understood at the molecular level. To address this issue, we performed DNA and RNA sequencing of sorted plasma cells from a large cohort of 90 newly diagnosed pPCL and compared with MM. We observed that pPCL presents a specific genomic landscape with a high prevalence of t(11;14) (about half) and high-risk genomic features such as del(17p), gain 1q, and del(1p32). In addition, pPCL displays a specific transcriptome when compared with MM. We then wanted to characterize specifically pPCL with t(11;14). We observed that this subentity displayed significantly fewer adverse cytogenetic abnormalities. This translated into better overall survival when compared with pPCL without t(11;14) (39.2 months vs 17.9 months, P = .002). Finally, pPCL with t(11;14) displayed a specific transcriptome, including differential expression of BCL2 family members. This study is the largest series of patients with pPCL reported so far.

Despite tremendous improvements in the outcome of patients with multiple myeloma in the past decade, high-risk patients have not benefited from the approval of novel drugs. The most important prognostic factor is the loss of parts of the short arm of chromosome 17, known as deletion 17p (del(17p)). A recent publication (on a small number of patients) suggested that these patients are at very high-risk only if del(17p) is associated with TP53 mutations, the so-called "double-hit" population. To validate this finding, we designed a much larger study on 121 patients presenting del(17p) in > 55% of their plasma cells, and homogeneously treated by an intensive approach. For these 121 patients, we performed deep next generation sequencing targeted on TP53. The outcome was then compared with a large control population (2505 patients lacking del(17p)). Our results confirmed that the "double hit" situation is the worst (median survival = 36 months), but that del(17p) alone also confers a poor outcome compared with the control cohort (median survival = 52.8 months vs 152.2 months, respectively). In conclusion, our study clearly confirms the extremely poor outcome of patients displaying "double hit," but also that del(17p) alone is still a very high-risk feature, confirming its value as a prognostic indicator for poor outcome.

CD8(+) T cells within the tumor microenvironment (TME) are exposed to various signals that ultimately determine functional outcomes. Here, we examined the role of the co-activating receptor CD226 (DNAM-1) in CD8(+) T cell function. The absence of CD226 expression identified a subset of dysfunctional CD8(+) T cells present in peripheral blood of healthy individuals. These cells exhibited reduced LFA-1 activation, altered TCR signaling, and a distinct transcriptomic program upon stimulation. CD226(neg) CD8(+) T cells accumulated in human and mouse tumors of diverse origin through an antigen-specific mechanism involving the transcriptional regulator Eomesodermin (Eomes). Despite similar expression of co-inhibitory receptors, CD8(+) tumor-infiltrating lymphocyte failed to respond to anti-PD-1 in the absence of CD226. Immune checkpoint blockade efficacy was hampered in Cd226(-/-) mice. Anti-CD137 (4-1 BB) agonists also stimulated Eomes-dependent CD226 loss that limited the anti-tumor efficacy of this treatment. Thus, CD226 loss restrains CD8(+) T cell function and limits the efficacy of cancer immunotherapy.

Background: Primary plasma cell leukemia (pPCL) is a rare and aggressive form of multiple myeloma (MM) with an extremely poor prognosis and distinct biological and clinical features. Because of its low incidence and its heterogeneity, biological knowledge about pPCL is lacking especially molecular process responsible for its aggressiveness. Here, we took advantage of a large series of pPCL to describe the genomic and transcriptomic landscape of pPCL, to identify potential driver mutations and pathways, and to determine their clinical impacts. Methods: To address these issues, we performed a targeted DNA sequencing and a RNA sequencing of sorted bone marrow plasma cells collected at the time of diagnosis from 96 patients with pPCL between 2014 and 2020. We compared their genomic profiles with those of 907 MM at diagnosis previously obtained in our laboratory and their transcriptomic profiles with those of 300 MM at diagnosis obtained from the IFM2009/DFCI trial (NCT01191060). Copy number aberrations (CNA), single nucleotide variants (SNV), translocations, mutations, gene expression (GE) and gene set enrichment were analyzed and correlated with clinical information (overall survival and progression-free survival). Results: Genome analysis highlighted a specific genomic profile of pPCL. Indeed, hyperdiploid karyotypes were less frequent in pPCL compared with MM (20% vs 57%, p<0,001). We found a high prevalence of translocations involving the heavy chain locus (IGH) in pPCL with higher incidences of t(11;14) (51% vs 23%, p<0,001) and t(14;16) (14% vs 3%, p<0,001), but an identical incidence of t(4;14) (11% vs 10%, p=0,7). pPCL presented more adverse cytogenetic abnormalities such as del(17p) (30% vs 9,5%, p<0,001), 1q gain (53% vs 32%, p<0,001) and del(1p32) (24% vs 9%, p<0,001). Among the 246 recurrently mutated genes in MM, mutations of TP53 (21% vs 5%, p<0,001) and IRF4 (11% vs 4%, p<0,005) were significantly more frequent in pPCL. Furthermore, pPCL presented high-risk genomic features with an increased proportion of Double Hit profiles (27% vs 5%, p<0,001) with more bi-allelic inactivation of TP53 (17% vs 3%, p<0,001) and more amp1q on the background of International Staging System III (11% vs 5%, p<0,005). Interestingly, by comparing genomic profiles from pPCL with and without t(11;14) we found two distinctive patterns. Indeed pPCL with t(11;14) showed more TP53 mutations and more bi-allelic inactivation of TP53. While pPCL without t(11;14) showed more adverse cytogenetic abnormalities such as trisomy 21, 1q gains and del(1p32). These results suggest two distinctive oncogenic mechanisms. RNA-seq analysis showed also a specific transcriptional landscape of pPCL. Indeed, unsupervised hierarchical clustering of gene expression profiles demonstrated two distinct clusters between pPCL and MM. Gene set enrichment analysis identified a significantly higher expression of genes involved in MYC Targets and G2M checkpoint, and a significantly lower expression of genes involved in P53 pathway, hypoxia and TNF alpha signaling via NF-κB. Furthermore, pPCL with and without t(11;14) presented two distinct transcriptomic patterns, in particular for genes implicated in the apoptotic machinery. Three members of the BCL2 family were differentially expressed with BCL2 and PMAIP1 [NOXA] significantly overexpressed and BCL2L1 significantly underexpressed in pPCL with t(11;14). Median PFS and OS of patients with pPCL were respectively at 11 and 15 months. Presence of TP53 mutations was associated with a significantly lower PFS (4 months, p<0,05) and OS (5 months, p<0,05). Neither the IgH translocations nor the ploidy status predicted for survival. Conclusion: To our knowledge, we present the study on the largest series of patients with pPCL. Our results provide new information on both genomic and transcriptomic landscape of pPCL. Despite their heterogeneity, pPCL present a specific mutational landscape with high prevalence of t(11;14) and high-risk genomic features. These results help to better understand oncogenicity and the aggressive behavior of pPCL and support the use of new treatment strategies such as BCL2 inhibitor (Venetoclax) for pPCL with t(11;14). Disclosures Perrot: Amgen, BMS/Celgene, Janssen, Sanofi, Takeda: Consultancy, Honoraria, Research Funding. Hulin:Celgene/Bristol-Myers Squibb, Janssen, GlaxoSmithKline, and Takeda: Honoraria.

Although anti-PD-1 and anti-CTLA-4 based immune checkpoint blockade (ICB) has represented a turning point in cancer care, clinical responses are observed only in a fraction of cancer patients. Most research focuses on the identification of additional inhibitory receptors restraining the anti-tumor functions of CD8+ T cells. By contrast, herein, we found that loss of the activating receptor CD226 (DNAM-1) was a critical mechanism affecting CD8+ T cell responsiveness to TCR stimulation. Using cancer patients’ samples and preclinical mouse models, we discovered that dysfunctional CD226-negative CD8+ T cells progressively accumulated in the tumor microenvironment through a mechanism involving the T-box transcription factor Eomesodermin (EOMES). More importantly, we demonstrated that CD226-negative tumor infiltrating lymphocytes had reduced anti-tumor functions and failed to respond to ICB. Altogether, our results revealed that CD226 loss is a critical immune escape mechanism restraining CD8+ T cell function and potentially affecting the therapeutic efficacy of cancer immunotherapy.

To the editor:Multiple myeloma (MM) is characterized by a large diversity of genetic abnormalities.[1][1][⇓][2]-[3][3] They can be classified in 3 categories: copy number changes, mutations, and translocations involving mainly the IGH gene at 14q32. Translocations are usually balanced

Clonal evolution drives tumor progression, chemoresistance and relapse in cancer. Little is known about clonal selection induced by therapeutic pressure in multiple myeloma. To address this issue, we performed large targeted sequencing of bone marrow plasma cells in 43 multiple myeloma patients at diagnosis and at relapse from exactly the same intensive treatment. The most frequently mutated genes at diagnosis were KRAS (35%), NRAS (28%), DIS3 (16%), BRAF, and LRP1B (12% each). At relapse, the mutational burden was unchanged. Many of the mutations were present at the subclonal level at both time points, including driver ones. According to patients and mutations, we observed different scenarios: selection of a very rare subclone present at diagnosis, appearance, or disappearance of mutations, but also stability. Our data highlight that chemoresistance and relapse could be induced by newly acquired mutations in myeloma drivers but also by (sub)clonal mutations preexisting to the treatment. Importantly, no specific mutation or rearrangement was observed at relapse, demonstrating that intensive treatment has a nonspecific effect on clonal selection in multiple myeloma. Finally, we identified 22 cases of biallelic event, including a double event deletion 17p/TP53mut.

Key Points There is no correlation between ctDNA and bone marrow for MRD by NGS using only immunoglobulin gene rearrangements in myeloma patients.

Multiple myeloma is a plasma cell malignancy characterized by recurrent IgH translocations and well described genomic heterogeneity. Although transcriptome profiles in multiple myeloma has been described, landscape of expressed fusion genes and their clinical impact remains unknown. To provide a comprehensive and detailed fusion gene cartography and suggest new mechanisms of tumorigenesis in multiple myeloma, we performed RNA sequencing in a cohort of 255 newly diagnosed and homogeneously treated multiple myeloma patients with long follow-up. Here, we report that patients have on average 5.5 expressed fusion genes. Kappa and lambda light chains and IgH genes are main partners in a third of all fusion genes. We also identify recurrent fusion genes that significantly impact both progression-free and overall survival and may act as drivers of the disease. Lastly, we find a correlation between the number of fusions, the age of patients and the clinical outcome, strongly suggesting that genomic instability drives prognosis of the disease.