Multiple myeloma (MM) is an incurable malignancy, primarily of the bone marrow (BM) caused by uncontrolled proliferation of neoplastic antibody-secreting plasma cells (PCs). It is the second most common haematological malignancy with >188 000 new cases diagnosed annually worldwide.1 Despite therapy advances, the progression-free survival (PFS) for MM patients over the last two decades has not improved and the 5-year overall survival (OS) has improved but remains poor at 54%.1, 2 Furthermore, the recent introduction of therapeutic risk stratification based on high-risk genomic lesions3 has not led to an improvement in OS.2, 4, 5 Rapid identification of high-risk patients at diagnosis would undoubtedly improve outcomes by informing early treatment selection. The presence of cytogenetic lesions, namely t(4;14), t(14;16) and deletion of 17p (del17p), are considered indicators of high-risk disease3 and are now incorporated into the Revised International Staging System (R-ISS), the most commonly implemented MM risk stratification strategy. This means that identification of these lesions is crucial prior to treatment initiation. Recently, 1q21 gain/amplification (1q+), mutations of TP53 and biallelic deletion of 1p32 (del1p) have also been identified to be prognostic for high-risk disease. However, some patients with standard cytogenetics can do poorly, suggesting the presence of additional biological or clinical factors that influence disease trajectory. Identifying these factors is critical to refining risk stratification and improving effective personalised therapeutic strategies in MM. We previously identified that desmoglein-2 (DSG2) expression is significantly elevated in the PCs of approximately 30% of MM patients and is prognostic for a fourfold increased risk of death.6 DSG2 is a cell surface expressed cadherin family adhesion protein that plays non-canonical roles in supporting proliferation and survival of non-desmosome-forming progenitor cells (endothelial, haematopoietic)7, 8 and has been increasingly implicated in solid tumour progression. The aim of this study was to elucidate the capacity of DSG2 expression as a prognostic marker in newly diagnosed (ND) MM within clinically defined cytogenetic cohorts for the first time and to explore the DSG2-associated gene expression profile that may contribute to disease progression. The level of DSG2 expression and its prognostic value were determined within contemporary cytogenetic subtypes using publicly available transcriptomic data from CD138+ malignant PCs from 678 NDMM patients (where cytogenetic/FISH and clinical outcome data were available) (Table S1), sourced from the MM research fund (MMRF)-coMMpass cohort (NCT01454297) (MMRF gateway). In this cohort, samples were stratified into quartiles based on DSG2 expression; MM patients within the top 25% for DSG2 (Q4) expression had significantly shorter PFS (median survival 27.8 vs. 37.2 months, HR 1.42, 95% CI 1.12–1.78, p = 0.026) (Figure 1A) and OS times (54.9 months vs. not reached, HR 1.81, 95% CI 1.37–2.41, p |2|), including 23 upregulated and 106 downregulated genes (Figure 1D; Table S3). Within the DSG2high cohort, upregulation of NSD2, KLF4, IGF1R, ROBO3 and HMGN5 (all FDR |2|) that were all upregulated in the DSG2high patient samples (Figure 2F; Figure S5F; Table S7). GSEA showed that elevated DSG2 expression in 1q+ (no translocations) NDMM-PCs was associated with downregulation of genes in pathways related to ribosomes, nonsense-mediated decay and translation (FDR q < 0.05) (Figure 2G; Figure S5G; Table S8). Downregulation of genes associated with ribosomal pathways has previously been reported to contribute to suboptimal response to bortezomib.15 Given the poor response rates of DSG2high NDMM patients to this proteasome inhibitor,6 the influence of ribosomal pathways warrants further investigation to elucidate the role(s) of DSG2 in high-risk MM. This study highlights the potential of DSG2 as a novel prognostic biomarker for high-risk NDMM, reveals that DSG2 expression is associated with chromosomal aberrations and identifies pathways associated with increased DSG2. Elevated DSG2 expression was predictive of poor OS among NDMM patients, including those with t(11;14) and/or 1q+, regardless of the presence of additional genomic alterations. Elevated DSG2 expression was associated with a distinct gene expression profile in NDMM-PCs and within patients with the 1q+ high-risk cytogenetic lesion. These findings implicate DSG2 as an underappreciated MM prognostic biomarker and provide the rationale for further investigation into its role(s) in MM, including whether targeting DSG2 may represent a novel approach for precision medicines. Moreover, this study suggests that DSG2 expression analysis would complement the current risk stratification strategies, aiding in the rapid identification of high-risk patients at diagnosis. Claudine S. Bonder: Investigation; writing – review and editing; supervision; funding acquisition. John Toubia: Investigation; writing – review and editing; methodology; formal analysis; data curation. Chung Hoow Kok: Conceptualization; investigation; writing – review and editing; supervision; data curation; formal analysis. Barbara J. McClure: Conceptualization; writing – original draft; formal analysis; investigation; data curation. Angelina Yong: Investigation; writing – review and editing. O. Giles Best: Writing – review and editing; investigation. Cindy H. Lee: Investigation; writing – review and editing. Charlotte E. J. Toomes: Writing – review and editing; investigation. These data were generated as part of the Multiple Myeloma Research Foundation Personalized Medicine Initiatives (https://research.themmrf.org and www.themmrf.org). The authors wish to acknowledge Ms. Leanne Winning for sharing her lived experience perspectives of smouldering myeloma. Open access publishing facilitated by Adelaide University, as part of the Wiley - Adelaide University agreement via the Council of Australasian University Librarians. This work was supported in part by funding to CSB from the National Health and Medical Research Council (NHMRC) (GNT2013460) and to BJM from the University of South Australia. CSB has received research funding from Carina Biotech Pty. CL is an Advisory Board member for Janssen, Takeda, Pfizer, Antengene and BMS and has received research funding from them. All authors declare no competing interests. This study analysed publicly available, de-identified datasets and did not involve the recruitment of new human subjects. Therefore, ethical approval was not required. Patients in the MMRF CoMMpass study (NCT01454297) provided written consent to participate in a longitudinal, clinical-genomic study. The consent was conducted as per the Declaration of Helsinki guidelines. These data were generated as part of the Multiple Myeloma Research Foundation Personalized Medicine Initiatives (https://research.themmrf.org and www.themmrf.org). Data were retrieved from the Genomic Data Commons (GDC) data portal on 28 July 2026 (https://portal.gdc.cancer.gov/). Figures S1–S5. Tables S1–S8. Data S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
McClure et al. (Mon,) studied this question.