AUTHORS
William Cain Pilcher Jr, BSc, Marina E Michaud, MSc2, Denis Ohlstrom, MSc3, Katherine E Ferguson4, Sarthak Satpathy5, Steven Foltz, PhD6, Chaitanya Acharya, PhD7, Immune Atlas Consortium8, Ioannis Vlachos, PhD9, David E. Avigan, MD10, Shaji Kumar, MD11, Ravi Vij, MD, MBA12, Li Ding, PhD13, Sacha Gnjatic, PhD14, Hearn Cho, MD, PhD15, Samir Parekh, MD16, George Mulligan, PhD17, Sagar Lonial, MD, FACP18, Ajay K Nooka, MD, MPH19, Manoj Bhasin, PhD20
AFFILIATIONS
1 Emory Winship Cancer Institute, Atlanta, Georgia, United States
2 Emory Winship Cancer Institute, Atlanta, Georgia, United States
3 Georgia Institute of Technology, Atlanta, Georgia, United States
4 Georgia Institute of Technology, Atlanta, Georgia, United States
5 Emory Winship Cancer Institute, Atlanta, Georgia, United States
6 Multiple Myeloma Research Foundation, Norwalk, Connecticut, United States
7 Multiple Myeloma Research Foundation, Norwalk, Connecticut, United States
8 Multiple Myeloma Research Foundation, Norwalk, Connecticut, United States
9 Beth Israel Deaconess Medical Center, Boston, Massachusetts, United States
10 Beth Israel Deaconess Medical Center, Boston, Massachusetts, United States
11 Mayo Clinic, Rochester, Minnesota, United States
12 Washington University School of Medicine in St. Louis, St Louis, Missouri, United States
13 Washington University School of Medicine in St. Louis, St Louis, Missouri, United States
14 Icahn School of Medicine at Mount Sinai, New York, New York, United States
15 Icahn School of Medicine at Mount Sinai, New York, New York, United States, Multiple Myeloma Research Foundation, Norwalk, Connecticut, United States
16 Icahn School of Medicine at Mount Sinai, New York, New York, United States
17 Multiple Myeloma Research Foundation, Norwalk, Connecticut, United States
18 Emory Winship Cancer Institute, Atlanta, Georgia, United States
19 Emory Winship Cancer Institute, Atlanta, Georgia, United States
20 Emory Winship Cancer Institute, Atlanta, Georgia, United States
ABSTRACT
Introduction: Functional high-risk (FHR) or rapid progression of myeloma, defined by progression within 18 months, remains poorly explained by conventional baseline risk assessment. Although updated IMS/IMWG consensus genomic staging (CGS) criteria refine high-risk stratification, it is unknown whether rapid progression arises through the same biological mechanisms across genomic risk groups. We investigated whether FHR/RP reflects immune-microenvironmental failure, tumor-intrinsic aggressiveness, or distinct combinations within standard- and high-risk myeloma.
Methods: We analyzed baseline single-cell RNA sequencing profiles from our MMRF Immune Atlas cohort (Pilcher et al., 2026), comparing rapid progressors (RP; PFS<18 months) with durable non-progressors (NP; PFS>4 years) within CGS standard-risk (SR; n=208) and high-risk (HR; n=78) groups. Multi-hit disease (n=15) was defined by two or more CGS high-risk features: TP53 mutation or del(17p13), a high-risk IgH translocation with gain(1q21) or del(1p32), gain(1q21) with del(1p32), biallelic del(1p32), or elevated β2-microglobulin with normal creatinine. Differences in immune-cell abundance and immune and tumor-cell transcriptional profiles were evaluated using multivariable models adjusted for clinical, demographic, and treatment covariates. Tumor-cell findings were independently evaluated using CD138⁺ bulk RNA sequencing from MMRF CoMMpass samples not included in the Immune Atlas (n=412). Numbat and cellSNP-lite were used to reconstruct plasma-cell clonal architecture from copy-number alterations and mitochondrial variants estimated from scRNA-seq data. Presence of t(4;14) in scRNA-seq data was estimated by NSD2 and FGFR3 expression.
Results: RP patients with SR cytogenetic profiles exhibited remodeling of anti-tumoral immune subpopulations, characterized by depletion of naïve CD4⁺ and CD8⁺ T cells, broad reductions across B-cell states, and enrichment of cytotoxic, HLA-expressing late activated CD8⁺ T cells (p<0.05). Despite early progression, these patients showed no clear enrichment of plasma-cell proliferation or established high-risk transcriptional signatures, revealing an immune-defined form of FHR not captured by genomic staging.
In contrast, RP patients with HR cytogenetic profiles showed limited immune remodeling but marked plasma-cell proliferation and enrichment of the UAMS70 high-risk signature (p<0.01). Further stratification of CGS HR patients showed that patients with two or more CGS high-risk features had substantially inferior progression-free and overall survival compared to single-hit and standard-risk patients (PFS: HR=3.73, p=1.32×10⁻⁵; OS: HR=5.08, p=1.27×10⁻⁶). Tumor cells from these multi-hit patients demonstrated the strongest proliferative and high-risk gene signatures. This is further supported through independent analysis of the CoMMpass cohort, confirming significantly higher proliferation, UAMS70, and EMC92 scores in patients with multi-hit disease compared to those with individual high-risk features (adjusted p<0.05).
Clonal reconstruction using innovative single-cell data analysis demonstrated that multi-hit disease has highly heterogeneous plasma-cell profiles. While some samples contained a single dominant plasma-cell population without discernible subclonal structure, others contained transcriptionally distinct clones with different cytogenetic alterations. In a representative patient, all clonotypes shared NSD2 translocation and del(17p13), whereas two major subclones harbored del(1p32) and gain(1q21), respectively, each with altered plasma-cell transcriptomes. Strikingly, enrichment of UAMS70 high-risk and proliferation signatures was concentrated within the gain(1q21) subclone plasma-cell cluster, directly linking the clonal genetic architecture to aggressive tumor-cell states.
Conclusions: These findings present at least two biologically distinct routes to early progression in FHR multiple myeloma: depletion of key anti-tumoral immune subpopulations in SR disease and activation of tumor-intrinsic proliferation pathways in HR disease. Multi-hit CGS identifies an exceptionally adverse subgroup in which aggressive transcriptional programs may be restricted to specific genetically defined subclones. These findings move FHR beyond a single clinical label and support the integration of genomic risk, immune state, and clonal architecture to identify mechanistically distinct patient groups that require different therapeutic strategies.
DISCLOSURES
William Cain Pilcher Jr, BSc: No relevant conflicts of interest to declare.