AUTHORS
Alessandro La Ferlita, Christina W Thai2, Kyla Viets Layng3, Lorenzo Grego4, Francesca Cottini, MD5
AFFILIATIONS
1 The Ohio State University, Columbus, Ohio, United States
2 Tempus AI, Inc, Chicago, Illinois, United States
3 Tempus AI, Inc, Chicago, Illinois, United States
4 Tempus AI, Inc, Chicago, Illinois, United States
5 The Ohio State University, Columbus, Ohio, United States
ABSTRACT
Background: Multiple Myeloma (MM) is a malignancy of plasma cells that disproportionately affects elderly adults. Clonal hematopoiesis of indeterminate potential (CHIP), defined by the presence of somatic clonal mutations in hematopoietic stem and progenitor cells without overt hematologic malignancy, occurs in at least 10% of adults over 70. Moreover, CHIP is more prevalent among individuals with cancer, and its presence is associated with an increased risk of hematologic malignancies, cardiovascular disease, and higher all-cause mortality. Limited data exist regarding the clinical characteristics, genomic features, and outcomes of patients (pts) with MM with or without concomitant CHIP.
Methods: Using the Tempus Lens Platform (Tempus AI, Inc., Chicago, IL) and its multimodal, de-identified database, we queried for pts with MM who underwent Tempus xT (DNA; 648 gene) and xR (RNA; whole-transcriptome) sequencing of their bone marrow (BM) aspirate at any point. We evaluated 86 genes previously associated with myeloid CHIP (M-CHIP) (Niroula et al., Nature Medicine, 2021). Pts with ≥1 M-CHIP mutation and VAF >2% were considered CHIP+, while those without M-CHIP mutations served as CHIP- controls. Pts with a concomitant diagnosis of myelodysplastic syndrome or acute leukemia were excluded. Clinical, demographic, cytogenetic, treatment, and molecular data as well as response rates and overall survival (OS) were collected. Fisher’s test and log-rank test were used for comparisons and survival analyses, respectively. Differential expression analysis of RNA-Seq data was performed using the LIMMA R package. Genes with Log2 Fold Change (Log2 FC) > 0.58 and adjusted p-value < 0.05 were considered differentially expressed. Gene set enrichment analysis (GSEA) was performed using the fgsea R package and MSigDB Hallmark as gene signatures.
Results: A total of 398 pts met study criteria. Of these, 177 were classified as MM with CHIP mutations (MM CHIP+), and 221 as MM without CHIP mutations (MM CHIP-). The most common CHIP-altered genes were DNMT3A (48/398, 12.06%), TP53 (42/398, 10.55%), KRAS (26/398, 6.53%), NRAS (24/398, 6.03%), TET2 (20/398, 5.02%), ASXL1 (16/398, 4.02%), SF3B1 (11/398, 2.76%), and PMM1D (6/398, 1.51%). No significant differences were observed in gender distribution, cytogenetic abnormalities, or type of treatment. However, pts with MM CHIP+ were older (median: 65 years [yrs] vs 60 yrs; p = 0.01), more frequently non-Hispanic White (p = 0.04), and had more frequent comorbidities such as back pain (p = 0.008) and chronic kidney disease (p = 0.02). Regarding outcomes, MM CHIP+ pts had an increased risk of disease progression (p < 0.0001) and inferior risk-adjusted OS (median 3.6 yrs vs 7.2 yrs; p = 0.006). RNA-Seq analysis was then performed between the two groups. GSEA identified the WNT signaling pathway as the most upregulated pathway in MM CHIP+ samples compared with MM CHIP- samples. Because bulk RNA-seq of bone marrow aspirates does not allow the identification of the specific cellular origin of differentially expressed genes, publicly available single-cell RNA sequencing (scRNA-seq) data from BM aspirates of 16 MM pts (8 CHIP+ and 8 CHIP-) (Lionetti et al., Blood. 2025) were analyzed to identify plasma cell-specific genes dysregulated in MM CHIP+ pts and compare them with bulk RNA-seq findings. 52 genes were upregulated in both the scRNA-seq plasma cells and bulk RNA-seq, out of 677 total upregulated genes in scRNA-seq and 676 in bulk RNA-seq. The genes upregulated in both datasets included DKK1, CCND1, TNFRSF17, TNFRSF13B, and NCAM1, all of which have established roles in MM biology. Higher expression of these individual genes was correlated with poorer OS. Finally, we evaluated single-gene mutations relative to the entire MM CHIP- cohort. Pts with TET2 and ASXL1 mutations had a shorter median OS compared to CHIP- (TET2: 2.2 yrs vs 7.2 yrs, p = 0.08; ASXL1: 1.4 yrs vs 7.2 yrs, p = 0.005) and an increased risk of progressive disease, suggesting that these genes may be key drivers of adverse outcomes.
Conclusions: The concomitant presence of CHIP mutations and MM might define a separate subset of pts with inferior outcomes and specific disease characteristics. These findings suggest that the use of unselected next-generation sequencing may be clinically informative for identifying these phenotypes. Further studies are warranted to validate these observations in additional cohorts.
DISCLOSURES
Francesca Cottini, MD: Pfizer - Research Funding; Sanofi - Consultancy (Includes expert testimony)