AUTHORS
Neela Easton, MD, Joshua Kassner, MD2, Roger Pearse, MD3, Victor Hugo Jimenez-Zepeda, MD4, Ruben Niesvizky, MD5, Mateo Mejia Saldarriaga, Assistant Professor6, Mark Bustoros, MD7
AFFILIATIONS
1 Weill Cornell Medicine, New York, New York, United States
2 Weill Cornell Medicine, New York, New York, United States
3 Weill Cornell Medicine, New York, New York, United States
4 Weill Cornell Medicine, New York, New York, United States
5 Weill Cornell Medicine, New York, New York, United States
6 Weill Cornell Medicine, New York, New York, United States
7 Weill Cornell Medicine, New York, New York, United States
ABSTRACT
Background
Light chain-only multiple myeloma (LCMM), defined by suppression of intact immunoglobulin production, comprises ~20% of newly diagnosed MM but remains understudied, with unclear molecular differences from standard MM (stdMM). Using the publicly available CoMMpass dataset of newly diagnosed MM patients, we characterized the clinical presentation, transcriptome, cytogenetics, and mutational landscape of LCMM relative to stdMM.
Methods
Patients were classified using IMWG definitions and baseline labs into LCMM (n=278: M-protein <1 g/dL, abnormal κ/λ, involved FLC ≥10 mg/dL) and stdMM (n=835). Baseline characteristics were compared by Wilcoxon rank-sum and chi-squared tests. Overall survival (OS)/progression-free survival (PFS) were modeled with Cox proportional hazards adjusted for age, transplant, induction regimen, and ISS. Baseline RNA-seq was analyzed with DESeq2 and Gene Set Enrichment Analysis (GSEA) (KEGG, GO, and a published MM molecular subclassification by Broyl et al.). Cytogenetic and somatic mutation data were compared by Fisher's exact test with Benjamini-Hochberg correction.
Results
Compared with stdMM, LCMM patients were similar in age at diagnosis (LCMM and stdMM median 63 years; p=0.68), sex (p=0.84), and race (p=0.72) but had higher rates of ISS stage III disease (34% vs 26%; p<0.001), renal insufficiency (29% vs 11%; p<0.001), lytic bone disease (68% vs 59%; p=0.01), hypercalcemia (7.9% vs 4.4%; p=0.04), and elevated β2-microglobulin (34% vs 26%; p=0.02). Adjusted PFS was worse in LCMM (HR 1.20; p=0.047) while adjusted OS did not differ (HR 1.21; p=0.10).
Transcriptome analysis showed 6008 genes that were differentially expressed (padj < 0.05). 180 genes met the threshold of padj<0.05 and |log2FC|>1, including 64 immunoglobulin-related genes reflecting downregulation of IgH and upregulation of IgK/IgL in LCMM. LCMM also showed upregulation of cell-cycle/DNA-replication and repair programs and a cancer-testis antigen (CTA) gene cluster (MYEOV, PAGE5, SSX6, FATE1), alongside downregulation of MAPK signaling, ribosomal/translational machinery, and NF-κB/chemokine pathways (MYD88, TLR2, TRAF6, CCL/CXCL families). GSEA against published MM transcription profiles corroborated these findings: LCMM was enriched for proliferation (PR; NES 2.34, padj<0.001) and CTA (NES 2.20, padj<0.001) signatures and depleted for the myeloid/innate immune signature (NES -1.88, padj=0.01). Individual differentially expressed genes also suggested extracellular matrix (ECM)/adhesion remodeling, including upregulation of the osteoblast-associated BGLAP and matrix proteins MFAP4, FMOD, and NID2, despite absent pathway-level enrichment.
Cytogenetically, LCMM was enriched for t(11;14) (odds ratio [OR] 2.4; 95% CI 1.6-3.5; padj<0.001), t(6;14) (OR 5.8; 95% CI 1.7-22.3; padj=0.006), del(17p13) (OR 1.9; 95% CI 1.1-3.1; padj=0.047) and del(14q23) (OR 2.5; 95% CI 1.7-3.5; padj<0.001) and had lower frequency of t(4;14) (OR 0.4; 95% CI 0.2-0.8; padj=0.007) and hyperdiploidy (OR 0.5; 95% CI 0.4-0.7; padj<0.001). Co-occurring del(14q23) and IgH translocation were more common in LCMM (OR 2.0; 95% CI 1.3-3.2; p=0.002). Among 92 pre-selected common MM mutations, only TP53 mutation was significant after BH correction (OR 3.03; padj=0.045). Biallelic TP53 disruption (del(17p13) + TP53 mut) was more common in LCMM (OR 3.7; 95% CI 1.6-8.6; p=0.001). Combined RAS-MAPK pathway mutations (KRAS/NRAS/BRAF/NF1/RASA2/PTPN11) were less common in LCMM (OR 0.67; 95% CI 0.49-0.92; p=0.01).
Conclusion
LCMM presents a higher-risk clinical phenotype, with more renal and bone disease at diagnosis and worse PFS than stdMM in the triplet era. LCMM is characterized by a distinct genomic architecture of CCND-family translocations, TP53 disruption, chromosome 14 deletion and translocation, and non-hyperdiploidy, coupled with downregulation of inflammatory/ribosomal transcriptional programs and upregulation of proliferative and cancer-testis antigen programs with altered ECM/adhesion gene expression. These findings support LCMM as a biologically distinct MM phenotype. Further characterization may inform risk stratification and reveal therapeutic vulnerabilities.
DISCLOSURES
Neela Easton, MD: No relevant conflicts of interest to declare.