Deciphering Variability in Light Chain Amyloidosis and Amyloidosis with Co-Occurring Multiple Myeloma By Single-Cell RNA Sequencing

Immunoglobulin light-chain amyloidosis (AL) and multiple myeloma (MM) are both caused by atypical plasma cell (PC) expansion. Notably, AL has been reported to be associated with MM in approximately 10-15% of cases with the two conditions potentially coexisting (Rajkumar, SV et al. 1998). Understanding the interaction between these diseases is vital for improving patient outcomes and developing targeted treatments. Investigating the tumor microenvironment (TME) in cases of AL with co-occurring MM or MGUS could enhance diagnosis and therapy, providing valuable insights into disease mechanisms and potential therapeutic targets for personalized treatment strategies.

The main objective of our study was to accurately understand the variability and heterogeneity at the cellular level in patients diagnosed with only AL and patients with AL_MM/MGUS using single cell RNA sequencing (scRNA-seq).

The Finnish Hematology Registry and Biobank provided 21 samples of viably frozen bone marrow mononuclear cells (BMMNC) from 20 AL patients with all samples taken at diagnosis (n=20) and one patient additionally providing a sample at relapse (n=1). Of these samples, 9 were from patients with co-occurring MM or monoclonal gammopathy of unknown significance (MGUS) and referred to as AL_MM/MGUS. Associated clinical data included cytogenetic information, which was obtained by fluorescent in situ hybridization. For scRNA-seq, BMMNCs were thawed and sorted based on CD138 expression and cell viability (7AAD dead cell marker). Viable CD138+ and CD138- cells were mixed in a ratio where CD138+ cells did not exceed 50%. For scRNA-seq, Chromium Single Cell 3’RNAseq and library preparation were done using Chromium™ Single Cell 3' Reagent version 3 chemistry. 10x Genomics Cell Ranger v3.0.1 pipelines were used for initial data processing. Quality control and down-stream analyses were performed using Seurat 5.1.0. To assign cell type identity to the clusters we used Azimuth (Stuart, T et al. 2019).

Our cohort was comprised of 13 male and 8 female patients with a median age of 66 years and median survival time of 1416.5 days. Twelve patients had AL and nine patients had AL_MM/MGUS. Cytogenetic analysis showed that 1q gain (6/21), t(11;14) (5/21) and trisomy 11 (5/21) were the most frequent abnormalities. A comparison of cytogenetic aberrations associated with AL and AL_MM/MGUS revealed that 1q gain was exclusively associated with the AL_MM/MGUS group (p<0.01).

From the integrated analysis of scRNA-seq data from 21 patient samples, we identified a total of 27 distinct clusters of cells belonging to 14 unique cell types. Out of the 27 clusters, 6 clusters were identified as plasma cell (PC) subpopulations. The proportions of PCs in the samples negatively correlated to the other major cell types such as CD4 memory (r=-0.57; p=0.0078) and CD14 monocytes (r=-0.49; p=0.025). Additionally, the different PC clusters exhibited distinct gene expression profiles revealing considerable heterogeneity at the inter and intra individual level.

Comparative analysis of the AL and AL_MM/MGUS sub-groups revealed significantly higher proportions of GMP, HSC, cDC2 and T cells in the AL group (p<0.05). Differential expression analysis of the PCs from the AL and AL_MM/MGUS sub-groups revealed 93 up-regulated and 133 down-regulated genes in AL_MM/MGUS samples, where 31/93 up-regulated genes were located on the chromosome 1q arm. Relative pathway enrichment analysis of the PC populations in the AL and AL_MM/MGUS samples using Single Cell Pathway Analysis (Bibby, J A et al. 2022) exposed significant differences in the multivariate distributions of pathways like apoptosis, TNF signaling via NFkB, p53 pathway, IL2 STAT5 signaling and MYC target V1 indicating distinct gene expression patterns of PCs from patients with AL and those from patients with AL_MM/MGUS.

In summary, our findings revealed exclusive association of 1q gain in AL_MM/MGUS, and distinct transcriptional profiles of PCs associated with AL or with AL_MM/MGUS. Comparison of the AL and AL_MM/MGUS TME showed similarities and differences, which could be useful for development of immune-based therapies. Overall, our study contributes to our understanding of AL and highlights the molecular mechanisms driving progression to AL with co-occurring MM/MGUS, which could support improved patient stratification and the design of targeted treatment strategies.