Clonal hematopoiesis is clonally unrelated to multiple myeloma and is associated with specific microenvironmental changes

Abstract

Multiple myeloma (MM) initiation is dictated by genomic events. However, its progression from asymptomatic stages to an aggressive disease that ultimately fails to respond to treatments is also dependent on changes of the tumor microenvironment (TME). Clonal hematopoiesis of indeterminate potential (CHIP) is a prevalent clonal condition of the hematopoietic stem cell whose presence is causally linked to a more inflamed microenvironment. Here, we demonstrate in 106 patients with MM that CHIP is frequently coexisting with MM at diagnosis, associates with a more advanced Revised International Staging System stage and higher age, and has a nonsignificant trend toward lower median hemoglobin. In our cohort, the 2 conditions do not share a clonal origin. Single-cell RNA sequencing in 16 patients with MM highlights significant TME changes when CHIP is present: decreased naive T cells, a proinflammatory TME, decreased antigen-presenting function by dendritic cells, and expression of exhaustion markers in CD8 cells. Inferred interactions between cell types in CHIP-positive TME suggested that especially monocytes, T cells, and clonal plasma cells may have a prominent role in mediating inflammation, immune evasion, and pro-survival signals in favor of MM cells. Altogether, our data reveal that, in the presence of CHIP, the TME of MM at diagnosis is significantly disrupted in line with what is usually found in more advanced disease, with potential translational implications. Our data highlight the relevance of this association and prompt for further studies on the modifier role of CHIP in the MM TME.

Graphical abstract

Figure 1.

Mutational spectrum of CHIP. (A) Oncoplot of mutations in CHIP-associated genes detected in the PB of patients with MM at diagnosis. Each column represents 1 tumor sample and each row a gene. The heat map is color coded according to (maximum) VAF at which each gene was found mutated in the corresponding sample, as depicted in the color-scale bar at the bottom. In case the variant was found also in DNA from BM PCs, the gene reveals a black contour. For mutations whose MM origin could not be excluded due to unavailability of PC DNA, genes are highlighted with a diagonal bar. The number and the types of mutations (as specified in the legend) identified in each gene are plotted on the right of the heat map, preceded by the indication of the percentage of mutated patients. The bar graph above the heat map represents the number of variants per patient. (B) Dynamics of CHIP-associated mutations detected in longitudinally analyzed patients, suggesting patterns of branching (MM16) or linear evolution (MM4), or stability, expansion, or reduction of subclones present both at diagnosis and during follow-up (MM21, MM27, MM33, and MM9). The bar charts represent the VAF of the variants (listed in the legend on the right) at each time point (color coded according to the disease phase, as detailed under the graphs).

Figure 2.

Clinical correlates of CHIP. (A) Box plots of age and hemoglobin levels in analyzed patients according to their CHIP status. (B) Stacked bar chart representing the number of CHIP-negative (CHIP-neg) and CHIP-positive (CHIP-pos) patients stratified by their Revised International Staging System (R-ISS) stage.

Figure 3.

Clonal interrelationships between MM and CHIP. Oncoplot representing the occurrence of mutations indicative of myeloid or lymphoid clonality in HSCs and PCs flow sorted from the BM of 8 CHIP-positive patients with MM. For each patient, the CHIP-defining mutated genes identified in peripheral WBCs are also reported, in the first column. The smaller orange square refers to a mutation at a VAF <2%. The abbreviation “n.a.” stands for not available information and regards the mutational status of MM driver genes, not sequenced in peripheral WBCs.

Figure 4.

Cellular composition of the BM microenvironment. (A) 2D UMAP representation of the 132 752 analyzed cells color coded by assigned cell type (left), patient (top right), or relative CHIP status (bottom right). (B) Box plot revealing the relative abundance of different T and NK cell subtypes between CHIP-negative (bright blue) and CHIP-positive (deep red) patients. ∗adjusted P value <.05; ∗∗∗adjusted P value <.001. gdT, gamma-delta T cells; MAIT, mucosa-associated invariant T cells; UMAP, uniform manifold approximation and projection.

Figure 5.

Single-cell transcriptomic characterization of the clonal BM microenvironment. (A) Cartoon depicting selected hallmark gene sets related to inflammatory response and significantly upregulated in ≥2 of the depicted cell types in CHIP-positive patients. Circle size corresponds to the number of cells in the category expressing the genes of interest, whereas the color represents the P value, as detailed in the legend of each of the GSEA analyses reported in supplemental Figure 3A-E. (B) Violin plot representing the distribution of peptide antigen binding score in myeloid DCs (Prog_DCs and cDC2) according to the occurrence of CHIP. (C) Violin plot representing the distribution of M1 macrophage polarization score (supplemental Methods) in CD14 and CD16 monocytes according to the occurrence of CHIP. ∗adjusted P value <.05; ∗∗∗adjusted P value <.001. GSEA, gene set enrichment analysis.

Figure 6.

Top 100 most differential cell-cell interactions between CHIP-positive and CHIP-negative groups. (A) Chord diagram visualization of the top 100 prioritized interactions in clonal (left) and nonclonal (right) TME. The arrowhead indicates the direction from sender to receiver cell type, and the color of the arrow indicates the sender cell type that expresses the ligand. (B) Representation of ligand activity, cell-type specificity of expression, and fraction of expression for a selected subset of ligand-receptor interactions.

Figure 7.

Expression of T-cell exhaustion markers in CD8 T cells according to CHIP status. (A) Dot plot displaying the expression of selected marker genes in CD8 T precursor exhausted cells of CHIP-positive and CHIP-negative samples. The X-axis lists gene names, whereas the Y-axis lists the CHIP status. Circle size corresponds to the percentage of cells in the category expressing the gene of interest, whereas shade correlates with the level of expression. (B) Violin plot depicting the T-cell dysfunction score computed in CD8 T-cell subpopulations of CHIP-negative and CHIP-positive samples. ∗adjusted P value <.05; ∗∗∗adjusted P value <.001.