Unraveling the Molecular Cross-talk in the Comorbidity of Multiple Myeloma and Systemic Light-Chain Amyloidosis through Multi-Dataset Analysis

Unraveling the Molecular Cross-talk in the Comorbidity of Multiple Myeloma and Systemic Light-Chain Amyloidosis through Multi-Dataset Analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Unraveling the Molecular Cross-talk in the Comorbidity of Multiple Myeloma and Systemic Light-Chain Amyloidosis through Multi-Dataset Analysis Xiaoyan Liu, Piaorong Zeng# This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7555168/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: The co-occurrence of multiple myeloma (MM) and light-chain amyloidosis (AL) accelerates disease progression, but their shared mechanisms remain unclear. Methods: We integrated bulk transcriptomic (GSE6477, GSE16558, GSE175384) and single-cell RNA-seq (GSE188222, GSE271107) data. Using WGCNA, PPI networks, and machine learning, we developed a prognostic signature validated in MMRF (N=859) and external cohorts. Immune infiltration and drug sensitivity were analyzed. Results: We identified 41 shared genes and established a 12-gene prognostic signature. High-risk patients showed distinct immune microenvironments and drug responses. Single-cell analysis revealed cell-type-specific expression patterns, with PTP4A3 emerging as a key regulator. Conclusions: This multi-omics study reveals shared molecular mechanisms in MM-AL comorbidity and provides a robust prognostic signature. PTP4A3 represents a potential therapeutic target, offering insights for precision medicine. Multiple myeloma Light-chain amyloidosis Prognostic signature Tumor microenvironment scRNA-seq PTP4A3 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 1. Introduction Multiple myeloma (MM), characterized by the clonal proliferation of malignant plasma cells in the bone marrow, is the second most common hematologic malignancy, with its incidence rising due to population aging 1 . Although novel therapeutic agents and autologous stem cell transplantation (ASCT) have significantly improved initial response rates, approximately 80% of patients eventually relapse. The median overall survival remains suboptimal at 5–7 years, highlighting the persistent challenges in achieving durable remission 2 . Existing prognostic models (e.g., R-ISS, R2-ISS) partially stratify risk but fail to fully address survival heterogeneity, particularly failing to identify subgroups driven by comorbidities 3 – 5 . These limitations underscore the urgent need to develop cross-disease molecular biomarkers. Immunoglobulin light-chain amyloidosis (AL), a rare hematologic disorder caused by misfolded immunoglobulin light-chain deposition in vital organs 6 , was formally incorporated into China's First Rare Disease Catalog in 2018 7 . The disease typically manifests with nonspecific symptoms such as persistent fatigue and peripheral edema, frequently leading to diagnostic delays and progressive damage to critical organs including the heart and kidneys 8 . Prognosis is profoundly influenced by organ involvement patterns, with cardiac AL exhibiting particularly dismal outcomes—median survival plummets to 3.5 months compared to 26.4 months in non-cardiac cases 9 . Although novel therapies like subcutaneous daratumumab have enhanced treatment adherence 10 , 11 , early detection through heightened clinical vigilance and multidisciplinary collaborative care remain cornerstone strategies for improving outcomes 12 . Current diagnosis relies on invasive organ biopsies or mass spectrometry for amyloid detection, which hinders early screening. Thus, noninvasive biomarkers and exploration of shared molecular mechanisms are critical for improving outcomes. Approximately 10–15% of MM patients develop concurrent light-chain amyloidosis, where organ involvement accelerates progression and reduces therapeutic response, exhibiting worse outcomes than isolated MM or AL 13 . MM-AL displays intermediate cytogenetic profiles, with t(11;14) prevalence between MM and AL, and fewer high-risk cytogenetic abnormalities (HRCA) 14 , 15 . These patients achieve lower hematologic responses and median overall survival, with early mortality 13 . Prolonged induction and MRD-guided maintenance improve event-free survival (EFS), yet targeted therapeutic strategies remain underdeveloped 16 . While previous studies have delineated distinct molecular profiles of MM and AL as isolated entities, the shared genetic architecture and prognostic regulatory networks underlying their comorbidity remain uncharted territory. This study pioneers the systematic dissection of core gene modules in MM-AL comorbidity through integrative analysis of large-scale transcriptomic datasets and systems biology approaches. By employing univariate and multivariate regression analyses, we establish a cross-disease prognostic model and elucidate the potential mechanisms by which key genes drive immune microenvironment remodeling and therapeutic resistance. Functional network mapping and drug sensitivity profiling further identify dual-functional therapeutic targets. Our work not only provides a novel molecular classification framework for MM-AL comorbidity but also deciphers the pathophysiological roles of shared genes, laying a theoretical foundation for developing precision therapeutic strategies across disease boundaries. 2. Methods 2.1 Literature Review and Database Search We conducted a comprehensive review of the existing literature to identify the pathophysiological characteristics of MM and AL potential therapeutic targets. Concurrently, we searched the Gene Expression Omnibus (GEO) database ( https://www.ncbi.nlm.nih.gov/geo/ ) 17 using the keywords " Multiple myeloma " and "Systemic Light Chain Amyloidosis" to identify relevant datasets related to AL and MM. 2.2 Data Collection MM training sets ( GSE6477, GSE16558 ) and AL training set ( GSE175384 ) were obtained from GEO. The baseline model incorporated 859 MM samples with survival data from TCGA-MMRF( https://www.cancer.gov/tcga ) 18 . External validation used GSE4581 (414 MM cases) and GSE73040 (AL diagnostic assessment). Immune regulatory genes (638 including 257 Th17/381 Treg-related from GeneCards) 19 (Supplementary Table 1). 2.3 Data Processing Raw datasets were processed using R language and Bioconductor packages, which included steps for background correction, log transformation, and normalization. For bulk RNA-seq data, we applied the transcripts per million (TPM) method to further normalize the gene expression data. To minimize technical variation between samples, we used the “ComBat” algorithm in the “sva” package to correct batch effects 20 . Identification of Differentially Expressed Genes (DEGs): We identified DEGs in AL and MM patients compared to healthy controls using the “limma” package. DEGs were selected based on statistical significance and biological relevance, with a significance threshold of an adjusted p -value 1 21 . 2.4 Functional Enrichment Analysis and Protein-Protein Interaction (PPI) Network Construction The “clusterProfiler” package was employed to perform Gene Ontology (GO) and KEGG pathway enrichment analyses on differentially expressed genes (DEGs) between AL/MM patients and healthy controls (threshold: |log2FC|>1 & FDR < 0.05). Top 10 significantly enriched pathways (FDR-adjusted p < 0.05) were visualized via “ggplot2” 22 . ssGSEA algorithm was used to calculate ssGSEA and GSVA score of MM patients by “GSVA” package in MM dataset 23 . The DEGs were submitted to the STRING database ( https://cn.string-db.org/ ) to compile the interactions of target proteins, with a medium confidence threshold set at 0.4. A comprehensive PPI network was constructed using Cytoscape software, and any proteins lacking connections were excluded 24 . GeneMANIA( https://genemania.org/ ). 2.5 Weighted Gene Go-expression Network Analysis (WGCNA) WGCNA is performed to identify modules of highly correlated genes, summarize the interconnections between modules and associations with external sample traits, and identify candidate biomarkers or therapeutic targets. In our research, WGCNA was constructed by the R package “WGCNA” to identify the modules with the highest relevance to diseases 3 . The R function “pickSoftThreshold” was utilized to determine the optimal soft threshold power (β) for the construction of a scale-free network. The weighted adjacency matrix was subsequently transformed into a topological overlap matrix (TOM) to assess the connectivity within the network. The dendrogram of the TOM matrix was constructed employing the average-linkage hierarchical clustering method. For this investigation, the minimum gene module size was set to 500 to ensure the identification of appropriate modules, and the threshold for merging similar modules was set at 0.2. The gene significance (GS) and module membership (MM, AL) were calculated to correlate modules with clinical traits. Among the identified modules, those requiring further analysis were selected based on their correlation coefficients (r) and P -values. 2.6 Screening of prognosis genes We performed univariate Cox proportional hazards regression analysis using the “survival” R package to identify genes significantly associated with overall survival in the MMRF cohort. Genes meeting the predefined significance threshold ( P 0.05). To mitigate overfitting and refine prognostic signatures, LASSO (least absolute shrinkage and selection operator) regression was implemented via the “glmnet” package, with penalty parameter (λ) optimization guided by 10 fold cross validation (minimum cross-validated error criterion). Survival disparities between risk-stratified groups were quantified using Kaplan-Meier estimators, with log-rank tests applied to evaluate statistical significance across both the discovery (MMRF) and external validation cohorts ( GSE4581 ). All analyses adhered to a two-sided α-level of 0.05 for hypothesis testing 25 . Following the identification of prognostic genes, we employed the Consensus Cluster Plus package to perform consensus clustering on MM samples based on the expression patterns of these genes. The MM samples were categorized into distinct subtypes, and a Kaplan-Meier survival analysis was conducted to compare the survival outcomes among these subtypes. 2.7 Risk Score Calculation and Stratification To calculate the risk score for each patient, we defined a linear predictor function that integrates the expression levels of selected genes weighted by their LASSO derived regression coefficients. Specifically, the risk score was computed as the sum of the product of each gene's expression value and its corresponding coefficient, implemented via the apply function in R to process the dataset. Subsequently, the LASSO regression model was applied to predict risk scores for all samples. Patients were stratified into high- and low-risk groups using the median risk score of the training cohort as the cutoff threshold: samples with scores above the median were classified as "high-risk", whereas those below were designated “low-risk”. Finally, Kaplan-Meier survival analysis with log-rank testing was performed to evaluate survival disparities between the risk-stratified subgroups within the training dataset. 2.8 Development and Validation of Nomograms In this study, we used a nomogram model to predict patients' survival probabilities and evaluated the model's performance via time-dependent ROC curve analysis. Initially, we applied a Cox proportional hazards regression model to patients' survival data to generate linear predictor scores (risk scores). Next, we extracted patients' survival times and event statuses, and performed time-dependent ROC analysis with the timeROC function to calculate AUC values at different time points, assessing the model's discriminatory ability. To visually present the model's performance, we plotted ROC curves in red, orange, and blue for 1, 2 and 3year predictions. 2.9 Methods for Mutation and tumor mutation burden (TMB) Analysis Somatic mutation data were extracted from the high-risk and low-risk groups. Waterfall plots were generated to visualize the top 20 most frequently mutated genes in each group. Additionally, we analyzed the TMB by calculating the total number of mutations per megabase of genome sequenced. Statistical comparisons of TMB between the high-risk and low-risk groups were performed using the Wilcoxon rank-sum test. All analyses were conducted using the “maftools” package in R, and visualizations were created with ggplot2. This comprehensive analysis provided insights into the mutational landscape and TMB differences between risk groups. 2.10 Immune Cell Infiltration and Correlation Analysis The abundance of different immune cell types in MM was quantified using the CIBERSORT algorithm. To assess the relationship between 22 types of immune cells and key genes, we performed Pearson correlation analysis 26 . Pearson correlation analysis was conducted to evaluate the relationships between immune cell fractions and key genes. Correlation heatmaps were generated to visualize these relationships, with significance assessed using the cor.mtest function. 2.11 Drug Sensitivity Analysis Drug sensitivity analysis was performed by predicting IC50 values for 251 anticancer compounds from the CGP2016 database using the “pRRophetic” package. Wilcoxon rank-sum tests ( p < 0.05) identified drugs with differential sensitivity between high- and low-risk groups, while Spearman correlation analysis ( p < 0.001) assessed associations between risk scores and IC50 values. Significant results were visualized through boxplots and scatter plots 27 . 2.12 Single-Gene Analysis and Functional Enrichment ROC analysis was performed using the “pROC” package to evaluate single-gene diagnostic performance. The area under the curve (AUC) with 95% confidence intervals (95% CI) was calculated via bootstrap resampling 28 . Patients were stratified into high/low expression groups based on optimal cutoffs. Kaplan-Meier curves with log-rank tests ( P < 0.05) assessed survival disparities, validated across independent cohorts. The expression trends of prognosis genes in the MM, AL and control groups in the GSE175384, GSE73040, GSE16558 and GSE6477 datasets were analyzed by wilcoxon test. Next, the expression of prognosis genes in different subtypes was compared by kruskal test. 2.13 Single Gene Differential Expression and Functional Enrichment Analysis We conducted an in-depth exploration of the expression differences and potential biological functions of core genes under specific biological conditions using single-gene analysis methods. Various bioinformatics tools in R, including ggplot2, limma, pheatmap, ggsci, clusterProfiler, enrichplot, and patchwork, were employed for differential expression analysis and KEGG pathway enrichment analysis of core genes. By comparing sample groups with high and low expression, we identified significant DEGs and analyzed their enrichment in relevant KEGG pathways. 2.14 Single-cell sequencing analysis In this study, we obtained single-cell RNA sequencing datasets related to AL ( GSE188222 ), and MM( GSE271107 ) from the GEO database. Utilizing the R programming language and the Seurat package from Bioconductor, we conducted quality control, normalization, identification of highly variable genes, dimensionality reduction through Principal Component Analysis (PCA), and clustering analysis. Visualization was performed using the Uniform Manifold Approximation and Projection (UMAP) algorithm. Additionally, we employed the “SingleR” package for automatic cell type annotation based on the Cancer Cell Line Encyclopedia (CCLE) database,Subtype-defining marker genes were then identified with FindAllMarkers using a log-fold-change threshold of 1 and adjusted P 1 as final cluster signatures 29 . Statistical analysis All statistical analyses were performed using R software. Unpaired two-tailed t-tests were used to calculate differences between two groups of data. Pearson or Spearman's rank correlation coefficient were applied depending on the distribution and nature of the data. 3. Results 3.1Differential Gene Selection Figure 1 depicts the study flowchart. DEGs were identified using the limma package. In the AL dataset, we identified 2529 upregulated genes and 902 downregulated genes (Fig. 2 A, B). In the MM dataset, there were 785 upregulated genes and 970 downregulated genes (Fig. 2 C, D). The intersection of these two datasets revealed 210 common DEGs, comprising 72 upregulated and 138 downregulated (Fig. 2 E, F). 3.2 Co-expression Network Module To systematically investigate the roles of Th17- and Treg-related genes in MM and AL, we curated gene sets associated with Th17 and Treg cells from the GeneCards database. GSVA and ssGSEA were employed to calculate pathway activity scores, followed by WGCNA to identify disease-associated gene modules. For the AL phenotype WGCNA analysis, we implemented standardized preprocessing procedures, including outlier removal and highly variable gene selection. A hierarchical clustering tree was constructed using TOM, and nine functional modules were obtained after parameter optimization (Fig. 2 G, I, K). Identification of key modules demonstrated the turquoise module (r = 0.92, p = 1e-200) exhibited significant positive correlations with the AL phenotype (Fig. 2 M). Consistent associations between genes and traits were further supported by GS analysis. Ultimately, 6,339 module-specific genes were identified(Fig. 2 O). In the WGCNA analysis for MM phenotype, by optimizing the soft-thresholding power parameter using the dynamic tree-cutting algorithm, seven functional modules were ultimately clustered (Fig. 2 H, J, L). Key module analysis revealed that the turquoise module (r = 0.7, p = 1e − 200) showed significant positive correlations with the MM phenotype. Scatter plots of gene significance (GS) versus trait correlations further validated these finding (Fig. 2 N). A total of 1,878 candidate functional genes were extracted from the core modules. The intersection of AL WGCNA, MM WGCNA, and DEGs genes encompasses a set of 41 central genes (Fig. 2 O). 3.3 Functional Enrichment (GO, KEGG) After performing GO and KEGG pathway enrichment analyses on the 41 shared genes between the two disease states, we found that these genes were significantly enriched in neuronal synaptic signaling, immune cell development, stress responses, and membrane receptor/channel functions ( Fig. 3 A, Supplementary Table 2) . The KEGG enriched pathways suggest critical roles in cancer progression (metastasis, drug resistance), infection immunity, and microenvironment regulation (matrix remodeling, cell adhesion) (Fig. 3 B, Supplementary Table 3 ). The PPI analysis of central genes revealed a network consisting of 52 interacting nodes and 133 edges, involving a total of 37 genes. The network exhibited an average node degree of 5.12 and an average local clustering coefficient of 0.412(Fig. 3 F). 3.4 Identification of Prognostic Genes and Construction of Prognostic Models Following initial identification of hub genes, we conducted univariate Cox proportional hazards modeling to screen for survival-associated biomarkers. Among the evaluated candidates, 23 genes demonstrated significant prognostic relevance (Fig. 4 A, Supplementary Table 4 ). After verifying proportional hazards assumptions (p > 0.05), 12 genes( HOMER3、OPN3、CLEC2D、PLA2G2D、PRR7、RCBTB2、EIF4EBP2、AZIN1、NFIL3、THOP1、PTP4A3、 and LSAMP )were retained for subsequent modeling( Supplementary Table 5) . LASSO regularization analysis confirmed these 12 genes as robust prognostic candidates (Fig. 4 B, C). Kaplan-Meier survival analysis revealed distinct prognostic implications for gene expression levels: elevated expression of HOMER3, OPN3, CLEC2D, PRR7, EIF4EBP2, AZIN1, NFIL3, THOP1 , and PTP4A3 was associated with significantly reduced overall survival, whereas high expression of RCBTB2, PLA2G2D, and LSAMP correlated with prolonged survival outcomes ( Supplementary Fig S2) . The survival curves (Fig. 4 D-G) and ROC curves (Fig. 4 H-K) not only demonstrate significant survival differences and predictive accuracy in the internal validation set but also yield consistent results in the external validation set. In the present study, we constructed heatmaps of prognostic genes for patients with MM to visualize the expression patterns associated with patient outcomes (Fig. 4 L, O). The distribution of risk scores in the training set demonstrated a clear dichotomy between high- and low-risk groups (Fig. 4 M, N). Survival analysis of the training set revealed a significant correlation between higher risk scores and decreased survival probability, as illustrated by the separation of the survival curves for high- and low-risk patients (Fig. 4 P, Q). 3.5 Identification of Prognostic Genes and Construction of Prognostic Models To assess the prognostic utility of the identified gene signature in combination with clinical parameters, we performed a comprehensive analysis. This began with univariate Cox regression analysis and evaluation of the proportional hazards (PH) assumption to identify potential independent prognostic factors. Further multivariate Cox regression analysis confirmed that the risk score, along with age, gender, and International Staging System (ISS) stage, serves as an independent prognostic indicator for MM patients (Fig. 5 A, B). All these factors were determined to be independent risk factors for MM, and a corresponding nomogram was constructed based on these elements (Fig. 5 C), as evidenced by the calibration curve (Fig. 5 D), which demonstrated close agreement between predicted and observed survival probabilities (Fig. 5 F-H ) . In the MMRF dataset, the DCA and calibration curve revealed a strong alignment between predicted and actual survival probabilities (Fig. 5 E). Moreover, the model demonstrated impressive discriminative ability with AUC values of 0.763 at 1 year, 0.798 at 2 years, and 0.780 at 3 years (Fig. 5 D). 3.6 Analysis of mutations To investigate the primary genetic mutations among different risk groups, we analyzed the somatic mutations in tumors from multiple myeloma patients in the MMRF dataset ( Fig. 6 E, F ) . The plots revealed that the top three genes with the highest mutation frequencies in the low-risk group were IGHV2-70, IGLV3-1, and KRAS (Fig. 6 B, D). In the high-risk group, the top three genes with the highest mutation frequencies were IGHV2-70, IGLV3-1, and IGHV2-70D (Fig. 6 A, C). 3.7 Comprehensive Analysis of Immune Cell Infiltration and Drug Sensitivity Analysis of 28 immune cell subsets revealed significant immune microenvironment remodeling between high- and low-risk groups (Fig. 7 A, B). Ten cell types—including resting dendritic cells, M1 macrophages, and resting mast cells—showed marked abundance differences (Fig. 7 D). Six immune subsets (e.g., naive B cells, neutrophils, activated dendritic cells) exhibited distinct infiltration patterns between treatment-naïve and relapsed cohorts (Fig. 7 E), indicating dynamic immune remodeling during disease progression. Correlation analysis identified significant associations between prognostic genes ( LSAMP, PTP4A3, THOP1 ) and immune subsets (Fig. 7 C, Supplementary Figure S3 ). According to statistical significance ( P < 0.05), high-risk group expression showed a negative correlation with drug sensitivity to Bortezomib, Lenalidomide, Doxorubicin, BMS-509744, and a positive correlation with sensitivity to TGX221, Cetuximab, Trametinib, and Navitoclax (Fig. 7 F, G ) . 3.8 Diagnostic Performance of Core Genes To evaluate how well core genes can distinguish AL from MM, we analyzed their diagnostic performance using ROC curves across multiple datasets. Genes with AUC values above 0.7 in independent datasets were considered strong diagnostic markers. In the discovery cohort ( GSE175384 ), all genes achieved AUCs exceeding 0.7( Fig. 8A) . Notably, five of these genes — AZIN1, PLA2G2D, EIF4EBP2, NFIL3 , and PTP4A3 — also demonstrated high diagnostic accuracy in the external validation cohort ( GSE73040 ) ( Fig. 8B) . For the MM specific classification, all genes showed consistent diagnostic superiority ( Fig. 8C) . In the GSE16558 dataset, except for THOP1 , the other genes reliably replicated their efficacy, further solidifying their potential as diagnostic tools ( Fig. 8D) . Figure 8 Diagnostic Performance of Core Genes Across Independent Cohorts. (A) ROC curve analysis evaluating the diagnostic accuracy of core genes for AL in the GSE175384 cohort. (B) Validation of core gene diagnostic efficacy in discriminating AL amyloidosis within the GSE73040 dataset. (C) ROC profiling of core genes for MM classification in the GSE175384 cohort. (D) Cross-dataset validation of core gene diagnostic specificity for MM in GSE16558 . 3.9 Cross-Cohort Validation of Core Gene Prognostic Efficacy We evaluated the prognostic performance of individual genes across independent cohorts using Kaplan-Meier survival analysis. In the MMRF cohort (Fig. 9 A), Kaplan-Meier survival curves of multiple core genes demonstrated significant stratification (log-rank P < 0.05), indicating strong associations with patient outcomes. In the independent GSE4581 cohort (Fig. 9 B), survival curves of OPN3, PRR7, RCBTB2, THOP1, PTP4A3 , and LSAMP showed consistent stratification patterns, supporting their robust prognostic generalizability across datasets. 3.10 Cross-Group Differential Expression Profiling of Core Genes To elucidate the conserved regulatory patterns of core genes across disease subtypes, we performed cross-group differential expression profiling to evaluate expression concordance between discovery and validation cohorts. Wilcoxon rank-sum tests revealed high concordance in expression trends of 12 prognostic genes between the AL training cohort and validation cohort (Fig. 10 B). Notably, AZIN1, PTP4A3 , and PLA2G2D showed significant differential expression across both cohorts ( P < 0.05). Consistent expression patterns were also observed in MM subgroups, indicating robust cross-cohort regulatory consistency (Fig. 10 C). 3.11 Overview of key gene expression at single-cell resolution Single-cell datasets GSE188222 and GSE271107 were downloaded and jointly analysed with Seurat. After quality control and Harmony batch correction, cells from AL, MM and control samples were merged (Fig. 11 A) and annotated against HumanPrimaryCellAtlasData using the SingleR package, yielding six major populations: CD4⁺ T cell, eosinophil, malignant plasma cell, microglial cell, natural killer cell and red blood cell (erythrocyte) (Fig. 11 B). Expression levels of twelve core genes across these six lineages are depicted in Fig. 11 C and D. 4. Discussion MM and AL are clinically closely related, with approximately 10–15% of MM patients developing secondary AL 30 . Once complicated by AL, particularly with involvement of vital organs such as the heart, patients exhibit sharply reduced treatment response rates, significantly increased early mortality, and extremely poor overall prognosis 31 . However, conventional prognostic models (e.g., R-ISS) are primarily based on tumor burden and cytogenetic characteristics of MM itself and fail to adequately capture the underlying molecular drivers responsible for this aggressive comorbid phenotype 32 . This pressing unmet clinical need is the starting point of our study. By integrating multi-omics data, we systematically identified for the first time the core molecular features shared by MM and AL, aiming to elucidate the molecular basis of their comorbidity and provide new tools to address this clinical challenge 33 . In this study, we identified and validated a robust 12-gene prognostic signature that effectively stratifies high-risk MM patients with significantly shortened overall survival. We propose that these high-risk patients represent a subgroup with an elevated propensity for progressing to secondary AL, or already harbor an aggressive molecular phenotype conducive to amyloid deposition. The signature’s strong association with adverse outcomes in MM, combined with marked overexpression in AL, highlights its potential as a predictive tool for identifying MM patients at risk of AL transformation. Future validation in longitudinal MM cohorts will be essential to confirm its utility in the early prediction of secondary AL. Functional enrichment analysis revealed that these 12 core genes are significantly involved in key pathways, including endoplasmic reticulum stress response, protein folding, and metabolic processes. Among them, PTP4A3/PRL-3 resides at the center of a complex regulatory network involving cytokines, transcription factors, kinases, and epigenetic regulators. It promotes cell migration, enhances survival, reprograms metabolism, and initiates positive-feedback loops, thereby comprehensively driving the malignant progression of multiple myeloma and representing a highly promising therapeutic target. Multiple studies have confirmed the pivotal role of PTP4A3/PRL-3 in the pathogenesis of MM 34 . Its overexpression not only directly increases myeloma cell migration and invasion but also profoundly modulates tumor cell survival, metabolism, and gene expression through multiple signaling pathways 35 . In MM, PRL-3 expression is sustained by an IL-6/STAT3 feedforward loop 36 , while it concurrently enhances glycolytic activity via STAT1/2 signaling 34 . Furthermore, PRL-3 activates SRC-family kinases (SFK) to promote migration and survival 37 . Epigenetically, the chromatin remodeler SMARCA2 collaborates with NSD2 to upregulate PRL-3, forming a self-reinforcing circuit particularly prominent in high-risk t(4;14) myeloma, solidifying PRL-3’s role as a central driver of disease progression 38 . Simultaneously, PTP4A3 may impair proteasomal function, further facilitating the aggregation of misfolded light chains. It may also remodel the tumor microenvironment by regulating cytokine signaling—such as increasing immunosuppressive cytokine secretion—thereby fostering a supportive niche for malignant plasma cells. Together, these synergistic mechanisms likely underlie the aggressive comorbidity of MM and AL, offering a molecular framework that explains the poor clinical outcomes in these patients. Our analysis further revealed profound immune microenvironment remodeling in high-risk patients, characterized by a dysfunctional, immunosuppressive state 39 . Drug sensitivity analysis provided clinically actionable insights: high-risk patients showed reduced sensitivity to first-line therapies such as bortezomib and lenalidomide, aligning with their observed treatment resistance and poor prognosis 40 . In contrast, increased sensitivity was observed toward targeted agents such as Trametinib (a MEK inhibitor) 41 . These findings offer a rationale for implementing risk-adapted therapy—early use of targeted combination strategies—to improve outcomes in this high-risk population. 5. Limitations Section Several limitations must be acknowledged. This study has several limitations that should be acknowledged. The prognostic model was primarily developed and validated in multiple myeloma cohorts due to the inherent challenge of obtaining large-scale AL amyloidosis datasets—a rare disease—with complete survival annotations. The analysis was based on existing datasets including : TCGA-MMRF (n = 859) and GEO datasets, and lacked independent experimental validation. While we confirmed the diagnostic accuracy of the signature in available AL cohorts, prospective validation in larger cohorts and functional studies to establish causality for identified genes like PTP4A3 are needed as future work. 6. Conclusion In conclusion, starting from the clinical challenge of MM-AL comorbidity, we have discovered a unifying molecular signature that not effectively identifies high-risk MM patients but also elucidates key biological processes related to proteotoxic stress and microenvironmental dysregulation. This signature provides both mechanistic insight and a practical tool for risk stratification, paving the way for future targeted therapeutic strategies aimed at improving outcomes for this refractory patient population. Declarations Ackowledgements The authors declare that no funding, technical assistance, or other external support was received for this study. Author contributions Liu, X. and Zeng, P. wrote the main manuscript text and prepared all figures. All authors read and approved the final manuscript. Funding No funding. Disclosure statement No potential conflict of interest was reported by the author(s). Data availability The datasets analysed during the current study are available in the Gene Expression Omnibus (GEO) repository [GSE188222, GSE271107, GSE6477, GSE16558, GSE175384, GSE4581, GSE73040] and The Cancer Genome Atlas (TCGA) repository [https://www.cancer.gov/tcga], but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of GEO and TCGA. Code availability Not applicable. Ethics, Consent to Participate, and Consent to Publish declarations Not applicable. Data availability statements References Malard, F. et al. Multiple myeloma. Nat. Rev. Dis. Primer 10 , 45 (2024). Dimopoulos, M. A. et al. Multiple myeloma: EHA-ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up†. Ann. Oncol. Off. J. Eur. Soc. Med. Oncol. 32 , 309–322 (2021). Hagen, P., Zhang, J. & Barton, K. High-risk disease in newly diagnosed multiple myeloma: beyond the R-ISS and IMWG definitions. Blood Cancer J. 12 , 83 (2022). D’Agostino, M. et al. Second Revision of the International Staging System (R2-ISS) for Overall Survival in Multiple Myeloma: A European Myeloma Network (EMN) Report Within the HARMONY Project. J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol. 40 , 3406–3418 (2022). Kuiper, R. et al. Prognostic and predictive performance of R-ISS with SKY92 in older patients with multiple myeloma: the HOVON-87/NMSG-18 trial. Blood Adv. 4 , 6298–6309 (2020). Sanchorawala, V. Systemic Light Chain Amyloidosis. N. Engl. J. Med. 390 , 2295–2307 (2024). Guo, J. et al. National Rare Diseases Registry System (NRDRS): China’s first nation-wide rare diseases demographic analyses. Orphanet J. Rare Dis. 16 , 515 (2021). Gertz, M. A. Immunoglobulin light chain amyloidosis: 2024 update on diagnosis, prognosis, and treatment. Am. J. Hematol. 99 , 309–324 (2024). Yu, C. et al. Light-chain amyloidosis with concomitant symptomatic myeloma (CRAB-SLiM features): clinical characteristics, cytogenetic abnormalities, and outcomes. BMC Cancer 24 , 1449 (2024). Alnasser, S. M., Alharbi, K. S., Almutairy, A. F., Almutairi, S. M. & Alolayan, A. M. Autologous Stem Cell Transplant in Hodgkin’s and Non-Hodgkin’s Lymphoma, Multiple Myeloma, and AL Amyloidosis. Cells 12 , 2855 (2023). Palladini, G. et al. Daratumumab plus CyBorD for patients with newly diagnosed AL amyloidosis: safety run-in results of ANDROMEDA. Blood 136 , 71–80 (2020). Zhao, L. et al. The clinical features and outcomes of systemic light chain amyloidosis with hepatic involvement. Ann. Med. 54 , 1226–1232 (2022). Huang, J. et al. Unveiling the ageing-related genes in diagnosing osteoarthritis with metabolic syndrome by integrated bioinformatics analysis and machine learning. Artif. Cells Nanomedicine Biotechnol. 53 , 57–68 (2025). Ozga, M. et al. AL amyloidosis: The effect of fluorescent in situ hybridization abnormalities on organ involvement and survival. Cancer Med. 10 , 965–973 (2021). Suzuki, K. Diagnosis and treatment of multiple myeloma and AL amyloidosis with focus on improvement of renal lesion. Clin. Exp. Nephrol. 16 , 659–671 (2012). Madan, S. et al. Clinical features and treatment response of light chain (AL) amyloidosis diagnosed in patients with previous diagnosis of multiple myeloma. Mayo Clin. Proc. 85 , 232–238 (2010). Barrett, T. et al. NCBI GEO: archive for functional genomics data sets--update. Nucleic Acids Res. 41 , D991-995 (2013). Tomczak, K., Czerwińska, P. & Wiznerowicz, M. The Cancer Genome Atlas (TCGA): an immeasurable source of knowledge. Contemp. Oncol. Poznan Pol. 19 , A68-77 (2015). Stelzer, G. et al. The GeneCards Suite: From Gene Data Mining to Disease Genome Sequence Analyses. Curr. Protoc. Bioinforma. 54 , 1.30.1-1.30.33 (2016). Huber, W. et al. Orchestrating high-throughput genomic analysis with Bioconductor. Nat. Methods 12 , 115–121 (2015). Smyth, G. K. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat. Appl. Genet. Mol. Biol. 3 , Article3 (2004). Kanehisa, M., Sato, Y., Kawashima, M., Furumichi, M. & Tanabe, M. KEGG as a reference resource for gene and protein annotation. Nucleic Acids Res. 44 , D457-462 (2016). Long, Q., Zhang, X., Ren, F., Wu, X. & Wang, Z.-M. Identification of novel biomarkers, shared molecular signatures and immune cell infiltration in heart and kidney failure by transcriptomics. Front. Immunol. 15 , 1456083 (2024). Szklarczyk, D. et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 51 , D638–D646 (2023). Díaz-Uriarte, R. & Alvarez de Andrés, S. Gene selection and classification of microarray data using random forest. BMC Bioinformatics 7 , 3 (2006). Chen, B., Khodadoust, M. S., Liu, C. L., Newman, A. M. & Alizadeh, A. A. Profiling Tumor Infiltrating Immune Cells with CIBERSORT. Methods Mol. Biol. Clifton NJ 1711 , 243–259 (2018). Geeleher, P., Cox, N. & Huang, R. S. pRRophetic: an R package for prediction of clinical chemotherapeutic response from tumor gene expression levels. PloS One 9 , e107468 (2014). Robin, X. et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics 12 , 77 (2011). Stuart, T. et al. Comprehensive Integration of Single-Cell Data. Cell 177 , 1888-1902.e21 (2019). Yu, C. et al. A Comparative Study of Clinical Characteristics, Cytogenetic Abnormalities and Outcomes Among Multiple Myeloma, Primary Light-Chain Amyloidosis and Multiple Myeloma with Concurrent AL Amyloidosis. Blood 142 , 2032–2032 (2023). Ríos-Tamayo, R. et al. AL Amyloidosis and Multiple Myeloma: A Complex Scenario in Which Cardiac Involvement Remains the Key Prognostic Factor. Life 13 , 1518 (2023). Zvida‐Bloch, T., Muchtar, E., Dispenzieri, A., Shpilberg, O. & Hershkovitz‐Rokah, O. The molecular landscape of AL amyloidosis. Br. J. Haematol. 206 , 1297–1311 (2025). Srivastava, A. et al. Cellular diversity and genetic alterations in light chain amyloidosis and amyloidosis with co-occurring multiple myeloma. Preprint at https://doi.org/10.1101/2025.06.06.25329039 (2025). Smith, C. N. & Blackburn, J. S. PRL-3 promotes a positive feedback loop between STAT1/2-induced gene expression and glycolysis in multiple myeloma. FEBS J. 288 , 6674–6676 (2021). Fagerli, U.-M. et al. Overexpression and involvement in migration by the metastasis-associated phosphatase PRL-3 in human myeloma cells. Blood 111 , 806–815 (2008). Chong, P. S. Y. et al. IL6 Promotes a STAT3-PRL3 Feedforward Loop via SHP2 Repression in Multiple Myeloma. Cancer Res. 79 , 4679–4688 (2019). Abdollahi, P. et al. Src Family Kinases Are Regulated in Multiple Myeloma Cells by Phosphatase of Regenerating Liver-3. Mol. Cancer Res. MCR 15 , 69–77 (2017). Chong, P. S. Y. et al. SMARCA2 Is a Novel Interactor of NSD2 and Regulates Prometastatic PTP4A3 through Chromatin Remodeling in t(4;14) Multiple Myeloma. Cancer Res. 81 , 2332–2344 (2021). Lu, J. et al. Myeloid-derived suppressor cells in cancer: therapeutic targets to overcome tumor immune evasion. Exp. Hematol. Oncol. 13 , 39 (2024). Kastritis, E. et al. Long‐term outcomes of primary systemic light chain (AL) amyloidosis in patients treated upfront with bortezomib or lenalidomide and the importance of risk adapted strategies. Am. J. Hematol. 90 , (2015). Hoffner, Msn, Anp-Bc, Aocnp, B. & Benchich, Msn, Np-C, Aocnp, K. Trametinib: A Targeted Therapy in Metastatic Melanoma. J. Adv. Pract. Oncol. 9 , (2018). Additional Declarations No competing interests reported. Supplementary Files Supplementarymaterials1.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7555168","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":536968934,"identity":"4f4d809d-cb43-4912-9891-f1bc07da54d1","order_by":0,"name":"Xiaoyan Liu","email":"","orcid":"","institution":"Loudi Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyan","middleName":"","lastName":"Liu","suffix":""},{"id":536968935,"identity":"74c2e048-95dc-480c-9456-de12e895dde2","order_by":1,"name":"Piaorong Zeng#","email":"data:image/png;base64,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","orcid":"","institution":"First Affiliated Hospital of Jinan University","correspondingAuthor":true,"prefix":"","firstName":"Piaorong","middleName":"","lastName":"Zeng#","suffix":""}],"badges":[],"createdAt":"2025-09-07 09:08:59","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7555168/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7555168/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95523969,"identity":"7858a8b6-710c-49d3-917d-635e77029354","added_by":"auto","created_at":"2025-11-10 10:01:43","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4765018,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscripts2.docx","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/17d71baa03ad7f296ea7761b.docx"},{"id":95314085,"identity":"1bcd1700-e57e-451f-9d08-7c3e2276e2a5","added_by":"auto","created_at":"2025-11-06 15:52:27","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4118,"visible":true,"origin":"","legend":"","description":"","filename":"1454d6975e9d463c8691a3915411f93b.json","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/ed92700845f19d839e0592c2.json"},{"id":95313709,"identity":"16de1a46-ae9a-484b-b84a-c957311a7569","added_by":"auto","created_at":"2025-11-06 15:51:53","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1089527,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterials1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/345b4552258ce342df7697ad.docx"},{"id":95314372,"identity":"9da1eff2-c238-45a4-b60a-67415c59a92e","added_by":"auto","created_at":"2025-11-06 15:52:43","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":106488,"visible":true,"origin":"","legend":"","description":"","filename":"1454d6975e9d463c8691a3915411f93b1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/7be93345d0054360ce0a74b5.xml"},{"id":95286047,"identity":"34fea7bf-2530-490c-ae14-39d10165eb3e","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":741208,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/bc8a55ba7424c7f40d3a27f8.jpeg"},{"id":95314237,"identity":"67570a5b-eab8-471d-83e7-952920897c0a","added_by":"auto","created_at":"2025-11-06 15:52:36","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":283769,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/b59ed5e9bc71a77c11beaea1.png"},{"id":95286068,"identity":"a5e02dbe-6746-4e14-a16d-7c5dead0de91","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":611816,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage11.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/d36288a9b0a14a98b62d5549.jpeg"},{"id":95313689,"identity":"3f8f8974-3417-4866-80b1-a710a3fd019f","added_by":"auto","created_at":"2025-11-06 15:51:52","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":494622,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/f13fdb72d02755eda9cce0d2.png"},{"id":95286055,"identity":"340c6161-6f38-4aca-9979-7cb352c9c771","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":469639,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/6b48652f9f6114e54a0258df.png"},{"id":95314390,"identity":"61a12f5e-d2b5-44f9-a56e-0708aa98dd76","added_by":"auto","created_at":"2025-11-06 15:52:46","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":355638,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/72377394b7900c1ba702be3f.png"},{"id":95286063,"identity":"034e6d7f-4286-4a56-a0e1-9f32b1ffe526","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124750,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/eadd651d0ef27bb0eee6511d.png"},{"id":95286065,"identity":"25911ad3-970f-4697-a445-eb32b087a22e","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":317113,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/5d169874fbed41e876da4296.png"},{"id":95314494,"identity":"0f9481b0-0e82-4b5e-9181-5cd0c21ee5a2","added_by":"auto","created_at":"2025-11-06 15:52:55","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":469779,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/dfefc5815b0805388adc828f.png"},{"id":95313918,"identity":"8e82ea59-3324-49ea-ba64-c09e6f761872","added_by":"auto","created_at":"2025-11-06 15:52:14","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":365900,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/9b0f7bfbc2afaaadfb9cef54.png"},{"id":95286059,"identity":"89ef2290-d3b8-41e9-922d-38ceda231106","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":354589,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/eacc35bdb04a8979e3a8ca06.png"},{"id":95314045,"identity":"3e3b7500-ba33-45c2-825e-07fd9cbc18f6","added_by":"auto","created_at":"2025-11-06 15:52:25","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":95238,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/d1232db70dfda5fc6b8e2891.png"},{"id":95314031,"identity":"ada67415-7284-420b-9874-2ca14153bf3b","added_by":"auto","created_at":"2025-11-06 15:52:23","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":86345,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/73cb13f87c07fbbda2495f5d.png"},{"id":95286071,"identity":"f54616d1-8f7c-4153-a3ee-e8ec3b0611a4","added_by":"auto","created_at":"2025-11-06 10:00:00","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":104948,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/37195ea195ade28f3405c1b9.png"},{"id":95314209,"identity":"ae5f5b99-3d9b-4931-a0d1-aaa10347caa9","added_by":"auto","created_at":"2025-11-06 15:52:35","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":119973,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/83332f5a3fc8bb6ad260dab1.png"},{"id":95286079,"identity":"8dccb502-d34e-4d51-a670-512705906858","added_by":"auto","created_at":"2025-11-06 10:00:00","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":87731,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/f6a79da083ae18ed8dd1a334.png"},{"id":95314312,"identity":"d7b34f96-d956-4aba-80b3-ab4023089f66","added_by":"auto","created_at":"2025-11-06 15:52:41","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":97907,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/b6f0352588456cfd55eb2a61.png"},{"id":95314087,"identity":"67167275-c1af-41e1-8a02-cd2a7cb415bf","added_by":"auto","created_at":"2025-11-06 15:52:27","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36182,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/51e863e7fd41dcb7b3065506.png"},{"id":95286062,"identity":"1d57ee65-1cc4-4dbe-89d9-cdd23181f6a8","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69204,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/a2db45b6373e79604cde7cca.png"},{"id":95314532,"identity":"f352b857-ecbc-4b39-b593-d9f6a6c585cd","added_by":"auto","created_at":"2025-11-06 15:52:58","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":125281,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/b273c8a7ad396c4609925e5c.png"},{"id":95286078,"identity":"74ef8ec7-a62c-4262-8b2d-839179d3c8fb","added_by":"auto","created_at":"2025-11-06 10:00:00","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":128530,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/e3e3ab26cc9fe8fe6795139f.png"},{"id":95286075,"identity":"bebf7dd3-96ce-4fda-915c-715305cf3046","added_by":"auto","created_at":"2025-11-06 10:00:00","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":107124,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/12f84b16e0c8fc8919120e27.png"},{"id":95286081,"identity":"4eb1a571-56cd-4c7c-97a9-4d86189c8aa5","added_by":"auto","created_at":"2025-11-06 10:00:00","extension":"xml","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":104936,"visible":true,"origin":"","legend":"","description":"","filename":"1454d6975e9d463c8691a3915411f93b1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/703150c085660713ddc89cf8.xml"},{"id":95314676,"identity":"0bd54671-663b-48bd-8e83-15efa0686e2d","added_by":"auto","created_at":"2025-11-06 15:53:10","extension":"html","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":116198,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/f0712811430bb6546ada4811.html"},{"id":95314112,"identity":"92e4a78e-f6c5-4517-a30c-39704511df3b","added_by":"auto","created_at":"2025-11-06 15:52:28","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":347027,"visible":true,"origin":"","legend":"\u003cp\u003edepicts the study flowchart\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/504902a3f3550e65bb52f50e.png"},{"id":95286041,"identity":"70c8eed0-4f6b-4758-ba59-64ed1be803a9","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":494622,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential Expression Analysis of Disease-Related Genes.\u003cstrong\u003e (A, B) \u003c/strong\u003eThe volcano plot and heatmap show DEGs between the AL group and control group.\u003cstrong\u003e (C, D)\u003c/strong\u003e The volcano plot and heatmap show DEGs between the MM group and control group. \u003cstrong\u003e(E, F）\u003c/strong\u003eThe Venn diagram illustrates the intersections between DEGs in MM and AL. \u003cstrong\u003e(G, H）\u003c/strong\u003eWGCNA analysis for GSE266124 and GSE211134 dataset. \u003cstrong\u003e(I, J)\u003c/strong\u003e Dendrogram showing the clustering of all DEGs. \u003cstrong\u003e(K, L)\u003c/strong\u003e Correlation between each module and diseases or controls. \u003cstrong\u003e(M, N)\u003c/strong\u003e Significant Modules. \u003cstrong\u003e(O) \u003c/strong\u003eVenn Diagram of Gene Overlaps Between AL WGCNA, MM WGCNA, and DEGs.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/cc760014a517d828316acfa1.png"},{"id":95314599,"identity":"204c5efd-c0e6-4f77-a521-f4567ecd6ba2","added_by":"auto","created_at":"2025-11-06 15:53:05","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":469639,"visible":true,"origin":"","legend":"\u003cp\u003eFunctional analysis for centralgenes.\u003cstrong\u003e (A)\u003c/strong\u003e Analysis of the centralgenes using GO. \u003cstrong\u003e(B)\u003c/strong\u003e An analysis of central genes using KEGG. \u003cstrong\u003e(C)\u003c/strong\u003e Gene-Biological Process Association Network. \u003cstrong\u003e(D)\u003c/strong\u003e The circle chart shows the genes enriched in different GO terms. \u003cstrong\u003e(E)\u003c/strong\u003e The circle chart shows the genes enriched in different KEGG terms. \u003cstrong\u003e(F)\u003c/strong\u003e PPI analysis of central genes. \u003cstrong\u003e(G)\u003c/strong\u003e Gene MANIA analysis outcomes.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/c0d1a21d402cfe4fa83981de.png"},{"id":95286045,"identity":"615b3136-b409-4903-ac55-8f4ec4198bd5","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":355638,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of Prognostic Genes and Construction of Prognostic Models. \u003cstrong\u003e(A)\u003c/strong\u003e Univariate cox proportional hazards modeling of hub genes. \u003cstrong\u003e(B)\u003c/strong\u003e LASSO regularization pathway analysis. \u003cstrong\u003e(C)\u003c/strong\u003e LASSO regularization Q-Q plot.\u003cstrong\u003e (D) \u003c/strong\u003eAll samples survival curves by risk group. \u003cstrong\u003e(E)\u003c/strong\u003e Training set survival curves by risk group.\u003cstrong\u003e (F)\u003c/strong\u003e Internal validation set survival curves by risk group. \u003cstrong\u003e(G)\u003c/strong\u003eExternal validation set survival curves by risk group. \u003cstrong\u003e(H)\u003c/strong\u003eAll samples roc curves at 1, 2, and 3 years. \u003cstrong\u003e(I) \u003c/strong\u003eTraining set roc curves at 1, 2, and 3 years.\u003cstrong\u003e (J)\u003c/strong\u003e Test set roc curves at 1, 2, and 3 years.\u003cstrong\u003e (K)\u003c/strong\u003e External validation set roc curves at 1, 2, and 3 years. \u003cstrong\u003e(L) \u003c/strong\u003eHeatmap of prognostic genes in training set. \u003cstrong\u003e(M)\u003c/strong\u003e Training set risk score distribution. \u003cstrong\u003e(N)\u003c/strong\u003e Training set survival analysis. \u003cstrong\u003e(O) \u003c/strong\u003eHeatmap of prognostic genes in validation set. \u003cstrong\u003e(P) \u003c/strong\u003eValidation set survival analysis.\u003cstrong\u003e (Q)\u003c/strong\u003eValidation set survival analysis.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/7057a1834991407bccb4156d.png"},{"id":95313675,"identity":"dff81071-3a00-4004-8bd3-0fb8332a7529","added_by":"auto","created_at":"2025-11-06 15:51:50","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":124750,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSurvival Analysis and Prognostic Model Evaluation (A)\u003c/strong\u003e Forest plot of univariate cox regression analysis. \u003cstrong\u003e(B)\u003c/strong\u003e Forest plot of multivariate cox regression analysis. \u003cstrong\u003e(C)\u003c/strong\u003e Nomogram for survival probability prediction. \u003cstrong\u003e(D)\u003c/strong\u003e Risk score distribution and survival status. \u003cstrong\u003e(E)\u003c/strong\u003e Decision Curve Analysis (DCA) for Survival Prediction Model. \u003cstrong\u003e(F)\u003c/strong\u003e Calibration curve for 1 year survival probability. \u003cstrong\u003e(G)\u003c/strong\u003e Calibration curve for 2-year survival probability. \u003cstrong\u003e(H)\u003c/strong\u003e Calibration curve for 3 -year survival probability.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/7f6c567750f95a16021fcdb5.png"},{"id":95286049,"identity":"36078c91-0710-42b9-a69b-617f9e473788","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":317113,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComprehensive Mutation Analysis Between High-Risk and Low-Risk Groups.\u003c/strong\u003e \u003cstrong\u003e(A)\u003c/strong\u003e High-Risk group mutation landscape. \u003cstrong\u003e(B)\u003c/strong\u003e Low-Risk group mutation landscape. \u003cstrong\u003e(C) \u003c/strong\u003eTop 20 mutated genes in high-risk group. \u003cstrong\u003e(D)\u003c/strong\u003e Top 20 mutated genes in low-risk group. \u003cstrong\u003e(E)\u003c/strong\u003e Mutation comparison between high-risk and low-risk groups. \u003cstrong\u003e(F)\u003c/strong\u003e Tumor mutation burden comparison between high-risk and low-risk groups.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/b6abc03a87a46d2e64fdbd96.png"},{"id":95286052,"identity":"21658f55-d9d7-4a85-b6d1-9c7ebd1b11b6","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":469779,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComprehensive Analysis of Immune Cell Infiltration and Drug Sensitivity. (A) \u003c/strong\u003eImmune cell composition analysis. \u003cstrong\u003e(B)\u003c/strong\u003eDistribution of immune cell types.\u003cstrong\u003e(C)\u003c/strong\u003e Heatmap of gene-immune cell correlation comparison. \u003cstrong\u003e(D)\u003c/strong\u003eCorrelation between risk score and immune cell fractions. \u003cstrong\u003e(E)\u003c/strong\u003e Comparative analysis of immune cell infiltration based on tumor type. \u003cstrong\u003e(F) \u003c/strong\u003eCorrelation of drug sensitivity with risk groups.\u003cstrong\u003e (G) \u003c/strong\u003eDrug sensitivity in different risk groups.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/19023ffde97ca633e98122df.png"},{"id":95286057,"identity":"ceaa3e37-7723-4853-8853-dd4c73ba6318","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":365900,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDiagnostic Performance of Core Genes Across Independent Cohorts.\u003c/strong\u003e \u003cstrong\u003e(A)\u003c/strong\u003e ROC curve analysis evaluating the diagnostic accuracy of core genes for AL in the \u003cem\u003eGSE175384\u003c/em\u003e cohort. \u003cstrong\u003e(B)\u003c/strong\u003e Validation of core gene diagnostic efficacy in discriminating AL amyloidosis within the \u003cem\u003eGSE73040\u003c/em\u003e dataset. \u003cstrong\u003e(C)\u003c/strong\u003e ROC profiling of core genes for MM classification in the \u003cem\u003eGSE175384\u003c/em\u003e cohort. \u003cstrong\u003e(D)\u003c/strong\u003e Cross-dataset validation of core gene diagnostic specificity for MM in \u003cem\u003eGSE16558\u003c/em\u003e.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/c4e69111db0584b6d7a5ec00.png"},{"id":95314272,"identity":"b6db23bb-6a18-476c-88ad-0783a3af0e7a","added_by":"auto","created_at":"2025-11-06 15:52:38","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":354589,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSingle-Gene Prognostic Analysis Across Independent Cohorts (A) Kaplan-Meier s\u003c/strong\u003eurvival curves for individual genes in the MMRF cohort. \u003cstrong\u003e(B)\u003c/strong\u003eKaplan-Meier survival curves for individual genes in the \u003cem\u003eGSE4581\u003c/em\u003e cohort.\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/27953d0336c29d10125aa8f8.png"},{"id":95286060,"identity":"4f92cb47-2426-41e6-8182-3a724a57614e","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":283769,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCross-Group Differential Expression Profiling of Core Genes\u003c/strong\u003e. \u003cstrong\u003e(A)\u003c/strong\u003e Comparative analysis of core gene expression patterns in the \u003cem\u003eGSE175384\u003c/em\u003e cohort. \u003cstrong\u003e(B)\u003c/strong\u003e Differential gene expression between the AL group and healthy controls. \u003cstrong\u003e(C)\u003c/strong\u003e Differential gene expression between the MM group and healthy controls. \u003cstrong\u003e(D)\u003c/strong\u003e Analysis of significantly differentially expressed genes across groups. Stars indicate the level of significance (\u003cem\u003ep\u003c/em\u003e\u0026lt; 0.05, \u003cem\u003ep \u003c/em\u003e\u0026lt; 0.01, *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001, **\u003cem\u003ep \u003c/em\u003e\u0026lt; 0.0001).\u003c/p\u003e","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/221006e415182970e3a65810.png"},{"id":95286054,"identity":"0c03aa1d-3263-402b-a327-7c5f0570f2ce","added_by":"auto","created_at":"2025-11-06 09:59:59","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":407597,"visible":true,"origin":"","legend":"\u003cp\u003e(A) UMAP visualization of the integrated single-cell atlas (GSE188222 + GSE271107) after Seurat quality control and Harmony batch correction. Cells from amyloid-light-chain (AL) amyloidosis, multiple-myeloma (MM) and healthy control (Ctrl) samples are co-embedded. (B) Cluster-based cell-type annotation performed with SingleR against HumanPrimaryCellAtlasData. Six major lineages are color-coded: CD4⁺ T cell, eosinophil, malignant plasma cell, microglial cell, natural-killer (NK) cell and red blood cell (erythrocyte). (C) Dot-plot showing mean scaled expression (dot color) and fraction of expressing cells (dot size) for twelve core lineage-defining genes across the six populations. (D) Split UMAPs displaying per-gene log-normalized expression levels for the same twelve genes; each small UMAP corresponds to one gene and is uniformly scaled from gray (low) to red (high).\u003c/p\u003e","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/05f384ac28bf770b1a66b5a6.png"},{"id":98440695,"identity":"d09cf32a-789a-417b-bb23-083122be2e30","added_by":"auto","created_at":"2025-12-17 17:04:12","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5361425,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/5c16fb38-7f25-40d7-9a10-618a9624e1cf.pdf"},{"id":95314286,"identity":"809d7f75-9b8a-4609-8afb-8763f1cd39fe","added_by":"auto","created_at":"2025-11-06 15:52:39","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1089527,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterials1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7555168/v1/ffcdfe890f3e28f221a29729.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Unraveling the Molecular Cross-talk in the Comorbidity of Multiple Myeloma and Systemic Light-Chain Amyloidosis through Multi-Dataset Analysis","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eMultiple myeloma (MM), characterized by the clonal proliferation of malignant plasma cells in the bone marrow, is the second most common hematologic malignancy, with its incidence rising due to population aging\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Although novel therapeutic agents and autologous stem cell transplantation (ASCT) have significantly improved initial response rates, approximately 80% of patients eventually relapse. The median overall survival remains suboptimal at 5\u0026ndash;7 years, highlighting the persistent challenges in achieving durable remission \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Existing prognostic models (e.g., R-ISS, R2-ISS) partially stratify risk but fail to fully address survival heterogeneity, particularly failing to identify subgroups driven by comorbidities \u003csup\u003e\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. These limitations underscore the urgent need to develop cross-disease molecular biomarkers.\u003c/p\u003e\u003cp\u003eImmunoglobulin light-chain amyloidosis (AL), a rare hematologic disorder caused by misfolded immunoglobulin light-chain deposition in vital organs\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e, was formally incorporated into China's First Rare Disease Catalog in 2018\u003csup\u003e7\u003c/sup\u003e. The disease typically manifests with nonspecific symptoms such as persistent fatigue and peripheral edema, frequently leading to diagnostic delays and progressive damage to critical organs including the heart and kidneys\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Prognosis is profoundly influenced by organ involvement patterns, with cardiac AL exhibiting particularly dismal outcomes\u0026mdash;median survival plummets to 3.5 months compared to 26.4 months in non-cardiac cases\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Although novel therapies like subcutaneous daratumumab have enhanced treatment adherence\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e, early detection through heightened clinical vigilance and multidisciplinary collaborative care remain cornerstone strategies for improving outcomes\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Current diagnosis relies on invasive organ biopsies or mass spectrometry for amyloid detection, which hinders early screening. Thus, noninvasive biomarkers and exploration of shared molecular mechanisms are critical for improving outcomes.\u003c/p\u003e\u003cp\u003eApproximately 10\u0026ndash;15% of MM patients develop concurrent light-chain amyloidosis, where organ involvement accelerates progression and reduces therapeutic response, exhibiting worse outcomes than isolated MM or AL\u003csup\u003e13\u003c/sup\u003e. MM-AL displays intermediate cytogenetic profiles, with t(11;14) prevalence between MM and AL, and fewer high-risk cytogenetic abnormalities (HRCA)\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. These patients achieve lower hematologic responses and median overall survival, with early mortality\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Prolonged induction and MRD-guided maintenance improve event-free survival (EFS), yet targeted therapeutic strategies remain underdeveloped\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eWhile previous studies have delineated distinct molecular profiles of MM and AL as isolated entities, the shared genetic architecture and prognostic regulatory networks underlying their comorbidity remain uncharted territory. This study pioneers the systematic dissection of core gene modules in MM-AL comorbidity through integrative analysis of large-scale transcriptomic datasets and systems biology approaches. By employing univariate and multivariate regression analyses, we establish a cross-disease prognostic model and elucidate the potential mechanisms by which key genes drive immune microenvironment remodeling and therapeutic resistance. Functional network mapping and drug sensitivity profiling further identify dual-functional therapeutic targets. Our work not only provides a novel molecular classification framework for MM-AL comorbidity but also deciphers the pathophysiological roles of shared genes, laying a theoretical foundation for developing precision therapeutic strategies across disease boundaries.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Literature Review and Database Search\u003c/h2\u003e\u003cp\u003eWe conducted a comprehensive review of the existing literature to identify the pathophysiological characteristics of MM and AL potential therapeutic targets. Concurrently, we searched the Gene Expression Omnibus (GEO) database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/geo/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/geo/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e)\u003csup\u003e17\u003c/sup\u003e using the keywords \" Multiple myeloma \" and \"Systemic Light Chain Amyloidosis\" to identify relevant datasets related to AL and MM.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Data Collection\u003c/h2\u003e\u003cp\u003eMM training sets (\u003cem\u003eGSE6477, GSE16558\u003c/em\u003e) and AL training set (\u003cem\u003eGSE175384\u003c/em\u003e) were obtained from GEO. The baseline model incorporated 859 MM samples with survival data from TCGA-MMRF(\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cancer.gov/tcga\u003c/span\u003e\u003cspan address=\"https://www.cancer.gov/tcga\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e)\u003csup\u003e18\u003c/sup\u003e. External validation used \u003cem\u003eGSE4581\u003c/em\u003e (414 MM cases) and \u003cem\u003eGSE73040\u003c/em\u003e (AL diagnostic assessment). Immune regulatory genes (638 including 257 Th17/381 Treg-related from GeneCards) \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e(Supplementary Table\u0026nbsp;1).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Data Processing\u003c/h2\u003e\u003cp\u003eRaw datasets were processed using R language and Bioconductor packages, which included steps for background correction, log transformation, and normalization. For bulk RNA-seq data, we applied the transcripts per million (TPM) method to further normalize the gene expression data. To minimize technical variation between samples, we used the \u0026ldquo;ComBat\u0026rdquo; algorithm in the \u0026ldquo;sva\u0026rdquo; package to correct batch effects\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Identification of Differentially Expressed Genes (DEGs): We identified DEGs in AL and MM patients compared to healthy controls using the \u0026ldquo;limma\u0026rdquo; package. DEGs were selected based on statistical significance and biological relevance, with a significance threshold of an adjusted \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and |log2FC| \u0026gt;1\u003csup\u003e21\u003c/sup\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Functional Enrichment Analysis and Protein-Protein Interaction (PPI) Network Construction\u003c/h2\u003e\u003cp\u003eThe \u0026ldquo;clusterProfiler\u0026rdquo; package was employed to perform Gene Ontology (GO) and KEGG pathway enrichment analyses on differentially expressed genes (DEGs) between AL/MM patients and healthy controls (threshold: |log2FC|\u0026gt;1 \u0026amp; FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Top 10 significantly enriched pathways (FDR-adjusted p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were visualized via \u0026ldquo;ggplot2\u0026rdquo; \u003csup\u003e22\u003c/sup\u003e. ssGSEA algorithm was used to calculate ssGSEA and GSVA score of MM patients by \u0026ldquo;GSVA\u0026rdquo; package in MM dataset\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. The DEGs were submitted to the STRING database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://cn.string-db.org/\u003c/span\u003e\u003cspan address=\"https://cn.string-db.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) to compile the interactions of target proteins, with a medium confidence threshold set at 0.4. A comprehensive PPI network was constructed using Cytoscape software, and any proteins lacking connections were excluded\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. GeneMANIA(\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://genemania.org/\u003c/span\u003e\u003cspan address=\"https://genemania.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Weighted Gene Go-expression Network Analysis (WGCNA)\u003c/h2\u003e\u003cp\u003eWGCNA is performed to identify modules of highly correlated genes, summarize the interconnections between modules and associations with external sample traits, and identify candidate biomarkers or therapeutic targets. In our research, WGCNA was constructed by the R package \u0026ldquo;WGCNA\u0026rdquo; to identify the modules with the highest relevance to diseases\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. The R function \u0026ldquo;pickSoftThreshold\u0026rdquo; was utilized to determine the optimal soft threshold power (β) for the construction of a scale-free network. The weighted adjacency matrix was subsequently transformed into a topological overlap matrix (TOM) to assess the connectivity within the network. The dendrogram of the TOM matrix was constructed employing the average-linkage hierarchical clustering method. For this investigation, the minimum gene module size was set to 500 to ensure the identification of appropriate modules, and the threshold for merging similar modules was set at 0.2. The gene significance (GS) and module membership (MM, AL) were calculated to correlate modules with clinical traits. Among the identified modules, those requiring further analysis were selected based on their correlation coefficients (r) and \u003cem\u003eP\u003c/em\u003e-values.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.6 Screening of prognosis genes\u003c/h2\u003e\u003cp\u003eWe performed univariate Cox proportional hazards regression analysis using the \u0026ldquo;survival\u0026rdquo; R package to identify genes significantly associated with overall survival in the MMRF cohort. Genes meeting the predefined significance threshold (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were retained for subsequent modeling. Proportional hazards assumptions were rigorously validated through Schoenfeld residual testing (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05). To mitigate overfitting and refine prognostic signatures, LASSO (least absolute shrinkage and selection operator) regression was implemented via the \u0026ldquo;glmnet\u0026rdquo; package, with penalty parameter (λ) optimization guided by 10 fold cross validation (minimum cross-validated error criterion). Survival disparities between risk-stratified groups were quantified using Kaplan-Meier estimators, with log-rank tests applied to evaluate statistical significance across both the discovery (MMRF) and external validation cohorts (\u003cem\u003eGSE4581\u003c/em\u003e). All analyses adhered to a two-sided α-level of 0.05 for hypothesis testing\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Following the identification of prognostic genes, we employed the Consensus Cluster Plus package to perform consensus clustering on MM samples based on the expression patterns of these genes. The MM samples were categorized into distinct subtypes, and a Kaplan-Meier survival analysis was conducted to compare the survival outcomes among these subtypes.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.7 Risk Score Calculation and Stratification\u003c/h2\u003e\u003cp\u003eTo calculate the risk score for each patient, we defined a linear predictor function that integrates the expression levels of selected genes weighted by their LASSO derived regression coefficients. Specifically, the risk score was computed as the sum of the product of each gene's expression value and its corresponding coefficient, implemented via the apply function in R to process the dataset. Subsequently, the LASSO regression model was applied to predict risk scores for all samples. Patients were stratified into high- and low-risk groups using the median risk score of the training cohort as the cutoff threshold: samples with scores above the median were classified as \"high-risk\", whereas those below were designated \u0026ldquo;low-risk\u0026rdquo;. Finally, Kaplan-Meier survival analysis with log-rank testing was performed to evaluate survival disparities between the risk-stratified subgroups within the training dataset.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e2.8 Development and Validation of Nomograms\u003c/h2\u003e\u003cp\u003eIn this study, we used a nomogram model to predict patients' survival probabilities and evaluated the model's performance via time-dependent ROC curve analysis. Initially, we applied a Cox proportional hazards regression model to patients' survival data to generate linear predictor scores (risk scores). Next, we extracted patients' survival times and event statuses, and performed time-dependent ROC analysis with the timeROC function to calculate AUC values at different time points, assessing the model's discriminatory ability. To visually present the model's performance, we plotted ROC curves in red, orange, and blue for 1, 2 and 3year predictions.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e2.9 Methods for Mutation and tumor mutation burden (TMB) Analysis\u003c/h2\u003e\u003cp\u003eSomatic mutation data were extracted from the high-risk and low-risk groups. Waterfall plots were generated to visualize the top 20 most frequently mutated genes in each group. Additionally, we analyzed the TMB by calculating the total number of mutations per megabase of genome sequenced. Statistical comparisons of TMB between the high-risk and low-risk groups were performed using the Wilcoxon rank-sum test. All analyses were conducted using the \u0026ldquo;maftools\u0026rdquo; package in R, and visualizations were created with ggplot2. This comprehensive analysis provided insights into the mutational landscape and TMB differences between risk groups.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e2.10 Immune Cell Infiltration and Correlation Analysis\u003c/h2\u003e\u003cp\u003eThe abundance of different immune cell types in MM was quantified using the CIBERSORT algorithm. To assess the relationship between 22 types of immune cells and key genes, we performed Pearson correlation analysis\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Pearson correlation analysis was conducted to evaluate the relationships between immune cell fractions and key genes. Correlation heatmaps were generated to visualize these relationships, with significance assessed using the cor.mtest function.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e2.11 Drug Sensitivity Analysis\u003c/h2\u003e\u003cp\u003eDrug sensitivity analysis was performed by predicting IC50 values for 251 anticancer compounds from the CGP2016 database using the \u0026ldquo;pRRophetic\u0026rdquo; package. Wilcoxon rank-sum tests (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) identified drugs with differential sensitivity between high- and low-risk groups, while Spearman correlation analysis (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) assessed associations between risk scores and IC50 values. Significant results were visualized through boxplots and scatter plots\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e2.12 Single-Gene Analysis and Functional Enrichment\u003c/h2\u003e\u003cp\u003eROC analysis was performed using the \u0026ldquo;pROC\u0026rdquo; package to evaluate single-gene diagnostic performance. The area under the curve (AUC) with 95% confidence intervals (95% CI) was calculated via bootstrap resampling\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Patients were stratified into high/low expression groups based on optimal cutoffs. Kaplan-Meier curves with log-rank tests (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) assessed survival disparities, validated across independent cohorts. The expression trends of prognosis genes in the MM, AL and control groups in the \u003cem\u003eGSE175384, GSE73040, GSE16558\u003c/em\u003e and \u003cem\u003eGSE6477\u003c/em\u003e datasets were analyzed by wilcoxon test. Next, the expression of prognosis genes in different subtypes was compared by kruskal test.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e2.13 Single Gene Differential Expression and Functional Enrichment Analysis\u003c/h2\u003e\u003cp\u003eWe conducted an in-depth exploration of the expression differences and potential biological functions of core genes under specific biological conditions using single-gene analysis methods. Various bioinformatics tools in R, including ggplot2, limma, pheatmap, ggsci, clusterProfiler, enrichplot, and patchwork, were employed for differential expression analysis and KEGG pathway enrichment analysis of core genes. By comparing sample groups with high and low expression, we identified significant DEGs and analyzed their enrichment in relevant KEGG pathways.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e2.14 Single-cell sequencing analysis\u003c/h2\u003e\u003cp\u003eIn this study, we obtained single-cell RNA sequencing datasets related to AL (\u003cem\u003eGSE188222\u003c/em\u003e), and MM(\u003cem\u003eGSE271107\u003c/em\u003e) from the GEO database. Utilizing the R programming language and the Seurat package from Bioconductor, we conducted quality control, normalization, identification of highly variable genes, dimensionality reduction through Principal Component Analysis (PCA), and clustering analysis. Visualization was performed using the Uniform Manifold Approximation and Projection (UMAP) algorithm. Additionally, we employed the \u0026ldquo;SingleR\u0026rdquo; package for automatic cell type annotation based on the Cancer Cell Line Encyclopedia (CCLE) database,Subtype-defining marker genes were then identified with FindAllMarkers using a log-fold-change threshold of 1 and adjusted P\u0026thinsp;\u0026lt;\u0026thinsp;0.05, retaining genes with |avg_log2FC| \u0026gt;1 as final cluster signatures\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003e\u003cb\u003eStatistical analysis\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAll statistical analyses were performed using R software. Unpaired two-tailed t-tests were used to calculate differences between two groups of data. Pearson or Spearman's rank correlation coefficient were applied depending on the distribution and nature of the data.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003e3.1Differential Gene Selection\u003c/h2\u003e\u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e depicts the study flowchart. DEGs were identified using the limma package. In the AL dataset, we identified 2529 upregulated genes and 902 downregulated genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, B). In the MM dataset, there were 785 upregulated genes and 970 downregulated genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC, D). The intersection of these two datasets revealed 210 common DEGs, comprising 72 upregulated and 138 downregulated (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE, F).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Co-expression Network Module\u003c/h2\u003e\u003cp\u003eTo systematically investigate the roles of Th17- and Treg-related genes in MM and AL, we curated gene sets associated with Th17 and Treg cells from the GeneCards database. GSVA and ssGSEA were employed to calculate pathway activity scores, followed by WGCNA to identify disease-associated gene modules. For the AL phenotype WGCNA analysis, we implemented standardized preprocessing procedures, including outlier removal and highly variable gene selection. A hierarchical clustering tree was constructed using TOM, and nine functional modules were obtained after parameter optimization (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG, I, K). Identification of key modules demonstrated the turquoise module (r\u0026thinsp;=\u0026thinsp;0.92, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1e-200) exhibited significant positive correlations with the AL phenotype (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eM). Consistent associations between genes and traits were further supported by GS analysis. Ultimately, 6,339 module-specific genes were identified(Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eO). In the WGCNA analysis for MM phenotype, by optimizing the soft-thresholding power parameter using the dynamic tree-cutting algorithm, seven functional modules were ultimately clustered (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH, J, L). Key module analysis revealed that the turquoise module (r\u0026thinsp;=\u0026thinsp;0.7, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1e\u0026thinsp;\u0026minus;\u0026thinsp;200) showed significant positive correlations with the MM phenotype. Scatter plots of gene significance (GS) versus trait correlations further validated these finding (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eN). A total of 1,878 candidate functional genes were extracted from the core modules. The intersection of AL WGCNA, MM WGCNA, and DEGs genes encompasses a set of 41 central genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eO).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Functional Enrichment (GO, KEGG)\u003c/h2\u003e\u003cp\u003eAfter performing GO and KEGG pathway enrichment analyses on the 41 shared genes between the two disease states, we found that these genes were significantly enriched in neuronal synaptic signaling, immune cell development, stress responses, and membrane receptor/channel functions \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, \u003cb\u003eSupplementary Table\u0026nbsp;2)\u003c/b\u003e. The KEGG enriched pathways suggest critical roles in cancer progression (metastasis, drug resistance), infection immunity, and microenvironment regulation (matrix remodeling, cell adhesion) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB, \u003cb\u003eSupplementary Table\u0026nbsp;3\u003c/b\u003e). The PPI analysis of central genes revealed a network consisting of 52 interacting nodes and 133 edges, involving a total of 37 genes. The network exhibited an average node degree of 5.12 and an average local clustering coefficient of 0.412(Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Identification of Prognostic Genes and Construction of Prognostic Models\u003c/h2\u003e\u003cp\u003eFollowing initial identification of hub genes, we conducted univariate Cox proportional hazards modeling to screen for survival-associated biomarkers. Among the evaluated candidates, 23 genes demonstrated significant prognostic relevance (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, \u003cb\u003eSupplementary Table\u0026nbsp;4\u003c/b\u003e). After verifying proportional hazards assumptions (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05), 12 genes(\u003cem\u003eHOMER3、OPN3、CLEC2D、PLA2G2D、PRR7、RCBTB2、EIF4EBP2、AZIN1、NFIL3、THOP1、PTP4A3、\u003c/em\u003eand \u003cem\u003eLSAMP\u003c/em\u003e)were retained for subsequent modeling(\u003cb\u003eSupplementary Table\u0026nbsp;5)\u003c/b\u003e. LASSO regularization analysis confirmed these 12 genes as robust prognostic candidates (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB, C). Kaplan-Meier survival analysis revealed distinct prognostic implications for gene expression levels: elevated expression of \u003cem\u003eHOMER3, OPN3, CLEC2D, PRR7, EIF4EBP2, AZIN1, NFIL3, THOP1\u003c/em\u003e, and \u003cem\u003ePTP4A3\u003c/em\u003e was associated with significantly reduced overall survival, whereas high expression of RCBTB2, PLA2G2D, and LSAMP correlated with prolonged survival outcomes (\u003cb\u003eSupplementary Fig S2)\u003c/b\u003e. The survival curves (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD-G) and ROC curves (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH-K) not only demonstrate significant survival differences and predictive accuracy in the internal validation set but also yield consistent results in the external validation set.\u003c/p\u003e\u003cp\u003eIn the present study, we constructed heatmaps of prognostic genes for patients with MM to visualize the expression patterns associated with patient outcomes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eL, O). The distribution of risk scores in the training set demonstrated a clear dichotomy between high- and low-risk groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eM, N). Survival analysis of the training set revealed a significant correlation between higher risk scores and decreased survival probability, as illustrated by the separation of the survival curves for high- and low-risk patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eP, Q).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003e3.5 Identification of Prognostic Genes and Construction of Prognostic Models\u003c/h2\u003e\u003cp\u003eTo assess the prognostic utility of the identified gene signature in combination with clinical parameters, we performed a comprehensive analysis. This began with univariate Cox regression analysis and evaluation of the proportional hazards (PH) assumption to identify potential independent prognostic factors. Further multivariate Cox regression analysis confirmed that the risk score, along with age, gender, and International Staging System (ISS) stage, serves as an independent prognostic indicator for MM patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, B). All these factors were determined to be independent risk factors for MM, and a corresponding nomogram was constructed based on these elements (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC), as evidenced by the calibration curve (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD), which demonstrated close agreement between predicted and observed survival probabilities (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF-H\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e\u003cp\u003eIn the MMRF dataset, the DCA and calibration curve revealed a strong alignment between predicted and actual survival probabilities (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE). Moreover, the model demonstrated impressive discriminative ability with AUC values of 0.763 at 1 year, 0.798 at 2 years, and 0.780 at 3 years (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec23\" class=\"Section2\"\u003e\u003ch2\u003e3.6 Analysis of mutations\u003c/h2\u003e\u003cp\u003eTo investigate the primary genetic mutations among different risk groups, we analyzed the somatic mutations in tumors from multiple myeloma patients in the MMRF dataset \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE, F\u003cb\u003e)\u003c/b\u003e. The plots revealed that the top three genes with the highest mutation frequencies in the low-risk group were IGHV2-70, IGLV3-1, and KRAS (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB, D). In the high-risk group, the top three genes with the highest mutation frequencies were IGHV2-70, IGLV3-1, and IGHV2-70D (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA, C).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\u003ch2\u003e3.7 \u003cb\u003eComprehensive Analysis of Immune Cell Infiltration and Drug Sensitivity\u003c/b\u003e\u003c/h2\u003e\u003cp\u003eAnalysis of 28 immune cell subsets revealed significant immune microenvironment remodeling between high- and low-risk groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA, B). Ten cell types\u0026mdash;including resting dendritic cells, M1 macrophages, and resting mast cells\u0026mdash;showed marked abundance differences (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD). Six immune subsets (e.g., naive B cells, neutrophils, activated dendritic cells) exhibited distinct infiltration patterns between treatment-na\u0026iuml;ve and relapsed cohorts (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eE), indicating dynamic immune remodeling during disease progression. Correlation analysis identified significant associations between prognostic genes (\u003cem\u003eLSAMP, PTP4A3, THOP1\u003c/em\u003e) and immune subsets (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC, \u003cb\u003eSupplementary Figure S3\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eAccording to statistical significance (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05), high-risk group expression showed a negative correlation with drug sensitivity to Bortezomib, Lenalidomide, Doxorubicin, BMS-509744, and a positive correlation with sensitivity to TGX221, Cetuximab, Trametinib, and Navitoclax (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eF, G\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec25\" class=\"Section2\"\u003e\u003ch2\u003e3.8 \u003cb\u003eDiagnostic Performance of Core Genes\u003c/b\u003e\u003c/h2\u003e\u003cp\u003eTo evaluate how well core genes can distinguish AL from MM, we analyzed their diagnostic performance using ROC curves across multiple datasets. Genes with AUC values above 0.7 in independent datasets were considered strong diagnostic markers. In the discovery cohort (\u003cem\u003eGSE175384\u003c/em\u003e), all genes achieved AUCs exceeding 0.7(\u003cb\u003eFig.\u0026nbsp;8A)\u003c/b\u003e. Notably, five of these genes \u0026mdash; \u003cem\u003eAZIN1, PLA2G2D, EIF4EBP2, NFIL3\u003c/em\u003e, and \u003cem\u003ePTP4A3\u003c/em\u003e \u0026mdash; also demonstrated high diagnostic accuracy in the external validation cohort (\u003cem\u003eGSE73040\u003c/em\u003e) (\u003cb\u003eFig.\u0026nbsp;8B)\u003c/b\u003e. For the MM specific classification, all genes showed consistent diagnostic superiority (\u003cb\u003eFig.\u0026nbsp;8C)\u003c/b\u003e. In the \u003cem\u003eGSE16558\u003c/em\u003e dataset, except for \u003cem\u003eTHOP1\u003c/em\u003e, the other genes reliably replicated their efficacy, further solidifying their potential as diagnostic tools (\u003cb\u003eFig.\u0026nbsp;8D)\u003c/b\u003e.\u003c/p\u003e\u003cp\u003eFigure\u0026nbsp;8 \u003cb\u003eDiagnostic Performance of Core Genes Across Independent Cohorts. (A)\u003c/b\u003e ROC curve analysis evaluating the diagnostic accuracy of core genes for AL in the \u003cem\u003eGSE175384\u003c/em\u003e cohort. \u003cb\u003e(B)\u003c/b\u003e Validation of core gene diagnostic efficacy in discriminating AL amyloidosis within the \u003cem\u003eGSE73040\u003c/em\u003e dataset. \u003cb\u003e(C)\u003c/b\u003e ROC profiling of core genes for MM classification in the \u003cem\u003eGSE175384\u003c/em\u003e cohort. \u003cb\u003e(D)\u003c/b\u003e Cross-dataset validation of core gene diagnostic specificity for MM in \u003cem\u003eGSE16558\u003c/em\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec26\" class=\"Section2\"\u003e\u003ch2\u003e3.9 \u003cb\u003eCross-Cohort Validation of Core Gene Prognostic Efficacy\u003c/b\u003e\u003c/h2\u003e\u003cp\u003eWe evaluated the prognostic performance of individual genes across independent cohorts using Kaplan-Meier survival analysis. In the \u003cem\u003eMMRF\u003c/em\u003e cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e9\u003c/span\u003eA), Kaplan-Meier survival curves of multiple core genes demonstrated significant stratification (log-rank \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05), indicating strong associations with patient outcomes. In the independent \u003cem\u003eGSE4581\u003c/em\u003e cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e9\u003c/span\u003eB), survival curves of \u003cem\u003eOPN3, PRR7, RCBTB2, THOP1, PTP4A3\u003c/em\u003e, and \u003cem\u003eLSAMP\u003c/em\u003e showed consistent stratification patterns, supporting their robust prognostic generalizability across datasets.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec27\" class=\"Section2\"\u003e\u003ch2\u003e3.10 \u003cb\u003eCross-Group Differential Expression Profiling of Core Genes\u003c/b\u003e\u003c/h2\u003e\u003cp\u003eTo elucidate the conserved regulatory patterns of core genes across disease subtypes, we performed cross-group differential expression profiling to evaluate expression concordance between discovery and validation cohorts. Wilcoxon rank-sum tests revealed high concordance in expression trends of 12 prognostic genes between the AL training cohort and validation cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e10\u003c/span\u003eB). Notably, \u003cem\u003eAZIN1, PTP4A3\u003c/em\u003e, and \u003cem\u003ePLA2G2D\u003c/em\u003e showed significant differential expression across both cohorts (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Consistent expression patterns were also observed in MM subgroups, indicating robust cross-cohort regulatory consistency (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e10\u003c/span\u003eC).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec28\" class=\"Section2\"\u003e\u003ch2\u003e3.11 Overview of key gene expression at single-cell resolution\u003c/h2\u003e\u003cp\u003eSingle-cell datasets \u003cem\u003eGSE188222\u003c/em\u003e and \u003cem\u003eGSE271107\u003c/em\u003e were downloaded and jointly analysed with Seurat. After quality control and Harmony batch correction, cells from AL, MM and control samples were merged (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e11\u003c/span\u003eA) and annotated against HumanPrimaryCellAtlasData using the SingleR package, yielding six major populations: CD4⁺ T cell, eosinophil, malignant plasma cell, microglial cell, natural killer cell and red blood cell (erythrocyte) (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e11\u003c/span\u003eB). Expression levels of twelve core genes across these six lineages are depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e11\u003c/span\u003eC and D.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eMM and AL are clinically closely related, with approximately 10\u0026ndash;15% of MM patients developing secondary AL\u003csup\u003e30\u003c/sup\u003e. Once complicated by AL, particularly with involvement of vital organs such as the heart, patients exhibit sharply reduced treatment response rates, significantly increased early mortality, and extremely poor overall prognosis\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. However, conventional prognostic models (e.g., R-ISS) are primarily based on tumor burden and cytogenetic characteristics of MM itself and fail to adequately capture the underlying molecular drivers responsible for this aggressive comorbid phenotype\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. This pressing unmet clinical need is the starting point of our study. By integrating multi-omics data, we systematically identified for the first time the core molecular features shared by MM and AL, aiming to elucidate the molecular basis of their comorbidity and provide new tools to address this clinical challenge\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn this study, we identified and validated a robust 12-gene prognostic signature that effectively stratifies high-risk MM patients with significantly shortened overall survival. We propose that these high-risk patients represent a subgroup with an elevated propensity for progressing to secondary AL, or already harbor an aggressive molecular phenotype conducive to amyloid deposition. The signature\u0026rsquo;s strong association with adverse outcomes in MM, combined with marked overexpression in AL, highlights its potential as a predictive tool for identifying MM patients at risk of AL transformation. Future validation in longitudinal MM cohorts will be essential to confirm its utility in the early prediction of secondary AL.\u003c/p\u003e\u003cp\u003eFunctional enrichment analysis revealed that these 12 core genes are significantly involved in key pathways, including endoplasmic reticulum stress response, protein folding, and metabolic processes. Among them, PTP4A3/PRL-3 resides at the center of a complex regulatory network involving cytokines, transcription factors, kinases, and epigenetic regulators. It promotes cell migration, enhances survival, reprograms metabolism, and initiates positive-feedback loops, thereby comprehensively driving the malignant progression of multiple myeloma and representing a highly promising therapeutic target. Multiple studies have confirmed the pivotal role of PTP4A3/PRL-3 in the pathogenesis of MM\u003csup\u003e34\u003c/sup\u003e. Its overexpression not only directly increases myeloma cell migration and invasion but also profoundly modulates tumor cell survival, metabolism, and gene expression through multiple signaling pathways\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. In MM, PRL-3 expression is sustained by an IL-6/STAT3 feedforward loop\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e, while it concurrently enhances glycolytic activity via STAT1/2 signaling\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Furthermore, PRL-3 activates SRC-family kinases (SFK) to promote migration and survival\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Epigenetically, the chromatin remodeler SMARCA2 collaborates with NSD2 to upregulate PRL-3, forming a self-reinforcing circuit particularly prominent in high-risk t(4;14) myeloma, solidifying PRL-3\u0026rsquo;s role as a central driver of disease progression\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Simultaneously, PTP4A3 may impair proteasomal function, further facilitating the aggregation of misfolded light chains. It may also remodel the tumor microenvironment by regulating cytokine signaling\u0026mdash;such as increasing immunosuppressive cytokine secretion\u0026mdash;thereby fostering a supportive niche for malignant plasma cells. Together, these synergistic mechanisms likely underlie the aggressive comorbidity of MM and AL, offering a molecular framework that explains the poor clinical outcomes in these patients.\u003c/p\u003e\u003cp\u003eOur analysis further revealed profound immune microenvironment remodeling in high-risk patients, characterized by a dysfunctional, immunosuppressive state\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Drug sensitivity analysis provided clinically actionable insights: high-risk patients showed reduced sensitivity to first-line therapies such as bortezomib and lenalidomide, aligning with their observed treatment resistance and poor prognosis\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. In contrast, increased sensitivity was observed toward targeted agents such as Trametinib (a MEK inhibitor)\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. These findings offer a rationale for implementing risk-adapted therapy\u0026mdash;early use of targeted combination strategies\u0026mdash;to improve outcomes in this high-risk population.\u003c/p\u003e"},{"header":"5. Limitations Section","content":"\u003cp\u003eSeveral limitations must be acknowledged. This study has several limitations that should be acknowledged. The prognostic model was primarily developed and validated in multiple myeloma cohorts due to the inherent challenge of obtaining large-scale AL amyloidosis datasets\u0026mdash;a rare disease\u0026mdash;with complete survival annotations. The analysis was based on existing datasets including : TCGA-MMRF (n\u0026thinsp;=\u0026thinsp;859) and GEO datasets, and lacked independent experimental validation. While we confirmed the diagnostic accuracy of the signature in available AL cohorts, prospective validation in larger cohorts and functional studies to establish causality for identified genes like PTP4A3 are needed as future work.\u003c/p\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003eIn conclusion, starting from the clinical challenge of MM-AL comorbidity, we have discovered a unifying molecular signature that not effectively identifies high-risk MM patients but also elucidates key biological processes related to proteotoxic stress and microenvironmental dysregulation. This signature provides both mechanistic insight and a practical tool for risk stratification, paving the way for future targeted therapeutic strategies aimed at improving outcomes for this refractory patient population.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAckowledgements\u003c/p\u003e\n\u003cp\u003eThe authors declare that no funding, technical assistance, or other external support was received for this study.\u003c/p\u003e\n\u003cp\u003eAuthor contributions\u003c/p\u003e\n\u003cp\u003eLiu, X. and Zeng, P. wrote the main manuscript text and prepared all figures. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eNo funding.\u003c/p\u003e\n\u003cp\u003eDisclosure statement\u003c/p\u003e\n\u003cp\u003eNo potential conflict of interest was reported by the author(s).\u003c/p\u003e\n\u003cp\u003eData availability\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe datasets analysed during the current study are available in the Gene Expression Omnibus (GEO) repository [GSE188222, GSE271107, GSE6477, GSE16558, GSE175384, GSE4581, GSE73040] and The Cancer Genome Atlas (TCGA) repository [https://www.cancer.gov/tcga], but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of GEO and TCGA.\u003c/p\u003e\n\u003cp\u003eCode availability\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003eEthics, Consent to Participate, and Consent to Publish declarations\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003eData availability statements\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMalard, F. \u003cem\u003eet al.\u003c/em\u003e Multiple myeloma. \u003cem\u003eNat. Rev. Dis. Primer\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 45 (2024).\u003c/li\u003e\n\u003cli\u003eDimopoulos, M. A. \u003cem\u003eet al.\u003c/em\u003e Multiple myeloma: EHA-ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up\u0026dagger;. \u003cem\u003eAnn. Oncol. Off. J. Eur. Soc. Med. Oncol.\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 309\u0026ndash;322 (2021).\u003c/li\u003e\n\u003cli\u003eHagen, P., Zhang, J. \u0026amp; Barton, K. High-risk disease in newly diagnosed multiple myeloma: beyond the R-ISS and IMWG definitions. \u003cem\u003eBlood Cancer J.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 83 (2022).\u003c/li\u003e\n\u003cli\u003eD\u0026rsquo;Agostino, M. \u003cem\u003eet al.\u003c/em\u003e Second Revision of the International Staging System (R2-ISS) for Overall Survival in Multiple Myeloma: A European Myeloma Network (EMN) Report Within the HARMONY Project. \u003cem\u003eJ. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol.\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 3406\u0026ndash;3418 (2022).\u003c/li\u003e\n\u003cli\u003eKuiper, R. \u003cem\u003eet al.\u003c/em\u003e Prognostic and predictive performance of R-ISS with SKY92 in older patients with multiple myeloma: the HOVON-87/NMSG-18 trial. \u003cem\u003eBlood Adv.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 6298\u0026ndash;6309 (2020).\u003c/li\u003e\n\u003cli\u003eSanchorawala, V. Systemic Light Chain Amyloidosis. \u003cem\u003eN. Engl. J. Med.\u003c/em\u003e \u003cstrong\u003e390\u003c/strong\u003e, 2295\u0026ndash;2307 (2024).\u003c/li\u003e\n\u003cli\u003eGuo, J. \u003cem\u003eet al.\u003c/em\u003e National Rare Diseases Registry System (NRDRS): China\u0026rsquo;s first nation-wide rare diseases demographic analyses. \u003cem\u003eOrphanet J. Rare Dis.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 515 (2021).\u003c/li\u003e\n\u003cli\u003eGertz, M. A. Immunoglobulin light chain amyloidosis: 2024 update on diagnosis, prognosis, and treatment. \u003cem\u003eAm. J. Hematol.\u003c/em\u003e \u003cstrong\u003e99\u003c/strong\u003e, 309\u0026ndash;324 (2024).\u003c/li\u003e\n\u003cli\u003eYu, C. \u003cem\u003eet al.\u003c/em\u003e Light-chain amyloidosis with concomitant symptomatic myeloma (CRAB-SLiM features): clinical characteristics, cytogenetic abnormalities, and outcomes. \u003cem\u003eBMC Cancer\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 1449 (2024).\u003c/li\u003e\n\u003cli\u003eAlnasser, S. M., Alharbi, K. S., Almutairy, A. F., Almutairi, S. M. \u0026amp; Alolayan, A. M. Autologous Stem Cell Transplant in Hodgkin\u0026rsquo;s and Non-Hodgkin\u0026rsquo;s Lymphoma, Multiple Myeloma, and AL Amyloidosis. \u003cem\u003eCells\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 2855 (2023).\u003c/li\u003e\n\u003cli\u003ePalladini, G. \u003cem\u003eet al.\u003c/em\u003e Daratumumab plus CyBorD for patients with newly diagnosed AL amyloidosis: safety run-in results of ANDROMEDA. \u003cem\u003eBlood\u003c/em\u003e \u003cstrong\u003e136\u003c/strong\u003e, 71\u0026ndash;80 (2020).\u003c/li\u003e\n\u003cli\u003eZhao, L. \u003cem\u003eet al.\u003c/em\u003e The clinical features and outcomes of systemic light chain amyloidosis with hepatic involvement. \u003cem\u003eAnn. Med.\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, 1226\u0026ndash;1232 (2022).\u003c/li\u003e\n\u003cli\u003eHuang, J. \u003cem\u003eet al.\u003c/em\u003e Unveiling the ageing-related genes in diagnosing osteoarthritis with metabolic syndrome by integrated bioinformatics analysis and machine learning. \u003cem\u003eArtif. Cells Nanomedicine Biotechnol.\u003c/em\u003e \u003cstrong\u003e53\u003c/strong\u003e, 57\u0026ndash;68 (2025).\u003c/li\u003e\n\u003cli\u003eOzga, M. \u003cem\u003eet al.\u003c/em\u003e AL amyloidosis: The effect of fluorescent in situ hybridization abnormalities on organ involvement and survival. \u003cem\u003eCancer Med.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 965\u0026ndash;973 (2021).\u003c/li\u003e\n\u003cli\u003eSuzuki, K. Diagnosis and treatment of multiple myeloma and AL amyloidosis with focus on improvement of renal lesion. \u003cem\u003eClin. Exp. Nephrol.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 659\u0026ndash;671 (2012).\u003c/li\u003e\n\u003cli\u003eMadan, S. \u003cem\u003eet al.\u003c/em\u003e Clinical features and treatment response of light chain (AL) amyloidosis diagnosed in patients with previous diagnosis of multiple myeloma. \u003cem\u003eMayo Clin. Proc.\u003c/em\u003e \u003cstrong\u003e85\u003c/strong\u003e, 232\u0026ndash;238 (2010).\u003c/li\u003e\n\u003cli\u003eBarrett, T. \u003cem\u003eet al.\u003c/em\u003e NCBI GEO: archive for functional genomics data sets--update. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, D991-995 (2013).\u003c/li\u003e\n\u003cli\u003eTomczak, K., Czerwińska, P. \u0026amp; Wiznerowicz, M. The Cancer Genome Atlas (TCGA): an immeasurable source of knowledge. \u003cem\u003eContemp. Oncol. Poznan Pol.\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, A68-77 (2015).\u003c/li\u003e\n\u003cli\u003eStelzer, G. \u003cem\u003eet al.\u003c/em\u003e The GeneCards Suite: From Gene Data Mining to Disease Genome Sequence Analyses. \u003cem\u003eCurr. Protoc. Bioinforma.\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, 1.30.1-1.30.33 (2016).\u003c/li\u003e\n\u003cli\u003eHuber, W. \u003cem\u003eet al.\u003c/em\u003e Orchestrating high-throughput genomic analysis with Bioconductor. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 115\u0026ndash;121 (2015).\u003c/li\u003e\n\u003cli\u003eSmyth, G. K. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. \u003cem\u003eStat. Appl. Genet. Mol. Biol.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, Article3 (2004).\u003c/li\u003e\n\u003cli\u003eKanehisa, M., Sato, Y., Kawashima, M., Furumichi, M. \u0026amp; Tanabe, M. KEGG as a reference resource for gene and protein annotation. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e, D457-462 (2016).\u003c/li\u003e\n\u003cli\u003eLong, Q., Zhang, X., Ren, F., Wu, X. \u0026amp; Wang, Z.-M. Identification of novel biomarkers, shared molecular signatures and immune cell infiltration in heart and kidney failure by transcriptomics. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1456083 (2024).\u003c/li\u003e\n\u003cli\u003eSzklarczyk, D. \u003cem\u003eet al.\u003c/em\u003e The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, D638\u0026ndash;D646 (2023).\u003c/li\u003e\n\u003cli\u003eD\u0026iacute;az-Uriarte, R. \u0026amp; Alvarez de Andr\u0026eacute;s, S. Gene selection and classification of microarray data using random forest. \u003cem\u003eBMC Bioinformatics\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 3 (2006).\u003c/li\u003e\n\u003cli\u003eChen, B., Khodadoust, M. S., Liu, C. L., Newman, A. M. \u0026amp; Alizadeh, A. A. Profiling Tumor Infiltrating Immune Cells with CIBERSORT. \u003cem\u003eMethods Mol. Biol. Clifton NJ\u003c/em\u003e \u003cstrong\u003e1711\u003c/strong\u003e, 243\u0026ndash;259 (2018).\u003c/li\u003e\n\u003cli\u003eGeeleher, P., Cox, N. \u0026amp; Huang, R. S. pRRophetic: an R package for prediction of clinical chemotherapeutic response from tumor gene expression levels. \u003cem\u003ePloS One\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, e107468 (2014).\u003c/li\u003e\n\u003cli\u003eRobin, X. \u003cem\u003eet al.\u003c/em\u003e pROC: an open-source package for R and S+ to analyze and compare ROC curves. \u003cem\u003eBMC Bioinformatics\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 77 (2011).\u003c/li\u003e\n\u003cli\u003eStuart, T. \u003cem\u003eet al.\u003c/em\u003e Comprehensive Integration of Single-Cell Data. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e177\u003c/strong\u003e, 1888-1902.e21 (2019).\u003c/li\u003e\n\u003cli\u003eYu, C. \u003cem\u003eet al.\u003c/em\u003e A Comparative Study of Clinical Characteristics, Cytogenetic Abnormalities and Outcomes Among Multiple Myeloma, Primary Light-Chain Amyloidosis and Multiple Myeloma with Concurrent AL Amyloidosis. \u003cem\u003eBlood\u003c/em\u003e \u003cstrong\u003e142\u003c/strong\u003e, 2032\u0026ndash;2032 (2023).\u003c/li\u003e\n\u003cli\u003eR\u0026iacute;os-Tamayo, R. \u003cem\u003eet al.\u003c/em\u003e AL Amyloidosis and Multiple Myeloma: A Complex Scenario in Which Cardiac Involvement Remains the Key Prognostic Factor. \u003cem\u003eLife\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 1518 (2023).\u003c/li\u003e\n\u003cli\u003eZvida‐Bloch, T., Muchtar, E., Dispenzieri, A., Shpilberg, O. \u0026amp; Hershkovitz‐Rokah, O. The molecular landscape of AL amyloidosis. \u003cem\u003eBr. J. Haematol.\u003c/em\u003e \u003cstrong\u003e206\u003c/strong\u003e, 1297\u0026ndash;1311 (2025).\u003c/li\u003e\n\u003cli\u003eSrivastava, A. \u003cem\u003eet al.\u003c/em\u003e Cellular diversity and genetic alterations in light chain amyloidosis and amyloidosis with co-occurring multiple myeloma. Preprint at https://doi.org/10.1101/2025.06.06.25329039 (2025).\u003c/li\u003e\n\u003cli\u003eSmith, C. N. \u0026amp; Blackburn, J. S. PRL-3 promotes a positive feedback loop between STAT1/2-induced gene expression and glycolysis in multiple myeloma. \u003cem\u003eFEBS J.\u003c/em\u003e \u003cstrong\u003e288\u003c/strong\u003e, 6674\u0026ndash;6676 (2021).\u003c/li\u003e\n\u003cli\u003eFagerli, U.-M. \u003cem\u003eet al.\u003c/em\u003e Overexpression and involvement in migration by the metastasis-associated phosphatase PRL-3 in human myeloma cells. \u003cem\u003eBlood\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, 806\u0026ndash;815 (2008).\u003c/li\u003e\n\u003cli\u003eChong, P. S. Y. \u003cem\u003eet al.\u003c/em\u003e IL6 Promotes a STAT3-PRL3 Feedforward Loop via SHP2 Repression in Multiple Myeloma. \u003cem\u003eCancer Res.\u003c/em\u003e \u003cstrong\u003e79\u003c/strong\u003e, 4679\u0026ndash;4688 (2019).\u003c/li\u003e\n\u003cli\u003eAbdollahi, P. \u003cem\u003eet al.\u003c/em\u003e Src Family Kinases Are Regulated in Multiple Myeloma Cells by Phosphatase of Regenerating Liver-3. \u003cem\u003eMol. Cancer Res. MCR\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 69\u0026ndash;77 (2017).\u003c/li\u003e\n\u003cli\u003eChong, P. S. Y. \u003cem\u003eet al.\u003c/em\u003e SMARCA2 Is a Novel Interactor of NSD2 and Regulates Prometastatic PTP4A3 through Chromatin Remodeling in t(4;14) Multiple Myeloma. \u003cem\u003eCancer Res.\u003c/em\u003e \u003cstrong\u003e81\u003c/strong\u003e, 2332\u0026ndash;2344 (2021).\u003c/li\u003e\n\u003cli\u003eLu, J. \u003cem\u003eet al.\u003c/em\u003e Myeloid-derived suppressor cells in cancer: therapeutic targets to overcome tumor immune evasion. \u003cem\u003eExp. Hematol. Oncol.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 39 (2024).\u003c/li\u003e\n\u003cli\u003eKastritis, E. \u003cem\u003eet al.\u003c/em\u003e Long‐term outcomes of primary systemic light chain (AL) amyloidosis in patients treated upfront with bortezomib or lenalidomide and the importance of risk adapted strategies. \u003cem\u003eAm. J. Hematol.\u003c/em\u003e \u003cstrong\u003e90\u003c/strong\u003e, (2015).\u003c/li\u003e\n\u003cli\u003eHoffner, Msn, Anp-Bc, Aocnp, B. \u0026amp; Benchich, Msn, Np-C, Aocnp, K. Trametinib: A Targeted Therapy in Metastatic Melanoma. \u003cem\u003eJ. Adv. Pract. Oncol.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, (2018).\u003cstrong\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Multiple myeloma, Light-chain amyloidosis, Prognostic signature, Tumor microenvironment, scRNA-seq, PTP4A3","lastPublishedDoi":"10.21203/rs.3.rs-7555168/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7555168/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e The co-occurrence of multiple myeloma (MM) and light-chain amyloidosis (AL) accelerates disease progression, but their shared mechanisms remain unclear.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e We integrated bulk transcriptomic (GSE6477, GSE16558, GSE175384) and single-cell RNA-seq (GSE188222, GSE271107) data. Using WGCNA, PPI networks, and machine learning, we developed a prognostic signature validated in MMRF (N=859) and external cohorts. Immune infiltration and drug sensitivity were analyzed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eWe identified 41 shared genes and established a 12-gene prognostic signature. High-risk patients showed distinct immune microenvironments and drug responses. Single-cell analysis revealed cell-type-specific expression patterns, with PTP4A3 emerging as a key regulator.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions:\u003c/strong\u003e This multi-omics study reveals shared molecular mechanisms in MM-AL comorbidity and provides a robust prognostic signature. PTP4A3 represents a potential therapeutic target, offering insights for precision medicine.\u003c/p\u003e","manuscriptTitle":"Unraveling the Molecular Cross-talk in the Comorbidity of Multiple Myeloma and Systemic Light-Chain Amyloidosis through Multi-Dataset Analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-06 09:59:54","doi":"10.21203/rs.3.rs-7555168/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4a9f75d7-a027-4de3-b7c0-9794b7fafac1","owner":[],"postedDate":"November 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-12-17T07:55:05+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-06 09:59:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7555168","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7555168","identity":"rs-7555168","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}