AI-derived five-gene signature predicts risk in multiple myeloma under bortezomib-based therapy

Sep 30, 2026

Multiple myeloma (MM) is an incurable plasma cell malignancy driven by clonal expansion in the bone marrow, with prognosis shaped not only by tumor-intrinsic factors but also by the bone marrow tumor microenvironment (TME). A robust prognostic model was built from transcriptomic profiling of the CD138-negative fraction, using bioinformatic deconvolution and a custom immune-stromal signature matrix. Feature selection and elastic-net penalized Cox regression produced the MM-5C five-gene model, enabling biologically distinct risk stratification and independent prognostic value beyond cytogenetics and clinical staging.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Abstract

Prognostic problem in multiple myeloma

TME and CD138-negative transcriptomics approach

Development of the MM-5C five-gene model

[www. nature.com/scientificreports](www. nature.com/scientificreports) OPEN AI-derived five-gene signature predicts risk in multiple myeloma under bortezomib-based therapy Grazia Gargano1,2􀀍, Susanna Anita Pappagallo2, Angela Maria Quinto2, Bernardo Rossini2, Doriana Gramegna2, Paolo Mondelli2, Maria CarmelaVegliante2, Giuseppina Opinto6, Flavia Esposito1, Gian Maria Zaccaria5, Vincenza Solli3, Orazio Palumbo4, Carolina Terragna3, Michele Cavo3, Nicoletta Del Buono1, Attilio Guarini2 & Sabino Ciavarella2 Multiple myeloma (MM) is a plasma cell malignancy characterized by clonal expansion within the bone marrow (BM) and remains largely incurable despite recent therapeutic advances. While prognostic indicators have predominantly focused on tumor-intrinsic features, the BM tumor microenvironment (TME) plays a critical role in disease progression and therapeutic resistance. Here, we present a robust prognostic model derived from transcriptomic profiling of the CD138-negative BM fraction, enabling precise risk stratification of patients undergoing bortezomib-based induction therapy. Using bioinformatic deconvolution and a custom immune-stromal signature matrix, we identified gene expression patterns representative of the MMTME. Through rigorous feature selection and elasticnet penalized Cox regression, we developed the MM-5C model, consisting of five genes (SOX11, METTL11B, C3, RBM10, and HOMEZ), which stratifies patients into biologically distinct risk groups and demonstrates prognostic independence from established cytogenetic and clinical staging systems. This model underscores the pivotal role ofTME components in shaping therapeutic outcomes and offers a scalable, clinically translatable tool for personalized risk stratification. Our findings highlight the necessity of integrating microenvironmental insights into MM prognostication and pave the way for microenvironment-informed therapeutic decision-making. Keywords Multiple myeloma, Tumor microenvironment, CD138-negative, Transcriptomic deconvolution, Prognostic model, Elastic-net Cox regression Abbreviations MM Multiple myeloma BM Bone marrow TME Tumor microenvironment ISS International Staging System R-ISS Revised International Staging System MPSS Myeloma Prognostic Score System LDH Lactate dehydrogenase R Responders NR Non-responders FISH Fluorescence in situ hybridization PFS Progression-free survival SCR Stringent complete response CR Complete response SD Stable disease PD Progressive disease NS Not specified 1Department of Mathematics, University of Bari Aldo Moro, Bari, Italy. 2Hematology and Cell Therapy Unit, IRCCSIstituto Tumori “Giovanni Paolo II”, Bari, Italy. 3Seràgnoli Institute of Hematology, IRCCS Azienda OspedalieroUniversitaria di Bologna, Bologna, Italy. 4UOC Genetica Medica, Fondazione IRCCS-Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy. 5Department of Electrical and Information Engineering (DEI), Polytechnic University of Bari, Bari, Italy. 6Clinical Pathology and Microbiology Unit, Bonomo Hospital, Andria, Italy. 􀀍 email: [grazia.gargano@uniba.it](grazia.gargano@uniba.it) [www. nature.com/scientificreports/](www. nature.com/scientificreports/) Multiple myeloma (MM) is a malignancy of plasma cells derived from terminally differentiated B cells. It is characterized by the clonal proliferation of abnormal plasma cells within the bone marrow (BM) . Over the last decades, the incidence of MM has been steadily increasing1,2, while therapeutic strategies evolved significantly, resulting in remarkable improvement of patient survival1,3. However, MM remains largely incurable, with considerable disease-related mortality4. Well-accepted prognostic factors predominantly rely on tumor-intrinsic features such as cytogenetic abnormalities, tumor burden, and clinical stage5–7, although significant efforts have been devoted to refining new prognostic models. Mao et al.8 , for instance, developed the Myeloma Prognostic Score

FAQs

What is the MM-5C model and what genes does it include?

MM-5C is a prognostic model developed from CD138-negative bone marrow transcriptomics. It consists of five genes: SOX11, METTL11B, C3, RBM10, and HOMEZ.

How does the study use the tumor microenvironment in risk prediction?

The model derives TME-representative gene expression patterns using bioinformatic deconvolution and a custom immune-stromal signature matrix, linking TME components to therapeutic outcomes and resistance.

Does the five-gene signature add prognostic value beyond standard staging?

Yes. The model demonstrates prognostic independence from established cytogenetic and clinical staging systems, supporting biologically distinct risk group stratification.

Sep 30, 2026

Uploaded by Lucas Martin