Machine Learning Models for Predicting Multiple Myeloma Staging and MGUS Progression Using Gene Expression Data DOI Creative Commons
Nestoras Karathanasis, George M. Spyrou

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Ноя. 15, 2024

Abstract In this study, we developed and evaluated Machine Learning (ML) models aimed at predicting the stage of multiple myeloma (MM) progression monoclonal gammopathy undetermined significance (MGUS) to MM. Accurate staging MM is critical for determining appropriate treatment strategies, our models, employing algorithms such as ElasticNet, Random Forest, Boosting, Support Vector Machines, demonstrated high efficacy in capturing biological differences across disease stages. Among these, ElasticNet model exhibited strong generalizability, achieving consistent multiclass AUC values various datasets data transformations. Predicting MGUS presents a significant challenge due scarcity cases that have progressed. We employed two-pronged approach address this: developing using limited dataset containing progressing patients training on combined datasets. The achieved slightly above 0.8, particularly with Boosting indicating their potential stratifying by risk. This study original integrating enhance predictive accuracy progression, offering novel methodology clinical applications patient monitoring early intervention. Our feature selection enrichment analyses further revealed identified genes are involved key signaling pathways, including PI3K-Akt, MAPK, Wnt, mTOR, all which play crucial roles pathogenesis. These findings align established knowledge, suggest possible therapeutic targets increase explainability models.

Язык: Английский

Insights into IL-6/JAK/STAT3 Signaling in the Tumor Microenvironment: Implications for Cancer Therapy DOI

Win Lwin Thuya,

Yang Cao,

Paul Chi-Lui Ho

и другие.

Cytokine & Growth Factor Reviews, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

3
Pathways regulating intestinal stem cells and potential therapeutic targets for radiation enteropathy DOI Creative Commons

Si-Min Chen,

Bing-Jie Guo,

A Lei Feng

и другие.

Molecular Biomedicine, Год журнала: 2024, Номер 5(1)

Опубликована: Окт. 10, 2024

Radiotherapy is a pivotal intervention for cancer patients, significantly impacting their treatment outcomes and survival prospects. Nevertheless, in the course of treating those with abdominal, pelvic, or retroperitoneal malignant tumors, procedure inadvertently exposes adjacent intestinal tissues to radiation, posing risks radiation-induced enteropathy upon reaching threshold doses. Stem cells within crypts, through controlled proliferation differentiation, support critical functions epithelium, ensuring efficient nutrient absorption while upholding its protective barrier properties. Intestinal stem (ISCs) regulation intricately orchestrated by diverse signaling pathways, among which are WNT, BMP, NOTCH, EGF, Hippo, Hedgehog NF-κB, each contributing complex control these cells' behavior. Complementing pathways additional regulators such as metabolic states, microbiota, all contribute fine-tuning ISCs behavior crypts. It harmonious interplay cascades modulating elements that preserves homeostasis epithelial (IECs), thereby gut's overall health function. This review delves into molecular underpinnings how respond context radiation enteropathy, aiming illuminate potential biological targets therapeutic intervention. Furthermore, we have compiled summary several current methodologies. By unraveling mechanisms methods, aspire furnish roadmap development novel therapeutics, advancing our capabilities mitigating damage.

Язык: Английский

Процитировано

4
Predicting the Progression from Asymptomatic to Symptomatic Multiple Myeloma and Stage Classification Using Gene Expression Data DOI Open Access
Nestoras Karathanasis, George M. Spyrou

Cancers, Год журнала: 2025, Номер 17(2), С. 332 - 332

Опубликована: Янв. 20, 2025

Background: The accurate staging of multiple myeloma (MM) is essential for optimizing treatment strategies, while predicting the progression asymptomatic patients, also referred to as monoclonal gammopathy undetermined significance (MGUS), symptomatic MM remains a significant challenge due limited data. This study aimed develop machine learning models enhance accuracy and stratify patients by their risk progression. Methods: We utilized gene expression microarray datasets models, combined with various data transformations. For staging, were trained on single dataset validated across five independent datasets, performance evaluated using multiclass area under curve (AUC) metrics. To predict in we employed two approaches: (1) training comprising who either progressed or remained stable without progressing myeloma, (2) combining samples then testing ability distinguish between that progressed. performed feature selection enrichment analyses identify key signaling pathways underlying disease stages Results: Multiple demonstrated high efficacy, ElasticNet achieving consistent AUC values 0.9 transformations, demonstrating robust generalizability. progression, both modeling approaches yielded similar results, exceeding 0.8 algorithms (ElasticNet, Boosting, Support Vector Machines), underscoring potential identifying risk. Enrichment revealed pathways, including PI3K-Akt, MAPK, Wnt, mTOR, central pathogenesis. Conclusions: best our knowledge, this first utilize classifying different integrate cases offering novel methodology clinical applications patient monitoring early intervention.

Язык: Английский

Процитировано

0
Detection of early relapse in multiple myeloma patients DOI Creative Commons

Tereza Růžičková,

Monika Vlachová,

Lukáš Pečínka

и другие.

Cell Division, Год журнала: 2025, Номер 20(1)

Опубликована: Янв. 29, 2025

Multiple myeloma (MM) represents the second most common hematological malignancy characterized by infiltration of bone marrow plasma cells that produce monoclonal immunoglobulin. While quality and length life MM patients have significantly increased, remains a hard-to-treat disease; almost all relapse. As is highly heterogenous, relapse at different times. It currently not possible to predict when will occur; numerous studies investigating dysregulation non-coding RNA molecules in cancer suggest microRNAs could be good markers Using small sequencing, we profiled microRNA expression peripheral blood three groups who relapsed intervals. In total, 24 were dysregulated among analyzed subgroups. Independent validation RT-qPCR confirmed changed levels miR-598-3p with times At same time, differences mass spectra between identified using matrix-assisted laser desorption/ionization time flight spectrometry. All results machine learning. Mass spectrometry coupled learning shows potential as reliable, rapid, cost-effective preliminary screening technique supplement current diagnostics.

Язык: Английский

Процитировано

0
Targeting ABCD1-ACOX1-MET/IGF1R axis suppresses multiple myeloma DOI
Zhannan Han, Zhibo Yan,

Zhehan Ma

и другие.

Leukemia, Год журнала: 2025, Номер unknown

Опубликована: Янв. 30, 2025

Язык: Английский

Процитировано

0
Insights into the Molecular Mechanisms and Nanoparticle-Based Therapies for Gastric Cancer: A review DOI Creative Commons

Smarika Chauhan,

Eswar Marcharla,

Sneha Hariharan

и другие.

Results in Engineering, Год журнала: 2025, Номер unknown, С. 104238 - 104238

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

0
MiRNA-21 promotes the migration and proliferation of prostate cancer cells via activating the JAK/STAT pathway DOI Creative Commons
Xin Han, Changjun Lv

Discover Oncology, Год журнала: 2025, Номер 16(1)

Опубликована: Фев. 12, 2025

This research explored the role of microRNA (miRNA)-21 in prostate cancer (PCa) cells, as well its regulation JAK/STAT pathway PCa cells. Quantitative real-time PCR was employed to examine miRNA-21 expression Cell viability and proliferation were detected by MTT colony formation assays. migration measured wound healing transwell The janus kinase/signal transducers activators transcription (JAK/STAT) pathway-related protein using western blot. results indicated that significantly up-regulated inhibition suppressed viability, Besides, lessened levels proteins both Additionally, Ruxolitinib treatment (an inhibitor pathway) could reverse elevated cell mimics-transfected Taken together, our study demonstrates promotes cells via activating JAK /STAT pathway.

Язык: Английский

Процитировано

0
PHLPP and LAMP2 predict favorable treatment response and survival in multiple myeloma patients who receive induction treatment with bortezomib DOI
Qingkun Han, Wei Han, Cuiping Li

и другие.

Irish Journal of Medical Science (1971 -), Год журнала: 2025, Номер unknown

Опубликована: Фев. 14, 2025

Язык: Английский

Процитировано

0
Dual CARM1- and IKZF3-targeting: a novel approach to multiple myeloma therapy Synergy between CARM1 inhibition and IMiDs DOI Creative Commons
Wei Ni, Swati Garg,

Basudev Chowdhury

и другие.

Deleted Journal, Год журнала: 2025, Номер 33(1), С. 200952 - 200952

Опубликована: Фев. 20, 2025

Advancements in the treatment of multiple myeloma (MM) have resulted an improvement survival rate. However, there continues to be urgent need for improved therapies. The protein arginine methyltransferase, CARM1 (coactivator associated methyltransferase 1), is emerging as a potential cancer therapy target and inhibitors been developed. MM cell lines are particularly dependent on survival. Here, we show that targeting through small molecule inhibition potentiates activity immunomodulatory drugs (IMiDs) line models MM. This likely occurs synergistic Aiolos (IKZF3) MYC expression. Rational design new molecule, 074, which consists inhibitor linked IMiD pomalidomide, was carried out with this agent led more potent killing cells than either or single agents. Importantly, 074 able override resistance. Taken together, our results demonstrate dual CARM1/IKZF3-targeting agents represent promising novel therapeutic strategy IMiD-resistant disease.

Язык: Английский

Процитировано

0
Downregulated expression of dual-specificity phosphatase-1 in multiple myeloma as a predictor of poor survival outcomes DOI Creative Commons

Dan Guo,

Jinfeng Lu, Lemin Hong

и другие.

Hematology, Год журнала: 2025, Номер 30(1)

Опубликована: Март 12, 2025

Multiple myeloma (MM) is an incurable hematological malignancy, Dual-specificity phosphatase-1 (DUSP1) plays a crucial role in the initiation and progression of various tumors. Here, we aim to elucidate DUSP1 MM. mRNA expression was analyzed based on public datasets, protein determined by immunohistochemistry. The association between clinicopathological characteristics, as well its impact survival, were investigated. Protein-protein interaction gene set enrichment analysis performed. Low detected MM it associated with elevated β2-microglobulin, C-reactive protein, creatinine, lactate dehydrogenase, plasma cell ratio, decreased hemoglobin levels. DUSP1high group exhibited superior outcomes across clinical endpoints. Univariate multivariate analyses indicated that low independent prognostic factor for poor OS (hazard ratio = 0.273). findings suggested related proto-oncogene c-Fos (FOS), heat shock family member 1a (HSPA1A), several members MAPK family, nuclear receptor subfamily 3, C, 1 (NR3C1), zinc finger 36 (ZFP36). levels positively correlated ribosomes negatively oocyte meiosis, one carbon pool folate, homologous recombination, base excision repair, pyrimidine metabolism pathways. potential mechanisms identified through PPI network could provide insight into how may influence be considered patients.

Язык: Английский

Процитировано

0