Dataset Information

Dynomics: A Novel and Promising Approach for Improved Breast Cancer Prognosis Prediction.

ABSTRACT: Traditional imaging techniques for breast cancer (BC) diagnosis and prediction, such as X-rays and magnetic resonance imaging (MRI), demonstrate varying sensitivity and specificity due to clinical and technological factors. Consequently, positron emission tomography (PET), capable of detecting abnormal metabolic activity, has emerged as a more effective tool, providing critical quantitative and qualitative tumor-related metabolic information. This study leverages a public clinical dataset of dynamic ¹⁸F-Fluorothymidine (FLT) PET scans from BC patients, extending conventional static radiomics methods to the time domain-termed as 'Dynomics'. Radiomic features were extracted from both static and dynamic PET images on lesion and reference tissue masks. The extracted features were used to train an XGBoost model for classifying tumor versus reference tissue and complete versus partial responders to neoadjuvant chemotherapy. The results underscored the superiority of dynamic and static radiomics over standard PET imaging, achieving accuracy of 94% in tumor tissue classification. Notably, in predicting BC prognosis, dynomics delivered the highest performance, achieving accuracy of 86%, thereby outperforming both static radiomics and standard PET data. This study illustrates the enhanced clinical utility of dynomics in yielding more precise and reliable information for BC diagnosis and prognosis, paving the way for improved treatment strategies.

SUBMITTER: Inglese M

PROVIDER: S-EPMC10303631 | biostudies-literature | 2023 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Dynomics: A Novel and Promising Approach for Improved Breast Cancer Prognosis Prediction.

Inglese Marianna M Ferrante Matteo M Boccato Tommaso T Conti Allegra A Pistolese Chiara A CA Buonomo Oreste C OC D'Angelillo Rolando M RM Toschi Nicola N

Journal of personalized medicine 20230615 6

Traditional imaging techniques for breast cancer (BC) diagnosis and prediction, such as X-rays and magnetic resonance imaging (MRI), demonstrate varying sensitivity and specificity due to clinical and technological factors. Consequently, positron emission tomography (PET), capable of detecting abnormal metabolic activity, has emerged as a more effective tool, providing critical quantitative and qualitative tumor-related metabolic information. This study leverages a public clinical dataset of dyn ...[more]

PMID: 37373993

Similar Datasets

Project description:The extensive and intricate relationships between circadian rhythm and cancer have been reported in numerous studies. However, in breast cancer (BC), the potential role of circadian clock-related genes (CCRGs) in prognosis prediction has not been fully clarified. The transcriptome data and clinical information were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus databases. A CCRGs-based risk signature was established by differential expression analysis, univariate, Lasso and multivariate Cox regression analyses. we conducted a gene set enrichment analysis (GSEA) between groups. A nomogram integrating independent clinical factors and risk score was generated and evaluated by calibration curves and decision curve analysis (DCA). Differentially expression analysis revealed 80 differentially expressed CCRGs, and 27 of them were significantly associated with the overall survival (OS) of BC. BC can be classified into 4 molecular subtypes with significant differences in prognosis based on the 27 CCRGs. Three prognostic CCRGs, including desmocollin 1 (DSC1), LEF1, and protocadherin 9 (PCDH9), were identified to be independent risk factors of BC prognosis and were used to construct a risk score model. BC patients were divided into high- and low-risk groups, and there were significant differences in prognosis between the 2 groups both in the training and validation cohorts. It was found that patients in different groups of race, status, or T stage had significant levels of risk score. Furthermore, patients of different risk levels exhibit varying degrees of sensitivity to vinorelbine, lapatinib, metformin, and vinblastine. GSEA showed that in the high-risk group, immune response-related activities were dramatically repressed whereas cilium-related processes were significantly stimulated. Cox regression analysis demonstrated that age, N stage, radiotherapy and the risk score were independent prognostic risk factors of BC, and a nomogram was established based on these variables. The nomogram exerted a favorable concordance index (0.798) as well as calibration performance, which strongly supports the clinical application of the nomogram. Our study indicated the disruption of the expression of CCRGs in BC and built a favorable prognostic risk model based on 3 independent prognostic CCRGs. These genes may be applied as candidate molecular targets for the diagnosis and therapy of BC.

Project description:BackgroundIn the ongoing battle against breast cancer, a leading cause of cancer-related mortality among women globally, the urgent need for innovative prognostic markers and therapeutic targets is undeniable. This study pioneers an advanced methodology by integrating machine learning techniques to unveil a vascular mimicry signature, offering predictive insights into breast cancer outcomes. Vascular mimicry refers to the phenomenon where cancer cells mimic blood vessel formation absent of endothelial cells, a trait associated with heightened tumor aggression and diminished response to conventional treatments.MethodsThe study's comprehensive analysis spanned data from over 6,000 breast cancer patients across 12 distinct datasets, incorporating both proprietary clinical data and single-cell data from 7 patients, accounting for a total of 43,095 cells. By employing an integrative strategy that utilized 10 machine learning algorithms across 108 unique combinations, the research scrutinized 100 existing breast cancer signatures. Empirical validation was sought through immunohistochemistry assays, alongside explorations into potential immunotherapeutic and chemotherapeutic avenues.ResultsThe investigation successfully identified six genes related to vascular mimicry from multi-center cohorts, laying the groundwork for a novel predictive model. This model outstripped the prognostic accuracy of traditional clinical and molecular indicators in forecasting recurrence and mortality risks. High-risk individuals identified by our model faced worse outcomes. Further validation through IHC assays in 30 patients underscored the model's extensive applicability. Notably, the model unveiled varying therapeutic responses; low-risk patients might achieve greater benefits from immunotherapy, whereas high-risk patients demonstrated a particular sensitivity to certain chemotherapies, such as ispinesib.ConclusionsThis model marks a significant step forward in the precise evaluation of breast cancer prognosis and therapeutic responses across different patient groups. It heralds the possibility of refining patient outcomes through tailored treatment strategies, accentuating the potential of machine learning in revolutionizing cancer prognosis and management.

Dataset Information

Dynomics: A Novel and Promising Approach for Improved Breast Cancer Prognosis Prediction.

Publications

Dynomics: A Novel and Promising Approach for Improved Breast Cancer Prognosis Prediction.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets