Machine learning enables detection of early-stage colorectal cancer by whole-genome sequencing of plasma cell-free DNA

Wan, Nathan; Weinberg, David S.; Liu, Tzu-Yu; Niehaus, Katherine E.; Ariazi, Eric A.; Delubac, Daniel; Kannan, Ajay; White, Brandon; Bailey, Mitch; Bertin, Marvin; Boley, Nathan; Bowen, Derek; Cregg, James M.; Drake, Adam; Ennis, Riley; Fransen, Signe; Gafni, Erik; Hansen, Larry G.; Liu, Yaping; Otte, Gabriel; Pecson, Jennifer; Rice, Brandon J.; Sanderson, Gabriel E.; Sharma, Aarushi; John, John St.; Tang, Catherina; Tzou, Abraham; Young, Leilani; Putcha, Girish; Haque, Imran S.

doi:10.1186/s12885-019-6003-8

Cited by 134 publications

(82 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous ML-based studies performed in ccfDNA materials have also shown promising results in CRC. Wan et al built an ML-based model for the early detection of CRC with an AUC of 0.92 (95% CI: 0.91–0.93) [ 76 ], while Luo et al constructed a predictive model that accurately discriminated CRC patients from healthy individuals (AUC: 0.96) [ 77 ].…”

Section: Discussionmentioning

confidence: 99%

Methylation Status of Corticotropin-Releasing Factor (CRF) Receptor Genes in Colorectal Cancer

Παναγοπούλου

Cheretaki

Karaglani

et al. 2021

JCM

View full text Add to dashboard Cite

The corticotropin-releasing factor (CRF) system has been strongly associated with gastrointestinal pathophysiology, including colorectal cancer (CRC). We previously showed that altered expression of CRF receptors (CRFRs) in the colon critically affects CRC progression and aggressiveness through regulation of colonic inflammation. Here, we aimed to assess the potential of CRFR methylation levels as putative biomarkers in CRC. In silico methylation analysis of CRF receptor 1 (CRFR1) and CRF receptor 2 (CRFR2) was performed using methylome data derived by CRC and Crohn’s disease (CD) tissues and CRC-derived circulating cell-free DNAs (ccfDNAs). In total, 32 and 33 differentially methylated sites of CpGs (DMCs) emerged in CRFR1 and CRFR2, respectively, between healthy and diseased tissues. The methylation patterns were verified in patient-derived ccfDNA samples by qMSP and associated with clinicopathological characteristics. An automated machine learning (AutoML) technology was applied to ccfDNA samples for classification analysis. In silico analysis revealed increased methylation of both CRFRs in CRC tissue and ccfDNA-derived datasets. CRFR1 hypermethylation was also noticed in gene body DMCs of CD patients. CRFR1 hypermethylation was further validated in CRC adjuvant-derived ccfDNA samples, whereas CRFR1 hypomethylation, observed in metastasis-derived ccfDNAs, was correlated to disease aggressiveness and adverse prognostic characteristics. AutoML analysis based on CRFRs methylation status revealed a three-feature high-performing biosignature for CRC diagnosis with an estimated AUC of 0.929. Monitoring of CRFRs methylation-based signature in CRC tissues and ccfDNAs may be of high diagnostic and prognostic significance in CRC.

show abstract

Section: Discussionmentioning

confidence: 99%

Methylation Status of Corticotropin-Releasing Factor (CRF) Receptor Genes in Colorectal Cancer

Παναγοπούλου

Cheretaki

Karaglani

et al. 2021

JCM

View full text Add to dashboard Cite

show abstract

“…Putcha et al performed a study on a ML-based approach to discover signatures in cell-free DNA to potentially improve the detection of colorectal cancer (NCT03688906; [ 66 , 67 ].…”

Section: Resultsmentioning

confidence: 99%

Rise of Clinical Studies in the Field of Machine Learning: A Review of Data Registered in ClinicalTrials.gov

Zippel

Bohnet-Joschko

2021

IJERPH

View full text Add to dashboard Cite

Although advances in machine-learning healthcare applications promise great potential for innovative medical care, few data are available on the translational status of these new technologies. We aimed to provide a comprehensive characterization of the development and status quo of clinical studies in the field of machine learning. For this purpose, we performed a registry-based analysis of machine-learning-related studies that were published and first available in the ClinicalTrials.gov database until 2020, using the database’s study classification. In total, n = 358 eligible studies could be included in the analysis. Of these, 82% were initiated by academic institutions/university (hospitals) and 18% by industry sponsors. A total of 96% were national and 4% international. About half of the studies (47%) had at least one recruiting location in a country in North America, followed by Europe (37%) and Asia (15%). Most of the studies reported were initiated in the medical field of imaging (12%), followed by cardiology, psychiatry, anesthesia/intensive care medicine (all 11%) and neurology (10%). Although the majority of the clinical studies were still initiated in an academic research context, the first industry-financed projects on machine-learning-based algorithms are becoming visible. The number of clinical studies with machine-learning-related applications and the variety of medical challenges addressed serve to indicate their increasing importance in future clinical care. Finally, they also set a time frame for the adjustment of medical device-related regulation and governance.

show abstract

“…Parallel analysis with RNAseq and MNase-seq of the matched primary tumor and blood samples may facilitate the discovery of correlations between cfDNA nucleosome positioning and relevant gene expression with the nucleosome occupancy of the genes. cfDNA contains DNA from both normal and tumor tissues in patients with breast cancer, and studies have found cfDNA derived from tissue-specific and tumor-specific open chromatin regions (NFR or NDR) 20 , 21 . Because the fractions of tumor- and non-tumor cfDNA vary among different patients 4 , a limitation of our study is that we failed to consider the two fractions of normal and tumor DNA.…”

Section: Discussionmentioning

confidence: 99%

Association between the nucleosome footprint of plasma DNA and neoadjuvant chemotherapy response for breast cancer

et al. 2021

View full text Add to dashboard Cite

Gene expression signatures have been used to predict the outcome of chemotherapy for breast cancer. The nucleosome footprint of cell-free DNA (cfDNA) carries gene expression information of the original tissues and thus may be used to predict the response to chemotherapy. Here we carried out the nucleosome positioning on cfDNA from 85 breast cancer patients and 85 healthy individuals and two cancer cell lines T-47D and MDA-MB-231 using low-coverage whole-genome sequencing (LCWGS) method. The patients showed distinct nucleosome footprints at Transcription Start Sites (TSSs) compared with normal donors. In order to identify the footprints of cfDNA corresponding with the responses to neoadjuvant chemotherapy in patients, we mapped on nucleosome positions on cfDNA of patients with different responses: responders (pretreatment, n = 28; post-1 cycle, post-3/4 cycles, and post-8 cycles of treatment, n = 12) and nonresponders (pretreatment, n = 10; post-1 cycle, post-3/4 cycles, and post-8 cycles of treatment, n = 10). The coverage depth near TSSs in plasma cfDNA differed significantly between responders and nonresponders at pretreatment, and also after neoadjuvant chemotherapy treatment cycles. We identified 232 TSSs with differential footprints at pretreatment and 321 after treatment and found enrichment in Gene Ontology terms such as cell growth inhibition, tumor suppressor, necrotic cell death, acute inflammatory response, T cell receptor signaling pathway, and positive regulation of vascular endothelial growth factor production. These results suggest that cfDNA nucleosome footprints may be used to predict the efficacy of neoadjuvant chemotherapy for breast cancer patients and thus may provide help in decision making for individual patients.

show abstract

Machine learning enables detection of early-stage colorectal cancer by whole-genome sequencing of plasma cell-free DNA

Cited by 134 publications

References 35 publications

Methylation Status of Corticotropin-Releasing Factor (CRF) Receptor Genes in Colorectal Cancer

Methylation Status of Corticotropin-Releasing Factor (CRF) Receptor Genes in Colorectal Cancer

Rise of Clinical Studies in the Field of Machine Learning: A Review of Data Registered in ClinicalTrials.gov

Association between the nucleosome footprint of plasma DNA and neoadjuvant chemotherapy response for breast cancer

Contact Info

Product

Resources

About