Bridging biological cfDNA features and machine learning approaches

Moser, Tina; Kühberger, Stefan; Lazzeri, Isaac; Vlachos, Georgios; Heitzer, Ellen

doi:10.1016/j.tig.2023.01.004

Cited by 40 publications

(26 citation statements)

References 184 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GCNN has the ability to explore the similarity and mutual representation among samples, therefore achieving great success in multi-class classification tasks (51,52). Unlike the reference-based deconvolution approaches (53,54), our GCNN approach is independent of a reference methylation atlas, which was developed from tissue or cell type specific methylation markers and thus may introduce bias due to discordance between the methylomes of tissue gDNA and plasma cfDNA (16,55). Although the methylation changes were reported as most predictive for TOO in previous studies (53,54), our results showed the contribution of each of the 9 features for TOO identification (Figure 8C).…”

Section: Discussionmentioning

confidence: 99%

“…In order to increase the sensitivity of early cancer detection while avoiding the high cost of deep sequencing, a screening test should survey a wide range of ctDNA signatures (16). Therefore, we utilized multiple ctDNA signatures to construct classification models for distinguishing cancer patients from healthy individuals.…”

Section: Spot-mas Assay Combining Different Features Of Cfdna To Enha...mentioning

confidence: 99%

“…Despite their great potential, there remain several challenges that these assays must solve to deliver accessible and reliable clinical adoption for the large population, including the low fraction of ctDNA in the blood of early stage cancer patients, the heterogeneity of ctDNA signatures from diverse cancer types, subtypes and stages (16), and the high sequencing depth required. To address these challenges, recent studies have focused on multi-analyte approachcombining genomic and nongenomic features such as methylomics and fragmentomics to increase the detection of ctDNA and accuracy for TOO identification (16)(17)(18). Advances in multimodal analysis approaches have led to the development of powerful screening tests that enable high sensitivity and cost-effectiveness.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multimodal analysis of methylomics and fragmentomics in plasma cell-free DNA for multi-cancer early detection and localization

Nguyen¹,

Nguyễn²,

Doan³

et al. 2023

Preprint

View full text Add to dashboard Cite

The non-invasive approach for early cancer detection promises a screening assay accessible for everyone. However, the delivery of this promise is limited due mostly to the high sequencing cost associated with available assays. Here, we developed a multimodal assay called SPOT-MAS (Screening for the Presence Of Tumor by Methylation And Size) to simultaneously profile methylomics, fragmentomics, copy number, and end motifs in a single workflow using targeted and shallow genome-wide sequencing of cell-free DNA. We applied SPOT-MAS to 738 nonmetastatic patients with breast, colorectal, gastric, lung and liver cancer, and 1,550 healthy controls. SPOT-MAS detected the five cancer types with a sensitivity of 72.4% and specificity of 97.0%, with AUC of 0.95 (95% CI 0.93-0.96). For tumor-of-origin, a graph convolutional neural network was adopted and could achieve an accuracy of 0.7. In conclusion, our study demonstrates comparable performance to other early cancer detection assays while requiring significantly lower sequencing depth, making it economically feasible for population-wide screening.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Spot-mas Assay Combining Different Features Of Cfdna To Enha...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multimodal analysis of methylomics and fragmentomics in plasma cell-free DNA for multi-cancer early detection and localization

Nguyen¹,

Nguyễn²,

Doan³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In addition to its expanding clinical applications coupled with improving accuracy, plasma ctDNA is being studied within the context of comprehensive multi-omics, such as WGS, methylation profiling, nucleosome mapping, and fragmentomics, as well as machine learning algorithms to uncover biological features from complex datasets, highlighting its potential to provide deeper insights into cancer biology [17,130].…”

Section: Future Directions and Perspectivesmentioning

confidence: 99%

Utilizing Plasma Circulating Tumor DNA Sequencing for Precision Medicine in the Management of Solid Cancers

Cha¹,

Kim²,

Han³

2023

Cancer Res Treat

View full text Add to dashboard Cite

Plasma circulating tumor DNA (ctDNA) sequencing has demonstrated clinical utility for tumor molecular profiling at initial diagnosis or tumor progression in advanced solid cancers and is being rapidly incorporated into the clinical practice guidelines, including non–small cell lung and breast cancer. Despite relatively low sensitivity, plasma ctDNA sequencing has several advantages over tissue-based assays, including ease of sampling, rapid turnaround time, repeatability, and the ability to overcome spatial heterogeneity, which makes it ideal for investigating acquired resistance and monitoring tumor evolution and dynamics. With technological advancement and declining costs, the clinical application of plasma ctDNA is expanding, and numerous ongoing clinical trials are examining its potential to guide the management of advanced, localized, and even preclinical cancers of various tumor types. The ability of plasma ctDNA analysis to detect minimal residual disease following curative treatment in the absence of clinical disease is among its most promising attributes. Plasma ctDNA sequencing can also facilitate the conduct of clinical trials and drug development, particularly in immunotherapy. In order to incorporate plasma ctDNA sequencing for clinical decision-making, it is important to understand the preanalytical and analytical factors that may affect its sensitivity and reliability.

show abstract

“…14,31 Due to their convinces in obtaining information of disease by gathering body fluids and standard PCR analysis, cfDNA has gained commercial availability for guiding parts of clinical decision-making processes like cancer immune-therapy monitoring and early diagnosis. 32,33 Several excellent aspects have been summarized to inspire the clinical application of cfDNA. However, one major challenge that needs to be addressed is the interference caused by non-specific cfDNA originating from normal tissues.…”

Section: Introductionmentioning

confidence: 99%

Recent advances of liquid biopsy: Interdisciplinary strategies toward clinical decision‐making

Bao,

Min,

et al. 2023

Interdisciplinary Medicine

View full text Add to dashboard Cite

Liquid biopsy has emerged as a promising avenue for non‐invasive and rapid retrieval of pathological information from patient body fluids. Over the years, liquid biopsy has garnered significant attention for clinically treating cancer by selecting appropriate biomarkers such as circulating tumor cells (CTCs) and extracellular vesicles (EVs). Further integration of advanced technologies has facilitated the efficient capture of biomarkers for liquid biopsy, revolutionizing clinical decision‐making in the multiple processes and stages of cancer patients. Underscoring the intersection of different disciplines, this review provides a holistic summary of recent breakthroughs specifically designed for the capture and application of distinctive biomarkers to blend real‐world clinical decision‐making with material science. Firstly, we focus on the main principles of liquid biopsy and recent technologies that facilitate the capture and release of biomarkers (e.g., CTCs and EVs), leveraging their physicochemical properties. Then, the clinical applications of biomarkers are summarized, highlighting their potential for providing comprehensive clinical information. Later, the incorporation of machine learning is also emphasized for enhancing clinical applications and enabling deeper insights in the design of next‐generation platforms for specific biomarker isolation. Finally, future opportunities for clinical decision‐making are explored by combining advanced nanotechnologies and artificial intelligence, thereby offering inconceivable possibilities for improving patient care.

show abstract

Bridging biological cfDNA features and machine learning approaches

Cited by 40 publications

References 184 publications

Multimodal analysis of methylomics and fragmentomics in plasma cell-free DNA for multi-cancer early detection and localization

Multimodal analysis of methylomics and fragmentomics in plasma cell-free DNA for multi-cancer early detection and localization

Utilizing Plasma Circulating Tumor DNA Sequencing for Precision Medicine in the Management of Solid Cancers

Recent advances of liquid biopsy: Interdisciplinary strategies toward clinical decision‐making

Contact Info

Product

Resources

About