Ultrasensitive plasma-based monitoring of tumor burden using machine-learning-guided signal enrichment.

In solid tumor oncology, circulating tumor DNA (ctDNA) is poised to transform care through accurate assessment of minimal residual disease (MRD) and therapeutic response monitoring. To overcome the sparsity of ctDNA fragments in low tumor fraction (TF) settings and increase MRD sensitivity, we previously leveraged genome-wide mutational integration through plasma whole-genome sequencing (WGS). Here we now introduce MRD-EDGE, a machine-learning-guided WGS ctDNA single-nucleotide variant (SNV) and copy-number variant (CNV) detection platform designed to increase signal enrichment. MRD-EDGESNV uses deep learning and a ctDNA-specific feature space to increase SNV signal-to-noise enrichment in WGS by ~300× compared to previous WGS error suppression. MRD-EDGECNV also reduces the degree of aneuploidy needed for ultrasensitive CNV detection through WGS from 1 Gb to 200 Mb, vastly expanding its applicability within solid tumors. We harness the improved performance to identify MRD following surgery in multiple cancer types, track changes in TF in response to neoadjuvant immunotherapy in lung cancer and demonstrate ctDNA shedding in precancerous colorectal adenomas. Finally, the radical signal-to-noise enrichment in MRD-EDGESNV enables plasma-only (non-tumor-informed) disease monitoring in advanced melanoma and lung cancer, yielding clinically informative TF monitoring for patients on immune-checkpoint inhibition.

Cilium Expression Score Predicts Glioma Survival.

The accurate classification, prognostication, and treatment of gliomas has been hindered by an existing cellular, genomic, and transcriptomic heterogeneity within individual tumors and their microenvironments. Traditional clustering is limited in its ability to distinguish heterogeneity in gliomas because the clusters are required to be exclusive and exhaustive. In contrast, biclustering can identify groups of co-regulated genes with respect to a subset of samples and vice versa. In this study, we analyzed 1,798 normal and tumor brain samples using an unsupervised biclustering approach. We identified co-regulated gene expression profiles that were linked to proximally located brain regions and detected upregulated genes in subsets of gliomas, associated with their histologic grade and clinical outcome. In particular, we present a cilium-associated signature that when upregulated in tumors is predictive of poor survival. We also introduce a risk score based on expression of 12 cilium-associated genes which is reproducibly informative of survival independent of other prognostic biomarkers. These results highlight the role of cilia in development and progression of gliomas and suggest potential therapeutic vulnerabilities for these highly aggressive tumors.