We previously described an approach called RealSeqS to evaluate aneuploidy in plasma cell-free DNA through the amplification of ~350,000 repeated elements with a single primer. We hypothesized that an unbiased evaluation of the large amount of sequencing data obtained with RealSeqS might reveal other differences between plasma samples from patients with and without cancer. This hypothesis was tested through the development of a machine learning approach called Alu Profile Learning Using Sequencing (A-PLUS) and its application to 7615 samples from 5178 individuals, 2073 with solid cancer and the remainder without cancer. Samples from patients with cancer and controls were prespecified into four cohorts used for model training, analyte integration, and threshold determination, validation, and reproducibility. A-PLUS alone provided a sensitivity of 40.5% across 11 different cancer types in the validation cohort, at a specificity of 98.5%. Combining A-PLUS with aneuploidy and eight common protein biomarkers detected 51% of the cancers at 98.9% specificity. We found that part of the power of A-PLUS could be ascribed to a single feature—the global reduction of AluS subfamily elements in the circulating DNA of patients with solid cancer. We confirmed this reduction through the analysis of another independent dataset obtained with a different approach (whole-genome sequencing). The evaluation of Alu elements may therefore have the potential to enhance the performance of several methods designed for the earlier detection of cancer.

Determining the etiologic basis of the mutations that are responsible for cancer is one of the fundamental challenges in modern cancer research. Different mutational processes induce different types of DNA mutations, providing ‘mutational signatures’ that have led to key insights into cancer etiology. The most widely used signatures for assessing genomic data are based on unsupervised patterns that are then retrospectively correlated with certain features of cancer. We show here that supervised machine-learning techniques can identify signatures, called SuperSigs, that are more predictive than those currently available. Surprisingly, we found that aging yields different SuperSigs in different tissues, and the same is true for environmental exposures. We were able to discover SuperSigs associated with obesity, the most important lifestyle factor contributing to cancer in Western populations.

Emerging data provide initial support for the concept that a single, minimally invasive liquid biopsy test, performed in conjunction with confirmatory radiologic or other diagnostic testing, when indicated, could be deployed on a broad scale to screen individuals for multiple types of cancer. Ideally, such a test could do this in a way that yields a clinically important percentage of true‐positive indications of cancer while minimizing false‐positive signals. Modern decision modeling approaches can and should be deployed to investigate the health and economic consequences of such multicancer early detection (MCED) testing within defined at‐risk populations. In this paper, through small‐scale analyses involving 3 hypothetical MCED‐detectible cancers, the authors illustrate the potential for MCED testing to be cost‐effective, along with the pivotal role of test‐induced stage shift on results. The time is ripe for additional, prospective investigations of the clinical value of MCED testing, the benefits versus the risks for screened populations, and the overall projected impact on health outcomes and costs over time.

Summary: In complex tissues containing cells that are difficult to dissociate, single-nucleus RNA-sequencing (snRNA-seq) has become the preferred experimental technology over single-cell RNA-sequencing (scRNA-seq) to measure gene expression. To accurately model these data in downstream analyses, previous work has shown that droplet-based scRNA-seq data are not zero-inflated, but whether droplet-based snRNA-seq data follow the same probability distributions has not been systematically evaluated. Using pseudonegative control data from nuclei in mouse cortex sequenced with the 10x Genomics Chromium system and mouse kidney sequenced with the DropSeq system, we found that droplet-based snRNA-seq data follow a negative binomial distribution, suggesting that parametric statistical models applied to scRNA-seq are transferable to snRNA-seq. Furthermore, we found that the quantification choices in adapting quantification mapping strategies from scRNA-seq to snRNA-seq can play a significant role in downstream analyses and biological interpretation. In particular, reference transcriptomes that do not include intronic regions result in significantly smaller library sizes and incongruous cell type classifications. We also confirmed the presence of a gene length bias in snRNA-seq data, which we show is present in both exonic and intronic reads, and investigate potential causes for the bias.

In complex tissues containing cells that are difficult to dissociate, single-nucleus RNA-sequencing (snRNA-seq) has become the preferred experimental technology over single-cell RNA-sequencing (scRNA-seq) to measure gene expression. To accurately model these data in downstream analyses, previous work has shown that droplet-based scRNA-seq data are not zero-inflated, but whether droplet-based snRNA-seq data follow the same probability distributions has not been systematically evaluated. Using pseudo-negative control data from nuclei in mouse cortex sequenced with the 10x Genomics Chromium system, we found that snRNA-seq data follow a negative binomial distribution, suggesting that parametric statistical models applied to scRNA-seq are transferable to snRNA-seq. Furthermore, we found that the quantification choices in adapting quantification mapping strategies from scRNA-seq to snRNA-seq can play a significant role in downstream analyses and biological interpretation. In particular, reference transcriptomes that do not include intronic regions result in significantly smaller library sizes and incongruous cell type classifications. We also confirmed the presence of a gene length bias in snRNA-seq data, which we show is present in both exonic and intronic reads, and investigate potential causes for the bias.

Screening for colorectal cancer with blood-based testing should detect advanced adenomas (AA), facilitating more effective cancer prevention. We evaluated four different methods to detect AAs in plasma: (i) a machine learning algorithm, Signatures of fragment Length (SignaL), based on cell-free DNA (cfDNA) fragmentation; (ii) a “Protein-17” assay measuring 17 cancer-associated proteins; (iii) a global aneuploidy score (GAS); and (iv) cfDNA mutation analysis querying 15 genes commonly mutated in colorectal cancer. Existing data from study populations with and without cancer were utilized to determine 99.5% specificity thresholds. We studied 40 AA cases and 32 colonoscopy-negative controls. SignaL detected 9/40 AAs [22.5%, 95% confidence interval (CI), 12.3%–37.5%] at 100% specificity (95% CI, 89.3%–100%). Protein-17 detected 5/40 AAs (12.5%, 95% CI, 5.5%–26.1%) including three cases not identified by SignaL, at 100% specificity (95% CI, 89.3%–100%). GAS detected 11/40 AAs (27.5%, 95% CI, 16.1%–42.8%) but resulted in 2/32 positive controls (93.8% specificity, 95% CI, 79.9%–98.3%). Combining SignaL, the Protein-17 assay, and GAS at the 99.5% specificity thresholds resulted in detection of 16/40 AAs (40%, 95% CI, 26.3%–55.4%) at 93.8% specificity (95% CI, 79.9%–98.3%). Of 32/40 evaluable AA plasma samples, only one was cfDNA mutation positive at level of significance of 0.01. A blood-based assay based on the analysis of repeated sequence elements plus proteins seems to be able to detect a considerable fraction of patients with AA at relatively high specificity. Large, prospective studies are required to determine whether this approach can add to the options currently available for screening patients for premalignant lesions of the colon. Significance: Blood-based screening for colorectal cancer could improve testing uptake and outcomes. We propose novel methods to detect AAs in plasma using cfDNA fragmentation patterns, cancer-associated proteins, and aneuploidy with high specificity. Larger studies are needed to validate clinical utility.