Normal Tissue Contamination Research Articles

348. IDENTIFYING NOVEL MOLECULAR SUBTYPES OF ESOPHAGEAL ADENOCARCINOMA USING LASER CAPTURE MICRODISSECTED RNA-SEQ SAMPLES

Abstract Background Treatment options for esophageal adenocarcinoma (EAC) are limited by a lack of disease stratification methods. In other cancers, gene expression profiling has been successfully used to identify prognostically-relevant and treatment-susceptible molecular subtypes. Previous efforts to subtype EAC have been hindered by low biopsy and resection tumour cellularity. We hypothesize that laser-capture microdissection (LCM) tumour cell enrichment will permit the classification EAC tumours into multiple molecular subtypes that will be important for future therapeutic strategies. Methods Treatment naïve patient samples (N = 52) were collected from primary biopsies (N = 37), resections (N = 10) and metastatic biopsies (N = 5). Samples were laser-capture microdissected to enrich tumour cells followed by total RNA-seq. Gene expression was quantified using salmon and non-negative matrix factorization with 10 components was used to identify gene expression programs. The components identified were validated on a publicly available normal tissue cohort (N = 13 esophagus, gastric, intestinal, Barrett’s samples) and the TCGA EAC cohort (N = 80). We applied NMF with 4 components to the normal tissue (K = cohort) to identify normal tissue gene signatures to validate our RNA-seq data. Results We verified that our RNA-seq samples were depleted for normal tissue using the four NMF signatures from normal tissue cohort and, as expected, our samples were depleted for gene expression from the gastric and esophagus tissues compared to TCGA. Next, we explored the gene expression programs from our tumour samples and identified seven tumour-intrinsic components and three tumour microenvironmental components. The top 50 genes from each tumour intrinsic components were used with consensus cluster plus to identify K = 6 subtypes (Figure 1). Interestingly, two of these subtypes would be difficult to separate from normal tissue contamination without the use of LCM. Conclusion We have successfully used LCM to deplete normal tissue gene expression from our esophageal adenocarcinoma cohort and identified six molecular subtypes. We are currently evaluating the treatment and clinical implications of these subtypes and aiming to accumulate an additional (N = 50) EAC RNA-seq samples.

Abstract 40: scBayes: A computational method to study tumor subclone-specific gene expression and chromatin accessibility using single-cell RNA sequencing and single-cell ATAC sequencing in combination of bulk DNA sequencing

Abstract Although critical for understanding the underlying cancer biology, the study of tumor cell phenotype and epigenetic state from bulk gene expression (RNA-seq) and chromatin accessibility (ATAC-seq) data is often challenging because tumor samples are typically mixtures of different cell populations due to normal tissue contamination, immune cell infiltration, and tumor subclonality. These cell populations may have vastly divergent transcriptomic and epigenetic characteristics, resulting in mixtures of signals that are often impossible to deconvolute. Here we present scBayes, a computational method that allows one to characterize the transcriptional behavior and/or epigenetic state corresponding to each tumor subclone present in a tumor sample or multiple serial or multisite samples from a given cancer patient. This tool starts with genomically defined subclones reconstructed from bulk DNA sequencing data, e.g., using our own published SubcloneSeeker algorithm. scBayes then layers single-cell gene expression (scRNA-seq) and/or chromatin accessibility (scATAC-seq) data on this genomic subclone “backbone” to assign individual cells to one of the tumor subclones or normal cell compartment, facilitating subclone-specific expression or epigenetic analysis. This allows (1) the identification of tumor expression or epigenetic signals without normal tissue interference, (2) the comparison between tumor and normal cells from the same sample, (3) the comparison of expression and epigenetic state across distinct tumor subclones, and (4) the study of transcriptional/epigenetic plasticity through the comparison of the cell states of the same genomically defined subclone across different time points or metastatic sites. scBayes achieves cell-to-subclone assignment by assessing the presence of somatic variants (CNVs or SNVs) that define cancer subclones in each individual cell from the single-cell sequencing reads. Because per-cell sequencing coverage is sparse, and therefore variant dropout frequent, scBayes utilizes a Bayesian probabilistic framework to calculate the likelihood of a particular cell originating from each subclone (including the normal clone, defined as having no somatic mutations) and makes a probabilistic assignment, maximizing the utilization of the sparse single-cell sequencing data for subclone assignment. We successfully applied this method to multiple longitudinal and metastatic cancer patient datasets, representing both bulk WES and WGS DNA sequencing, as well as single-cell datasets collected using the 10X Chromium and Fluidigm C1 platforms. Our algorithmic approach enables comparative gene expression and pathway analysis, and open chromatin accessibility assessment, across subclones. scBayes is open source and freely available at https://github.com/yiq/scBayes. Citation Format: Yi Qiao, Xiaomeng Huang, Gabor Marth. scBayes: A computational method to study tumor subclone-specific gene expression and chromatin accessibility using single-cell RNA sequencing and single-cell ATAC sequencing in combination of bulk DNA sequencing [abstract]. In: Proceedings of the AACR Special Conference on Advancing Precision Medicine Drug Development: Incorporation of Real-World Data and Other Novel Strategies; Jan 9-12, 2020; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(12_Suppl_1):Abstract nr 40.

Abstract 2910: Multi-sites rapid-autopsy data reveals aggressive metastatic colonization and metastasis phenotypes

Abstract Metastatic breast cancer is a deadly disease with low 5-year survival rate (22%). Because it is extremely difficult to track metastatic spread in a living patient, the patterns of new organ colonization are currently poorly understood. In this study, we have collected tumor biopsies at initial diagnosis and at mastectomy necessitated by the patient’s relapse, and autopsies from twenty-six metastatic tumors across seven organs and two skin tissues as normal controls via a rapid autopsy procedure within hours after the patient’s death. All samples were subjected to 60X Illumina whole genome sequencing and bulk RNA sequencing. We used copy number variation (CNV) and short variant (SNP, INDEL) data to reconstruct the phylogenetic relationships among the tumor samples, and established how the tumor evolved at the subclone level across different metastatic sites, using our SubcloneSeeker algorithm. We selected tumor samples with the lowest normal tissue contamination for transcriptomic analysis. Differentially expressed gene and pathways were identified from RNA-seq data. Our analysis revealed that the cancer first escaped to the lung, an event that occurred before the time of initial diagnosis. This was followed by four distinct ‘waves’ (G1, G2, G3 and G4) of massive metastatic expansion, beginning with the abdominal organs (involving samples in G1), then the lymph nodes (G2 and G4), brain and bones (G3). We see clear evidence for metastatic “recolonization” where further evolved subclones invaded already established, earlier metastatic sites. Subclone analysis also revealed a bifurcating pattern of evolution in the primary tumor, in which two lineages gave rise to the G1 and G2-G4 metastases respectively. Consistent with the genomic findings, unsupervised clustering based on the gene expression data showed that samples in G1, and samples in G2-G4 formed distinct clusters. Comparing gene expression profiles between G1 (n=4) and G3 (n=9) samples showed that samples in G1 exhibited enrichment in Integrin activation pathways and samples in G3 presented a signature for chromosome 17 amplification. These distinct metastatic phenotypes indicate different survival mechanisms for G1 and G3 metastatic tumors. The high number of sampled sites in this study, 30 in all, allowed us to reconstruct the evolution of the disease in this patient with unprecedented resolution, identify novel patterns of metastatic colonization, and identify distinct metastatic phenotypes. This study suggests the need for more comprehensive collection and evaluation of metastatic sites to ensure that therapies chosen to target specific genomic aberrations are able to combat the tumor at all metastatic sites. Citation Format: Xiaomeng Huang, Yi Qiao, Samuel W. Brady, Adam Cohen, Andrea H. Bild, Gabor T. Marth. Multi-sites rapid-autopsy data reveals aggressive metastatic colonization and metastasis phenotypes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2910.

Abstract 209: Laser capture microdissection of metastatic colon adenocarcinoma samples enables RNA sequencing analysis of tumor and stromal compartments

Abstract Gene expression data from whole tumor samples represent averaged expression of all the tumor cells, stromal cells and normal adjacent cells present in a given sample. Identification of colon tumor-specific genes may be hampered by contamination from surrounding normal adjacent tissue or stroma. In order to better understand gene expression derived specifically from colon tumor cells we performed laser-capture microdissection to obtain separate samples of tumor and stromal/normal adjacent tissue. We generated data from 12 samples of well-differentiated colon adenocarcinoma metastasized to the liver, lung or ovary. Snap-frozen samples were sectioned, stained with an RNA-preserving haematoxylin and eosin staining procedure and tumor cell and stromal cell containing areas were defined with the support of a pathologist. Tumor and non-tumor samples (containing both stroma and normal adjacent tissue) were captured separately and total RNA isolated from these alongside whole (non-dissected) sections. RNA-sequencing was performed with Kapa HyperPrep Riboerase library preparation and Illumina sequencing at UCL Genomics, London, UK and FPKM values were calculated. For tumors metastasized to liver, analysis of liver-specific genes (including ALB, HP and ORM1) showed that in 6/9 samples these genes were present in the whole section, but absent in the microdissected tumor cells. Some samples (3/9) showed very low expression of liver-specific markers in either the whole section or the dissected non-tumor compartment, indicating low levels of normal adjacent tissue contamination, confirmed by pathology analysis. Similarly in tumors metastasized to the lung, we found absence of lung-specific genes including surfactants in the microdissected tumor cells (2/2 samples), despite high expression in whole sections and the non-tumor compartment. We were able to demonstrate localization of known and novel colon tumor genes in the tumor cell compartment. For example, a cancer testis antigen expressed in 13% of colon adenocarcinoma samples in the TCGA dataset was localized specifically to the tumor compartment in 3/12 microdissected samples and was absent from the non-tumor fraction. We were also able to identify genes derived from immune infiltrates and cancer-associated fibroblasts in the non-tumor compartment. This unique dataset can be used to identify genes expressed in metastatic colon adenocarcinoma without interference from stromal cell, immune infiltrate or normal adjacent tissue. Citation Format: Stephanie Littlewood, Rebekka Krumbach, Gemma Corr, Laura Collins, Francesco Carbone, Adriana Gambardella, Emma Hickman. Laser capture microdissection of metastatic colon adenocarcinoma samples enables RNA sequencing analysis of tumor and stromal compartments [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 209.

Toward Signaling-Driven Biomarkers Immune to Normal Tissue Contamination

The goal of this study was to discover a minimally invasive pathway-specific biomarker that is immune to normal cell mRNA contamination for diagnosing head and neck squamous cell carcinoma (HNSCC). Using Elsevier’s MedScan natural language processing component of the Pathway Studio software and the TRANSFAC database, we produced a curated set of genes regulated by the signaling networks driving the development of HNSCC. The network and its gene targets provided prior probabilities for gene expression, which guided our CoGAPS matrix factorization algorithm to isolate patterns related to HNSCC signaling activity from a microarray-based study. Using patterns that distinguished normal from tumor samples, we identified a reduced set of genes to analyze with Top Scoring Pair in order to produce a potential biomarker for HNSCC. Our proposed biomarker comprises targets of the transcription factor (TF) HIF1A and the FOXO family of TFs coupled with genes that show remarkable stability across all normal tissues. Based on validation with novel data from The Cancer Genome Atlas (TCGA), measured by RNAseq, and bootstrap sampling, the biomarker for normal vs. tumor has an accuracy of 0.77, a Matthews correlation coefficient of 0.54, and an area under the curve (AUC) of 0.82.

Development and validation of a microRNA based diagnostic assay for primary tumor site classification of liver core biopsies

Identification of the primary tumor site in patients with metastatic cancer is clinically important, but remains a challenge. Hence, efforts have been made towards establishing new diagnostic tools. Molecular profiling is a promising diagnostic approach, but tissue heterogeneity and inadequacy may negatively affect the accuracy and usability of molecular classifiers. We have developed and validated a microRNA-based classifier, which predicts the primary tumor site of liver biopsies, containing a limited number of tumor cells. Concurrently we explored the influence of surrounding normal tissue on classification. MicroRNA profiling was performed using quantitative Real-Time PCR on formalin-fixed paraffin-embedded samples. 278 primary tumors and liver metastases, representing nine primary tumor classes, as well as normal liver samples were used as a training set. A statistical model was applied to adjust for normal liver tissue contamination. Performance was estimated by cross-validation, followed by independent validation on 55 liver core biopsies with a tumor content as low as 10%. A microRNA classifier developed, using the statistical contamination model, showed an overall classification accuracy of 74.5% upon independent validation. Two-thirds of the samples were classified with high-confidence, with an accuracy of 92% on high-confidence predictions. A classifier trained without adjusting for liver tissue contamination, showed a classification accuracy of 38.2%. Our results indicate that surrounding normal tissue from the biopsy site may critically influence molecular classification. A significant improvement in classification accuracy was obtained when the influence of normal tissue was limited by application of a statistical contamination model.

Computational purification of individual tumor gene expression profiles leads to significant improvements in prognostic prediction

Tumor heterogeneity is a limiting factor in cancer treatment and in the discovery of biomarkers to personalize it. We describe a computational purification tool, ISOpure, which directly addresses the effects of variable contamination by normal tissue in clinical tumor specimens. ISOpure uses a set of tumor expression profiles and a panel of healthy tissue expression profiles to generate a purified cancer profile for each tumor sample, and an estimate of the proportion of RNA originating from cancerous cells. Applying ISOpure before identifying gene signatures leads to significant improvements in the prediction of prognosis and other clinical variables in lung and prostate cancer.

Identification of novel melanoma gene expression signatures using a Bayesian approach to filter out gene expression noise from normal tissues.

e19053 Background: Identification of pertinent gene expression signatures may be confounded by failure to account for admixture of tumor specimens with normal tissue. We propose a novel method for filtering out gene expression “noise” from normal tissues in a set of melanoma samples derived from the NCBI Gene Expression Omnibus (GEO) public database. Methods: Using customized software implemented in MATLAB, we performed a preliminary search of GEO using the query term “melanoma”. We manually identified 695 clinical samples within 19 different gene expression studies (median 43 samples per series). Gene expression results were merged across different platforms using rank normalization for the target gene of the respective probes. Standardized gene expression signatures for 36 normal tissues were identified in GEO (GSE2361). We performed a modified sparse Bayesian factor regression analysis that incorporated signatures of normal tissues. Additional regression allowed for discovery of latent factors characteristic of melanoma subtypes. Pathway analysis was performed using Ingenuity Pathway Analysis (IPA) on characteristic genes for inferred melanoma-specific latent factors. Results: Normal tissue contributed significantly to the reported gene expression profiles, explaining a median of 11.5% of specimen variability (range 1.3-55.0%). In a subset analysis of 40 primary or in situ melanoma lesions versus 16 metastatic lesions (GSE7553), primary lesions were more likely contaminated with genes characteristic of normal skin (p<0.001). The top 5 latent factors explained a median of 17.0% of residual sample variance (range 0.6-64%). On IPA analysis, latent factor 1 was strongly associated with dermatologic disease (p = 7.46 e-36), latent factor 2 was associated with inflammatory response (p = 9.71e-25), and latent factor 3 was associated with upregulation of the MITF transcription factor regulation (p = 4.16e-19). Conclusions: Appropriate methods that filter out normal tissue contamination of melanoma specimens can improve our understanding of gene expression patterns in melanoma subtypes, and potentially improve future translational studies.

Abstract 4947: Analysis of structural mutations in cancer genomes

Abstract In our group, we have extensively sequenced various cancer genomes (breast cancer, ovarian cancer and melanoma) using a range of genome-wide sequencing approaches, including DNA-PET, RNA-SEQ and exome-capture sequencing. Taking each tumor as a system of its own, we aim to integrate the different outputs of these technologies to explore the composite of hard-wired changes that drive tumorgenesis in different cancers. One type of hard-wired changes is copy number alterations, where the cancer genome tends to gain oncogenes and lose tumor suppressors by DNA amplifications and deletions respectively. While copy number variation data are now a standard output of the DNA-PET pipeline, the accurate prediction of copy number gains and losses is further complicated by several issues, including normal tissue contamination, tumor heterogeneity and aneuploidy. I will present my current work to deconvolute the contribution of these factors to the copy number profile. I will also discuss the classification of our sequenced cancer genomes into various chromotypes based on their unique copy number signatures. I will show how different chromotypes are enriched in specific types of structural variations, which may underlay alternative mechanisms of genomic instability promoting cancer evolution. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 4947. doi:1538-7445.AM2012-4947

Systematic Bias in Genomic Classification Due to Contaminating Non-neoplastic Tissue in Breast Tumor Samples

BackgroundGenomic tests are available to predict breast cancer recurrence and to guide clinical decision making. These predictors provide recurrence risk scores along with a measure of uncertainty, usually a confidence interval. The confidence interval conveys random error and not systematic bias. Standard tumor sampling methods make this problematic, as it is common to have a substantial proportion (typically 30-50%) of a tumor sample comprised of histologically benign tissue. This "normal" tissue could represent a source of non-random error or systematic bias in genomic classification.MethodsTo assess the performance characteristics of genomic classification to systematic error from normal contamination, we collected 55 tumor samples and paired tumor-adjacent normal tissue. Using genomic signatures from the tumor and paired normal, we evaluated how increasing normal contamination altered recurrence risk scores for various genomic predictors.ResultsSimulations of normal tissue contamination caused misclassification of tumors in all predictors evaluated, but different breast cancer predictors showed different types of vulnerability to normal tissue bias. While two predictors had unpredictable direction of bias (either higher or lower risk of relapse resulted from normal contamination), one signature showed predictable direction of normal tissue effects. Due to this predictable direction of effect, this signature (the PAM50) was adjusted for normal tissue contamination and these corrections improved sensitivity and negative predictive value. For all three assays quality control standards and/or appropriate bias adjustment strategies can be used to improve assay reliability.ConclusionsNormal tissue sampled concurrently with tumor is an important source of bias in breast genomic predictors. All genomic predictors show some sensitivity to normal tissue contamination and ideal strategies for mitigating this bias vary depending upon the particular genes and computational methods used in the predictor.

BACOM: in silico detection of genomic deletion types and correction of normal cell contamination in copy number data

Identification of somatic DNA copy number alterations (CNAs) and significant consensus events (SCEs) in cancer genomes is a main task in discovering potential cancer-driving genes such as oncogenes and tumor suppressors. The recent development of SNP array technology has facilitated studies on copy number changes at a genome-wide scale with high resolution. However, existing copy number analysis methods are oblivious to normal cell contamination and cannot distinguish between contributions of cancerous and normal cells to the measured copy number signals. This contamination could significantly confound downstream analysis of CNAs and affect the power to detect SCEs in clinical samples. We report here a statistically principled in silico approach, Bayesian Analysis of COpy number Mixtures (BACOM), to accurately estimate genomic deletion type and normal tissue contamination, and accordingly recover the true copy number profile in cancer cells. We tested the proposed method on two simulated datasets, two prostate cancer datasets and The Cancer Genome Atlas high-grade ovarian dataset, and obtained very promising results supported by the ground truth and biological plausibility. Moreover, based on a large number of comparative simulation studies, the proposed method gives significantly improved power to detect SCEs after in silico correction of normal tissue contamination. We develop a cross-platform open-source Java application that implements the whole pipeline of copy number analysis of heterogeneous cancer tissues including relevant processing steps. We also provide an R interface, bacomR, for running BACOM within the R environment, making it straightforward to include in existing data pipelines. The cross-platform, stand-alone Java application, BACOM, the R interface, bacomR, all source code and the simulation data used in this article are freely available at authors' web site: http://www.cbil.ece.vt.edu/software.htm.

Model-Integrated Estimation of Normal Tissue Contamination for Cancer SNP Allelic Copy Number Data

SNP allelic copy number data provides intensity measurements for the two different alleles separately. We present a method that estimates the number of copies of each allele at each SNP position, using a continuous-index hidden Markov model. The method is especially suited for cancer data, since it includes the fraction of normal tissue contamination, often present when studying data from cancer tumors, into the model. The continuous-index structure takes into account the distances between the SNPs, and is thereby appropriate also when SNPs are unequally spaced. In a simulation study we show that the method performs favorably compared to previous methods even with as much as 70% normal contamination. We also provide results from applications to clinical data produced using the Affymetrix genome-wide SNP 6.0 platform.

Analysing DNA copy number changes using Fused Margin Regression

DNA copy number change is an important form of structural variations in human genomes. Detecting copy number changes using DNA array data is a challenging task due to high density genomic loci, low signal to noise ratios, and normal tissue contamination. We propose Fused Margin Regression (FMR) method that combines a variable fusion rule and robust ε-insensitive loss criterion to estimate piecewise constant segments of the underlying copy number profile. We tested FMR method on both simulation and real CGH and SNP array datasets, and observed competitively improved performance as compared to several widely-adopted existing methods.

Allelic loss analysis by capillary electrophoresis: an accurate, automated method for detection of deletions in solid tumors.

In this report, we describe a fast and accurate capillary electrophoresis, PCR-based method for detecting loss of allelic heterozygosity in solid tumor samples. This automated method requires small sample sizes, and data can be obtained in less than 15 min. The method is particularly powerful for uncovering deletions in tumor sample preparations containing significant normal tissue contamination.

Loss of heterozygosity in tumor cells requires re-evaluation: the data are biased by the size-dependent differential sensitivity of allele detection

Normal tissue contamination of tumors may eclipse the detection of loss of heterozygosity (LOH) by microsatellite analysis and may also hamper isolation of tumor suppressor genes. To test the potential impact of this problem, we prepared artificial mixtures of mouse-human microcell hybrid lines that carried different alleles of the same chromosome 3 marker. After performing an allele titration assay, we found a consistent difference between the LOH of a high molecular weight (H) allele and the LOH of a low molecular weight (L) allele of the same CA repeat marker. It follows that normal tissue admixtures will be less of a problem when LOH affects a H allele than with a L allele. Random screening of 100 papers published between 1994 and 1999 revealed that the loss of a L allele was recorded at about half the frequency (52%) of loss of a H allele. To avoid this bias, we have developed rules for the evaluation of LOH data. We suggest that the loss of a L allele should be given more weight than the loss of a H allele in LOH studies using microsatellite markers.