Clarifying the longitudinal factor structure, temporal stability, and construct validity of Go/No-Go task-related neural activation across adolescence and young adulthood

This study aimed to clarify the psychometric properties and development of Go/No-Go (GNG) task-related neural activation across critical periods of neurobiological maturation by examining its longitudinal stability, factor structure, developmental change, and associations with a computational index of task-general cognitive control. A longitudinal sample (N=289) of adolescents from the Michigan Longitudinal Study was assessed at four time-points (mean number of timepoints per participant=2.05; standard deviation=0.89) spanning early adolescence (ages 10-13) to young adulthood (22-25). Results suggested that regional neural activations from the "successful inhibition" (SI>GO) and "failed inhibition" (FI>GO; error-monitoring) contrasts are each described well by a single general factor. Neural activity across both contrasts showed developmental increases throughout adolescence that plateau in young adulthood. Neural activity metrics evidenced low temporal stability across this period of marked developmental change, and the SI>GO factor showed no relations with a behavioral index of cognitive control. The FI>GO factor displayed stronger criterion validity in the form of significant, positive associations with behaviorally measured cognitive control. Findings emphasize the utility of well-validated psychometric methods and longitudinal data for clarifying the measurement properties of functional neuroimaging metrics and improving measurement practices in developmental cognitive neuroscience.

HMGA1 induces FGF19 to drive pancreatic carcinogenesis and stroma formation

High mobility group A1 (HMGA1) chromatin regulators are upregulated in diverse tumors where they portend adverse outcomes, although how they function in cancer remains unclear. Pancreatic ductal adenocarcinomas (PDACs) are highly lethal tumors characterized by dense desmoplastic stroma composed predominantly of cancer-associated fibroblasts and fibrotic tissue. Here, we uncover an epigenetic program whereby HMGA1 upregulates FGF19 during tumor progression and stroma formation. HMGA1 deficiency disrupts oncogenic properties in vitro while impairing tumor inception and progression in KPC mice and subcutaneous or orthotopic models of PDAC. RNA sequencing revealed HMGA1 transcriptional networks governing proliferation and tumor-stroma interactions, including the FGF19 gene. HMGA1 directly induces FGF19 expression and increases its protein secretion by recruiting active histone marks (H3K4me3, H3K27Ac). Surprisingly, disrupting FGF19 via gene silencing or the FGFR4 inhibitor BLU9931 recapitulates most phenotypes observed with HMGA1 deficiency, decreasing tumor growth and formation of a desmoplastic stroma in mouse models of PDAC. In human PDAC, overexpression of HMGA1 and FGF19 defines a subset of tumors with extremely poor outcomes. Our results reveal what we believe is a new paradigm whereby HMGA1 and FGF19 drive tumor progression and stroma formation, thus illuminating FGF19 as a rational therapeutic target for a molecularly defined PDAC subtype.

A mechanical task for measuring sign- and goal-tracking in humans: A proof-of-concept study

Cue-based associative learning (i.e., Pavlovian conditioning) is a foundational component of behavior in almost all forms of animal life and may provide insight into individual differences in addiction liability. Cues can take on incentive-motivational properties (i.e., incentive salience) through Pavlovian learning. Extensive testing with non-human animals (primarily rats) has demonstrated significant variation among individuals in the behaviors this type of learning evokes. So-named "sign-trackers" and "goal-trackers" have been examined in many studies of non-human animals, but this work in humans is still a nascent area of research. In the present proof-of-concept study, we used a Pavlovian conditioned approach task to investigate human sign- and goal-tracking in emerging adults. Conditioned behaviors that developed over the course of the task were directed toward the reward-cue and toward the reward location. Participants' eye-gaze and behavior during the task were submitted to a latent profile analysis, which revealed three groups defined as sign-trackers (n = 10), goal-trackers (n = 4), and intermediate responders (n = 36). Impulsivity was a significant predictor of the sign-tracking group relative to the goal-tracking group. The present study provides preliminary evidence that a simple procedure can produce learned Pavlovian conditioned approach behavior in humans. Though further investigation is required, findings provide a promising step toward the long-term goal of translating important insights gleaned from basic research into treatment strategies that can be applied to clinical populations.

Development of an automated liquid biopsy assay for methylated markers in advanced breast cancer

Current molecular liquid biopsy assays to detect recurrence or monitor response to treatment require sophisticated technology, highly trained personnel, and a turnaround time of weeks. We describe the development and technical validation of an automated Liquid Biopsy for Breast Cancer Methylation (LBx-BCM) prototype, a DNA methylation detection cartridge assay that is simple to perform and quantitatively detects nine methylated markers within 4.5 h. LBx-BCM demonstrated high interassay reproducibility when analyzing exogenous methylated DNA (75-300 DNA copies) spiked into plasma (Coefficient of Variation, CV = 7.1 - 10.9%) and serum (CV = 19.1 - 36.1%). It also demonstrated high interuser reproducibility (Spearman r = 0.887, P < 0.0001) when samples of metastatic breast cancer (MBC, N = 11) and normal control (N = 4) were evaluated independently by two users. Analyses of interplatform reproducibility indicated very high concordance between LBx-BCM and the reference assay, cMethDNA, among 66 paired plasma samples (MBC N = 40, controls N = 26; Spearman r = 0.891; 95% CI = 0.825 - 0.933, P< 0.0001). LBx-BCM achieved a ROC AUC = 0.909 (95% CI = 0.836 - 0.982), 83% sensitivity and 92% specificity; cMethDNA achieved a ROC AUC = 0.896 (95% CI = 0.817 - 0.974), 83% sensitivity and 92% specificity in test set samples. The automated LBx-BCM cartridge prototype is fast, with performance levels equivalent to the highly sensitive, manual cMethDNA method. Future prospective clinical studies will evaluate LBx-BCM detection sensitivity and its ability to monitor therapeutic response during treatment for advanced breast cancer.

Automated and rapid detection of cancer in suspicious axillary lymph nodes in patients with breast cancer

Preoperative staging of suspicious axillary lymph nodes (ALNs) allows patients to be triaged to ALN dissection or to sentinel lymph node biopsy (SLNB). Ultrasound-guided fine needle aspiration (FNA) and cytology of ALN is moderately sensitive but its clinical utility relies heavily on the cytologist's experience. We proposed that the 5-h automated GeneXpert system-based prototype breast cancer detection assay (BCDA) that quantitatively measures DNA methylation in ten tumor-specific gene markers could provide a facile, accurate test for detecting cancer in FNA of enlarged lymph nodes. We validated the assay in ALN-FNA samples from a prospective study of patients (N = 230) undergoing SLNB. In a blinded analysis of 218 evaluable LN-FNAs from 108 malignant and 110 benign LNs by histology, BCDA displayed a sensitivity of 90.7% and specificity of 99.1%, achieving an area under the ROC curve, AUC of 0.958 (95% CI: 0.928-0.989; P < 0.0001). Next, we conducted a study of archival FNAs of ipsilateral palpable LNs (malignant, N = 72, benign, N = 53 by cytology) collected in the outpatient setting prior to neoadjuvant chemotherapy (NAC). Using the ROC-threshold determined in the prospective study, compared to cytology, BCDA achieved a sensitivity of 94.4% and a specificity of 92.5% with a ROC-AUC = 0.977 (95% CI: 0.953-1.000; P < 0.0001). Our study shows that the automated assay detects cancer in suspicious lymph nodes with a high level of accuracy within 5 h. This cancer detection assay, scalable for analysis to scores of LN FNAs, could assist in determining eligibility of patients to different treatment regimens.

Methylated markers accurately distinguish primary central nervous system lymphomas (PCNSL) from other CNS tumors

BACKGROUND

Definitive diagnosis of primary central nervous system lymphoma (PCNSL) requires invasive surgical brain biopsy, causing treatment delays. In this paper, we identified and validated tumor-specific markers that can distinguish PCNSL from other CNS tumors in tissues. In a pilot study, we tested these newly identified markers in plasma.

RESULTS

The Methylation Outlier Detector program was used to identify markers in TCGA dataset of 48 diffuse large B-cell lymphoma (DLBCL) and 656 glioblastomas and lower-grade gliomas. Eight methylated markers clearly distinguished DLBCL from gliomas. Marker performance was verified (ROC-AUC of ≥ 0.989) in samples from several GEO datasets (95 PCNSL; 2112 other primary CNS tumors of 11 types). Next, we developed a novel, efficient assay called Tailed Amplicon Multiplexed-Methylation-Specific PCR (TAM-MSP), which uses two of the methylation markers, cg0504 and SCG3 triplexed with ACTB. FFPE tissue sections (25 cases each) of PCNSL and eight types of other primary CNS tumors were analyzed using TAM-MSP. TAM-MSP distinguished PCNSL from the other primary CNS tumors with 100% accuracy (AUC = 1.00, 95% CI 0.95-1.00, P < 0.001). The TAM-MSP assay also detected as few as 5 copies of fully methylated plasma DNA spiked into 0.5 ml of healthy plasma. In a pilot study of plasma from 15 PCNSL, 5 other CNS tumors and 6 healthy individuals, methylation in cg0504 and SCG3 was detectable in 3/15 PCNSL samples (20%).

CONCLUSION

The Methylation Outlier Detector program identified methylated markers that distinguish PCNSL from other CNS tumors with accuracy. The high level of accuracy achieved by these markers was validated in tissues by a novel method, TAM-MSP. These studies lay a strong foundation for a liquid biopsy-based test to detect PCNSL-specific circulating tumor DNA.

Abstract PS4-03: An automated DNA methylation assay for monitoring treatment response in patients with metastatic breast cancer

BACKGROUND: Previously, we demonstrated the clinical validity of cMethDNA, a circulating methylated tumor DNA (ctDNA) assay, in serum samples from TBCRC 005 (J Clin Oncol, 2017; 35:751-758) to predict progression free- and overall-survival, and to monitor therapeutic response in patients with stage IV breast cancer. Here, Johns Hopkins (JH) and Cepheid partnered to develop an automated GeneXpert (GX) cartridge-based system to provide quantitative measures of DNA methylation within 5 hours.METHODS: With a goal of discriminating stage IV breast cancer from healthy and benign breast disease with high sensitivity and specificity, we evaluated breast cancer-specific DNA methylation markers (selected through comprehensive methylome analysis) in STRECK tube plasma of 46 patients with metastatic breast cancer enrolled in Individualized Molecular Analyses Guide Efforts in Breast Cancer (IMAGE II trial), 17 benign breast disease and 9 healthy normal controls (J0888 repository). Blood from IMAGE II participants was collected upon disease progression. A newly designed GX Breast Cancer Monitoring Assay for research use only (RUO*) first converted unmethylated CpG sites in ctDNA from 1 ml plasma with bisulfite. The sample was then split into two methylation detection cartridges, which quantitated DNA methylation of 9 markers along with an ACTB reference. Cumulative methylation (CM) of the 9-gene panel was calculated using a novel algorithm. Performance was assessed based on Receiver Operating Characteristic (ROC) curves and Mann-Whitney analyses.RESULTS: The GX Breast Cancer Monitoring Assay (RUO)* showed that the 9-gene panel was significantly more methylated in cancer compared to normal/benign plasma samples (median for cancer: 428.0 CM units versus for benign: 0.0 CM units; P&lt; 0.0001), and revealed a sensitivity of 85% and specificity of 92%, using a cumulative methylation threshold of 35.5 units based on ROC area under the curve (AUC) = 0.909 (95% CI 0.836 – 0.982, P&lt;0.0001). We will present comparisons of the GX results to cMethDNA, the gold standard assay, which reported 85-90% sensitivity at 90% specificity.CONCLUSIONS: We identified a panel of methylated DNA markers that discriminates stage IV breast from benign breast disease and healthy normal subjects using ctDNA. Our automated cartridge-based assay prototype demonstrates high sensitivity and specificity for detecting invasive breast cancer. Its ability to assess changes in DNA methylation will be tested next with clinical trial samples collected longitudinally during treatment. This assay has potential clinical utility in monitoring therapeutic response and predicting disease recurrence.* For Research Use Only. Not for use in diagnostic procedures. Not reviewed by any regulatory body.

Citation Format: Mary Jo Fackler, Suzana Tulac, Neesha Venkatesan, Adam J. Aslam, Leslie M. Cope, Jennifer Lehman, Rita Denbow, Jeffrey Reynolds, Morgan Buckley, Bradley M. Downs, Kala Visvanathan, Christopher B. Umbricht, Antonio C. Wolff, Vered Stearns, Edwin W. Lai, Saraswati Sukumar. An automated DNA methylation assay for monitoring treatment response in patients with metastatic breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS4-03.

DNA Methylation Markers for Breast Cancer Detection in the Developing World

PURPOSE

An unmet need in low-resource countries is an automated breast cancer detection assay to prioritize women who should undergo core breast biopsy and pathologic review. Therefore, we sought to identify and validate a panel of methylated DNA markers to discriminate between cancer and benign breast lesions using cells obtained by fine-needle aspiration (FNA).Experimental Design: Two case-control studies were conducted comparing cancer and benign breast tissue identified from clinical repositories in the United States, China, and South Africa for marker selection/training (N = 226) and testing (N = 246). Twenty-five methylated markers were assayed by Quantitative Multiplex-Methylation-Specific PCR (QM-MSP) to select and test a cancer-specific panel. Next, a pilot study was conducted on archival FNAs (49 benign, 24 invasive) from women with mammographically suspicious lesions using a newly developed, 5-hour, quantitative, automated cartridge system. We calculated sensitivity, specificity, and area under the receiver-operating characteristic curve (AUC) compared with histopathology for the marker panel.

RESULTS

In the discovery cohort, 10 of 25 markers were selected that were highly methylated in breast cancer compared with benign tissues by QM-MSP. In the independent test cohort, this panel yielded an AUC of 0.937 (95% CI = 0.900-0.970). In the FNA pilot, we achieved an AUC of 0.960 (95% CI = 0.883-1.0) using the automated cartridge system.

CONCLUSIONS

We developed and piloted a fast and accurate methylation marker-based automated cartridge system to detect breast cancer in FNA samples. This quick ancillary test has the potential to prioritize cancer over benign tissues for expedited pathologic evaluation in poorly resourced countries.

Abstract 1388: A novel Hyper-Methylation Outlier method for blood biomarker discovery

Introduction: DNA methylation analysis of circulating cell-free DNA in blood enables minimally invasive detection of screening early stage cancer, and monitoring of changes in tumor burden. However, there is no accepted best practice for developing blood cancer markers. In this study, we tested the ability of different methods to discover serum methylation biomarkers from breast tissue methylation arrays. We compare the ability of two previously published biomarker discovery programs, Limma and iEVORA, with a novel Hyper-methylation Outlier method to identify makers from breast tissues that can detect tumor cfDNA in serum.

Experimental Design: The Hyper-Methylation Outlier method approach was designed to be robust to loss of signal associated with dilution from non-tumor DNA in blood. Candidate CpG sites were selected to meet two criteria in training data from primary tissue: A) At least 95% of healthy control samples have methylation beta values &lt;0.10 and, B) at least one tumor sample is a distinct outlier, with beta value &gt;0.30, and prioritized according to the number of outlying tumor samples. A serum sample is called positive if it is an outlier for at least one marker, using a threshold of beta &gt; 0.20 to help account for the expected loss of signal. The three marker discovery methods were used to discover cancer markers using two sets of breast tissue samples. Set 1 included 103 primary breast cancers, 6 normal-adjacent breast tissues, and 15 normal breast organoids, while set 2 included 236 primary breast cancer and 27 normal breast tissue samples. While the Hyper-Methylation Outlier method has a built-in classifier, Limma and iEVORA were paired with logistic regression and random forest, to develop classifiers. Selected markers were tested in a simulation in which 6,400 breast cancer serum samples were generated in silico by mixing tumor DNA methylation profiles with profiles from normal serum at various dilutions. We then compared the performance of the various methods, an independent set of normal serum (n= 67) along with a small number (n=6) of metastatic breast cancer serum samples.

Results: Limma paired with random forest, iEVORA paired with logistic regression and the Hyper-Methylation Outlier method performed similarly in the in silico cancer serum set. All three methods exhibited a sensitivity of over 90%, at dilutions in the range of 80-50% breast cancer to normal serum. In the independent set of normal and metastatic breast cancer serum samples, the Hyper-Methylation Outlier method achieved an AUC of 0.819 and a classifier built by pairing iEVORA paired with logistic regression set achieved an AUC of 0.820.

Conclusions: We have shown that two discovery methods that identify outlier markers, iEVORA and the Hyper-Methylated Outlier method, have the ability to discover blood cancer markers from cancer tissue arrays. However, these markers need further technical validation before their true utility is fully understood.

Citation Format: Bradley M. Downs, Juanjuan Li, Mary Jo Fackler, Antonio C. Wolff, Saraswati Sukumar, Chris B. Umbricht, Leslie M. Cope. A novel Hyper-Methylation Outlier method for blood biomarker discovery [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 1388.

Measuring DNA Copy Number Variation Using High-Density Methylation Microarrays

Genetic and epigenetic changes drive carcinogenesis, and their integrated analysis provides insights into mechanisms of cancer development. Computational methods have been developed to measure copy number variation (CNV) from methylation array data, including ChAMP-CNV, CN450K, and, introduced here, Epicopy. Using paired single nucleotide polymorphism (SNP) and methylation array data from the public The Cancer Genome Atlas repository, we optimized CNV calling and benchmarked the performance of these methods. We optimized the thresholds of all three methods and showed comparable performance across methods. Using Epicopy as a representative analysis of Illumina450K array, we show that Illumina450K-derived CNV methods achieve a sensitivity of 0.7 and a positive predictive value of 0.75 in identifying CNVs, which is similar to results achieved when comparing competing SNP microarray platforms with each other.

Abstract P5-12-04: A new method of data analysis to derive DNA methylation signatures that stratify risk of recurrence in triple negative breast cancer

Background: Triple negative breast cancer (TNBC) accounts for 10-17% of all breast cancer and is more likely to be of higher histological grade, poorly differentiated, associated with a higher recurrence rate and with decreased overall survival. The clinical course of a TNBC patient remains difficult to predict, as tumors with homogenous morphological characteristics may vary in response to therapy and have divergent outcomes. Therefore, additional analytical methods are needed to better classify TNBC. Our goal is to refine the analysis of methylome datasets to derive reliable molecular signatures that can distinguish TNBC patients with good outcomes who may benefit from less aggressive treatment, from those with poor outcomes who would be candidates for more aggressive treatments.

Methods: Our laboratory has conducted and reported, in this meeting, results from analysis of 450k methylation array data on a discovery set of 53 high-risk TNBC cases and 62 low-risk controls treated by locoregional therapy alone, as well as 5 normal breast tissue samples. High-risk cases were defined as patients that relapsed within 0.5 to 6.5 years from the time of diagnosis, while low-risk controls had no relapse and &gt;4 year recurrence-free intervals (RFI). In this work, we devised and applied a novel methylation biomarker discovery program named Hypermethylated Outlier Detector (HOD) that emphasizes the selection of highly methylated markers in cases compared to controls, to find a high-risk signature in the TNBC discovery set. The methylation signature identified by HOD was interrogated in a test set of 50 TNBCs (with 16 recurrences) that did not receive chemotherapy, and in a second test set of 131 TNBCs (with 33 recurrences) that did receive chemotherapy.

Results: HOD identified 39 hypermethylated markers (beta &gt;0.20) that could accurately distinguish between the high-risk cases and the low-risk controls in the discovery set of TNBCs (n=115) treated with locoregional therapy alone. In the test set of TNBC (n=50) with no chemotherapy the 39 markers distinguished high from low risk individuals (likelihood ratio test P=0.049). In a second test set of TNBC (n=131) that received chemotherapy the 39 hypermethylated markers again distinguished high from low risk individuals (likelihood ratio test P=0.0043).

Conclusions: We have presented evidence that a methylation signature identified by HOD can be used to identify TNBC patients that have a high-risk of relapse regardless of receiving chemotherapy. This methylation signature could potentially be used to inform physician decisions on therapeutic strategies for TNBC patients. This could ultimately lead to less aggressive treatment given to patients possessing a methylation profile consistent with a better prognosis. Conversely, patients with hypermethylation in the 39 markers will likely benefit from a more aggressive course of treatment.

Citation Format: Downs BM, Cope LM, Fackler MJ, Cho S, Wolff AC, Regan MM, Sukumar S, Umbricht CB. A new method of data analysis to derive DNA methylation signatures that stratify risk of recurrence in triple negative breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P5-12-04.

Abstract P4-08-09: DNA methylation markers predict recurrence-free interval in triple-negative breast cancer

BACKGROUND. Chemotherapy remains the treatment mainstay for triple-negative breast cancer (TNBC). Nevertheless, randomized trials have shown that not all TNBC require it, nor does it benefit all patients that receive it. Molecular tools to risk-stratify TNBC are currently lacking. In light of the importance of epigenetic processes modulating gene expression, we performed an array-based genome-wide DNA methylation search in well-documented institutional and clinical trial cohorts of TNBC for markers that can distinguish breast cancers with a favorable natural history from those with a high risk of recurrence.

METHODS. We performed an array-based genome-wide DNA methylation survey of well-documented institutional and clinical trial cohorts of TNBC and conducted molecular marker discovery on institutional TNBCs (115 patient samples; 53 recurrences) treated by locoregional therapy (LRT) alone. The identified hypermethylated gene signatures were then tested in a TNBC cohort (50 patient samples; 16 recurrences) from the no chemotherapy arms of IBCSG trials VIII and IX, and in a separate combined cohort of TNBCs (131 patient samples; 33 recurrences) treated with chemotherapy from an institutional repository and from IBCSG trials VIII and IX. Cross platform validation was conducted using quantitative multiplexed methylation specific PCR (QM-MSP) on hypermethylated markers in samples from both the Discovery Set and IBCSG LRT Test Set.

RESULTS. We identified methylation signatures in the discovery cohort consisting of 100 or 30 CpG probes that discriminated patients who remained recurrence-free from those with recurrent disease. These signatures were then tested in the IBCSG no chemotherapy cohort, and we found that hypermethylation was associated with shorter recurrence-free interval (RFI). A significant association of both 100 CpG (P&lt;0.0001) and 30 CpG (P=0.0021) signatures with shorter RFI was found in the combined institutional and IBCSG chemotherapy cohort. We observed an enrichment of methylation probes residing on chromosome 19, particularly within 19q13.41-43, that significantly correlated with RFI following chemotherapy. QM-MSP results reflected that of the methylation array [Spearman correlation coefficient of r = 0.495 (P = 0.0009)] indicating that the relationship between high methylation and short RFI is detectable independent of analytical platform. We also observed enrichment for Chromosome 19-specific probes within the 100 and 30 probe sets. While only 5% of all CpG markers are located within Chr19, 15% of the 100 CpG set, 37% of the 30 CpG set, and 47% of the 17 CpGs that are statistically significantly correlated with RFI in the chemotherapy group reside on the Chr19, mostly within 19q13.41-43.

CONCLUSIONS. Methylation markers may be of prognostic importance in TNBC and our findings should be validated in additional clinical trial cohorts.

Citation Format: Fackler MJ, Cho SS, Cope LM, Gabrielson E, Wilsbach K, Lynch C, Marks JR, Geradts J, Regan MM, Viale G, Wolff AC, Umbricht CB, Sukumar S. DNA methylation markers predict recurrence-free interval in triple-negative breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-08-09.

Methylomic Analysis of Ovarian Cancers Identifies Tumor-Specific Alterations Readily Detectable in Early Precursor Lesions

PURPOSE

High-grade serous ovarian carcinoma (HGSOC) typically remains undiagnosed until advanced stages when peritoneal dissemination has already occurred. Here, we sought to identify HGSOC-specific alterations in DNA methylation and assess their potential to provide sensitive and specific detection of HGSOC at its earliest stages.

EXPERIMENTAL DESIGN

MethylationEPIC genome-wide methylation analysis was performed on a discovery cohort comprising 23 HGSOC, 37 non-HGSOC malignant, and 36 histologically unremarkable gynecologic tissue samples. The resulting data were processed using selective bioinformatic criteria to identify regions of high-confidence HGSOC-specific differential methylation. Quantitative methylation-specific real-time PCR (qMSP) assays were then developed for 8 of the top-performing regions and analytically validated in a cohort of 90 tissue samples. Lastly, qMSP assays were used to assess and compare methylation in 30 laser-capture microdissected (LCM) fallopian tube epithelia samples obtained from cancer-free and serous tubal intraepithelial carcinoma (STIC) positive women.

RESULTS

Bioinformatic selection identified 91 regions of robust, HGSOC-specific hypermethylation, 23 of which exhibited an area under the receiver-operator curve (AUC) value ≥ 0.9 in the discovery cohort. Seven of 8 top-performing regions demonstrated AUC values between 0.838 and 0.968 when analytically validated by qMSP in a 90-patient cohort. A panel of the 3 top-performing genes (c17orf64, IRX2, and TUBB6) was able to perfectly discriminate HGSOC (AUC 1.0). Hypermethylation within these loci was found exclusively in LCM fallopian tube epithelia from women with STIC lesions, but not in cancer-free fallopian tubes.

CONCLUSIONS

A panel of methylation biomarkers can be used to accurately identify HGSOC, even at precursor stages of the disease.

Abstract LB-220: An automated breast cancer detection assay for screening in the developing world

Introduction: Breast cancer incidence is rapidly increasing globally with 1.7 million new cases and 600,000 deaths due to the disease annually. Death due to breast cancer is four to six times more frequent in the developing world compared to the USA due to lack of early detection, definitive diagnosis, and limited access to treatment. Ultrasound imaging for early detection and ultrasound guided fine needle aspiration (FNA) of suspicious lesions followed by an automated, cartridge-based analysis of methylated genes could triage malignancies for faster pathology assessment and treatment, thus enabling better use of scarce resources.

Experimental Procedures: To identify a panel of methylated DNA markers to enable sensitive and specific detection of malignant breast lesions, we screened 24 methylated genes (known to be frequently methylated in malignant breast tissue and unmethylated in normal breast tissue) by Quantitative Multiplex Methylation-Specific PCR (QM-MSP) analysis. Formalin-fixed paraffin-embedded (FFPE) sections of biopsies of both benign and malignant breast lesions from the USA, China and South Africa were analyzed. Samples were divided into Training and Test sets. The Training set consisted of 206 tissues [66 invasive ductal carcinoma (IDC), 30 ductal carcinoma in situ (DCIS), 99 benign breast disease (BBD) and 11 normal breast (NB)]. The gene panel selected in the Training set was examined in an independent Test set of tissues (n=204) [65 IDC, 29 DCIS, 99 BBD and 11 NB]. Further, we optimized the technical performance of an automated, prototype breast cancer detection cartridge system for quantitative assessment of gene methylation and tested it in pilot study using FNA samples from breast cancers.

Results: Analysis of the tissues in the Training set (n=206) led to the selection of a panel of 10 genes highly methylated in malignant lesions with little or no methylation in benign lesions. For the 10-gene panel to achieve a sensitivity greater than 90%, a laboratory cutoff of 14.5 cumulative methylation (CM) units (out of a possible 10,000 units) was set. In the Training set, the 10-gene panel achieved a sensitivity of 90% and a specificity of 85% with receiver operating characteristic (ROC) statistics: ROC, AUC= 0.947. In a blinded Test set of tissue samples (n=204), with a laboratory cutoff of 14.5 CM units, the 10-gene panel achieved a sensitivity of 87% and a specificity of 89%, with ROC statistics: p&lt;0.001, AUC= 0.936, and provided significant accuracy for breast cancers from three countries and all molecular subtypes. In a pilot study of FNA samples using the cartridge system, robust methylation in all ten genes was detected in malignant tumors.

Conclusions: QM-MSP of breast lesions led to the selection of a panel of 10 genes methylated for detection of breast cancer. We have validated the technical performance of an automated, prototype cartridge system. The study reveals the potential of methylation markers to provide fast, accurate and automated cancer detection at a low cost in developing regions globally.

Research Use Only. Not for diagnostic tests.

A sponsored research agreement from Cepheid to Dr. Sukumar's Lab at Johns Hopkins University.

Citation Format: Bradley M. Downs, Mary Jo Fackler, Claudia Mercado-Rodriguez, Ashley Cimino-Mathews, Chuang Chen, Jing-Ping Yuan, Eunice van den Berg, Leslie M. Cope, Susan C. Harvey, Syed Z. Ali, Suzana Tulac, Kriszten J. Kocmond, Edwin W. Lai, Brian Rhees, Mike Bates, Saraswati Sukumar. An automated breast cancer detection assay for screening in the developing world [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 LB-220.

Abstract P6-03-07: An automated DNA methylation assay (QM-MSP) for rapid breast cancer diagnosis in underdeveloped countries

BACKGROUND: Underdeveloped countries reported 882,900 new cases of breast cancer and 324,000 deaths in 2012, likely to be a gross underestimation according to recent reports. Often, mammography screening is not available, primary care services are limited, and pathology and treatment services are available only in the regional hospitals. Because of the lack of access to diagnostic and treatment services, it is estimated that more than 90% of patients with breast cancer never present for medical treatment. To address this situation, an accurate, easy-to-perform diagnostic test appropriate for use in remote clinics is desperately needed. Johns Hopkins (JH) and Cepheid partnered to translate a robust Quantitative Multiplex Methylation-Specific PCR (QM-MSP) assay to an automated, cartridge-based system that provides quantitative measures of DNA methylation within hours of fine needle aspiration or core biopsy of image-detected suspicious lesions.

METHODS: With a goal of discriminating malignant from benign breast disease with high sensitivity and specificity, we evaluated 24 breast cancer-specific DNA methylation markers (selected through comprehensive methylome analysis) in 119 invasive ductal carcinomas and 186 benign breast tissues. QM-MSP was performed on sections of formalin-fixed paraffin-embedded (FFPE) tissues to quantify DNA methylation. The dynamic range and performance of quantitative methylation detection was tested using a subset of 9 genes in the cartridge-based system.

RESULTS: QM-MSP was performed in a Training set consisting of 93 tissues [n=43 IDC, n=50 benign lesions (25 usual ductal hyperplasia, UDH, and 25 papilloma)] from the US. We selected 9 DNA markers significantly (p&lt;0.05) more methylated in malignant compared to benign lesions, which had low or no methylation. An independent Test set consisted of benign (n=26) and malignant (n=10) tissues (mostly Caucasian; JH Test Set). As a panel, the 9 markers were significantly more methylated in malignant than benign tissue (p&lt;0.001), revealing a sensitivity of 90% and specificity of 92%, using a laboratory cutoff of 9.5 CMI units (900 unit scale) based on receiver operator characteristic statistics (ROC; p&lt;0.0001, AUC=0.977). To determine if the markers characterized in the JH Test Set could perform as well in samples from a different geography, the panel was tested on 176 tissues from Wuhan, China (China Test Set). In this cohort (66 IDC and 110 benign tissues - 49 fibroadenoma, 19 benign cyst, 12 UDH, 30 papilloma), sensitivity was 89% and specificity was 89% for detection of breast cancer with ROC AUC=0.945. An advanced version of the cartridge with up to 12 methylated DNA markers is under development, thus far showing robust signals in cancer and low background in benign tissues. Current work at JH is focused on optimizing the technical performance of the cartridge.

CONCLUSIONS: We identified a panel of methylated DNA markers that discriminate malignant from benign breast lesions and built a prototype automated cartridge-based assay with promising sensitivity and specificity for breast cancer. Such an assay has the potential to aid in specimen triage in the pathology lab and provide fast, low cost and accurate diagnosis of breast cancer in LMIC settings.

Citation Format: Fackler MJ, Downs BM, Mercado-Rodriguez C, Cimino-Mathews A, Chen C, Yuan J, Cope LM, Kohlway A, Kocmond K, Lai E, Weidler J, Visvanathan K, Umbricht CB, Harvey S, Wolff AC, Bates M, Sukumar S. An automated DNA methylation assay (QM-MSP) for rapid breast cancer diagnosis in underdeveloped countries [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P6-03-07.

Monitoring of Serum DNA Methylation as an Early Independent Marker of Response and Survival in Metastatic Breast Cancer: TBCRC 005 Prospective Biomarker Study

Purpose Epigenetic alterations measured in blood may help guide breast cancer treatment. The multisite prospective study TBCRC 005 was conducted to examine the ability of a novel panel of cell-free DNA methylation markers to predict survival outcomes in metastatic breast cancer (MBC) using a new quantitative multiplex assay (cMethDNA). Patients and Methods Ten genes were tested in duplicate serum samples from 141 women at baseline, at week 4, and at first restaging. A cumulative methylation index (CMI) was generated on the basis of six of the 10 genes tested. Methylation cut points were selected to maximize the log-rank statistic, and cross-validation was used to obtain unbiased point estimates. Logistic regression or Cox proportional hazard models were used to test associations between the CMI and progression-free survival (PFS), overall survival (OS), and disease status at first restaging. The added value of the CMI in predicting survival outcomes was evaluated and compared with circulating tumor cells (CellSearch). Results Median PFS and OS were significantly shorter in women with a high CMI (PFS, 2.1 months; OS, 12.3 months) versus a low CMI (PFS, 5.8 months; OS, 21.7 months). In multivariable models, among women with MBC, a high versus low CMI at week 4 was independently associated with worse PFS (hazard ratio, 1.79; 95% CI, 1.23 to 2.60; P = .002) and OS (hazard ratio, 1.75; 95% CI, 1.21 to 2.54; P = .003). An increase in the CMI from baseline to week 4 was associated with worse PFS ( P < .001) and progressive disease at first restaging ( P < .001). Week 4 CMI was a strong predictor of PFS, even in the presence of circulating tumor cells ( P = .004). Conclusion Methylation of this gene panel is a strong predictor of survival outcomes in MBC and may have clinical usefulness in risk stratification and disease monitoring.

Fecal Metabolome in Hmga1 Transgenic Mice with Polyposis: Evidence for Potential Screen for Early Detection of Precursor Lesions in Colorectal Cancer

Because colorectal cancer (CRC) remains a leading cause of cancer mortality worldwide, more accessible screening tests are urgently needed to identify early stage lesions. We hypothesized that highly sensitive, metabolic profile analysis of stool samples will identify metabolites associated with early stage lesions and could serve as a noninvasive screening test. We therefore applied traveling wave ion mobility mass spectrometry (TWIMMS) coupled with ultraperformance liquid chromatography (UPLC) to investigate metabolic aberrations in stool samples in a transgenic model of premalignant polyposis aberrantly expressing the gene encoding the high mobility group A (Hmga1) chromatin remodeling protein. Here, we report for the first time that the fecal metabolome of Hmga1 mice is distinct from that of control mice and includes metabolites previously identified in human CRC. Significant alterations were observed in fatty acid metabolites and metabolites associated with bile acids (hypoxanthine xanthine, taurine) in Hmga1 mice compared to controls. Surprisingly, a marked increase in the levels of distinctive short, arginine-enriched, tetra-peptide fragments was observed in the transgenic mice. Together these findings suggest that specific metabolites are associated with Hmga1-induced polyposis and abnormal proliferation in intestinal epithelium. Although further studies are needed, these data provide a compelling rationale to develop fecal metabolomic analysis as a noninvasive screening tool to detect early precursor lesions to CRC in humans.

Comprehensive Pan-Genomic Characterization of Adrenocortical Carcinoma

We describe a comprehensive genomic characterization of adrenocortical carcinoma (ACC). Using this dataset, we expand the catalogue of known ACC driver genes to include PRKAR1A, RPL22, TERF2, CCNE1, and NF1. Genome wide DNA copy-number analysis revealed frequent occurrence of massive DNA loss followed by whole-genome doubling (WGD), which was associated with aggressive clinical course, suggesting WGD is a hallmark of disease progression. Corroborating this hypothesis were increased TERT expression, decreased telomere length, and activation of cell-cycle programs. Integrated subtype analysis identified three ACC subtypes with distinct clinical outcome and molecular alterations which could be captured by a 68-CpG probe DNA-methylation signature, proposing a strategy for clinical stratification of patients based on molecular markers.

Abstract B32: Inhibiting DNA methylation causes an interferon response in cancer via dsRNA including endogenous retroviruses

DNA methyltransferase inhibitors (DNMTis) upregulate immune attraction, including the interferon response, in solid tumors. We now define viral defense signaling as one mechanism for this. In epithelial ovarian cancer cells DNMTis upregulate viral defense by cytosolic sensing of double-stranded RNA (dsRNA), triggering a Type I Interferon response, upregulation of downstream interferon response genes, and increased apoptosis. Knockdown of the dsRNA sensors TLR3 and MAVS and inhibition of the interferon alpha/beta receptor blunt the DNMTi induced dsRNA response. DNMTis cause apoptosis of cancer cells, which is partially rescued by inhibiting the interferon alpha/beta receptor. We observe upregulation and demethylation of hypermethylated endogenous retroviruses (ERVs) and overexpression of individual ERVs whose sense and anti-sense transcripts may be key candidates for triggering the above signaling. Overexpression of ERVs alone is sufficient to trigger an interferon response in the absence of DNMTis. Basal levels of ERV and viral defense gene expression significantly correlate in primary OC and basal expression of the viral defense signature separates primary TCGA samples for multiple tumor types into low versus high expression groups. In melanoma patients treated with an immune checkpoint therapy, high viral defense signature expression in tumors significantly associates with durable clinical response and DNMTi treatment sensitizes to anti-CTLA4 therapy in a pre-clinical melanoma model. We thus define a major mechanism for how DNMTis may induce cancer cells to increase immune attraction and possibly sensitize patients to immunotherapy. Experiments determining which Aza-upregulated molecules on tumor cells are necessary for attraction and activation of host immune cells are ongoing.

Citation Format: Katherine B. Chiappinelli, Pamela L. Strissel, Alexis Desrichard, Huili Li, Christine Henke, Benjamin Akman, Alexander Hein, Neal S. Rote, Leslie M. Cope, Alexandra Snyder, Vladimir Makarov, Sadna Budhu, Jedd Wolchok, Cynthia A. Zahnow, Taha Mergoub, Timothy A. Chan, Reiner Strick, Stephen B. Baylin. Inhibiting DNA methylation causes an interferon response in cancer via dsRNA including endogenous retroviruses. [abstract]. In: Proceedings of the AACR Special Conference on Chromatin and Epigenetics in Cancer; Sep 24-27, 2015; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2016;76(2 Suppl):Abstract nr B32.

Inhibiting DNA Methylation Causes an Interferon Response in Cancer via dsRNA Including Endogenous Retroviruses

We show that DNA methyltransferase inhibitors (DNMTis) upregulate immune signaling in cancer through the viral defense pathway. In ovarian cancer (OC), DNMTis trigger cytosolic sensing of double-stranded RNA (dsRNA) causing a type I interferon response and apoptosis. Knocking down dsRNA sensors TLR3 and MAVS reduces this response 2-fold and blocking interferon beta or its receptor abrogates it. Upregulation of hypermethylated endogenous retrovirus (ERV) genes accompanies the response and ERV overexpression activates the response. Basal levels of ERV and viral defense gene expression significantly correlate in primary OC and the latter signature separates primary samples for multiple tumor types from The Cancer Genome Atlas into low versus high expression groups. In melanoma patients treated with an immune checkpoint therapy, high viral defense signature expression in tumors significantly associates with durable clinical response and DNMTi treatment sensitizes to anti-CTLA4 therapy in a pre-clinical melanoma model.

Abstract 4873: Epicopy: Measuring DNA copy number variation using high-density methylation microarrays

Background:

Cancers exploit changes in both genomic copy number and DNA methylation to promote growth and escape tumor-suppressor pathways. Under the 2-hit hypothesis, a single gene is likely to be altered by multiple mechanisms at the same time and integrated analysis can sharpen the focus on the most likely drivers. Requiring larger amounts of input material and imposing additional costs may limit multiplatform analysis, especially in studies using archival tissues with long clinical follow-up information where the nucleic acids are degraded and yields are generally lower than are obtained from fresh tissues. Taking advantage of similarities between methylation arrays and SNP arrays, we developed Epicopy, a robust computational method to identify DNA copy number variation (CNV) using high-density Illumina Human Methylation 450K methylation microarrays, thereby delivering two complementary genetic and epigenetic profiles from a single chip.

Methods:

Epicopy was developed using data from thyroid carcinoma samples arrayed by The Cancer Genome Atlas (TCGA) and subsequently validated on breast and lung small cell carcinoma TCGA datasets. Using Epicopy, we identified circumstances where CNV information can be reliably measured by methylation microarrays.

Results:

Using TCGA SNP microarrays as the gold standard to assess the performance of methylation derived CNV data from the thyroid, breast, and lung small cell carcinoma datasets, we showed that Epicopy is able to detect CNVs identified by SNP arrays at a sensitivity of 0.69 and specificity of 0.90. Frequently occurring CNVs identified using Genomic Identification of Significant Targets in Cancer (GISTIC), were identified with even higher accuracy.

Conclusion:

Epicopy provides a robust method to obtain both copy number and methylation information from a single methylation microarray experiment and will add value to methylation microarrays at no additional cost to the user. Tools to highlight regions of high sensitivity and specificity will also be provided to help users decide on the feasibility of using Epicopy to identify CNVs in regions of interest. Epicopy is implemented in the R statistical language and will be made available as a freestanding package as part of the Bioconductor bioinformatics software project.

Citation Format: Soonweng Cho, Hyun-seok Kim, Leslie M. Cope, Christopher B. Umbricht. Epicopy: Measuring DNA copy number variation using high-density methylation microarrays. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4873. doi:10.1158/1538-7445.AM2015-4873

Abnormal brain structure in youth who commit homicide

BACKGROUND

Violence that leads to homicide results in an extreme financial and emotional burden on society. Juveniles who commit homicide are often tried in adult court and typically spend the majority of their lives in prison. Despite the enormous costs associated with homicidal behavior, there have been no serious neuroscientific studies examining youth who commit homicide.

METHODS

Here we use neuroimaging and voxel-based morphometry to examine brain gray matter in incarcerated male adolescents who committed homicide (n = 20) compared with incarcerated offenders who did not commit homicide (n = 135). Two additional control groups were used to understand further the nature of gray matter differences: incarcerated offenders who did not commit homicide matched on important demographic and psychometric variables (n = 20) and healthy participants from the community (n = 21).

RESULTS

Compared with incarcerated adolescents who did not commit homicide (n = 135), incarcerated homicide offenders had reduced gray matter volumes in the medial and lateral temporal lobes, including the hippocampus and posterior insula. Feature selection and support vector machine learning classified offenders into the homicide and non-homicide groups with 81% overall accuracy.

CONCLUSIONS

Our results indicate that brain structural differences may help identify those at the highest risk for committing serious violent offenses.

Modular network construction using eQTL data: an analysis of computational costs and benefits

Background: In this paper, we consider analytic methods for the integrated analysis of genomic DNA variation and mRNA expression (also named as eQTL data), to discover genetic networks that are associated with a complex trait of interest. Our focus is the systematic evaluation of the trade-off between network size and network search efficiency in the construction of these networks.

Results: We developed a modular approach to network construction, building from smaller networks to larger ones, thereby reducing the search space while including more variables in the analysis. The goal is achieving a lower computational cost while maintaining high confidence in the resulting networks. As demonstrated in our simulation results, networks built in this way have low node/edge false discovery rate (FDR) and high edge sensitivity comparing to greedy search. We further demonstrate our method in a data set of cellular responses to two chemotherapeutic agents: docetaxel and 5-fluorouracil (5-FU), and identify biologically plausible networks that might describe resistances to these drugs.

Conclusion: In this study, we suggest that guided comprehensive searches for parsimonious networks should be considered as an alternative to greedy network searches.

Exploring massive, genome scale datasets with the GenometriCorr package

We have created a statistically grounded tool for determining the correlation of genomewide data with other datasets or known biological features, intended to guide biological exploration of high-dimensional datasets, rather than providing immediate answers. The software enables several biologically motivated approaches to these data and here we describe the rationale and implementation for each approach. Our models and statistics are implemented in an R package that efficiently calculates the spatial correlation between two sets of genomic intervals (data and/or annotated features), for use as a metric of functional interaction. The software handles any type of pointwise or interval data and instead of running analyses with predefined metrics, it computes the significance and direction of several types of spatial association; this is intended to suggest potentially relevant relationships between the datasets.

Availability and implementation: The package, GenometriCorr, can be freely downloaded at http://genometricorr.sourceforge.net/. Installation guidelines and examples are available from the sourceforge repository. The package is pending submission to Bioconductor.

Genome-wide methylation analysis identifies genes specific to breast cancer hormone receptor status and risk of recurrence

To better understand the biology of hormone receptor-positive and-negative breast cancer and to identify methylated gene markers of disease progression, we carried out a genome-wide methylation array analysis on 103 primary invasive breast cancers and 21 normal breast samples, using the Illumina Infinium HumanMethylation27 array that queried 27,578 CpG loci. Estrogen and/or progesterone receptor-positive tumors displayed more hypermethylated loci than estrogen receptor (ER)-negative tumors. However, the hypermethylated loci in ER-negative tumors were clustered closer to the transcriptional start site compared with ER-positive tumors. An ER-classifier set of CpG loci was identified, which independently partitioned primary tumors into ER subtypes. A total of 40 (32 novel and 8 previously known) CpG loci showed differential methylation specific to either ER-positive or ER-negative tumors. Each of the 40 ER subtype-specific loci was validated in silico, using an independent, publicly available methylome dataset from the Cancer Genome Atlas. In addition, we identified 100 methylated CpG loci that were significantly associated with disease progression; the majority of these loci were informative particularly in ER-negative breast cancer. Overall, the set was highly enriched in homeobox containing genes. This pilot study shows the robustness of the breast cancer methylome and illustrates its potential to stratify and reveal biological differences between ER subtypes of breast cancer. Furthermore, it defines candidate ER-specific markers and identifies potential markers predictive of outcome within ER subgroups.

Modeling liquid association

In 2002, Ker-Chau Li introduced the liquid association measure to characterize three-way interactions between genes, and developed a computationally efficient estimator that can be used to screen gene expression microarray data for such interactions. That study, and others published since then, have established the biological validity of the method, and clearly demonstrated it to be a useful tool for the analysis of genomic data sets. To build on this work, we have sought a parametric family of multivariate distributions with the flexibility to model the full range of trivariate dependencies encompassed by liquid association. Such a model could situate liquid association within a formal inferential theory. In this article, we describe such a family of distributions, a trivariate, conditional normal model having Gaussian univariate marginal distributions, and in fact including the trivariate Gaussian family as a special case. Perhaps the most interesting feature of the distribution is that the parameterization naturally parses the three-way dependence structure into a number of distinct, interpretable components. One of these components is very closely aligned to liquid association, and is developed as a measure we call modified liquid association. We develop two methods for estimating this quantity, and propose statistical tests for the existence of this type of dependence. We evaluate these inferential methods in a set of simulations and illustrate their use in the analysis of publicly available experimental data.

Identification of a CpG island methylator phenotype that defines a distinct subgroup of glioma

We have profiled promoter DNA methylation alterations in 272 glioblastoma tumors in the context of The Cancer Genome Atlas (TCGA). We found that a distinct subset of samples displays concerted hypermethylation at a large number of loci, indicating the existence of a glioma-CpG island methylator phenotype (G-CIMP). We validated G-CIMP in a set of non-TCGA glioblastomas and low-grade gliomas. G-CIMP tumors belong to the proneural subgroup, are more prevalent among lower-grade gliomas, display distinct copy-number alterations, and are tightly associated with IDH1 somatic mutations. Patients with G-CIMP tumors are younger at the time of diagnosis and experience significantly improved outcome. These findings identify G-CIMP as a distinct subset of human gliomas on molecular and clinical grounds.

Feature-level exploration of a published Affymetrix GeneChip control dataset

A comment on Preferred analysis methods for Affymetrix GeneChips revealed by a wholly defined control dataset by SE Choe, M Boutros, AM Michelson, GM Church and MS Halfon. Genome Biology 2005, 6:R16.

A benchmark for Affymetrix GeneChip expression measures

MOTIVATION

The defining feature of oligonucleotide expression arrays is the use of several probes to assay each targeted transcript. This is a bonanza for the statistical geneticist, who can create probeset summaries with specific characteristics. There are now several methods available for summarizing probe level data from the popular Affymetrix GeneChips, but it is difficult to identify the best method for a given inquiry.

RESULTS

We have developed a graphical tool to evaluate summaries of Affymetrix probe level data. Plots and summary statistics offer a picture of how an expression measure performs in several important areas. This picture facilitates the comparison of competing expression measures and the selection of methods suitable for a specific investigation. The key is a benchmark data set consisting of a dilution study and a spike-in study. Because the truth is known for these data, we can identify statistical features of the data for which the expected outcome is known in advance. Those features highlighted in our suite of graphs are justified by questions of biological interest and motivated by the presence of appropriate data.

Summaries of Affymetrix GeneChip probe level data

High density oligonucleotide array technology is widely used in many areas of biomedical research for quantitative and highly parallel measurements of gene expression. Affymetrix GeneChip arrays are the most popular. In this technology each gene is typically represented by a set of 11-20 pairs of probes. In order to obtain expression measures it is necessary to summarize the probe level data. Using two extensive spike-in studies and a dilution study, we developed a set of tools for assessing the effectiveness of expression measures. We found that the performance of the current version of the default expression measure provided by Affymetrix Microarray Suite can be significantly improved by the use of probe level summaries derived from empirically motivated statistical models. In particular, improvements in the ability to detect differentially expressed genes are demonstrated.

Can intraductal breast carcinoma be excised completely by local excision? Clinical and pathologic predictors

BACKGROUND

Microscopic evaluation of excised intraductal breast carcinoma (DCIS) specimens using a serial subgross technique reveals that in many patients the lesion is larger than expected, often making complete excision impossible with less than a true quadrantectomy. Data is presented on 181 patients with DCIS in whom the initial biopsy was performed using a more cosmetic wide local excision rather than a true quadrantectomy.

METHODS

Clear margins were defined as no tumor within 1 mm of any inked or dyed margin. All of these patients subsequently underwent mastectomy or reexcision of the initial biopsy site. This allowed pathologic evaluation for residual disease.

RESULTS

At mastectomy or reexcision, 76% of patients with initially involved margins had residual DCIS, as did 43% of patients with initially clear margins (P < 0.0001). Larger tumor size was a statistically significant predictor of initial margin involvement and residual DCIS (P < 0.05). Patients with comedo-DCIS had a greater tendency toward positive initial histologic margins and residual DCIS, but this trend was not statistically significant (P < 0.1).

CONCLUSION

DCIS presents major problems to both surgeons and pathologists. It is difficult to excise completely using a wide local excision. Histologically negative margins do not guarantee that residual DCIS has not been left behind. Inadequate excision of the primary lesions may be the most important cause of local failure after conservative treatment for intraductal breast carcinoma.