Jennie Jeong

Whole Transcriptome RNA-Seq Analysis of Breast Cancer Recurrence Risk Using Formalin-Fixed Paraffin-Embedded Tumor Tissue

RNA biomarkers discovered by RT-PCR-based gene expression profiling of archival formalin-fixed paraffin-embedded (FFPE) tissue form the basis for widely used clinical diagnostic tests; however, RT-PCR is practically constrained in the number of transcripts that can be interrogated. We have developed and optimized RNA-Seq library chemistry as well as bioinformatics and biostatistical methods for whole transcriptome profiling from FFPE tissue. The chemistry accommodates low RNA inputs and sample multiplexing. These methods both enable rediscovery of RNA biomarkers for disease recurrence risk that were previously identified by RT-PCR analysis of a cohort of 136 patients, and also identify a high percentage of recurrence risk markers that were previously discovered using DNA microarrays in a separate cohort of patients, evidence that this RNA-Seq technology has sufficient precision and sensitivity for biomarker discovery. More than two thousand RNAs are strongly associated with breast cancer recurrence risk in the 136 patient cohort (FDR <10%). Many of these are intronic RNAs for which corresponding exons are not also associated with disease recurrence. A number of the RNAs associated with recurrence risk belong to novel RNA networks. It will be important to test the validity of these novel associations in whole transcriptome RNA-Seq screens of other breast cancer cohorts.

Fusion transcript discovery in formalin-fixed paraffin-embedded human breast cancer tissues reveals a link to tumor progression.

The identification of gene fusions promises to play an important role in personalized cancer treatment decisions. Many rare gene fusion events have been identified in fresh frozen solid tumors from common cancers employing next-generation sequencing technology. However the ability to detect transcripts from gene fusions in RNA isolated from formalin-fixed paraffin-embedded (FFPE) tumor tissues, which exist in very large sample repositories for which disease outcome is known, is still limited due to the low complexity of FFPE libraries and the lack of appropriate bioinformatics methods. We sought to develop a bioinformatics method, named gFuse, to detect fusion transcripts in FFPE tumor tissues. An integrated, cohort based strategy has been used in gFuse to examine single-end 50 base pair (bp) reads generated from FFPE RNA-Sequencing (RNA-Seq) datasets employing two breast cancer cohorts of 136 and 76 patients. In total, 118 fusion events were detected transcriptome-wide at base-pair resolution across the 212 samples. We selected 77 candidate fusions based on their biological relevance to cancer and supported 61% of these using TaqMan assays. Direct sequencing of 19 of the fusion sequences identified by TaqMan confirmed them. Three unique fused gene pairs were recurrent across the 212 patients with 6, 3, 2 individuals harboring these fusions respectively. We show here that a high frequency of fusion transcripts detected at the whole transcriptome level correlates with poor outcome (P<0.0005) in human breast cancer patients. This study demonstrates the ability to detect fusion transcripts as biomarkers from archival FFPE tissues, and the potential prognostic value of the fusion transcripts detected.

RT-PCR-based gene expression profiling for cancer biomarker discovery from fixed, paraffin-embedded tissues.

A molecular test providing clear identification of individuals at highest risk for developing metastatic disease from among early stage breast cancer patients has proven to be of great benefit in breast cancer treatment planning and therapeutic management. Patients with high risk of disease recurrence can also get an estimate of the magnitude of benefit to be gained by adding chemotherapy to surgery and hormonal therapy. Developing this clinical test was made possible by the availability of technologies capable of identifying molecular biomarkers from the gene expression profiles of preserved surgical specimens. Molecular tests such as the Oncotype DX(®) breast cancer test are proving to be more effective tools for individualized patient stratification and treatment planning than traditional methods such as patient demographic variables and histopathology indicators.Molecular biomarkers must be clinically validated before they can be effectively applied toward patient management in clinical practice. The most effective and efficient means of clinical validation is to use archived surgical specimens annotated with well-characterized clinical outcomes. However, carrying out this type of clinical study requires optimization of traditional molecular expression profiling techniques to analyze RNA from fixed, paraffin-embedded (FPE) tissues. In order to develop our clinically validated breast cancer assay, we modified molecular methods for RNA extraction, RNA quantitation, reverse transcription, and quantitative PCR to work optimally in archived clinical samples. Here, we present an updated description of current best practices for isolating both mRNA and microRNA from FPE tissues for RT-PCR-based expression profiling.

Rt-PCR gene expression profiling of RNA from paraffin-embedded tissues prepared using a range of different fixatives and conditions.

Although RNA is isolated from archival fixed tissues routinely for reverse transcription polymerase chain reaction (RT-PCR) and microarray analyses to identify biomarkers of cancer prognosis and therapeutic response prediction, the sensitivity of these molecular profiling methods to variability in pathology tissue processing has not been described in depth. As increasing numbers of expression analysis studies using fixed archival tumor specimens are reported, it is important to examine how dependent these results are on tissue-processing methods.We carried out a series of studies to systematically evaluate the effects of various tissue-fixation reagents and protocols on RNA quality and RT-PCR gene expression profiles. Human placenta was selected as a model specimen for these studies since it is relatively easily obtained and has proliferative and invasive qualities similar to solid tumors. In addition, each specimen is relatively homogeneous and large enough to provide sufficient tissue to systematically compare a range of fixation conditions and reagents, thereby avoiding the variability inherent in studying collections of tumor tissue specimens. Since anatomical pathology laboratories generally offer hundreds of different tissue-fixation protocols, we focused on fixation reagents and conditions used to process the most common solid tumors for primary cancer diagnosis. Fresh placentas donated under an IRB-approved protocol were collected at delivery and immediately submerged in cold saline for transport to a central pathology laboratory for processing. RNA was extracted from each specimen, quantified, and analyzed for size distribution and analytical performance using a panel of 24 RT-PCR gene expression assays. We found that different tissue-fixation reagents and tissue-processing conditions resulted in widely varying RNA extraction yields and extents of RNA fragmentation. However, the RNA extraction method and RT-PCR assays could be optimized to achieve successful gene expression analysis for nearly all fixation conditions represented in these studies.

Tumor marker discovery by expression profiling RNA from formalin fixed paraffin embedded tissues.

Clear identification among early-stage cancer patients of those at highest risk of having metastatic disease would be of great benefit in treatment planning and management. Considerable additional benefit would accrue to high-risk patients if their responses to specific therapeutic alternatives could be predicted. Molecular biomarkers in the form of gene expression profiles are proving to be more effective tools for both prognostic and predictive patient stratification than more traditional methods such as patient demographics and histopathology indicators. Such biomarkers must be clinically validated before they can be effectively used to manage patients in clinical studies or clinical practice. This can be most efficiently accomplished by analyzing archived clinical samples with well-characterized clinical outcomes. Doing studies of this type requires reoptimization of traditional molecular expression profiling techniques to analyze RNA from fixed paraffin-embedded tissues. We have modified molecular methods for RNA extraction, RNA quantification, reverse transcription, and quantitative PCR to work optimally in archived clinical samples in order to develop a clinically validated assay for breast cancer prognosis and prediction of patient response to hormonal and chemotherapy.

Abstract P4-02-08: Global quantitative measures using next-generation sequencing for breast cancer presence outperform individual tumor markers in plasma

Background: Analytically and clinically validated non-invasive blood tests that quantify breast cancer burden and clinical drug response/resistance are greatly needed. Many groups have successfully detected tumor markers in blood using a variety of technologies, including next generation sequencing (NGS). We performed a comprehensive NGS study on a small number of patients to evaluate the value of global versus individual markers for the quantitation of tumor-derived cell free DNA (cfDNA) in plasma. Methods: DNA isolated from formalin-fixed primary tumor, buffy coat cells, and plasma from 2 patients with metastatic breast cancer were characterized simultaneously for copy number aberrations (CNAs) and differentially methylated regions (DMRs) using whole genome bisulfite sequencing (WBGS), and targeted sequencing-based genotyping of 346 cancer-associated single nucleotide variations (SNVs). CNA and DMR regions were identified from log normalized, GC content corrected counts and DMR data using Poisson and binomial distribution theory and false discovery rate controlling methods. Percent tumor in cfDNA was estimated from the normalized ratio (plasma: primary tumor) of CNA or DMR compared to buffy coat, aggregating over genomic regions. Sample sets from 8 non-metastatic patients were also profiled using the targeted SNV panel in order to compare SNVs between samples and estimate percent tumor cfDNA. Results: WGBS detected tumor specific alterations in each primary tumor compared to buffy coat. By analyzing the genome using 100 Kb bins, we observed over 1000 bins with detectable CNA signal and, among 56 million CpG sites, over 30,000 DMRs. As expected, 5 or fewer informative somatic SNVs were detected in each patient. Analysis of these somatic changes in plasma revealed that the tumor fraction estimated from SNV detected in cfDNA varied widely between sites originally discovered in the patient’s primary tumor. In contrast, similar estimates of tumor fraction in cfDNA were obtained using CNA and DMR profiles within each patient; both methods yielded similar estimates of over 50% in one patient and less than 10% in the other. For the patient with high tumor fraction, both CNA and DMR profiles contained examples of individual large genomic regions that displayed additional clear aberrations in the plasma compared to the original tumor, such as a striking loss of a >25 Mb region of chromosome 4. Conclusions: Although individual somatic SNV in cfDNA can be detected in metastatic disease, calculated allelic fraction based on individual SNVs varies greatly within the same patient. Measuring and integrating CNA or DMR across the genome provided more consistent and reliable estimates of tumor DNA fraction in plasma, and also revealed alterations in plasma from patients with metastatic disease that were not prominent in the primary tumor. Citation Format: Ellen M Beasley, Richard D Abramson, Gregory E Alexander, David Chan, Kristen Bradley, Francois Collin, Michael Crager, Andrew Dei Rossi, Joseph Dorado, Adam Friedman, William J Gibb, Jennie Jeong, Col Jones, C J Ku, Yan Ma, John Morlan, Kunbin Qu, Aibing Rao, Aaron Scott, Haluk Tezcan. Global quantitative measures using next-generation sequencing for breast cancer presence outperform individual tumor markers in plasma [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P4-02-08.

Abstract 4859: Tumor and normal classification of formalin-fixed, paraffin-embedded (FFPE) specimens by transcriptome RNA-seq

We have used RNA-seq to profile and compare normal and cancerous human breast tissue. FFPE breast specimens from a total of 24 patients, 12 normal (N) and 12 tumor (T) specimens from surgical resections, were analyzed on an Illumina9s GA IIx sequencer. Whole transcriptome RNA-Seq libraries were prepared after depletion of ribosomal RNA by a protocol developed at Genomic Health Inc. (GHI). The analysis was multiplexed across two flow cells using barcoding, with two specimens per sequencing lane (1 T and 1 closely age-matched N library from a different patient). FFPE tissue archive times ranged from 10 to 13 years and they were also closely matched within each lane. To evaluate reproducibility, triplicate libraries were created from 4 of the specimens and analyzed within and across flow cells. Libraries yielded, on average, 19 million 51 bp sequences. R 2 values obtained from replicate libraries prepared from the same patient RNA were > 0.9 within and between flow cells. More than 80% of known genes in the human genome were detected in all patients. Several thousand intergenic transcripts were identified by an algorithm developed at GHI. A negative binomial model with tag-wise estimates of dispersion was applied to the known genes and intergenic regions. Inter-patient count variance is generally higher in the set of intergenic sequences than in the set of gene (RefSeq) sequences. Thousands of gene (RefSeq) and intergenic sequences were found to be differentially expressed between T and N tissues. We sought to build classifiers based on flow cell #1 data that could stratify T and N tissues when applied to flow cell #2 data. Sets of genes and intergenic regions were selected for analysis based on high inter-patient count variance. Support vector machine classifiers were trained and then applied to the data from flow cell #2, and also to another GHI tumor/normal RNA-Seq study. Either a set of 100 genes (RefSeq), or a set of 70 intergenic sequences accurately distinguished tumor and normal tissues. Our results offer further evidence of the potential of RNA-Seq for discovery of biomarkers. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4859. doi:10.1158/1538-7445.AM2011-4859

Abstract B48: Immunohistochemistry and RT-PCR evaluation of fixative effects in a model tissue system

The standard practice in hospital pathology laboratories is to preserve patient clinical tissue specimens as fixed, paraffin-embedded (FPE) tissue. FPE specimens are used for routine pathologic examination, immunohistochemistry (IHC) studies, and a variety of molecular diagnostic assays. The results of these studies assist in determining the patient9s clinical status and in therapeutic decision making. However, the methodology for tissue fixation is not standardized across laboratories and a number of different tissue fixatives are currently commercially available. The use of different tissue fixatives may significantly affect the performance of IHC and molecular diagnostic assays. The results of nine common tissue fixatives and their effects on both IHC- and RNA-based molecular assays are reported. Using human placenta as a model tissue system, nine common fixatives (B5, Bouin9s, ethyl alcohol 70%, formalin, Hollandes, Penfix, Prefer, Zenker9s, and zinc formalin) were compared for effects on six IHC assays and a panel of 42 gene targets by RT-PCR assays, as well as performance relative to fresh (RNAlater®) or frozen (OCT) unfixed tissue. The 42-gene panel assessed by RT-PCR included the six genes assessed by IHC. The IHC assays were scored using a semiquantitative method. For RT-PCR, raw assay scores were derived and subsequently normalized. Different fixatives resulted in varying effects on IHC and molecular assay performance. Per gene across each fixative, mRNA expression levels assessed by RT-PCR assays demonstrated wide variation, which could be largely corrected for by normalization. Variation in immunoreactivity as a function of tissue fixative was also observed with IHC assays. Compared to IHC, RT-PCR assays demonstrated greater sensitivity and were able to detect lower levels of gene expression, when the IHC assay gave negative results. Interestingly, fixative related effects were not always similar between IHC and RT-PCR assays. Therefore, it is recommended that the effects of tissue fixation be taken into consideration when performing data analysis and making comparisons between IHC and molecular diagnostic assays. Citation Information: Clin Cancer Res 2010;16(14 Suppl):B48.

Abstract 2160: New prognostic breast cancer biomarkers selected from formalin fixed tissue RNA samples after whole transcriptome amplification

RT-PCR-based gene expression profiling in archival FFPE tissues with associated clinical records is clinically important in oncology, as evidenced by the wide use of the breast cancer prognostic and predictive Oncotype DX ® 21-gene test 1 . The strength of evidence underlying this test relies on the use of landmark clinical trial patient cohorts and other valuable clinical specimens used for its development and validation. However, the limited amount of RNA available from these FFPE specimens restricts the number of candidate biomarkers that can be tested in discovery studies. To compensate for this limitation, we have developed a method to amplify FFPE RNA which preserves the pre-amplified RNA expression profiles. We have now applied this method to evaluate more than 300 previously untested candidate genes in two of the key clinical cohorts used to develop the Oncotype DX assay. In total, 214 patient RNA samples were amplified and their expression levels for new prognostic markers were analyzed. Genes discovered to be associated with prognosis in the original studies were included in the amplified RNA study as positive controls to confirm that gene expression profiles in the amplified RNA remained consistent with those in the original RNA extracts. 2,3 New genes were found that correlate with the risk of breast cancer recurrence across both clinical populations. A number of these genes co-express in a “metabolism-related” gene signature which includes ENO1, the gene encoding enolase 1. After stratifying the patients into ER+ and ER- cohorts, additional prognostic gene expression biomarkers were identified in the ER- patients that are consistent with a TGFbeta-related “stromal response” signature. Newly identified genes were confirmed for association with clinical outcome using publicly available microarray-based data sets. 4 Most of the newly identified genes and the previously validated biomarker genes were confirmed as being significantly associated with patient outcome using these data. It will be important to confirm these findings in additional separate patient cohorts. 1. Paik S, et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med 2004;351:2817 2. Cobleigh, Ml, etl. al.. (2005) Tumor gene expression predicts distant recurrence-free survival in breast cancer patients with 10 or more positive nodes: High throughput RT-PCR assay of paraffin-embedded tumor tissues. Clin Cancer Res 11: 8623 3. Cronin, M, et. al., (2004) Measurement of Gene Expression in Archival Paraffin-embedded Tissues: Development and Performance of a 92 Gene RT-PCR Assay. American Journal of Pathology, 164: 35 4. Wirapati, P., et. al., (2008) Meta-Anlayis of gene expression profiles in breast cancer: toward a unified understanding of breast cancer subtyping and prognosis signatures. Breast Cancer Research, 10: R65 (doi:10.1186/bcr2124) Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2160.