Sinnwell, JP

Abstract Breast cancer is a heterogeneous disease, and unsupervised clustering approaches using gene expression data have identified 3-6 distinct subtypes of triple negative breast cancer (TNBC). A genomically and clinically distinct subtype of TNBC is referred to as LAR (Luminal Androgen Receptor). Tumors with this subtype typically express high levels of the AR and exhibit alterations within genes involved in the PI3K pathway (e.g. PIK3CA mutations). Prospective studies are underway using drugs that target the AR alone or in combination with PI3K and CDK 4/6 inhibitors. Given the importance of accurately identifying this subtype, we sought to develop an online tool that uses submitted gene expression data to confidently characterize LAR samples by corroborating the classification with previously published clustering approaches. Methods: We have investigated TNBC RNA-Seq data from The Cancer Genome Atlas (TCGA) breast cancer study (N=123 samples) by cluster analysis. Analysis of the average silhouette width in both biased and unbiased K-means clustering approaches demonstrated LAR and basal as two distinct and significant clusters. A shrunken centroid model of 426 differentially expressed genes, named as CABAL (Clustering Among BAsal and Luminal androgen receptor), was constructed by comparing LAR and basal subtypes. Results: We applied the CABAL model to classify the four TNBC microarray datasets that were previously used in clustering experiments as well as an independent RNA-Seq data cohort. Non-negative matrix factorization (NMF) and fuzzy clustering were applied to the samples (N=1046). Clustering similarity among the methods was assessed with the adjusted rand index, and CABAL demonstrated significant similarity with both fuzzy and NMF clustering methods. Similarly, hierarchical clustering analysis performed on the pooled cohort of 1046 samples recapitulated the CABAL classification with an area under the receiver operating curve of 0.91. Conclusions: Confident and robust identification of samples with the LAR phenotype is paramount in the assessment of clinical associations and therapeutic efficacy. To facilitate LAR identification, we have provided a web-based prediction tool of the CABAL classification, integrated with the NMF and fuzzy clustering results to identify candidate LAR samples. The end user is provided with the pair-wise adjusted rand indexes, thus reinforcing in the clustering characterizations. Further, our online LAR depiction tool provides a set of graphical and tabular summaries, which will be illustrated, while providing additional molecular characterizations of the PAM50 and Metabric classifications. The availability of this tool could advance the genomic research and treatment of TNBC patients. Citation Format: Thompson KJ, Alaparthi T, Sinnwell JP, Carlson EE, Tang X, Bockol M, Vedell PT, Ingle JN, Suman V, Weinshilboum RM, Wang L, Boughey JC, Kalari KR, Goetz MP. Molecular subtyping of androgen receptor-positive patients using gene expression profiles [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 P1-03-04.

Abstract Introduction: The vast majority of cancer patients continue to receive treatments that are minimally informed by omics data. In the case of breast cancer, only ER and HER2 are routinely used for treatment selection. There is a particular need for personalized treatment in individuals with primary and secondary drug resistance or aggressive breast cancers. Emerging bioinformatics and statistical methods have made a fundamental impact on cancer research. However, challenges remains with regard to patient-centric data analysis and providing genomic data guidance to oncologists. There exists a large number of FDA approved anti-neoplastic drugs used to treat cancers other than breast and the development of innovative informatics methods and algorithms to repurpose those drugs should benefit breast cancer patients. Methods and Results: We have developed precision care systems (such as PANOPLY and CORPUS) to identify personalized therapies for an individual patient and to deliver genomic reports in a standard, searchable format so that a researcher or an oncologist can quickly navigate through molecular data and obtain prioritized drugs and targets.The PANOPLY (Precision cancer genomics report: single sample inventory) algorithm applies machine learning and topology-based network analysis methods to integrate multi-omics profiles and clinical data; individual-specific molecular alterations are identified and compared with a set of matched-controls having similar clinical data. Since there is a lack of a “gold standard” dataset to test such algorithms, we simulated 500 case-control sets and evaluated drug predictions across multiple simulation scenarios. We applied the PANOPLY algorithm to The Cancer Genome Atlas (TCGA) breast cancer cohort, which consists of multi-omics data and clinical data. In addition, PANOPLY was also applied to an in-house neoadjuvant breast cancer study (BEAUTY) that consists of multi-omics data, clinical data, and patient-derived xenografts (PDXs). In the TCGA breast cancer study we obtained survival data to determine the cases and matched-controls; and in the BEAUTY, we used pathologic complete response (pCR) as an outcome to determine responders and non-responders. Recurrent targetable alterations were not enriched in patients without pCR in the BEAUTY study. We have applied the PANOPLY to non-responder patients to identify individual specific alterations, dysregulated networks, drug targets, and drugs for each patient and stored them as case reports in CORPUS (Computational Oncology Reports and Precision therapeUticS), a web-based repository that allows clinicians to review genomic reports. Using comprehensive “omic” data derived from a triple negative breast cancer patient who had pre and post-neoadjuvant chemotherapy PDXs, PANOPLY prioritized the PARP inhibitors as the top class of drug. Using the PDX models available from this patient, we tested olaparib and confirmed the in vivo antitumor activity (more effective than vehicle with a p-value < 0.05 in the PDXs). Further studies to confirm PANOPLY findings are currently underway. Conclusions: In summary, the PANOPLY and CORPUS systems incorporate molecular data together with clinical data to provide genomic reports with proposed drug targets to advance or enable precision breast cancer care. Citation Format: Kalari KR, Sinnwell JP, Thompson KJ, Tang X, Carlson EE, Alaparthi T, Yu J, Vedell PT, Kalmbach MT, Bockol MA, Hossain A, Weinshilboum RM, Boughey JC, Wang L, Suman VJ, Goetz MP. Multiscale modeling of omics data for precision breast cancer treatment [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 P3-06-10.

Abstract Background: Tumor-infiltrating lymphocytes (TIL) have prognostic and potentially predictive significance in the (neo)adjuvant treatment of high-risk breast cancer. However, quantitative TIL measurement is not routinely performed. It is unclear why some tumors attract large quantities of TIL while others do not. We sought to confirm the association between TIL and pathologic complete response rate (pCR) and to further use next generation sequencing (NGS) to identify genes and gene pathways associated with the presence/absence of TIL. Methods: We studied 140 women with high risk stage I-III breast cancer, enrolled in the Breast Cancer Genome Guided Therapy Study (BEAUTY), obtaining serial biopsies for DNA/RNA sequencing and MRI imaging to assess response to neoadjuvant chemotherapy (NAC) with taxane (+/- trastuzumab+/-pertuzumab for HER2+ disease) followed by AC or (F)EC. Diagnostic pre-NAC core needle biopsies and surgical resection specimens post-NAC were available from 110 patients. Stromal TIL were semi-quantitated on a scale of 1-4 (with 1: ≤10/hpf, 2: subtle infiltrate >10/hpf, 3: moderate infiltrate readily visible at low power magnification, 4: dense infiltrate with innumerable lymphocytes). For this analysis, low TIL was defined as scores of 1-2 vs. high defined as 3-4. Using pre-NAC biopsies, RNAseq was performed using the Illumina HiSeq2000 and the Mayo Analysis Pipeline for RNAseq (MAP-Rseq) for quality control, sequence alignment, and gene counts. The quantity of TIL was associated with transcripts across the transcriptome after conditional quantile normalization. Differentially expressed genes were obtained using EdgeR analysis, using a false discovery rate of 0.05, and pathways were evaluated using GAGE methods. Results: The pCR and residual cancer burden (RCB)-0/I rates by stromal TIL status within each molecular subtype are presented in the table. A diverse spectrum of 1344 genes with differential expression between tumors with high vs. low stromal TIL was identified. The genes with >2.0-fold change (FC) and p<1e-09 included S100A7 (4.49 FC), LCN2 (2.48 FC), and ART3 (2.82 FC) (genes known to be involved in immune regulation), as well as TDRD1 (2.71 FC) (a gene related to ERG [ETS-related gene] expression). In addition, the "regulation of actin cytoskeleton" pathway was upregulated in tumors with high TIL, while the "Hedgehog signaling" and "Wnt signaling" pathways were downregulated. Molecular SubtypeStromal TILspCR rate n (%)RCB-0/I rateLuminal AHigh------Luminal ALow0/9 (0%)0/9 (0%)Luminal BHigh1/9 (11.1%)1/8 (12.5%)Luminal BLow3/24 (12.5%)6/23 (26.1%)ER+/HER2+High3/9 (33.3%)4/9 (44.4%)ER+/HER2+Low1/6 (16.7%)1/6 (16.7%)ER-/HER2+High8/9 (88.9%)7/7 (100%)ER-/HER2+Low4/8 (50.0%)6/8 (75.0%)Triple NegativeHigh10/19 (52.6%)13/19 (68.4%)Triple NegativeLow7/14 (50.0%)9/13 (69.2%) Conclusions: We identified genes and gene pathways associated with high TIL expression in breast tumors prior to NAC that provide insight into the interactions between TIL and tumors. TIL can be easily semi-quantitated on H&E and along with these novel biomarkers, may contribute to the personalization of breast cancer therapy. Citation Format: Moyer AM, Boughey JC, Kalari KR, Suman VJ, McLaughlin SA, Moreno-Aspitia A, Northfelt DW, Gray RJ, Sinnwell JP, Carlson EE, Dockter TJ, Jones KN, Felten SJ, Conners AL, Wieben ED, Ingle JN, Wang L, Weinshilboum RM, Visscher DW, Goetz MP. Differential mRNA expression patterns in breast tumors with high vs. low quantity of stromal tumor–Infiltrating lymphocytes. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P4-04-05.

Abstract Background: Tumorigenesis involves both genetic and epigenetic changes. Epigenetic alterations are reversible and are promising cancer therapeutic targets. Decitabine (5-aza-2'-deoxycytidine), a DNA methyltransferase inhibitor, is FDA approved for hematological malignancies. However, the effect of decitabine in breast cancer is not completely understood. Previous reports indicated that one decitabine mechanism involves regulation of protein levels for DNMT1, the major DNA methyltransferase that methylates hemimethylated CpG di-nucleotides in DNA. However, the E3 ligase involved in this process has not been identified. Whether decitabine also regulates DNMT3A and 3B in a similar fashion remains unclear. Therefore, our goals were to 1) understand mechanisms underlying decitabine action, 2) test the antitumor activity of decitabine in breast cancer models and 3) identify biomarkers associated with response to decitabine. Methods and Results: Western blots of breast cancer cell lines showed that DNMT1, DNMT3A, and DNMT3B protein levels decreased following decitabine treatment without a reduction in mRNA levels. Bioinformatic analysis of DNA methyltransferase sequences revealed a potential TRAF6 binding motif, and the interaction with TRAF6 (TNF receptor-associated factor 6) was confirmed by IP. TRAF6 functions as an E3 ligase. To determine whether TRAF6 might be the E3 ligase responsible for the degradation of DNMTs after decitabine treatment, we knocked down TRAF6 by RNA interference or knocked out the TRAF6 gene by CRISPR/Cas9. Down regulation of TRAF6 attenuated DNMT ubiquitination and increased DNMT protein levels, suggesting that TRAF6 might mediate proteasome-dependent degradation of all three DNMTs. This was further confirmed by reconstituting the knockout cells with WT and a TRAF6-C70A mutant, followed by assessing DNMT protein levels. Global DNA methylation was also increased after TRAF6 depletion and was confirmed in TRAF6 knock out cells in which DNMT levels were unaffected by decitabine. Cell cytotoxicity and colony forming assays showed that TRAF6 knockout cells were resistant to decitabine, suggesting that a major decitabine mechanism of action is through the regulation of TRAF6 which, in turn, degrades DNMTs, leading to decreased global methylation. Finally, decitabine significantly induced TRAF6 at both mRNA and protein levels, a process that might create positive feedback leading to increased degradation of DNMT proteins upon decitabine treatment. Based on these results, we further hypothesized that levels of the three DNMTs might influence decitabine response. Using 18 breast cancer patient derived xenograft (PDX) models, we found a wide range of DNMT protein levels regardless of ER/HER2 status. DNMT levels in the PDX models were directly associated with sensitivity to decitabine treatment, confirming our hypothesis. Conclusion: Our data showed that decitabine might be an effective agent for treating breast cancer and revealed a novel mechanism underlying decitabine treatment. Baseline DNMT protein levels may serve as a biomarker for predicting decitabine drug response. Citation Format: Yu J, Qin B, Boughey JC, Moyer AM, Visscher DW, Sinnwell JP, Yin P, Thompson KJ, Docter TJ, Kalari KR, Suman VJ, Wieben ED, Felten SJ, Conners AL, Jones KN, McLaughlin SA, Copland JA III, Moreno Aspitia A, Northfelt DW, Gray RJ, Ingle JN, Lou Z, Weinshilboum R, Goetz MP, Wang L. Regulation of DNA methyltransferases via TRAF6 determines breast cancer response to decitabine. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P3-07-51.

Abstract Background: Based upon the association between pathologic response and disease free survival, the neoadjuvant setting is increasingly being used for drug development. NGS has identified unique and recurrent genetic alterations in breast cancer (BC) that are potentially targetable; however, the clinical implications are mostly unknown. We developed a prospective neoadjuvant study (BEAUTY) in high risk BC patients (pts) using weekly T followed by anthracycline-based chemo wherein percutaneous tumor biopsies (PTB) are obtained before/during/after chemo for NGS and PDX. Our goal is to identify novel biomarkers/pathways and develop PDX to test new therapeutic approaches. Methods: Pts underwent PTB at baseline and after 12 wks of T. Response to T was defined based upon 12 week Ki-67: responder (<15%) vs non-responder (≥15%). Pts with histologic response and absence of invasive BC at 12 wks were classified as responders. NGS was performed using PTB/blood DNA (exome) and PTB (RNA seq). MRI response was classified using RECIST criteria. NGS data were used to identify somatic copy number variants (cnvs) and expressed single nucleotide variants (eSNVs). Non-SCID mice (estrogen supplemented) were implanted ≤ 30 minutes with PTB samples. Results: Of the first 78 pts, 44 have completed T. Here we focus on 18 pts with either triple negative or luminal B BC. Clinical characteristics according to Ki-67 response are shown in Table 1. Comparison of genomic alterations in BEAUTY pts with TCGA identified a greater overlap with copy number gains (73%) compared to deletions (40%), along with similar observations of mutations in TP53, PTEN, RYR2, and AKT1 genes. Association analysis of CNVs and eSNVs between responders/non-responders identified 33 genes (predominantly located in chromosomes 1, 8, 13) and 580 eSNVs (corresponding to 497 genes) with a p < 0.05. Differential gene expression (DGE) analysis of responders/non-responders identified 198 genes with a p-value < 0.05. Integrated analysis of 539 genes (CNVs, eSNVs and DGE) identified pathways such as TGF-beta, Jak-Stat, WNT and NOTCH signalling. PDX take rate was 44% [triple negative (6/10); Luminal B (2/8)]. PDX growth rate was significantly associated with clinical baseline Ki-67 (p = 0.00014). Conclusion: This is the first prospective study to demonstrate the feasibility of using PTB to obtain both NGS data and PDX in the neoadjuvant setting. PDX take rate is associated with BC subtype and baseline Ki-67. Studies are ongoing to 1) validate genes/pathways associated with treatment response in subsequent BEAUTY pts; 2) genomically characterize and assess PDX in vivo response to T and 3) Use NGS data to prioritize new drugs/drug combinations in PDX. Funded by Mayo Clinic Center for Individualized Medicine and MC Cancer Center. Clinical CharacteristicsOverallResponders: 12 week Ki-67 < 15% (n = 9)Non-Responders: 12 week Ki-67 ≥ 15% (n = 9)Median Age495345T stage T2/T314 (78%)7 (78%)7 (78%)Node Positive8 (44%)4 (44%)4 (44%)Triple negative10 (56%)6 (67%)4 (44%)Luminal B8 (44%)3 (33%)5 (56%)Ki-67 after 12 Weeks of T Median 5% (0-11%)Median 35% (17-60%)Complete/Partial MRI Response after T 6 (67%)2 (22%) Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-08-10.