Citations

Abstract Purpose: To determine if the degree of estrogen suppression with aromatase inhibitors (AI: anastrozole, exemestane, letrozole) is associated with efficacy in early-stage breast cancer, and to examine for differences in the mechanism of action between the three AIs. Experimental Design: Matched case–control studies [247 matched sets from MA.27 (anastrozole vs. exemestane) and PreFace (letrozole) trials] were undertaken to assess whether estrone (E1) or estradiol (E2) concentrations after 6 months of adjuvant therapy were associated with risk of an early breast cancer event (EBCE). Preclinical laboratory studies included luciferase activity, cell proliferation, radio-labeled ligand estrogen receptor binding, surface plasmon resonance ligand receptor binding, and nuclear magnetic resonance assays. Results: Women with E1 ≥1.3 pg/mL and E2 ≥0.5 pg/mL after 6 months of AI treatment had a 2.2-fold increase in risk (P = 0.0005) of an EBCE, and in the anastrozole subgroup, the increase in risk of an EBCE was 3.0-fold (P = 0.001). Preclinical laboratory studies examined mechanisms of action in addition to aromatase inhibition and showed that only anastrozole could directly bind to estrogen receptor α (ERα), activate estrogen response element-dependent transcription, and stimulate growth of an aromatase-deficient CYP19A1−/− T47D breast cancer cell line. Conclusions: This matched case–control clinical study revealed that levels of estrone and estradiol above identified thresholds after 6 months of adjuvant anastrozole treatment were associated with increased risk of an EBCE. Preclinical laboratory studies revealed that anastrozole, but not exemestane or letrozole, is a ligand for ERα. These findings represent potential steps towards individualized anastrozole therapy.

Single nucleotide variants in the open reading frames (ORFs) of pharmacogenes are important causes of interindividual variability in drug response. The functional characterization of variants of unknown significance within ORFs remains a major challenge for pharmacogenomics. Deep mutational scanning (DMS) is a high‐throughput technique that makes it possible to analyze the functional effect of hundreds of variants in a parallel and scalable fashion. We adapted a “landing pad” DMS system to study the function of missense variants in the ORFs of cytochrome P450 family 2 subfamily C member 9 ( CYP2C9 ) and cytochrome P450 family 2 subfamily C member 19 ( CYP2C19 ). We studied 230 observed missense variants in the CYP2C9 and CYP2C19 ORFs and found that 19 of 109 CYP2C9 and 36 of 121 CYP2C19 variants displayed less than ~ 25% of the wild‐type protein expression, a level that may have clinical relevance. Our results support DMS as an efficient method for the identification of damaging ORF variants that might have potential clinical pharmacogenomic application.

Abstract Purpose: We hypothesized that integrated analysis of cancer types from different lineages would reveal novel molecularly defined subgroups with unique therapeutic vulnerabilities. On the basis of the molecular similarities between subgroups of breast and ovarian cancers, we analyzed these cancers as a single cohort to test our hypothesis. Experimental Design: Identification of transcriptional subgroups of cancers and drug sensitivity analyses were performed using mined data. Cell line sensitivity to Hsp90 inhibitors (Hsp90i) was tested in vitro. The ability of a transcriptional signature to predict Hsp90i sensitivity was validated using cell lines, and cell line- and patient-derived xenograft (PDX) models. Mechanisms of Hsp90i sensitivity were uncovered using immunoblot and RNAi. Results: Transcriptomic analyses of breast and ovarian cancer cell lines uncovered two mixed subgroups comprised primarily of triple-negative breast and multiple ovarian cancer subtypes. Drug sensitivity analyses revealed that cells of one mixed subgroup are significantly more sensitive to Hsp90i compared with cells from all other cancer lineages evaluated. A gene expression classifier was generated that predicted Hsp90i sensitivity in vitro, and in cell line- and PDXs. Cells from the Hsp90i-sensitive subgroup underwent apoptosis mediated by Hsp90i-induced upregulation of the proapoptotic proteins Bim and PUMA. Conclusions: Our findings identify Hsp90i as a potential therapeutic strategy for a transcriptionally defined subgroup of ovarian and breast cancers. This study demonstrates that gene expression profiles may be useful to identify therapeutic vulnerabilities in tumor types with limited targetable genetic alterations, and to identify molecularly definable cancer subgroups that transcend lineage.

Abstract Motivation Estimation of isoform-level gene expression from RNA-seq data depends on simplifying assumptions, such as uniform read distribution, that are easily violated in real data. Such violations typically lead to biased estimates. Most existing methods provide bias correction step(s), which is based on biological considerations—such as GC content—and applied in single samples separately. The main problem is that not all biases are known. Results We have developed a novel method called XAEM based on a more flexible and robust statistical model. Existing methods are essentially based on a linear model Xβ, where the design matrix X is known and is computed based on the simplifying assumptions. In contrast XAEM considers Xβ as a bilinear model with both X and β unknown. Joint estimation of X and β is made possible by a simultaneous analysis of multi-sample RNA-seq data. Compared to existing methods, XAEM automatically performs empirical correction of potentially unknown biases. We use an alternating expectation-maximization (AEM) algorithm, alternating between estimation of X and β. For speed XAEM utilizes quasi-mapping for read alignment, thus leading to a fast algorithm. Overall XAEM performs favorably compared to recent advanced methods. For simulated datasets, XAEM obtains higher accuracy for multiple-isoform genes. In a differential-expression analysis of a real single-cell RNA-seq dataset, XAEM achieves substantially better rediscovery rates in independent validation sets. Availability and implementation The method and pipeline are implemented as a tool and freely available for use at http://fafner.meb.ki.se/biostatwiki/xaem/. Supplementary information Supplementary data are available at Bioinformatics online.

Abstract Molecular and cellular diversity in cancer cells enables evolutionary processes that govern tumor progression and contributes to treatment failures, but the mechanisms governing this variability remain incompletely understood. We previously described tumor-specific epigenetic alterations in DNA devoid of canonical CpG methylation sites throughout the genomes of spontaneous breast tumors which arise in genetically uniform, yet heterozygous, [FVB X BALB/c MMTV-NeuT+] F1 mice. Those changes were detected in purified tumor DNA using a template-assisted quantitative ligation assay employing locus-specific and SNP-specific oligonucleotide genotyping probes, and the assay was subsequently shown to be sensitive to epigenetic modifications of the templates. We now describe losses of heterozygosity similarly distributed throughout the transcriptomes of tumors in this same model. We quantified expressed alleles across a population of clonal tumors and tested each expressed variant for the occurrence of allelic ratio outliers. The results revealed variegated patterns of expression in hundreds of genes, highly significantly enriched in pathways typically co-opted by tumors (Molecular Mechanisms of Cancer, P < 10-10, IPA analysis). An astounding 87% (2634 of 3044) of expressed genes that could be measured using this approach were represented in at least one tumor. Furthermore, the frequency of these outliers in any one individual tumor was found to correlate strongly with the transcriptional repression of an additional large set of non-polymorphic genes. These findings reveal underappreciated, latent mechanisms driving sporadic errors in epigenetic programming that promote repression of a multiple cis-linked transcripts. This genome-wide molecular chaos presents as dysregulated cellular homeostasis and is the foundation for heritable diversity as tumors evolve. Citation Format: Sara J. Felts, Xiaojia Tang, Virginia P. Van Keulen, Krishna R. Kalari, Larry R. Pease. Stochastic variation in gene expression is selected during clonal evolution of spontaneous mouse mammary tumors [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 2898.

e16559 Background: ARVs that develop early during treatment with abiraterone acetate/ prednisone (AA/P) may play a role in treatment resistance. We evaluated metastatic site and whole blood mRNA expression of full-length AR (AR-FL) and ARVs to characterize treatment-emergent ARVs (TE-ARVs) from men with mCRPC pre- and post- 12 weeks of AA/P collected in a prospective clinical trial (NCT#01953640). Methods: RNAseq performed on paired metastatic site biopsy (n = 40) and whole blood (n = 25) passed quality control. Reads were aligned to the GRCh38 reference genome with the spliced-alignment TopHat2 package. AR-FL or ARVs were detected if ≥ 2 splice reads aligned to unique splice junctions for AR-FL, AR-8, AR-45, AR-23, AR-V3, AR-V5V6, AR-V7, AR-V8, AR-V9, AR-V10, AR-V12, AR-V13, and AR-V14, and normalized to splice reads per million (SRPM). Cox proportional hazard regression analysis was performed on AR-FL and AR-Vs with ≥1 SRPM for association with time to treatment change (TTTC). Results: In metastatic site biopsies post-AA/P, the average number of splice reads was 27,376,541 (range 7,753,998 to 62,456,773). The median number of ARVs detected was 2 (range 0-8), with a total of total of 110 ARVs identified (Table). Dynamic shifts in ARV profiles were observed post-AAP, with 41 TE-ARVs identified in 17 unique patients. The most common TE-ARV was AR-8 (n = 8), followed by AR-45 (n = 5), AR-23 (n = 5), and AR-V7 (n = 4). The presence of AR-V7 post-AAP was adversely associated with TTTC (hazard ratio 2.46, p = 0.013). The identification of early TE- ARV was not associated with TTTC. In whole blood samples, post-AA/P detection of AR-FL was low (n = 3) and no ARVs were detected. Conclusions: No specific patterns were observed in ARV profiles obtained in the metastatic biopsies after 12 weeks of treatment with AA/P. Clinical trial information: 01953640. [Table: see text]

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 Background: The LAR subtype is a genomically distinct subset of TNBC. Using a large cohort of non-metastatic TNBC patients (pts) with long term follow-up, we sought to further characterize the clinicopathologic features and outcomes of LAR vs non-LAR TNBC. Methods: From a cohort of 9982 women with surgically-treated non-metastatic breast cancer, 605 met criteria for TNBC (ER/PR<1% and HER2-negative) by central pathology. RNA extracted from 304 FFPE tumor specimens using the HighPure RNA extraction kit was subjected to TruSeq RNA Access library preparation and sequencing on a HiSeq2500. Adequate RNA was available for 283 pts. Tumors were classified as LAR or non-LAR using a shrunken centroid model, CABAL (Clustering Among BAsal and Luminal androgen receptor). In addition to previously described analyses [Leon-Ferre et al, Breast Cancer Res Treat 2017], immunohistochemical (IHC) androgen receptor (AR) staining was performed and the impact of various parameters on invasive disease-free survival (IDFS) and overall survival (OS) was assessed using Cox proportional hazards models. Results: 58 (20%) tumors were classified as LAR and 225 (80%) as non-LAR. Compared to non-LAR, LAR pts were older (mean age 65 vs 54) and more often postmenopausal (79%vs53%), both p=0.01. Apocrine histology was more common among LAR tumors (21%vs0%), which were also lower grade (grade3: 69%vs95%) and had lower Ki-67 (Ki-67>15%: 64%vs82%), all p<0.01. Additionally, LAR tumors had lower median stromal tumor infiltrating lymphocytes (TILs, 20%vs25%) and were less frequently lymphocyte-predominant [≥50% stromal or intratumoral TILs (19%vs32%)], although neither reached statistical significance. AR IHC was available for 223 of 283 tumors. Median AR IHC score in LAR was 65% (range 0-100%) vs 0% (range 0-90%) in non-LAR. T/N stage, surgery type, and receipt of adjuvant chemotherapy (AdjCT) or radiotherapy were similar between LAR and non-LAR. LAR pts had shorter IDFS and OS compared to non-LAR (5.6 vs 11.8 yrs and 10.8 vs 20.8 yrs, respectively), although this did not reach statistical significance. Test of proportional hazard assumption was not significant for IDFS or OS (p = 0.30 and 0.09). IDFS estimates were numerically higher in LAR vs non-LAR (80.2%vs70.5%,p = 0.92) at 3yrs post-diagnosis; whereas the opposite was true (40.9%vs55.6%,p = 0.07) after 10yrs. OS estimates at 3 and 5yrs were similar between LAR and non-LAR, but at 10yrs OS was inferior in LAR (40.9%vs66.4%,p = 0.24). In a univariate analysis including both LAR and non-LAR, older age, higher N stage, lower TILs and absence of AdjCT were associated with poorer IDFS and OS. In a multivariate analysis, higher N stage and absence of AdjCT remained associated with both poorer IDFS and OS; while lower stromal TILs were associated with poorer IDFS (p=0.01), and with a trend towards poorer OS (p=0.07). Conclusions: LAR TNBCs occurred in older women, were lower grade, and had lower TIL density than nonLAR tumors. While significant differences in IDFS or OS were not demonstrated, LAR pts exhibited a numerically lower risk of a disease event at 3yrs, but higher risk by 10yrs compared to nonLAR pts. In the entire cohort, higher N stage, absence of AdjCT and lower TILs were independently associated with poorer outcomes. Citation Format: Leon-Ferre RA, Polley M-Y, Liu H, Kalari KR, Boughey JC, Liu MC, Cafourek V, Negron V, Ingle JN, Thompson KJ, Tang X, Barman P, Carlson E, Visscher DW, Carter JC, Couch FJ, Goetz MP. Characteristics, outcomes and prognostic factors of luminal androgen receptor (LAR) triple-negative breast cancer (TNBC) [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 P3-08-01.