Salendra Singh

Network-Based Enriched Gene Subnetwork Identification: A Game-Theoretic Approach

Identifying subsets of genes that jointly mediate cancer etiology, progression, or therapy response remains a challenging problem due to the complexity and heterogeneity in cancer biology, a problem further exacerbated by the relatively small number of cancer samples profiled as compared with the sheer number of potential molecular factors involved. Pure data-driven methods that merely rely on multiomics data have been successful in discovering potentially functional genes but suffer from high false-positive rates and tend to report subsets of genes whose biological interrelationships are unclear. Recently, integrative data-driven models have been developed to integrate multiomics data with signaling pathway networks in order to identify pathways associated with clinical or biological phenotypes. However, these approaches suffer from an important drawback of being restricted to previously discovered pathway structures and miss novel genomic interactions as well as potential crosstalk among the pathways. In this article, we propose a novel coalition-based game-theoretic approach to overcome the challenge of identifying biologically relevant gene subnetworks associated with disease phenotypes. The algorithm starts from a set of seed genes and traverses a protein–protein interaction network to identify modulated subnetworks. The optimal set of modulated subnetworks is identified using Shapley value that accounts for both individual and collective utility of the subnetwork of genes. The algorithm is applied to two illustrative applications, including the identification of subnetworks associated with (i) disease progression risk in response to platinum-based therapy in ovarian cancer and (ii) immune infiltration in triple-negative breast cancer. The results demonstrate an improved predictive power of the proposed method when compared with state-of-the-art feature selection methods, with the added advantage of identifying novel potentially functional gene subnetworks that may provide insights into the mechanisms underlying cancer progression.

Radiogenomic analysis of hypoxia pathway is predictive of overall survival in Glioblastoma

Hypoxia, a characteristic trait of Glioblastoma (GBM), is known to cause resistance to chemo-radiation treatment and is linked with poor survival. There is hence an urgent need to non-invasively characterize tumor hypoxia to improve GBM management. We hypothesized that (a) radiomic texture descriptors can capture tumor heterogeneity manifested as a result of molecular variations in tumor hypoxia, on routine treatment naïve MRI, and (b) these imaging based texture surrogate markers of hypoxia can discriminate GBM patients as short-term (STS), mid-term (MTS), and long-term survivors (LTS). 115 studies (33 STS, 41 MTS, 41 LTS) with gadolinium-enhanced T1-weighted MRI (Gd-T1w) and T2-weighted (T2w) and FLAIR MRI protocols and the corresponding RNA sequences were obtained. After expert segmentation of necrotic, enhancing, and edematous/nonenhancing tumor regions for every study, 30 radiomic texture descriptors were extracted from every region across every MRI protocol. Using the expression profile of 21 hypoxia-associated genes, a hypoxia enrichment score (HES) was obtained for the training cohort of 85 cases. Mutual information score was used to identify a subset of radiomic features that were most informative of HES within 3-fold cross-validation to categorize studies as STS, MTS, and LTS. When validated on an additional cohort of 30 studies (11 STS, 9 MTS, 10 LTS), our results revealed that the most discriminative features of HES were also able to distinguish STS from LTS (p = 0.003).

InFlo: a novel systems biology framework identifies cAMP-CREB1 axis as a key modulator of platinum resistance in ovarian cancer

Characterizing the complex interplay of cellular processes in cancer would enable the discovery of key mechanisms underlying its development and progression. Published approaches to decipher driver mechanisms do not explicitly model tissue-specific changes in pathway networks and the regulatory disruptions related to genomic aberrations in cancers. We therefore developed InFlo, a novel systems biology approach for characterizing complex biological processes using a unique multidimensional framework integrating transcriptomic, genomic and/or epigenomic profiles for any given cancer sample. We show that InFlo robustly characterizes tissue-specific differences in activities of signalling networks on a genome scale using unique probabilistic models of molecular interactions on a per-sample basis. Using large-scale multi-omics cancer datasets, we show that InFlo exhibits higher sensitivity and specificity in detecting pathway networks associated with specific disease states when compared to published pathway network modelling approaches. Furthermore, InFlo’s ability to infer the activity of unmeasured signalling network components was also validated using orthogonal gene expression signatures. We then evaluated multi-omics profiles of primary high-grade serous ovarian cancer tumours (N=357) to delineate mechanisms underlying resistance to frontline platinum-based chemotherapy. InFlo was the only algorithm to identify hyperactivation of the cAMP-CREB1 axis as a key mechanism associated with resistance to platinum-based therapy, a finding that we subsequently experimentally validated. We confirmed that inhibition of CREB1 phosphorylation potently sensitized resistant cells to platinum therapy and was effective in killing ovarian cancer stem cells that contribute to both platinum-resistance and tumour recurrence. Thus, we propose InFlo to be a scalable and widely applicable and robust integrative network modelling framework for the discovery of evidence-based biomarkers and therapeutic targets.

Mitotic vulnerability in triple-negative breast cancer associated with LIN9 is targetable with BET inhibitors

Triple-negative breast cancers (TNBC) are highly aggressive, lack FDA-approved targeted therapies, and frequently recur, making the discovery of novel therapeutic targets for this disease imperative. Our previous analysis of the molecular mechanisms of action of bromodomain and extraterminal protein inhibitors (BETi) in TNBC revealed these drugs cause multinucleation, indicating BET proteins are essential for efficient mitosis and cytokinesis. Here, using live cell imaging, we show that BET inhibition prolonged mitotic progression and induced mitotic cell death, both of which are indicative of mitotic catastrophe. Mechanistically, the mitosis regulator LIN9 was a direct target of BET proteins that mediated the effects of BET proteins on mitosis in TNBC. Although BETi have been proposed to function by dismantling super-enhancers (SE), the LIN9 gene lacks an SE but was amplified or overexpressed in the majority of TNBCs. In addition, its mRNA expression predicted poor outcome across breast cancer subtypes. Together, these results provide a mechanism for cancer selectivity of BETi that extends beyond modulation of SE-associated genes and suggest that cancers dependent upon LIN9 overexpression may be particularly vulnerable to BETi. Cancer Res; 77(19); 5395-408. ©2017 AACR.

Abstract P4-02-07: Radiogenomic analysis of HER2+ breast cancer reveals MRI features correlated with genomic immune index are predictive of neoadjuvant chemotherapy response

Background: It has been previously shown that computer-extracted heterogeneity features calculated within the peritumoral region of a breast cancer from baseline dynamic contrast-enhanced MRI [DCE-MRI] can predict treatment outcome. This approach may due to microenvironmental signatures of elevated immune response, a predictor of favorable response to neoadjuvant chemotherapy [NAC] in HER2+ breast cancer. We assessed the correlation between peritumoral radiomic features and a tissue-derived genomic immune index [II] in HER2+ breast cancer and whether these immune-correlated imaging features are predictive of pathologic response. Methods: 33 HER2+ patients with both 1.5 or 3 T DCE-MRI imaging and targeted RNA sequencing of biopsy samples collected prior to NAC from a multicenter trial [BrUOG 211B, n=26] and The Cancer Genome Atlas-Breast Cancer project [TCGA-BRCA, n=7] were retrospectively analyzed. II was derived from expression of a 140-gene immune signature using the ESTIMATE algorithm. An attending breast radiologist annotated lesion boundaries on the DCE-MRI phase of peak contrast enhancement. Beyond this intratumoral region, 5 annular peritumoral regions in 3 mm increments out to a maximum radius of 15 mm were analyzed. Computer-extracted heterogeneity descriptors computed within the intratumoral and peritumoral regions were summarized by first order statistics. Redundancy was reduced by eliminating correlated imaging features (R2>.6). From the remaining features, the 5 features that were collectively best correlated with II were selected by feed forward, leave-one-out multilinear regression. The regression model was applied to an independent test set of 28 HER2+ patients with post NAC surgical specimens. The estimated II was assessed for its ability to differentiate patients who achieved a pathologic complete response in the breast [pCR, ypT0/is] (n=16) and those who did not (n=12) by 2-sided Wilcoxon rank sum test of median and area under the receiver operating characteristic curve (AUC). Results: The set of top features that significantly correlated (p Conclusions: From a set of quantitative features characterizing heterogeneity within the peritumoral region on DCE-MRI, we identified peritumoral imaging features correlated with a genomic index of immune response in HER2+ breast cancer and were predictive of pathologic response in an independent testing set. Our findings suggest that the predictive capability of peritumoral radiomics may be tied to a patient9s immune response to the cancer. In addition to providing insight to the biological basis of peritumoral radiomics, imaging signatures of immune response themselves possess clinical value as a potential means for the non-invasive prediction of HER2+ cancer biology and treatment outcome. Additional independent validation is needed on a larger test set to confirm our preliminary findings. Citation Format: Braman N, Prasanna P, Singh S, Beig N, Gilmore H, Etesami M, Bates D, Gallagher K, Bloch BN, Somlo G, Sikov W, Harris L, Plecha D, Varadan V, Madabhushi A. Radiogenomic analysis of HER2+ breast cancer reveals MRI features correlated with genomic immune index are predictive of neoadjuvant chemotherapy response [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 P4-02-07.

Abstract 419: Integrative analysis of multi-omics tumor profiles identifies pathways associated with resistance to anti-HER2 therapy in early stage breast cancer

Background: HER2 positive breast cancers are heterogeneous at both the clinical and molecular levels, with the HER2-enriched subtype exhibiting increased levels of immune infiltration signatures and the highest rate of pathologic complete response (pCR), while the HER2-Basal subtype is resistant to anti-HER2 therapy (Varadan et al, CCR 2016). Here we aim to characterize the molecular underpinnings of response and their interaction with increased immune infiltration across these HER2-subtypes using integrative analyses of genomic and transcriptomic data from two multicenter trials (DFCI 03-311 and BrUOG 211B). Methods: Fresh tumor core biopsies were taken at baseline and a 2-week time point after a single dose of trastuzumab. 80 HER2+ early breast cancer (EBC) patients were enrolled in the 03-311 trial, and 60 patients in the 211B trial. Biopsy samples were profiled for gene expression (Microrrays:03-311; RNAseq:211B), somatic mutations (Whole-exome sequencing: 211B; Targeted Sequencing: 03-311) and somatic copy-number aberrations (SNP-arrays: 03-311; Whole-exome Sequencing: 211B). Subtyping was performed using gene expression data and tumors were classified into HER2-Enriched, HER2-Luminal and HER2-Basal subtypes. Integrative analysis of gene expression and copy-number data to infer signaling network activities per sample, was performed using the recently developed InFlo framework (Dimitrova et al, Oncogene 2016). Results: HER2-Basal tumors exhibited lower average copy number for HER2 and were less likely to have high-level amplifications of co-amplicons (e.g. 11q13, 20q13). In the 211B and 03-311 trials, respectively, 62% and 63% of somatic mutations persisted after one dose of therapy, while 21% and 19% of mutations were undetectable after one dose of therapy. Tumors harboring amplifications in the 8p11 (FGFR1) genomic locus exhibited higher indices immune signatures associated with macrophages (P=0.0073) and T-cells (P = 0.0493) in 211B, but this association did not achieve significance in the 03-311 trial. Integrative InFlo-based analysis of tumor gene expression and copy-number profiles after one dose of trastuzumab in the 211B trial revealed significantly higher activity of signaling pathways associated with CD4+ T-cells in the responders (P=0.008), while higher activity of mTOR pathway was observed in non-responders (P=0.0014). Conclusions: Changes in mutational profiles over time may either be related to therapy-induced alterations of clonal architecture or the consequence of intra-tumor heterogeneity, thus warranting further exploration. Integrative analysis of gene expression and copy-number profiles reveal signaling pathways associated with response and resistance, enabling the discovery of biomarkers of response to anti-HER2 therapy. Citation Format: Salendra Singh, Hannah Gilmore, Maysa Abu-Khalaf, George Somlo, William Sikov, Lyndsay Harris, Vinay Varadan. Integrative analysis of multi-omics tumor profiles identifies pathways associated with resistance to anti-HER2 therapy in early stage breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 419. doi:10.1158/1538-7445.AM2017-419

Abstract P1-05-09: Association of co-amplicons with immune infiltration in subtypes of HER2-Positive breast cancer

Background : HER2+ breast cancers are heterogeneous at both clinical and molecular levels. We and others have determined that the HER2-Enriched subtype exhibits the highest rate of pathologic complete response (pCR) to neoadjuvant chemotherapy and trastuzumab (T), while the HER2-Basal subtype is resistant to anti-HER2 therapy (Carey et al, JCO 2015;Varadan et al, CCR 2016). Additionally, we reported that signatures of immune cell infiltration and immune cell subsets evaluated after one dose of T can predict pCR to preoperative T and chemotherapy (Varadan et al, CCR 2016). Given recent evidence for improved immune response with increasing mutational load, we chose to characterize the association of somatic mutations and copy-number alterations with subtypes of HER2+ breast cancer and immune modulation after one dose of T. Methods: Fresh tumor core biopsies were taken at baseline and 2 weeks after one dose of either T or nab-paclitaxel (N) from 60 patients with stage II-III HER2+ cancers enrolled on a multicenter trial (BrUOG 211B). All patients then received 18 weeks of T+N+carboplatin. PAM50 subtyping was performed using gene expression data from patient tumor biopsies and tumors were classified into HER2-Enriched, HER2-Luminal and HER2-Basal subtypes. Whole-exome sequencing (WES) was performed on a total of 86 samples (49 baseline, 37 brief-exposure), sequenced at an average depth of 90X. Somatic mutations were detected by applying multiple mutation-detection algorithms on the WES data, followed by stringent quality control using public and in-house variant databases, and mutation data curated from 11,000 tumors sequenced by the TCGA. Somatic copy-number alterations were estimated using a published algorithm, ENVE (Varadan et al, Genome Med 2015) that robustly detects somatic copy-number alterations in WES tumor profiles. We employed previously defined gene-expression signatures (Varadan et al, CCR 2016) of total immune infiltration and immune cell subsets, to assess for association with genomic aberrations. Results : HER2-Basal tumors exhibited lower average copy number for HER2 and were less likely to have high-level amplifications of co-amplicons (e.g. 11q13, 20q13) with the exception of the MYC amplicon (8q24). They also exhibited a non-significant (P=0.33) trend towards higher mutational burden (Avg=85) compared to HER2-Luminals (Avg=79). A majority of somatic mutations (62%, 2282/3666) persisted after a single-dose of either T or N, while 17% (624/3666) were not detectable after brief-exposure. There was no association between immune infiltration and mutational burden in any HER2 subtype. Tumors harboring FGFR1 (8p11) amplifications exhibited higher gene-signature levels for macrophages ( P =0.0073) and T-cells ( P =0.0493) but not B-cells ( P =0.213). Conclusions: The HER2-Basal subtype is less likely to respond to trastuzumab-based neoadjuvant therapy and exhibits lower numbers of common amplicons. The disappearance of mutations after brief-exposure to therapy may be due to either tumor heterogeneity/sampling or clonal selection. The association of 8p11 amplifications with increased T-cell infiltration suggests that this amplicon may play an immunogenic role in HER2+ breast cancer. These results warrant further investigation in larger cohorts. Citation Format: Singh S, Gilmore H, Somlo G, Abu-Khalaf M, Sikov W, Harris L, Varadan V. Association of co-amplicons with immune infiltration in subtypes of HER2-Positive breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P1-05-09.

Abstract 304: HER2 basal tumors have frequent mutation of HER3 and are more resistant to HER2-targeted therapy

Background: HER2 tumors are heterogeneous at both the clinical and molecular levels (Harris et al 2007, Varadan, AACR 2015). We and others have identified a distinct molecular subtype termed ‘HER2 basal’ that demonstrates lower response to preoperative trastuzumab and is characterized by overexpression of the ‘basal-like’ signature genes (Harris, CCR 2007; Carey, JCO 2015). We sought to better understand the molecular underpinnings of the HER2 basal phenotype by performing whole exome sequencing (WES) in a cohort of HER2 tumors treated with trastuzumab-containing preoperative therapy. Methods: A multicenter trial (BrUOG 211B) was conducted to determine predictors of pathologic complete response (pCR) to trastuzumab (T)-containing preoperative therapy. Sixty HER2+ Stage II-III patients were enrolled and assigned to either a single dose of T or nab-paclitaxel (N) followed by T+N+carboplatin (TNC). Fresh tumor core biopsies were taken at baseline and 2 weeks after the initial dose of T or N. PAM50 subtyping was performed using gene expression data from patient tumor biopsies and tumors were classified into HER2-Enriched, HER2-Luminal and HER2-Basal subtypes based on ER/PR IHC values and relative expression of the proliferation-associated genes within the PAM50 gene list. WES was performed on a total of 86 samples (49 baseline and 37 from the post brief-exposure timepoint), with each sample sequenced at an average depth of 90X. Somatic mutations (single nucleotide variants and short indels) were assessed using Genome Analysis Toolkit (GATK) UnifiedGenotyper followed by a robust pipeline comparing multiple databases to eliminate germline variants. Copy-number estimation was performed on the WES data using a novel algorithm, ‘ENVE’ (Varadan, Genome Med, 2015) that robustly identifies somatic copy-number alterations. Results: WES profiles of HER2 basal tumors (n = 6) revealed lower average copy number for HER2 (2.5 vs 5.8) and were less likely to have high-level amplification (>4 copies per cell) of other amplicons (e.g. 11q13, 20q13) with the exception of 8q24. The so-called ‘Myc’ amplicon was co-amplified in all subtypes relatively equally (3/6 in HER2-B, 11/19 in HER2-E, 10/23 in HER2-L). HER2-B tumors were also characterized by higher frequency of mutations in Rb (2/6 versus 1/42; P = 0.0376) and HER3 (4/6 vs 2/42; P = 0.0011) compared with other subtypes and frequent alteration of p53 (4/6 versus 17/42; P = 0.22). pCR to preop TNC was HER2-B 3/8, HER2-E 10/20, HER2-L 6/21, not significant in this small cohort. The presence of HER3 mutations suggests a potential mechanism for the previously reported poor clinical response to preoperative anti-HER2 therapy in HER2 basal tumors. Conclusions: The HER2 basal subtype is more likely to exhibit mutations in HER3 and is less likely to respond to trastuzumab-based therapy. Larger cohorts are required to confirm these findings and trials of other HER2 targeted agents in this subtype should be considered. Citation Format: Vinay Varadan, Salendra Singh, Hannah Gilmore, Shikha Parsai, George Somlo, Maysa Abu-Khalaf, William Sikov, Lyndsay N. Harris. HER2 basal tumors have frequent mutation of HER3 and are more resistant to HER2-targeted therapy. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 304.

ENVE: a novel computational framework characterizes copy-number mutational landscapes in colorectal cancers from African American patients

Reliable detection of somatic copy-number alterations (sCNAs) in tumors using whole-exome sequencing (WES) remains challenging owing to technical (inherent noise) and sample-associated variability in WES data. We present a novel computational framework, ENVE, which models inherent noise in any WES dataset, enabling robust detection of sCNAs across WES platforms. ENVE achieved high concordance with orthogonal sCNA assessments across two colorectal cancer (CRC) WES datasets, and consistently outperformed a best-in-class algorithm, Control-FREEC. We subsequently used ENVE to characterize global sCNA landscapes in African American CRCs, identifying genomic aberrations potentially associated with CRC pathogenesis in this population. ENVE is downloadable at https://github.com/ENVE-Tools/ENVE.