The NIH Common Fund Data Ecosystem (CFDE) integrates data resources from 18 NIH Common Fund programs for discovery and integrative analysis. These programs generate valuable but heterogeneous datasets that can be difficult to discover, access, and reuse. CFDE aims to provide a collaborative, community-built infrastructure that links and enriches Common Fund programs. We describe the evolution, structure, and core technologies of CFDE, including practical approaches that support submission, integration, visualization, and public release of multimodal data. Training programs and workforce initiatives lower barriers to adoption. CFDE has devised solutions to critical issues facing cross-program initiatives, including data scale and heterogeneity, dataset integration, and long-term sustainability. We demonstrate the utility of linking Common Fund resources through integrative tools and cross-dataset queries to yield insights that would otherwise be infeasible. Collectively, CFDE shows that a standards-driven, federated approach enhances and unifies cross-disciplinary resources, fostering collaboration and data-driven discovery.

Combination chemotherapy remains essential for clinical management of triple-negative breast cancer (TNBC), making it impossible to assess responses to multiple agents in a single patient. Herein, we conduct multi-omic analyses of TNBC patient-derived xenografts (PDXs) treated with single agent carboplatin and docetaxel, or the combination, to identify candidate mechanisms of resistance, as well as predictive biomarkers to individual treatments, and to develop therapeutic strategies to overcome resistance. Genomic, transcriptomic, and proteomic profiles of baseline tumors from 50 TNBC PDXs were associated with responses to human equivalent doses of either single agent carboplatin, docetaxel, and the combination. Integration of external TNBC PDX and clinical datasets was performed using ComBat. Protein Marker Selection (ProMS) tool was used to integrate both RNA and protein data to select 5-10 RNA feature combinations for optimized prediction of chemotherapy response in a logistic regression model. Combination responses were generally no better than the best single agent, with enhanced response in only ∼13% of PDX, and apparent antagonism in a comparable percentage. Single ome comparisons showed largely non-overlapping results between genes associated with single agent and combination treatment that validated in independent patient cohorts. Multi-omic analyses of PDXs identified agent-specific treatment-associated biomarkers and predictive biomarker combinations. Notably, integrating proteomic with mRNA data improved machine learning models that achieved AUROC performance of 0.85 to predict pathologic complete response to combination chemotherapy. These models await further evaluation in datasets that may become available, including data from the BEAUTY (NCT02022202), TBCRC 030 (NCT01982448), and RESPONSE (NCT05020860) clinical trials, or in future experimental settings using PDX/PDX organoids. In PDXs responsive to any treatment, several basal cytokeratins were among the top upregulated proteins compared to PDXs resistant to all treatments. KRT5 was validated by IHC in PDX tumors, achieving an AUROC of 0.83 for predicting responsiveness, indicating its potential as a future chemoresponse marker in TNBC. PDXs refractory to all treatments showed dysregulated mitochondrial function. Treatment with romidepsin, an HDAC inhibitor that targets this process indirectly, increased DNA damage, and enhanced carboplatin response, in a chemoresistant PDX model with high abundance of HDAC proteins. Multi-omic characterization identifies molecular mechanisms and predictive biomarkers for stratifying TNBC tumors for single or combination chemotherapy treatments, suggests targeted therapies to augment chemotherapy response, and provides a valuable resource for researchers and clinicians. Jonathan T. Lei, Lacey E. Dobrolecki, Chen Huang, Ramakrishnan R. Srinivasan, Suhas V. Vasaikar, Alaina N. Lewis, Christina Sallas, Na Zhao, Jin Cao, John D. Landua, Chang I. Moon, Yuxing Liao, Susan G. Hilsenbeck, C K. Osborne, Mothaffar F. Rimawi, Matthew J. Ellis, Varduhi Petrosyan, Bo Wen, Kai Li, Alexander B. Saltzman, Antrix Jain, Anna Malovannaya, Gerburg M. Wulf, Elisabetta Marangoni, Shunqiang Li, Daniel C. Kraushaar, Tao Wang, Senthil Damodaran, Xiaofeng Zheng, Funda Meric-Bernstam, Gloria V. Echeverria, Meenakshi Anurag, Xi Chen, Bryan E. Welm, Alana L. Welm, Bing Zhang, Michael T. Lewis. Patient-derived xenografts allow deconvolution and prediction of chemotherapy responses [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 739.

The Playbook Workflow Builder (PWB) is a web-based platform to dynamically construct and execute bioinformatics workflows by utilizing a growing network of input datasets, semantically annotated API endpoints, and data visualization tools contributed by an ecosystem of collaborators. Via a user-friendly user interface, workflows can be constructed from contributed building-blocks without technical expertise. The output of each step of the workflow is added into reports containing textual descriptions, figures, tables, and references. To construct workflows, users can click on cards that represent each step in a workflow, or construct workflows via a chat interface that is assisted by a large language model (LLM). Completed workflows are compatible with Common Workflow Language (CWL) and can be published as research publications, slideshows, and posters. To demonstrate how the PWB generates meaningful hypotheses that draw knowledge from across multiple resources, we present several use cases. For example, one of these use cases prioritizes drug targets for individual cancer patients using data from the NIH Common Fund programs GTEx, LINCS, Metabolomics, GlyGen, and ExRNA. The workflows created with PWB can be repurposed to tackle similar use cases using different inputs. The PWB platform is available from: https://playbook-workflow-builder.cloud/.

Combination chemotherapy remains essential for clinical management of triple-negative breast cancer (TNBC). Consequently, responses to multiple single agents cannot be delineated at the single patient level, even though some patients might not require all drugs in the combination. Herein, we conduct multi-omic analyses of orthotopic TNBC patient-derived xenografts (PDXs) treated with single agent carboplatin, docetaxel, or the combination. Combination responses were usually no better than the best single agent, with enhanced response in only ~13% of PDX, and apparent antagonism in a comparable percentage. Single-omic comparisons showed largely non-overlapping results between genes associated with single agent and combination treatments that could be validated in independent patient cohorts. Multi-omic analyses of PDXs identified agent-specific biomarkers/biomarker combinations, nominating high Cytokeratin-5 (KRT5) as a general marker of responsiveness. Notably, integrating proteomic with transcriptomic data improved predictive modeling of pathologic complete response to combination chemotherapy. PDXs refractory to all treatments were enriched for signatures of dysregulated mitochondrial function. Targeting this process indirectly in a PDX with HDAC inhibition plus chemotherapy in vivo overcomes chemoresistance. These results suggest possible resistance mechanisms and therapeutic strategies in TNBC to overcome chemoresistance, and potentially allow optimization of chemotherapeutic regimens.

Introduction: Triple-negative breast cancer (TNBC) patients frequently receive combination chemotherapy treatment, including most recently taxane/platinum combinations. However, treatment is not biomarker-guided. As such, it is not known which patient will respond to one or the other agent, or indeed which patients actually require the combination for the most effective treatment. We hypothesized that optimization of chemotherapy may be possible if molecular mechanisms and biomarkers underlying response to individual treatments can be identified. We evaluated this hypothesis in a preclinical trial using a cohort of 50 patient-derived xenograft (PDX) models of TNBC treated with either single-agent docetaxel or carboplatin, or their combination. Methods: 50 TNBC PDXs were evaluated for response to four weekly treatments with either single agent docetaxel (20 mg/kg), or carboplatin (50 mg/kg); 42 of these were also treated with their combination. Multi-omics profiling (genomics, transcriptomics, proteomics) was conducted before treatment. Gene level associations by treatment type were used to construct consensus gene sets by integrating our data with external TNBC PDX and patient cohorts which were then used to build XGBoost models to predict treatment-specific responses. Pathway analysis was performed using signed -log10 p-values from gene level results as input for Gene Set Enrichment Analysis. Results: Direct comparison of responses to carboplatin, docetaxel, and their combination showed that combination treatment was largely ineffective at generating enhanced responses over the best single agent. Only 13% of the 42 PDX showed enhanced responses in combination, with a comparable percentage (12%) showing antagonism between docetaxel and carboplatin, a phenomenon observed previously in vitro. Proteogenomic profiles revealed distinct genes associated with responses to each agent and their combination, suggesting different molecular mechanisms underlying each treatment response. A substantial number of genes linked to single-agent and combination treatments were validated in multiple independent PDX and patient cohorts receiving platinum and taxane-containing neoadjuvant therapy, confirming the clinical relevance of our PDX panel. Chemotherapy-specific predictors for pathological complete response (pCR)/CR achieved AUROCs of 0.79, 0.67, and 0.70 for platinum, taxane, and combination treatments, respectively. The single-agent platinum model was the most effective in predicting platinum response, with a similar trend for taxane and platinum+taxane predictors. These findings reinforce the observation that distinct gene sets are linked to responses to different chemotherapy treatments. Since the predictors used treatment-associated genes found in both PDX and clinical samples, these results suggest biomarker combinations for translation and clinical development to select TNBC tumors that may respond to specific regimens. In PDXs responsive to any treatment, several basal cytokeratins were among the top upregulated proteins compared to PDXs resistant to all treatments. KRT5 was validated by IHC in PDX tumors, achieving an AUROC of 0.83 for predicting responsiveness, indicating its potential as a future chemoresponse marker in TNBC. PDXs refractory to all treatment arms had higher levels of mitochondrial and cancer stem-cell-related pathways. Treatment with romidepsin, an HDAC inhibitor that targets these pathways and increases DNA damage, enhanced carboplatin response in a chemoresistant PDX model with high abundance of HDAC proteins. Conclusion: Proteogenomic characterization identifies molecular mechanisms and putative biomarkers for stratifying TNBC tumors for single or combination chemotherapy treatments, suggests targeted therapies to augment chemotherapy response, and provides a valuable resource for researchers and clinicians. Citation Format: Jonathan Lei, Lacey E. Dobrolecki, Chen Huang, Ramakrishnan R. Srinivasan, Suhas Vasaikar, Alaina N. Lewis, Christina Sallas, John D. Landua, Chang In Moon, Yuxing Liao, Na Zhao, Jin Cao, Susan G. Hilsenbeck, C. Kent Osborne, Mothaffar F. Rimawi, Matthew J. Ellis, Varduhi Petrosyan, Bo Wen, Kai Li, Alexander B. Saltzman, Antrix Jain, Anna Malovannaya, Gerburg Wulf, Shunqiang Li, Daniel C. Kraushaar, Elisabetta Marangoni, Tao Wang, Senthil Damodaran, Xiaofeng Zheng, Funda Meric-Bernstam, Bryan E. Welm, Alana L. Welm, Xi Chen, Gloria V. Echeverria, Meenakshi Anurag, Bing Zhang, Michael T. Lewis. Patient-derived Xenografts (PDX) Allow Deconvolution of Combination Chemotherapy Response [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr PS4-02.

Here, we introduce the ID-GBA (Information Distance Guilt By Association) method to expand highly connected sets of nodes by deploying a novel algorithm for subgraph extension based on the guilt-by-association principle and information distance. In this study, ID-GBA was utilized to expand disease clusters, and identify novel disease genes. We first validate its ability to expand related disease sets from disease/disease graphs built using Open Targets’ gene association scores. We then analyze disease/control gene expression networks and show that ID-GBA recaptures known disease genes in nine disease/control graphs. Compared to existing methods such as Random Walk with Restarts and Personalized PageRank, ID-GBA achieves significantly higher Normalized Discounted Cumulative Gain scores, which indicates superior predictive performance at capturing known disease genes. Additionally, unlike other approaches that require users to specify either a threshold parameter or a fixed number of nodes to include in the extended subgraph, ID-GBA includes a built-in, automated, and data-driven thresholding mechanism. These results establish ID-GBA as a novel open-source tool to uncover hidden relationships in gene/gene, disease/disease, and other complex networks.

Abstract We have developed CTD2 (an algorithm to "Connect the Dots") to capture biological signals and identify candidate genes in complex networks. CTD2 extends the functionality of its predecessor CTD by providing ranked lists of genes that are “guilty by association” and significantly linked to other genes of interest. With the explosion of large scale multi-omics studies in the cancer field, novel approaches are needed to interpret these valuable datasets. Both CTD and CTD2 are information-theoretic algorithms that allow identification of highly connected sets of genes in complex networks without the need for permutation testing. CTD has been previously used to interpret perturbations in different subtypes of breast cancer. Additionally CTD has been used to identify biomarkers of chemotherapy response in Triple Negative Breast Cancer (TNBC) murine PDX models to both platinum and taxane agents. These small multigene biomarkers of response were shown to be informative for the response of both patients and PDXs. CTD2 was developed to expand the utility of the CTD package by also capturing genes that are "guilty by association". These genes are significantly connected to genes of interest (such as disease genes), and are ranked by their connectedness to these informative gene sets. To demonstrate its utility, we investigated if CTD2 could identify known breast cancer genes. Using TCGA breast cancer expression data, we built case/control graphs over 5,000 variable genes for each subtype of breast cancer. Genes previously associated with breast cancer were identified with DisGeNET and split into a training set discovered pre-2015 and test set discovered post-2015. The genes that were discovered pre-2015 that overlapped with our networks (n = 680) were then used as an input for CTD2 along with the case/control graphs. We then ranked the connectedness of all the genes in these graphs to the training set and found that the test set of breast cancer genes that were discovered post-2015 were significantly enriched in these ranked lists.We have shown that CTD2 and CTD can be utilized to discover biologically informative signals in complex networks. Furthermore we have deployed these tools on the Cancer Genomics Cloud to make them easily accessible for users without a bioinformatics background. The democratization of these in silico tools will allow for their adaptation by a wider audience and aid in the interpretation of large multi-omic datasets. Citation Format: Varduhi Petrosyan, Vladimir Kovacevic, Predrag Obradovic, Cera Fisher, Zelia Worman, Divya Sain, Jack DiGiovanna, Brandi Davis-Dusenberry, Aleksandar Milosavljevic. CTD2 "Connects the Dots" to capture disease genes in complex networks and its application on the Cancer Genomics Cloud [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2330.

Many biomedical research projects produce large-scale datasets that may serve as resources for the research community for hypothesis generation, facilitating diverse use cases. Towards the goal of developing infrastructure to support the findability, accessibility, interoperability, and reusability (FAIR) of biomedical digital objects and maximally extracting knowledge from data, complex queries that span across data and tools from multiple resources are currently not easily possible. By utilizing existing FAIR application programming interfaces (APIs) that serve knowledge from many repositories and bioinformatics tools, different types of complex queries and workflows can be created by using these APIs together. The Playbook Workflow Builder (PWB) is a web-based platform that facilitates interactive construction of workflows by enabling users to utilize an ever-growing network of input datasets, semantically annotated API endpoints, and data visualization tools contributed by an ecosystem. Via a user-friendly web-based user interface (UI), workflows can be constructed from contributed building-blocks without technical expertise. The output of each step of the workflows are provided in reports containing textual descriptions, as well as interactive and downloadable figures and tables. To demonstrate the ability of the PWB to generate meaningful hypotheses that draw knowledge from across multiple resources, we present several use cases. For example, one of these use cases sieves novel targets for individual cancer patients using data from the GTEx, LINCS, Metabolomics, GlyGen, and the ExRNA Communication Consortium (ERCC) Common Fund (CF) Data Coordination Centers (DCCs). The workflows created with the PWB can be published and repurposed to tackle similar use cases using different inputs. The PWB platform is available from: . ### Competing Interest Statement The authors have declared no competing interest.

Proper characterization of cancer cell states within the tumor microenvironment is a key to accurately identifying matching experimental models and the development of precision therapies. To reconstruct this information from bulk RNA-seq profiles, we developed the XDec Simplex Mapping (XDec-SM) reference-optional deconvolution method that maps tumors and the states of constituent cells onto a biologically interpretable low-dimensional space. The method identifies gene sets informative for deconvolution from relevant single-cell profiling data when such profiles are available. When applied to breast tumors in The Cancer Genome Atlas (TCGA), XDec-SM infers the identity of constituent cell types and their proportions. XDec-SM also infers cancer cells states within individual tumors that associate with DNA methylation patterns, driver somatic mutations, pathway activation and metabolic coupling between stromal and breast cancer cells. By projecting tumors, cancer cell lines, and PDX models onto the same map, we identify in vitro and in vivo models with matching cancer cell states. Map position is also predictive of therapy response, thus opening the prospects for precision therapy informed by experiments in model systems matched to tumors in vivo by cancer cell state.

Although systemic chemotherapy remains the standard of care for TNBC, even combination chemotherapy is often ineffective. The identification of biomarkers for differential chemotherapy response would allow for the selection of responsive patients, thus maximizing efficacy and minimizing toxicities. Here, we leverage TNBC PDXs to identify biomarkers of response. To demonstrate their ability to function as a preclinical cohort, PDXs were characterized using DNA sequencing, transcriptomics, and proteomics to show consistency with clinical samples. We then developed a network-based approach (CTD/WGCNA) to identify biomarkers of response to carboplatin (MSI1, TMSB15A, ARHGDIB, GGT1, SV2A, SEC14L2, SERPINI1, ADAMTS20, DGKQ) and docetaxel (c, MAGED4, CERS1, ST8SIA2, KIF24, PARPBP). CTD/WGCNA multigene biomarkers are predictive in PDX datasets (RNAseq and Affymetrix) for both taxane- (docetaxel or paclitaxel) and platinum-based (carboplatin or cisplatin) response, thereby demonstrating cross-expression platform and cross-drug class robustness. These biomarkers were also predictive in clinical datasets, thus demonstrating translational potential.

Background: Triple-negative breast cancer (TNBC) patients frequently receive combination chemotherapy treatment, but a direct comparison of response to carboplatin, docetaxel, and their combination in 50 TNBC patient-derived xenografts (PDXs) showed that combination treatment was largely ineffective at generating enhanced responses over the best single agent. This suggests de-escalation of chemotherapy may be possible if molecular mechanisms and biomarkers underlying response to individual treatments can be identified. To this end, we performed multi-omics profiling for the 50 TNBC PDXs. Methods: Orthotopic TNBC PDXs were treated with four weekly cycles of docetaxel, carboplatin, or the combination. Changes in tumor volume after 4 weeks of treatment were assessed quantitatively and by modified RECIST criteria. Genomic, transcriptomic, and mass-spectrometry-based proteomic profiling were performed on baseline tumors prior to treatments to identify associations with chemotherapy response at the gene and pathway level. ProMS was used to integrate both RNA and protein data to select a 5 RNA feature combination for optimized prediction of carboplatin response in a logistic regression model. Publicly available neoadjuvant chemotherapy clinical datasets with transcriptomic data and response information used for validation/testing included TNBC samples from: GSE18864, I-SPY2 (GSE194040), and BrighTNess (GSE164458). Results: Proteogenomic profiles revealed distinct genes associated with response to each agent and their combination, respectively, suggesting distinct molecular mechanisms underlying response to each treatment. A substantial number of genes associated with single agent and combination treatment were validated in multiple independent patient cohorts receiving platinum and taxane containing neoadjuvant therapy, confirming clinical relevance of our PDX panel. For the same treatment, different types of molecular data identified distinct sets of associated genes, providing highly complementary information. At the pathway level, RNA and protein data converged to metabolic and E2F/G2M related pathways which were upregulated in PDXs resistant or responsive to all treatment types, respectively, while variable levels of MYC-related proliferation pathways were observed across all treatments suggesting pathways that are common across and unique to different treatments. Several individual genes found to be higher in PDXs with better response to either single-agent had discriminatory power in external clinical TNBC datasets treated with similar neoadjuvant chemotherapy regimens. In addition, a logistic regression-based carboplatin response prediction model trained to select a group of 5 RNA markers (TKT, MAGI2, ATF6B, MCM7, LRP6) using both RNA and protein data performed the best in predicting response to cisplatin in a clinical TNBC dataset vs predicting response to other datasets with taxane and platinum + taxane combination containing chemotherapy regimens, demonstrating specificity of the prediction model. These results suggest potential individual biomarkers or biomarker combinations to select TNBC tumors that may respond to either single agent carboplatin, docetaxel, or their combination. PDXs refractory to all treatment arms had higher levels of proteostasis-related pathways including proteasome degradation and the unfolded protein response (UPR) related to endoplasmic reticulum stress and altered levels of chromatin regulation. Subsequent pharmacological targeting of the UPR pathway and targeting HDACs enhanced chemotherapy response. Conclusion: Proteogenomic characterization identifies molecular mechanisms and putative biomarkers for stratifying TNBC tumors for single or combination chemotherapy treatments, suggests targeted therapies to augment chemotherapy response, and provides a valuable resource for researchers and clinicians. Citation Format: Jonathan T. Lei, Chen Huang, Ramakrishnan R. Srinivasan, Suhas Vasaikar, Lacey E. Dobrolecki, Alaina N. Lewis, Na Zhao, Jin Cao, Susan G. Hilsenbeck, C. Kent Osborne, Mothaffar Rimawi, Matthew J. Ellis, Varduhi Petrosyan, Alexander B. Saltzman, Anna Malovannaya, John D. Landua, Bo Wen, Antrix Jain, Gerburg M. Wulf, Shunqiang Li, Daniel C. Kraushaar, Tao Wang, Xi Chen, Gloria V. Echeverria, Meenakshi Anurag, Bing Zhang, Michael T. Lewis. Patient-derived xenografts allow deconvolution of single agent and combination chemotherapy responses in triple-negative breast cancer [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P2-23-01.

TNBC is a heterogeneous subtype of breast cancer, and only a subset of TNBC can be established as PDXs. Here, we show that there is an engraftment bias toward TNBC with low levels of immune cell infiltration. Additionally, TNBC that failed to engraft show gene expression consistent with a cancer-promoting immunological state, leading us to hypothesize that the immunological state of the tumor and possibly the state of the immune system of the host may be essential for engraftment.

Background: Chemotherapy is essential for the management of patients with triple-negative breast cancer (TNBC). Identification of biomarkers that may indicate treatment efficacy will be critical to improve patient stratification prior to treatment. To elucidate molecular determinants underlying chemotherapy response, we conducted a proteogenomic study using TNBC patient-derived xenografts (PDXs) treated with chemotherapy. Methods: Orthotopic TNBC PDXs were treated with four weekly cycles of docetaxel, carboplatin, or the combination. Changes in tumor volume after 4 weeks of treatment were evaluated. Genomic, transcriptomic, and mass-spectrometry-based proteomic profiling were performed on baseline tumors prior to treatments to identify associations with chemotherapy response (https://pdxportal.research.bcm.edu). Fisher’s exact test was used to identify significant associations between mutation and copy number events and dichotomized treatment response. For gene level analyses, Spearman’s correlation was calculated between mRNA or protein abundance and log2 fold change in tumor volumes after treatment. Signed -log10 p-values from Spearman’s correlation analysis were used as input for Gene Set Enrichment Analysis. Validation in external datasets was performed using METABRIC (PMID: 22522925) and BrighTNess clinical trial (PMID: 29501363) human datasets, an independent PDX dataset (PMID: 32546838), and shRNA screen data from DepMap (PMID: 30389920). Results: Combination carboplatin and docetaxel was largely ineffective at generating enhanced responses over the best single agent, suggesting de-escalation of chemotherapy may be possible. Genomic aberrations in BRCA2 and BCL9 were enriched in carboplatin-responsive PDXs while aberrations in RAF1 were enriched in docetaxel-resistant PDXs. Genes with gene-drug response correlations supported by both mRNA and protein measurements, but not mRNA or protein alone, for both carboplatin and docetaxel treatment in PDXs were associated with prognosis from basal human breast tumors receiving any chemotherapy from the METABRIC dataset. These data suggest that the combination of mRNA and protein data provided increased accuracy in identifying genes associated with clinical outcome in TNBC. Some of the top genes with genomic aberrations and/or overexpression at both mRNA and protein levels in chemoresistant PDXs, many of which have not been evaluated for their ability to augment response to taxane- or platinum-based chemotherapies, were validated in independent datasets with PDXs and TNBC patients receiving carboplatin and taxane combination. Further, some were found to be dependencies in TNBC cell lines from a publicly available genetic perturbation dataset. At the pathway level, both mRNA and protein data associated models resistant to both agents with enhanced oxidative phosphorylation and proteostatic pathways including proteosome degradation and the unfolded protein response (UPR) pathway related to endoplasmic reticulum stress. Pharmacological targeting the UPR pathway in combination with docetaxel showed activity in PDX models resistant to single-agent docetaxel. These results suggest targeting the UPR pathway as a novel therapeutic strategy to overcome chemotherapy resistance in TNBC. Conclusion: Proteogenomic analysis of PDX tumors identified diverse genes and pathways associated with chemotherapy resistance and response and further suggests potential therapeutic opportunities in TNBC. Citation Format: Jonathan T Lei, Chen Huang, Ramakrishnan R Srinivasan, Suhas Vasaikar, Lacey E Dobrolecki, Alaina N Lewis, Christina Sallas, Susan G Hilsenbeck, C. Kent Osborne, Mothaffar F Rimawi, Matthew J Ellis, Varduhi Petrosyan, Alexander B Saltzman, Anna Malovannaya, Gerburg Wulf, Daniel C Kraushaar, Tao Wang, Xi Chen, Gloria V Echeverria, Meenakshi Anurag, Bing Zhang, Michael T Lewis. Proteogenomic analysis of differential chemotherapy responses in patient-derived xenografts of triple-negative breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P5-07-01.

SUMMARYThe characterization of cancer cell states within the tumor microenvironment is a key to understanding tumor biology and an important step toward the development of precision therapies. To reconstruct this information from bulk RNA-seq profiles, we developed the XDec Simplex Mapping (XDec-SM) approach, a reference-optional deconvolution method that leverages single-cell information, when such information is available, to map tumors and the states of constituent cells onto a biologically interpretable, low-dimensional space. When applied to breast tumors in The Cancer Genome Atlas (TCGA), XDec-SM infers the identity of constituent cell types and their proportions. XDec-SM also infers cancer cells states within individual tumors that associate with DNA methylation patterns, driver somatic mutations, pathway activation and metabolic coupling between stromal and breast cancer cells. By projecting tumors, cancer cell lines, and PDX models onto the same map, we identify both in vitro and in vivo models with matching cancer cell states. Map position is also predictive of therapy response, thus opening the prospects for precision therapy informed by experiments in model systems matched to tumors in vivo by cancer cell state.

Most cancers harbor a diverse collection of cell types including a typically heterogeneous cancer cell fraction. To reconstruct cell-intrinsic and heterotypic interactions driving tumor progression, we combine the XDec deconvolution method with cell-type-specific gene expression correlation analysis into the XDec-CHI method. XDec-CHI identifies intra-and inter-cellular pathways using correlation and places them in the context of specific tumor subtypes, as defined by the state of constituent cancer cells. We make the method web-accessible for analysis of publicly accessible pancreatic ductal adenocarcinoma, breast, head and neck, glioblastoma, and glioma tumors. We apply the method to TCGA and ICGC datasets to identify immune-suppressive interactions within PDAC tumors that are relevant for immunotherapies targeting PD-L1. Subtype-specific interactions derived from correlative analyses validated in co-culture experiments suggest PDAC subtypes have distinct therapeutic weaknesses, with Basal-like and MSLN-high Classical B tumors most likely to respond to therapies targeting PD-L1.

SummaryTriple negative breast cancer (TNBC) is a highly heterogeneous set of diseases that has, until recently, lacked any FDA-approved, molecularly targeted therapeutics. Thus, systemic chemotherapy regimens remain the standard of care for many. Unfortunately, even combination chemotherapy is ineffective for many TNBC patients, and side-effects can be severe or lethal. Identification of predictive biomarkers for chemotherapy response would allow for the prospective selection of responsive patients, thereby maximizing efficacy and minimizing unwanted toxicities. Here, we leverage a cohort of TNBC PDX models with responses to single-agent docetaxel or carboplatin to identify biomarkers predictive for differential response to these two drugs. To demonstrate their ability to function as a preclinical cohort, PDX were molecularly characterized using whole-exome DNA sequencing, RNAseq transcriptomics, and mass spectrometry-based total proteomics to show proteogenomic consistency with TCGA and CPTAC clinical samples. Focusing first on the transcriptome, we describe a network-based computational approach to identify candidate epithelial and stromal biomarkers of response to carboplatin (MSI1, TMSB15A, ARHGDIB, GGT1, SV2A, SEC14L2, SERPINI1, ADAMTS20, DGKQ) and docetaxel (ITGA7, MAGED4, CERS1, ST8SIA2, KIF24, PARPBP). Biomarker panels are predictive in PDX expression datasets (RNAseq and Affymetrix) for both taxane (docetaxel or paclitaxel) and platinum-based (carboplatin or cisplatin) response, thereby demonstrating both cross expression platform and cross drug class robustness. Biomarker panels were also predictive in clinical datasets with response to cisplatin or paclitaxel, thus demonstrating translational potential of PDX-based preclinical trials. This network-based approach is highly adaptable and can be used to evaluate biomarkers of response to other agents.

[This corrects the article DOI: 10.1371/journal.pcbi.1008550.].

Abstract Background: Chemotherapy is essential for the management of patients with triple-negative breast cancer (TNBC). Identification of biomarkers that may indicate treatment efficacy will be critical to improve patient stratification prior to treatment. To elucidate molecular determinants underlying chemotherapy response, we conducted a proteogenomic study using TNBC patient-derived xenografts (PDXs) treated with chemotherapy. Approach: 50 TNBC PDXs were treated with either docetaxel or carboplatin. Changes in tumor volume after 4 weeks from baseline were evaluated. Genomic, transcriptomic, and mass-spectrometry-based proteomic profiling were performed on baseline tumors prior to treatment to identify associations with chemotherapy response. Fisher's exact tests were used to test for significant enrichment of mutation and copy number events (p<0.05). Gene Set Enrichment Analysis was performed for pathway analyses. Results: At the DNA level, genomic aberrations in BRCA2 and BCL2 were enriched in carboplatin-responsive PDXs, while ARID1B aberrations were enriched in docetaxel-responsive PDXs. Gene-drug response correlations supported by both mRNA and protein-based measurements, but not mRNA or protein alone, for both carboplatin and docetaxel treatment in PDXs were associated with prognosis from basal and claudin-low human breast tumors in receipt of any chemotherapy from the METABRIC dataset. These data suggest that the combination of mRNA and protein data increased power to identify genes related to clinical outcome in TNBC. Some of the top genes overexpressed at both mRNA and protein levels in chemoresistant PDXs are targets of approved drugs, many of which have not been evaluated for their ability to augment response to taxane- or platinum-based chemotherapies. These genes are being investigated as therapeutic targets as well as markers of chemotherapy response. At the pathway level, both RNA and protein data associated models resistant to both agents with enhanced oxidative phosphorylation and translation regulation. Protein data further associated resistant models with elevated cytoplasmic ribosomal proteins. In contrast, both RNA and protein data associated tumors sensitive to both agents with genes involved in the E2F-Rb axis and cell cycle progression. Moreover, DNA mismatch repair and mRNA processing pathways were uniquely associated with carboplatin and docetaxel sensitivity, respectively, while amino acid metabolism and MAPK signaling pathways were uniquely associated with carboplatin and docetaxel resistance, respectively. Conclusion: Taken together, proteogenomic analysis of PDX tumors identifies diverse genes and pathways associated with chemotherapy response and further suggests potential therapeutic opportunities in TNBC. Citation Format: Jonathan T. Lei, Chen Huang, Ramakrishnan R. Srinivasan, Suhas Vasaikar, Lacey E. Dobrolecki, Alaina N. Lewis, Christina Sallas, Susan G. Hilsenbeck, C Kent Osborne, Mothaffar F. Rimawi, Matthew J. Ellis, Varduhi Petrosyan, Alexander B. Saltzman, Anna Malovannaya, Gerburg Wulf, Daniel C. Kraushaar, Tao Wang, Gloria V. Echeverria, Bing Zhang, Michael T. Lewis. Proteogenomic characterization of triple-negative breast cancer patient-derived xenografts reveals molecular correlates of differential chemotherapy response and potential therapeutic targets to overcome resistance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2992.

Abstract Background: Epithelial-stromal interactions play a critical role in treatment resistance, and the state of the stroma can be an indicator for a tumor’s resistance profile. We have developed and characterized a large group of breast cancer PDX models (n=69) (Zhang et al.,2013) and have imported PDX from other sources for inclusion into preclinical drug testing trials. These PDX represent the three main subtypes defined clinically by IHC/FISH (i.e. Estrogen Receptor positive, HER2 positive, and “Triple-negative”). Because the epithelium and stroma in PDX are from two different species (human and mouse, respectively) mRNA (and protein) expression levels can be assigned in a species-specific manner and interrogated computationally to divine interactions that may be important for PDX behaviors including treatment response or engraftment behavior. Methods: Fifty “triple negative” PDX were treated with four weekly cycles of either docetaxel (20mg/kg, IP) or carboplatin (50mg/kg, IP) vs. control, and evaluated quantitatively for the change in tumor volume from baseline (~200mm3) after four weeks. Using deep RNA-seq data (~200M reads/sample), we identified interactions between the epithelial and stromal in the PDX models using CASTIN (Komura et al.,2016) and variations in cancer specific pathways. We then employed a machine learning approach to identify ligand-receptor interactions whose differential expression correlated with treatment response to each agent. We also used epigenomic deconvolution (Onuchic et al., 2016) to devise a novel classification method based on the cancer-cell specific epigenomic profiles of TCGA samples, and then used it to classify the PDX with the intent to correlate with treatment response. Results: PDX showed a full range of responses to each agent, from total resistance to complete response. However, several PDX showed differential responses to either docetaxel or carboplatin. We identified variation in the expression patterns of cancer-associated pathways in the epithelial cell fraction of the tumors, including Wnt signaling, as well as pathways involved in hepatic fibrosis. Informative interactions included bidirectional Eph receptor-ephrin signaling which has previously been found to be overexpressed in breast cancer (Vaught et al.,2008). In the epigenomic deconvolution analysis, our classification scheme divided the basal-like samples in TCGA into two groups: a “hot” immune profile-enriched group, and a “cold” or immune profile-deficient group. By applying this classification to the PDX models, unexpectedly, an overwhelming majority of the triple-negative PDX fell into the “cold” basal group. Thus, classification in this manner showed no association with treatment response in the PDX. Conclusions: These results indicate that tumors may respond uniquely to a given chemotherapeutic, and suggest that differential expression of signaling pathways as well as specific ligand-receptor pairs may prove predictive of resistance and could allow for the development of novel therapies targeting these tissue interactions. While our epigenomic deconvolution results showed no association with treatment response, they unexpectedly suggested that success of PDX engraftment may depend on the stromal immune system composition of the primary tumor such that “hot” primary tumors have lower engraftment efficiency in the “cold” stromal environment of immunocompromised mice. Citation Format: Varduhi Petrosyan, Chen Huang, Ramakrishnan R Srinivasan, Lacey E Dobrolecki, Christina Sallas, Alaina N Lewis, Tao Wang, Bing Zhang, Aleksandar Milosavljevic, Michael T Lewis. Histoepigenetic characterization of breast cancer patient derived xenografts (PDX) implicates epithelial-stromal interactions in differential chemotherapy resistance and PDX engraftment [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P1-03-02.