Blockade of tumor cell-intrinsic PD-L1 signaling enhances AURKA-targeted therapy in triple negative breast cancer.

Triple negative breast cancer (TNBC) accounts for 15-20% of all breast cancers and mainly affects pre-menopausal and minority women. Because of the lack of ER, PR or HER2 expression in TNBC, there are limited options for tailored therapies. While TNBCs respond initially to standard of care chemotherapy, tumor recurrence commonly occurs within 1 to 3 years post-chemotherapy and is associated with early organ metastasis and a high incidence of mortality. One of the major mechanisms responsible for drug resistance and emergence of organ metastasis is activation of epithelial to mesenchymal transition (EMT) reprogramming. EMT-mediated cancer cell plasticity also promotes the enrichment of cancer cells with a CD44high/CD24low and/or ALDHhigh cancer stem-like phenotype [cancer stem cells (CSCs)], characterized by an increased capacity for tumor self-renewal, intrinsic drug resistance, immune evasion and metastasis. In this study we demonstrate for the first time a positive feedback loop between AURKA and intra-tumoral PD-L1 oncogenic pathways in TNBC. Genetic targeting of intra-tumoral PD-L1 expression impairs the enrichment of ALDHhigh CSCs and enhances the therapeutic efficacy of AURKA-targeted therapy. Moreover, dual AURKA and PD-L1 pharmacological blockade resulted in the strongest inhibition of tumor growth and organ metastatic burden. Taken together, our findings provide a compelling preclinical rationale for the development of novel combinatorial therapeutic strategies aimed to inhibit cancer cell plasticity, immune evasion capacity and organ metastasis in patients with advanced TNBC.

Pre-treatment peripheral blood immunophenotyping and response to neoadjuvant chemotherapy in operable breast cancer.

BACKGROUND: Tumor immune infiltration and peripheral blood immune signatures have prognostic and predictive value in breast cancer. Whether distinct peripheral blood immune phenotypes are associated with response to neoadjuvant chemotherapy (NAC) remains understudied. METHODS: Peripheral blood mononuclear cells from 126 breast cancer patients enrolled in a prospective clinical trial (NCT02022202) were analyzed using Cytometry by time-of-flight with a panel of 29 immune cell surface protein markers. Kruskal-Wallis tests or Wilcoxon rank-sum tests were used to evaluate differences in immune cell subpopulations according to breast cancer subtype and response to NAC. RESULTS: There were 122 evaluable samples: 47 (38.5%) from patients with hormone receptor-positive, 39 (32%) triple-negative (TNBC), and 36 (29.5%) HER2-positive breast cancer. The relative abundances of pre-treatment peripheral blood T, B, myeloid, NK, and unclassified cells did not differ according to breast cancer subtype. In TNBC, higher pre-treatment myeloid cells were associated with lower pathologic complete response (pCR) rates. In hormone receptor-positive breast cancer, lower pre-treatment CD8 + naïve and CD4 + effector memory cells re-expressing CD45RA (TEMRA) T cells were associated with more extensive residual disease after NAC. In HER2 + breast cancer, the peripheral blood immune phenotype did not differ according to NAC response. CONCLUSIONS: Pre-treatment peripheral blood immune cell populations (myeloid in TNBC; CD8 + naïve T cells and CD4 + TEMRA cells in luminal breast cancer) were associated with response to NAC in early-stage TNBC and hormone receptor-positive breast cancers, but not in HER2 + breast cancer. TRIAL REGISTRATION: NCT02022202 . Registered 20 December 2013.

ReCorDE: a framework for identifying drug classes targeting shared vulnerabilities with applications to synergistic drug discovery.

Cancer is typically treated with combinatorial therapy, and such combinations may be synergistic. However, discovery of these combinations has proven difficult as brute force combinatorial screening approaches are both logistically complex and resource-intensive. Therefore, computational approaches to augment synergistic drug discovery are of interest, but current approaches are limited by their dependencies on combinatorial drug screening training data or molecular profiling data. These dataset dependencies can limit the number and diversity of drugs for which these approaches can make inferences. Herein, we describe a novel computational framework, ReCorDE (Recurrent Correlation of Drugs with Enrichment), that uses publicly-available cell line-derived monotherapy cytotoxicity datasets to identify drug classes targeting shared vulnerabilities across multiple cancer lineages; and we show how these inferences can be used to augment synergistic drug combination discovery. Additionally, we demonstrate in preclinical models that a drug class combination predicted by ReCorDE to target shared vulnerabilities (PARP inhibitors and Aurora kinase inhibitors) exhibits class-class synergy across lineages. ReCorDE functions independently of combinatorial drug screening and molecular profiling data, using only extensive monotherapy cytotoxicity datasets as its input. This allows ReCorDE to make robust inferences for a large, diverse array of drugs. In conclusion, we have described a novel framework for the identification of drug classes targeting shared vulnerabilities using monotherapy cytotoxicity datasets, and we showed how these inferences can be used to aid discovery of novel synergistic drug combinations.

Anastrozole dose escalation for optimal estrogen suppression in postmenopausal early-stage breast cancer: A prospective trial.

BACKGROUND: We previously reported that postmenopausal women with ER+ breast cancer (BC) receiving adjuvant anastrozole 1 mg/day (ANA1) with estrone (E1) ≥1.3 pg/mL and estradiol (E2) ≥0.5 (inadequate estrogen suppression [IES]) had a 3.0-fold increased risk of a BC event. The objective of this study was to determine if increasing anastrozole to 10 mg/day (ANA10) could result in adequate estrogen suppression (AES: E1 <1.3 pg/mL and/or E2 <0.5) among those with IES on ANA1. METHODS: Postmenopausal women with ER+ BC planning to receive adjuvant ANA1 were eligible. E1 and E2 were assessed pre- and post-8-10 weeks of ANA1. Those with IES were switched to 8-10 week cycles of ANA10 followed by letrozole 2.5 mg/day. E1 and E2 were assessed after each cycle. Anastrozole concentrations were measured post-ANA1 and post-ANA10. Primary analyses included patients who documented taking at least 80% of planned treatment (adherent cohort). RESULTS: 132 (84.6%) of 156 eligible patients were ANA1-adherent. IES occurred in 40 (30.3%) adherent patients. 25 (78.1%) of 32 patients who began ANA10 were adherent, and AES was achieved in 19 (76.0%; 90%CI: 58.1-89.0%) patients. Anastrozole concentrations post-ANA1 and post-ANA10 did not differ by estrogen suppression status among adherent patients. AES was maintained/attained in 21 (91.3%) of 23 letrozole-adherent patients. CONCLUSIONS: Approximately 30% of ANA1-adherent patients had IES. Among those who switched to ANA10 and were adherent, 76% had AES. Further studies are required to validate emerging data that ANA1 results in IES for some patients and to determine the clinical benefit of switching to ANA10 or an alternative AI.

NetActivity enhances transcriptional signals by combining gene expression into robust gene set activity scores through interpretable autoencoders.

Grouping gene expression into gene set activity scores (GSAS) provides better biological insights than studying individual genes. However, existing gene set projection methods cannot return representative, robust, and interpretable GSAS. We developed NetActivity, a machine learning framework that generates GSAS based on a sparsely-connected autoencoder, where each neuron in the inner layer represents a gene set. We proposed a three-tier training that yielded representative, robust, and interpretable GSAS. NetActivity model was trained with 1518 GO biological processes terms and KEGG pathways and all GTEx samples. NetActivity generates GSAS robust to the initialization parameters and representative of the original transcriptome, and assigned higher importance to more biologically relevant genes. Moreover, NetActivity returns GSAS with a more consistent definition and higher interpretability than GSVA and hipathia, state-of-the-art gene set projection methods. Finally, NetActivity enables combining bulk RNA-seq and microarray datasets in a meta-analysis of prostate cancer progression, highlighting gene sets related to cell division, key for disease progression. When applied to metastatic prostate cancer, gene sets associated with cancer progression were also altered due to drug resistance, while a classical enrichment analysis identified gene sets irrelevant to the phenotype. NetActivity is publicly available in Bioconductor and GitHub.

Syk-dependent homologous recombination activation promotes cancer resistance to DNA targeted therapy.

Enhanced DNA repair is an important mechanism of inherent and acquired resistance to DNA targeted therapies, including poly ADP ribose polymerase (PARP) inhibition. Spleen associated tyrosine kinase (Syk) is a non-receptor tyrosine kinase acknowledged for its regulatory roles in immune cell function, cell adhesion, and vascular development. This study presents evidence indicating that Syk expression in high-grade serous ovarian cancer and triple-negative breast cancers promotes DNA double-strand break resection, homologous recombination (HR), and subsequent therapeutic resistance. Our investigations reveal that Syk is activated by ATM following DNA damage and is recruited to DNA double-strand breaks by NBS1. Once localized to the break site, Syk phosphorylates CtIP, a pivotal mediator of resection and HR, at Thr-847 to promote repair activity, particularly in Syk-expressing cancer cells. Inhibition of Syk or its genetic deletion impedes CtIP Thr-847 phosphorylation and overcomes the resistant phenotype. Collectively, our findings suggest a model wherein Syk fosters therapeutic resistance by promoting DNA resection and HR through a hitherto uncharacterized ATM-Syk-CtIP pathway. Moreover, Syk emerges as a promising tumor-specific target to sensitize Syk-expressing tumors to PARP inhibitors, radiation and other DNA-targeted therapies.

OmicsFootPrint: a framework to integrate and interpret multi-omics data using circular images and deep neural networks.

The OmicsFootPrint framework addresses the need for advanced multi-omics data analysis methodologies by transforming data into intuitive two-dimensional circular images and facilitating the interpretation of complex diseases. Utilizing Deep Neural Networks and incorporating the SHapley Additive exPlanations (SHAP) algorithm, the framework enhances model interpretability. Tested with The Cancer Genome Atlas (TCGA) data, OmicsFootPrint effectively classified lung and breast cancer subtypes, achieving high Area Under Curve (AUC) scores - 0.98±0.02 for lung cancer subtype differentiation, 0.83±0.07 for breast cancer PAM50 subtypes, and successfully distinguishe between invasive lobular and ductal carcinomas in breast cancer, showcasing its robustness. It also demonstrated notable performance in predicting drug responses in cancer cell lines, with a median AUC of 0.74, surpassing existing algorithms. Furthermore, its effectiveness persists even with reduced training sample sizes. OmicsFootPrint marks an enhancement in multi-omics research, offering a novel, efficient, and interpretable approach that contributes to a deeper understanding of disease mechanisms.

Identification of a Notch transcriptomic signature for breast cancer.

BACKGROUND: Dysregulated Notch signalling contributes to breast cancer development and progression, but validated tools to measure the level of Notch signalling in breast cancer subtypes and in response to systemic therapy are largely lacking. A transcriptomic signature of Notch signalling would be warranted, for example to monitor the effects of future Notch-targeting therapies and to learn whether altered Notch signalling is an off-target effect of current breast cancer therapies. In this report, we have established such a classifier. METHODS: To generate the signature, we first identified Notch-regulated genes from six basal-like breast cancer cell lines subjected to elevated or reduced Notch signalling by culturing on immobilized Notch ligand Jagged1 or blockade of Notch by γ-secretase inhibitors, respectively. From this cadre of Notch-regulated genes, we developed candidate transcriptomic signatures that were trained on a breast cancer patient dataset (the TCGA-BRCA cohort) and a broader breast cancer cell line cohort and sought to validate in independent datasets. RESULTS: An optimal 20-gene transcriptomic signature was selected. We validated the signature on two independent patient datasets (METABRIC and Oslo2), and it showed an improved coherence score and tumour specificity compared with previously published signatures. Furthermore, the signature score was particularly high for basal-like breast cancer, indicating an enhanced level of Notch signalling in this subtype. The signature score was increased after neoadjuvant treatment in the PROMIX and BEAUTY patient cohorts, and a lower signature score generally correlated with better clinical outcome. CONCLUSIONS: The 20-gene transcriptional signature will be a valuable tool to evaluate the response of future Notch-targeting therapies for breast cancer, to learn about potential effects on Notch signalling from conventional breast cancer therapies and to better stratify patients for therapy considerations.

Patient-specific signaling signatures predict optimal therapeutic combinations for triple negative breast cancer.

Triple negative breast cancer (TNBC) is a heterogeneous group of tumors which lack estrogen receptor, progesterone receptor, and HER2 expression. Targeted therapies have limited success in treating TNBC, thus a strategy enabling effective targeted combinations is an unmet need. To tackle these challenges and discover individualized targeted combination therapies for TNBC, we integrated phosphoproteomic analysis of altered signaling networks with patient-specific signaling signature (PaSSS) analysis using an information-theoretic, thermodynamic-based approach. Using this method on a large number of TNBC patient-derived tumors (PDX), we were able to thoroughly characterize each PDX by computing a patient-specific set of unbalanced signaling processes and assigning a personalized therapy based on them. We discovered that each tumor has an average of two separate processes, and that, consistent with prior research, EGFR is a major core target in at least one of them in half of the tumors analyzed. However, anti-EGFR monotherapies were predicted to be ineffective, thus we developed personalized combination treatments based on PaSSS. These were predicted to induce anti-EGFR responses or to be used to develop an alternative therapy if EGFR was not present.In-vivo experimental validation of the predicted therapy showed that PaSSS predictions were more accurate than other therapies. Thus, we suggest that a detailed identification of molecular imbalances is necessary to tailor therapy for each TNBC. In summary, we propose a new strategy to design personalized therapy for TNBC using pY proteomics and PaSSS analysis. This method can be applied to different cancer types to improve response to the biomarker-based treatment.

Age-Dependent Sex Differences in the Prevalence of Selective Serotonin Reuptake Inhibitor Treatment: A Retrospective Cohort Analysis.

Background: Antidepressants are among the most prescribed medications in the United States. The aim of this study was to explore the prevalence of antidepressant prescriptions and investigate sex differences and age-sex interactions in adults enrolled in the Right Drug, Right Dose, Right Time: Using Genomic Data to Individualize Treatment (RIGHT) study. Materials and Methods: We conducted a retrospective analysis of the RIGHT study. Using electronic prescriptions, we assessed 12-month prevalence of antidepressant treatment. Sex differences and age-sex interactions were evaluated using multivariable logistic regression and flexible recursive smoothing splines. Results: The sample consisted of 11,087 participants (60% women). Antidepressant prescription prevalence was 22.24% (27.96% women, 13.58% men). After adjusting for age and enrollment year, women had significantly greater odds of antidepressant prescription (odds ratio = 2.29; 95% confidence interval = 2.07, 2.54). Furthermore, selective serotonin reuptake inhibitors (SSRIs) had a significant age-sex interaction. While SSRI prescriptions in men showed a sustained decrease with age, there was no such decline for women until after reaching ∼50 years of age. There are important limitations to consider in this study. Electronic prescription data were cross-sectional; information on treatment duration or adherence was not collected; this cohort is not nationally representative; and enrollment occurred over a broad period, introducing confounding by changes in temporal prescribing practices. Conclusions: Underscored by the significant interaction between age and sex on odds of SSRI prescription, our results warrant age to be incorporated as a mediator when investigating sex differences in mental illness, especially mood disorders and their treatment.

Artificial Intelligence-based Efficacy Prediction of Phase 3 Clinical Trial for Repurposing Heart Failure Therapies.

Introduction: Drug repurposing involves finding new therapeutic uses for already approved drugs, which can save costs as their pharmacokinetics and pharmacodynamics are already known. Predicting efficacy based on clinical endpoints is valuable for designing phase 3 trials and making Go/No-Go decisions, given the potential for confounding effects in phase 2. Objectives: This study aims to predict the efficacy of the repurposed Heart Failure (HF) drugs for the Phase 3 Clinical Trial. Methods: Our study presents a comprehensive framework for predicting drug efficacy in phase 3 trials, which combines drug-target prediction using biomedical knowledgebases with statistical analysis of real-world data. We developed a novel drug-target prediction model that uses low-dimensional representations of drug chemical structures and gene sequences, and biomedical knowledgebase. Furthermore, we conducted statistical analyses of electronic health records to assess the effectiveness of repurposed drugs in relation to clinical measurements (e.g., NT-proBNP). Results: We identified 24 repurposed drugs (9 with a positive effect and 15 with a non-positive) for heart failure from 266 phase 3 clinical trials. We used 25 genes related to heart failure for drug-target prediction, as well as electronic health records (EHR) from the Mayo Clinic for screening, which contained over 58,000 heart failure patients treated with various drugs and categorized by heart failure subtypes. Our proposed drug-target predictive model performed exceptionally well in all seven tests in the BETA benchmark compared to the six cutting-edge baseline methods (i.e., best performed in 266 out of 404 tasks). For the overall prediction of the 24 drugs, our model achieved an AUCROC of 82.59% and PRAUC (average precision) of 73.39%. Conclusion: The study demonstrated exceptional results in predicting the efficacy of repurposed drugs for phase 3 clinical trials, highlighting the potential of this method to facilitate computational drug repurposing.

Leveraging multi-source to resolve inconsistency across pharmacogenomic datasets in drug sensitivity prediction.

Pharmacogenomics datasets have been generated for various purposes, such as investigating different biomarkers. However, when studying the same cell line with the same drugs, differences in drug responses exist between studies. These variations arise from factors such as inter-tumoral heterogeneity, experimental standardization, and the complexity of cell subtypes. Consequently, drug response prediction suffers from limited generalizability. To address these challenges, we propose a computational model based on Federated Learning (FL) for drug response prediction. By leveraging three pharmacogenomics datasets (CCLE, GDSC2, and gCSI), we evaluate the performance of our model across diverse cell line-based databases. Our results demonstrate superior predictive performance compared to baseline methods and traditional FL approaches through various experimental tests. This study underscores the potential of employing FL to leverage multiple data sources, enabling the development of generalized models that account for inconsistencies among pharmacogenomics datasets. By addressing the limitations of low generalizability, our approach contributes to advancing drug response prediction in precision oncology.

Syk-dependent alternative homologous recombination activation promotes cancer resistance to DNA targeted therapy.

Enhanced DNA repair is an important mechanism of inherent and acquired resistance to DNA targeted therapies, including poly ADP ribose polymerase inhibition. Spleen associated tyrosine kinase (Syk) is a non-receptor tyrosine kinase known to regulate immune cell function, cell adhesion, and vascular development. Here, we report that Syk can be expressed in high grade serous ovarian cancer and triple negative breast cancers and promotes DNA double strand break resection, homologous recombination (HR) and therapeutic resistance. We found that Syk is activated by ATM following DNA damage and is recruited to DNA double strand breaks by NBS1. Once at the break site, Syk phosphorylates CtIP, a key mediator of resection and HR, at Thr-847 to promote repair activity, specifically in Syk expressing cancer cells. Syk inhibition or genetic deletion abolished CtIP Thr-847 phosphorylation and overcame the resistant phenotype. Collectively, our findings suggest that Syk drives therapeutic resistance by promoting DNA resection and HR through a novel ATM-Syk-CtIP pathway, and that Syk is a new tumor-specific target to sensitize Syk-expressing tumors to PARPi and other DNA targeted therapy.

Integration of multiomics data shows down regulation of mismatch repair and tubulin pathways in triple-negative chemotherapy-resistant breast tumors.

BACKGROUND: Triple-negative breast cancer (TNBC) is the most aggressive breast cancer subtype. Patients with TNBC are primarily treated with neoadjuvant chemotherapy (NAC). The response to NAC is prognostic, with reductions in overall survival and disease-free survival rates in those patients who do not achieve a pathological complete response (pCR). Based on this premise, we hypothesized that paired analysis of primary and residual TNBC tumors following NAC could identify unique biomarkers associated with post-NAC recurrence. METHODS AND RESULTS: We investigated 24 samples from 12 non-LAR TNBC patients with paired pre- and post-NAC data, including four patients with recurrence shortly after surgery (< 24 months) and eight who remained recurrence-free (> 48 months). These tumors were collected from a prospective NAC breast cancer study (BEAUTY) conducted at the Mayo Clinic. Differential expression analysis of pre-NAC biopsies showed minimal gene expression differences between early recurrent and nonrecurrent TNBC tumors; however, post-NAC samples demonstrated significant alterations in expression patterns in response to intervention. Topological-level differences associated with early recurrence were implicated in 251 gene sets, and an independent assessment of microarray gene expression data from the 9 paired non-LAR samples available in the NAC I-SPY1 trial confirmed 56 gene sets. Within these 56 gene sets, 113 genes were observed to be differentially expressed in the I-SPY1 and BEAUTY post-NAC studies. An independent (n = 392) breast cancer dataset with relapse-free survival (RFS) data was used to refine our gene list to a 17-gene signature. A threefold cross-validation analysis of the gene signature with the combined BEAUTY and I-SPY1 data yielded an average AUC of 0.88 for six machine-learning models. Due to the limited number of studies with pre- and post-NAC TNBC tumor data, further validation of the signature is needed. CONCLUSION: Analysis of multiomics data from post-NAC TNBC chemoresistant tumors showed down regulation of mismatch repair and tubulin pathways. Additionally, we identified a 17-gene signature in TNBC associated with post-NAC recurrence enriched with down-regulated immune genes.

A ribosomal gene panel predicting a novel synthetic lethality in non-BRCAness tumors.

Poly (ADP-ribose) polymerase (PARP) inhibitors are one of the most exciting classes of targeted therapy agents for cancers with homologous recombination (HR) deficiency. However, many patients without apparent HR defects also respond well to PARP inhibitors/cisplatin. The biomarker responsible for this mechanism remains unclear. Here, we identified a set of ribosomal genes that predict response to PARP inhibitors/cisplatin in HR-proficient patients. PARP inhibitor/cisplatin selectively eliminates cells with high expression of the eight genes in the identified panel via DNA damage (ATM) signaling-induced pro-apoptotic ribosomal stress, which along with ATM signaling-induced pro-survival HR repair constitutes a new model to balance the cell fate in response to DNA damage. Therefore, the combined examination of the gene panel along with HR status would allow for more precise predictions of clinical response to PARP inhibitor/cisplatin. The gene panel as an independent biomarker was validated by multiple published clinical datasets, as well as by an ovarian cancer organoids library we established. More importantly, its predictive value was further verified in a cohort of PARP inhibitor-treated ovarian cancer patients with both RNA-seq and WGS data. Furthermore, we identified several marketed drugs capable of upregulating the expression of the genes in the panel without causing HR deficiency in PARP inhibitor/cisplatin-resistant cell lines. These drugs enhance PARP inhibitor/cisplatin sensitivity in both intrinsically resistant organoids and cell lines with acquired resistance. Together, our study identifies a marker gene panel for HR-proficient patients and reveals a broader application of PARP inhibitor/cisplatin in cancer therapy.

Expression and Therapeutic Targeting of TROP-2 in Treatment-Resistant Prostate Cancer.

PURPOSE: Men with metastatic castration-resistant prostate cancer (mCRPC) frequently develop resistance to androgen receptor signaling inhibitor (ARSI) treatment; therefore, new therapies are needed. Trophoblastic cell-surface antigen (TROP-2) is a transmembrane protein identified in prostate cancer and overexpressed in multiple malignancies. TROP-2 is a therapeutic target for antibody-drug conjugates (ADC). EXPERIMENTAL DESIGN: TROP-2 gene (TACSTD2) expression and markers of treatment resistance from prostate biopsies were analyzed using data from four previously curated cohorts of mCRPC (n = 634) and the PROMOTE study (dbGaP accession phs001141.v1.p1, n = 88). EPCAM or TROP-2-positive circulating tumor cells (CTC) were captured from peripheral blood for comparison of protein (n = 15) and gene expression signatures of treatment resistance (n = 40). We assessed the efficacy of TROP-2-targeting agents in a mouse xenograft model generated from prostate cancer cell lines. RESULTS: We demonstrated that TACSTD2 is expressed in mCRPC from luminal and basal tumors but at lower levels in patients with neuroendocrine prostate cancer. Patients previously treated with ARSI showed no significant difference in TACSTD2 expression, whereas patients with detectable AR-V7 expression showed increased expression. We observed that TROP-2 can serve as a cell surface target for isolating CTCs, which may serve as a predictive biomarker for ADCs. We also demonstrated that prostate cancer cell line xenografts can be targeted specifically by labeled anti-TROP-2 agents in vivo. CONCLUSIONS: These results support further studies on TROP-2 as a therapeutic and diagnostic target for mCRPC.

Dynamic assessment of serum chromogranin A and treatment response with abiraterone acetate in metastatic castration-resistant prostate cancer.

OBJECTIVE: Elevated serum chromogranin A (CGA) is associated with intrinsic or treatment-related neuroendocrine differentiation (NED) in men with metastatic castration-resistant prostate cancer (mCRPC). Fluctuations in serum CGA during treatment of mCRPC have had conflicting results. We analyzed the impact of (i) rising serum CGA and (ii) baseline CGA/PSA ratio during treatment to identify associations with abiraterone acetate (AA) therapy. METHODS: Between June 2013 and August 2015, 92 men with mCRPC were enrolled in a prospective trial with uniform serum CGA processing performed before initiating abiraterone acetate/prednisone (AA/P) and serially after 12 weeks of AA/P treatments. Serum CGA was measured using a homogenous automated immunofluorescent assay. Patients receiving proton pump inhibitors or with abnormal renal function were excluded due to possible false elevations of serum CGA (n = 21 excluded), therefore 71 patients were analyzed. All patients underwent a composite response assessment at 12-weeks. Kaplan-Meier estimates and Cox Regression models were used to calculate the association with time-to-treatment failure analyses and overall survival. RESULTS: An increase in chromogranin was associated with a lower risk of treatment failure (hazard ratio [HR]: 0.52, p = 0.0181). The median CGA/PSA ratio was 7.8 (2.6-16.0) and an elevated pretreatment CGA/PSA ratio above the median was associated with a lower risk of treatment failure (HR: 0.54 p value = 0.0185). An increase in CGA was not found to be associated with OS (HR: 0.71, 95% CI: 0.42-1.21, p = 0.207). An elevated baseline CGA/PSA ratio was not associated with OS (HR: 0.62, 95% CI: 0.37-1.03, p = 0.062). An increase in PSA after 12 weeks of treatment was associated with an increased risk of treatment failure (HR: 4.14, CI: 2.21-7.73, p = < 0.0001) and worse OS (HR: 2.93, CI: 1.57-4.45, p = < 0.0001). CONCLUSIONS: We show that an increasing chromogranin on AA/P and an elevated baseline CGA/PSA in patients with mCRPC were associated with a favorable response to AA/P with no changes in survival. There may be limited clinical utility in serum CGA testing to evaluate for lethal NED as AA/P did not induce lethal NED in this cohort. This highlights that not all patients with an increasing CGA have a worse OS.

SLFN5-mediated chromatin dynamics sculpt higher-order DNA repair topology.

Repair of DNA double-strand breaks (DSBs) elicits three-dimensional (3D) chromatin topological changes. A recent finding reveals that 53BP1 assembles into a 3D chromatin topology pattern around DSBs. How this formation of a higher-order structure is configured and regulated remains enigmatic. Here, we report that SLFN5 is a critical factor for 53BP1 topological arrangement at DSBs. Using super-resolution imaging, we find that SLFN5 binds to 53BP1 chromatin domains to assemble a higher-order microdomain architecture by driving damaged chromatin dynamics at both DSBs and deprotected telomeres. Mechanistically, we propose that 53BP1 topology is shaped by two processes: (1) chromatin mobility driven by the SLFN5-LINC-microtubule axis and (2) the assembly of 53BP1 oligomers mediated by SLFN5. In mammals, SLFN5 deficiency disrupts the DSB repair topology and impairs non-homologous end joining, telomere fusions, class switch recombination, and sensitivity to poly (ADP-ribose) polymerase inhibitor. We establish a molecular mechanism that shapes higher-order chromatin topologies to safeguard genomic stability.

Preclinical-to-clinical Anti-cancer Drug Response Prediction and Biomarker Identification Using TINDL.

Prediction of the response of cancer patients to different treatments and identification of biomarkers of drug response are two major goals of individualized medicine. Here, we developed a deep learning framework called TINDL, completely trained on preclinical cancer cell lines (CCLs), to predict the response of cancer patients to different treatments. TINDL utilizes a tissue-informed normalization to account for the tissue type and cancer type of the tumors and to reduce the statistical discrepancies between CCLs and patient tumors. Moreover, by making the deep learning black box interpretable, this model identifies a small set of genes whose expression levels are predictive of drug response in the trained model, enabling identification of biomarkers of drug response. Using data from two large databases of CCLs and cancer tumors, we showed that this model can distinguish between sensitive and resistant tumors for 10 (out of 14) drugs, outperforming various other machine learning models. In addition, our small interfering RNA (siRNA) knockdown experiments on 10 genes identified by this model for one of the drugs (tamoxifen) confirmed that tamoxifen sensitivity is substantially influenced by all of these genes in MCF7 cells, and seven of these genes in T47D cells. Furthermore, genes implicated for multiple drugs pointed to shared mechanism of action among drugs and suggested several important signaling pathways. In summary, this study provides a powerful deep learning framework for prediction of drug response and identification of biomarkers of drug response in cancer. The code can be accessed at https://github.com/ddhostallero/tindl.

Bayesian Machine Learning Enables Identification of Transcriptional Network Disruptions Associated with Drug-Resistant Prostate Cancer.

Survival rates of patients with metastatic castration-resistant prostate cancer (mCRPC) are low due to lack of response or acquired resistance to available therapies, such as abiraterone (Abi). A better understanding of the underlying molecular mechanisms is needed to identify effective targets to overcome resistance. Given the complexity of the transcriptional dynamics in cells, differential gene expression analysis of bulk transcriptomics data cannot provide sufficient detailed insights into resistance mechanisms. Incorporating network structures could overcome this limitation to provide a global and functional perspective of Abi resistance in mCRPC. Here, we developed TraRe, a computational method using sparse Bayesian models to examine phenotypically driven transcriptional mechanistic differences at three distinct levels: transcriptional networks, specific regulons, and individual transcription factors (TF). TraRe was applied to transcriptomic data from 46 patients with mCRPC with Abi-response clinical data and uncovered abrogated immune response transcriptional modules that showed strong differential regulation in Abi-responsive compared with Abi-resistant patients. These modules were replicated in an independent mCRPC study. Furthermore, key rewiring predictions and their associated TFs were experimentally validated in two prostate cancer cell lines with different Abi-resistance features. Among them, ELK3, MXD1, and MYB played a differential role in cell survival in Abi-sensitive and Abi-resistant cells. Moreover, ELK3 regulated cell migration capacity, which could have a direct impact on mCRPC. Collectively, these findings shed light on the underlying transcriptional mechanisms driving Abi response, demonstrating that TraRe is a promising tool for generating novel hypotheses based on identified transcriptional network disruptions. SIGNIFICANCE: The computational method TraRe built on Bayesian machine learning models for investigating transcriptional network structures shows that disruption of ELK3, MXD1, and MYB signaling cascades impacts abiraterone resistance in prostate cancer.