BMS-754807

Drug combination therapies are well-established strategies for the treatment of cancer with low toxicity and fewer adverse effects. Computational drug synergy prediction approaches can accelerate the discovery of novel combination therapies, but the existing methods do not explicitly consider the key role of important substructures in producing synergistic effects. To this end, we propose a significant substructure-aware anticancer drug synergy prediction method, named SDDSynergy, to adaptively identify critical functional groups in drug synergy. SDDSynergy splits the task of predicting drug synergy into predicting the effect of individual substructures on cancer cell lines and highlights the impact of important substructures through a novel drug-cell line attention mechanism. And a substructure pair attention mechanism is incorporated to capture the information on internal substructure pairs interaction in drug combinations, which aids in predicting synergy. The substructures of different sizes and shapes are directly obtained from the molecular graph of the drugs by multilayer substructure information passing networks. Extensive experiments on three real-world data sets demonstrate that SDDSynergy outperforms other state-of-the-art methods. We also verify that many of the novel drug combinations predicted by SDDSynergy are supported by previous studies or clinical trials through an in-depth literature survey.

Breast cancer (BC), a widespread and lethal neoplasm, is irrespective of the subtype of BC. Metastasis remains a crucial determinant for unfavorable outcome. The identification of novel diagnostic markers is instrumental in optimizing the treatment regime for BC. The direct correlation between anoikis and the progression/outcome of BC is well established. Nevertheless, the contribution of anoikis-related genes (ARGs) in BC remains obscure at present. We implemented the METABRIC dataset to scrutinize and assess differentially expressed ARGs in BC versus healthy breast tissues. An unsupervised consensus clustering approach for ARGs was employed to classify patients into diverse subtypes. ESTIMATE algorithms were utilized to assess immune infiltrative patterns. Prognostic gene expression patterns were derived from LASSO regression and univariate COX regression analysis. Subsequently, these signatures underwent examination via use of the Kaplan-Meier survival curve. 6 pairs of fresh tissue specimens (tumor and adjacent non-tumor) were employed to assess the expression of 7 ARGs genes via qPCR. Notably, DCN and FOS were not expressed in BC tissue, which had been excluded in our subsequent experiments. Also, among remaining 5 ARGs, solely the expression of ADH1A demonstrated a statistically remarkable disparity between freshly collected cancer tissues and the adjacent ones. ADH1A-overexpressed and ADH1A-sh vectors were transfected into MCF-7 and MCF-7-AR cell lines, respectively. The expression status of FABP4, CALML5, ADH1A, C1orf106, CIDEC, β-catenin, N-cadherin, and Vimentin in the clinical samples were scrutinized using RT-qPCR and western blotting techniques. Migration and invasion through transwell chambers were employed to assess the migratory and invasive potential of the cells. Detailed evaluation of cell proliferation was conducted utilizing a Cell Counting Kit-8 (CCK-8) assay. The apoptotic index of the cells was determined by flow cytometry analysis. An innovative anoikis-associated signature consisting of seven genes, namely ADH1A, DCN, CIEDC, FABP4, FOS, CALML5, and C1orf106, was devised to stratify BC patients into high- and low-risk cohorts. This unique risk assessment model, formulated via the distinctive signature approach, has been validated as an independent prognostic indicator. Additional analysis demonstrated that distinct risk subtypes manifested variances in the tumor microenvironment and drug sensitivities. Suppression of ADH1A enhanced the migratory and invasive capacities and reduced these tumorigenesis-related protein levels, underscoring the prognostic role of ADH1A in the progression of BC. Through our meticulous study, we have elucidated the possible molecular markers and clinical implications of ARGs in BC. Our model, which incorporate seven ARGs, has proven to accurately forecast the survival outcomes of BC patients. Moreover, the thorough molecular study of ADH1A has augmented our comprehension of ARGs in BC and opened a novel avenue for guiding personalized and precise therapeutic interventions for BC patients.