Comprehensive machine learning-based preoperative blood features predict the prognosis for ovarian cancer.

PURPOSE: Significant advancements in improving ovarian cancer (OC) outcomes have been limited over the past decade. To predict prognosis and improve outcomes of OC, we plan to develop and validate a robust prognosis signature based on blood features. METHODS: We screened age and 33 blood features from 331 OC patients. Using ten machine learning algorithms, 88 combinations were generated, from which one was selected to construct a blood risk score (BRS) according to the highest C-index in the test dataset. RESULTS: Stepcox (both) and Enet (alpha = 0.7) performed the best in the test dataset with a C-index of 0.711. Meanwhile, the low RBS group possessed observably prolonged survival in this model. Compared to traditional prognostic-related features such as age, stage, grade, and CA125, our combined model had the highest AUC values at 3, 5, and 7 years. According to the results of the model, BRS can provide accurate predictions of OC prognosis. BRS was also capable of identifying various prognostic stratifications in different stages and grades. Importantly, developing the nomogram may improve performance by combining BRS and stage. CONCLUSION: This study provides a valuable combined machine-learning model that can be used for predicting the individualized prognosis of OC patients.

Clinical significance and immune infiltration analyses of a novel coagulation-related signature in ovarian cancer.

Ovarian cancer (OV) is the most lethal gynecological malignancies worldwide. The coagulation cascade could induce tumor cell infiltration and contribute to OV progression. However, coagulation-related gene (CRG) signature for OV prognosis hasn't been determined yet. In this study, we evaluated the prognostic value of coagulation scores through receiver operating characteristics (ROC) analysis and K-M curves, among OV patients at our institution. Based on the transcriptome data of TCGA-OV cohort, we stratified two coagulation-related subtypes with distinct differences in prognosis and tumor immune microenvironment (p < 0.05). Moreover, from the 6406 differentially-expressed genes (DEGs) between the GTEx (n = 180) and TCGA-OV cohorts (n = 376), we identified 138 potential CRGs. Through LASSO-Cox algorithm, we finally distinguished a 3-gene signature (SERPINA10, CD38, and ZBTB16), with promising prognostic ability in both TCGA (p < 0.001) and ICGC cohorts (p = 0.040). Stepwise, we constructed a nomogram based on the clinical features and coagulation-related signature for overall survival prediction, with the C-index of 0.6761, which was evaluated by calibration curves. Especially, based on tissue microarrays analysis, Quantitative real-time fluorescence PCR (qRT-PCR), and Western Blot, we found that aberrant upregulation of CRGs was related to poor prognosis in OV at both mRNA and protein level (p < 0.05). Collectively, the coagulation-related signature was a robust prognostic biomarker, which could provide therapeutic benefits for chemotherapy/immunotherapy and assist clinical decision in OV patients.

Treatment of Retinoblastoma 1-Intact Hepatocellular Carcinoma With Cyclin-Dependent Kinase 4/6 Inhibitor Combination Therapy.

BACKGROUND AND AIMS: Synthetic cyclin-dependent kinase (CDK) 4/6 inhibitors exert antitumor effects by forcing RB1 in unphosphorylated status, causing not only cell cycle arrest but also cellular senescence, apoptosis, and increased immunogenicity. These agents currently have an indication in advanced breast cancers and are in clinical trials for many other solid tumors. HCC is one of promising targets of CDK4/6 inhibitors. RB family dysfunction is often associated with the initiation of HCC; however, this is revivable, as RB family members are not frequently mutated or deleted in this malignancy. APPROACH AND RESULTS: Loss of all Rb family members in transformation related protein 53 (Trp53)-/- mouse liver resulted in liver tumor reminiscent of human HCC, and re-expression of RB1 sensitized these tumors to a CDK4/6 inhibitor, palbociclib. Introduction of an unphosphorylatable form of RB1 (RB7LP) into multiple liver tumor cell lines induced effects similar to palbociclib. By screening for compounds that enhance the efficacy of RB7LP, we identified an I kappa B kinase (IKK)ß inhibitor Bay 11-7082. Consistently, RB7LP expression and treatment with palbociclib enhanced IKKα/ß phosphorylation and NF-κB activation. Combination therapy using palbociclib with Bay 11-7082 was significantly more effective in hepatoblastoma and HCC treatment than single administration. Moreover, blockade of IKK-NF-κB or AKT pathway enhanced effects of palbociclib on RB1-intact KRAS Kirsten rat sarcoma viral oncogene homolog mutated lung and colon cancers. CONCLUSIONS: In conclusion, CDK4/6 inhibitors have a potential to treat a wide variety of RB1-intact cancers including HCC when combined with an appropriate kinase inhibitor.

Identification of a Prognostic Signature for Ovarian Cancer Based on Ubiquitin-Related Genes Suggesting a Potential Role for FBXO9.

BACKGROUND: Ovarian cancer (OV) is associated with high mortality and poses challenges in diagnosis and prognosis prediction. Ubiquitin-related genes (UbRGs) are involved in the initiation and progression of cancers, but have still not been utilized for diagnosis and prognosis of OV. METHODS: K48-linked ubiquitination in ovarian tissues from our OV and control cohort was assessed using immunohistochemistry. UbRGs, including ubiquitin and ubiquitin-like regulators, were screened based on the TCGA-OV and GTEx database. Univariate Cox regression analysis identified survival-associated UbRGs. A risk model was established using the LASSO regression and multivariate Cox regression analysis. The relationship between UbRGs and immune cell infiltration, tumor mutational burden, drug sensitivity, and immune checkpoint was determined using the CIBERSORT, ESTIMATE, and Maftools algorithms, based on the Genomics of Drug Sensitivity in Cancer and TCGA-OV databases. GEPIA2.0 was used to analyze the correlation between FBXO9/UBD and DNA damage repair-related genes. Finally, FBXO9 and UBD were accessed in tissues or cells using immunohistochemistry, qPCR, and Western blot. RESULTS: We confirmed the crucial role for ubiquitination in OV as a significant decrease of K48-linked ubiquitination was observed in primary OV lesions. We identified a prognostic signature utilizing two specific UbRGs, FBXO9 and UBD. The risk score obtained from this signature accurately predicted the overall survival of TCGA-OV training dataset and GSE32062 validation dataset. Furthermore, this risk score also showed association with immunocyte infiltration and drug sensitivity, revealing potential mechanisms for ubiquitination mediated OV risk. In addition, FBXO9, but not UBD, was found to be downregulated in OV and positively correlated with DNA damage repair pathways, suggesting FBXO9 as a potential cancer suppressor, likely via facilitating DNA damage repair. CONCLUSIONS: We identified and validated a signature of UbRGs that accurately predicts the prognosis, offers valuable guidance for optimizing chemotherapy and targeted therapies, and suggests a potential role for FBXO9 in OV.

CSGALNACT2 restricts ovarian cancer migration and invasion by modulating MAPK/ERK pathway through DUSP1.

PURPOSE: Ovarian cancer is one of the leading causes of cancer-related death among women. CSGALNACT2 is a vital Golgi transferase and is related to a variety of human diseases. However, its expression pattern and function in ovarian cancer remain uncertain. METHODS: The Cancer Genome Atlas and GEPIA databases were used to assess the expression of CSGALNACT2 in ovarian cancer patients. RNA-seq, qRT-PCR, and IHC were used to verify the expression of CSGALNACT2 in ovarian cancer tissues. Then, in vivo and in vitro experiments were conducted to evaluate the role of CSGALNACT2 in the progression of ovarian cancer. RNA-seq and GSEA were used to reveal the potential biological function and oncogenic pathways of CSGALNACT2. RESULTS: We demonstrated that the mRNA expression and protein level of CSGALNACT2 were significantly downregulated in ovarian cancer and ovarian cancer metastatic tissues. CSGALNACT2 can significantly inhibit the migration, invasion, and clonogenic growth of ovarian cancer in vitro and is progressively lost during ovarian cancer progression in vivo. CSGALNACT2 suppresses ovarian cancer migration and invasion via DUSP1 modulation of the MAPK/ERK pathway through RNA-seq, KEGG analysis, and Western blotting. Moreover, CSGALNACT2 expression was correlated with immune cell infiltration and had prognostic value in different immune cell-enriched or decreased ovarian cancer. In addition, patients with CSGALNACT2 downregulation are less likely to benefit from immunotherapy. CONCLUSION: As an ovarian cancer suppressor gene, CSGALNACT2 inhibits the development of ovarian cancer, and it might be used as a prognostic biomarker in patients with ovarian cancer.

Targeting RB1 Loss in Cancers.

Retinoblastoma protein 1 (RB1) is encoded by a tumor suppressor gene that was discovered more than 30 years ago. Almost all mitogenic signals promote cell cycle progression by braking on the function of RB1 protein through mono- and subsequent hyper-phosphorylation mediated by cyclin-CDK complexes. The loss of RB1 function drives tumorigenesis in limited types of malignancies including retinoblastoma and small cell lung cancer. In a majority of human cancers, RB1 function is suppressed during tumor progression through various mechanisms. The latter gives rise to the acquisition of various phenotypes that confer malignant progression. The RB1-targeted molecules involved in such phenotypic changes are good quarries for cancer therapy. Indeed, a variety of novel therapies have been proposed to target RB1 loss. In particular, the inhibition of a number of mitotic kinases appeared to be synthetic lethal with RB1 deficiency. A recent study focusing on a neighboring gene that is often collaterally deleted together with RB1 revealed a pharmacologically targetable vulnerability in RB1-deficient cancers. Here we summarize current understanding on possible therapeutic approaches targeting functional or genomic aberration of RB1 in cancers.