No abstract

Background: Cervical cancer (CC) is a major health threat to females, and distal metastasis is common in patients with advanced CC. Anoikis is necessary for the development of distal metastases. Understanding the mechanisms associated with anoikis in CC is essential to improve its survival rate.Methods: The expression matrix of long non-coding RNAs (lncRNAs) from cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC) patients was extracted from The Cancer Genome Atlas (TCGA), and highly relevant anoikis-related lncRNAs (ARLs) were identified by the single sample gene set enrichment analysis (ssGSEA) method. ARLs-related molecular subtypes were discerned based on prognosis-related ARLs. ARLs-related prognostic risk score (APR_Score) was calculated and risk model was constructed using LASSO COX and COX models. In addition, we also assessed immune cell activity in the immune microenvironment (TME) for both subtypes and APR_Score groups. A nomogram was utilized for predicting improved clinical outcome. Finally, this study also discussed the potential of ARLs-related signatures in predicting response to immunotherapy and small molecular drugs.Results: Three ARLs-related subtypes were identified from TCGA-CESC (AC1, AC2, and AC3), with AC3 patients having the highest ARG scores, higher angiogenesis scores, and the worst prognosis. AC3 had lower immune cell scores in TME but higher immune checkpoint gene expression and higher potential for immune escape. Next, we constructed a prognostic risk model consisting of 7-ARLs. The APR_Score exhibited a greater robustness as an independent prognostic indicator in predicting prognosis, and the nomogram was a valuable tool for survival prediction. ARLs-related signatures emerged as a potential novel indicator for immunotherapy and small molecular drug selection.Conclusion: We firstly constructed novel ARLs-related signatures capable of predicting prognosis and offered novel ideas for therapy response in CC patients.

Microsatellite instability (MSI) has emerged as an important predictor of sensitivity for immunotherapy-based strategies. β-2-Microglobulin (B2M) contains microsatellites within the coding regions and is prone to somatic changes in MSI/mismatch repair deficiency (MSI/dMMR) tumors. To delineate prevalence and associations of B2M mutations in MSI-H/dMMR cancers, we investigated the mutational profile of B2M and clinical and pathological features in gastric cancer (GC), colorectal cancer (CRC), and endometrial cancer (EC) with a high incidence of microsatellite instability-high (MSI-H)/dMMR. Formalin-fixed paraffin-embedded (FFPE) tumor tissues along with matched normal tissues were collected from 108 MSI/dMMR patients with GC, CRC, and EC. Genomic profiling of tissue and blood samples were assessed next-generation sequencing (NGS). Immunohistochemistry (IHC) was used to examine the presence or absence of B2M protein. Alternations in the exonic microsatellite regions of B2M were observed at various but high frequencies (57.5% in CRC, 23.9% in GC, and 13.6% in EC) and in different forms. NGS assay revealed that genes involved in chromatin regulation, the PI3K pathway, the WNT pathway, and mismatch repair were extensively altered in the MSI-H cohort. Signature 6 and 26, 2 of 4 mutational signatures associated with defective DNA mismatch repair, featured with high numbers of small insertion/deletions (INDEL) dominated in all 3 types of cancer. Alternations in the exonic microsatellite regions of B2M were observed at various but high frequencies (57.5% in CRC, 23.9% in GC, and 13.6% in EC) and in different forms. Tumor mutational burden (TMB) was significantly higher in the patients carrying MSI-H/dMMR tumors with B2M mutation than that in patients with wild-type B2M (P = .026).The frame shift alteration occurring at the exonic microsatellite sties caused loss of function of B2M gene. In addition, a case with CRC carrying indels in B2M gene resisted the ICI treatment was reported. In conclusion, patients carrying MSI-H/dMMR tumors with B2M mutation showed significantly higher TMB. Prescription of ICIs should be thoroughly evaluated for these patients.

Background Breast cancer is the most common malignancy in women worldwide, which seriously threatens women's physical and mental health, but currently, there is no classification method for tumor samples based on gene expression profiles for faster breast cancer diagnosis. The study aimed to establish a novel genetic model to distinguish breast cancer patients from the normal population.Results We utilized published expression profiles of breast cancer patients (GSE15852, GSE70905) to identify potential predictive gene panels. A total of 7 differentially expressed genes were identified as predictors. Random forest algorithm and artificial neural network were applied to screen the predictive features and build a model to predict breast cancer. In parallel, we validated this prediction model using expression profiling of a completely independent set of breast cancer patients(GSE70947). The new model was successfully built based on the molecular prognostic scoring system and showed significant predictive value in the training group (AUC = 0.991), which was simultaneously validated in an independent dataset (AUC = 0.817).Conclusions Random forest algorithm combined with artificial neural network successfully constructed a prediction model for breast cancer. The new model can predict breast cancer patients, which is helpful for the diagnosis of breast cancer in the clinic.

Background. Cervical cancer (CC) is one of the most frequent female malignancy. Cancer stem cells (CSCs) positively affect survival outcomes in cancer patients, but in cervical cancer, the mechanism of tumor stem cells is still uncertain. Methods. RNA-seq data and related clinical follow-up of patients suffering from CC were from TCGA. Consensus clustering screened prognostic mRNAsi-related genes and identified molecular subtypes for CC. Based on the overlapping differentially expressed genes (DEGs) in subtypes, we employed LASSO and multivariate Cox regression to screen prognostic-related genes and established the RiskScore system. The patients were grouped by RiskScore, the prognosis was analyzed by the Kaplan-Meier (K-M) curve among the various groups, and the precision of the RiskScore was assessed by the ROC curve. Finally, the potential worth of RiskScore in immunotherapy/chemotherapy response was assessed by evaluating TIDE scores and chemotherapy drug IC 50 values. Results. We noticed that patients with low mRNAsi had a shorter survival and then identified three molecular subtypes (C1-3), with the C1 having the worst prognosis and the lowest mRNAsi. Finally, we identified 7 prognostic-related genes (SPRY4, PPP1R14A, MT1A, DES, SEZ6L2, SLC22A3, and CXCL8) via LASSO and Cox regression analysis. We established a 7-gene model defined RiskScore to predict the prognosis of CC patients. K-M curve indicated that low RiskScore patients had improved prognosis, and ROC curves indicated that RiskScore could precisely direct the prognostic evaluation for those suffering from the cancer. This was also confirmed in the GSE44001 and GSE52903 external cohorts. Patients were more sensitive to immunotherapy if with low RiskScore, and RiskScore exhibited precise assessment ability in predicting response to immunological therapy in CC patients. Conclusion. CC stemness is associated with patient prognosis, and the RiskScore constructed based on stemness characteristics is an independent prognostic index, which is expected to be a guide for immunotherapy, providing a new idea for CC clinical practice.