A Six-Gene Prognostic and Predictive Radiotherapy-Based Signature for Early and Locally Advanced Stages in Non-Small-Cell Lung Cancer.

Non-small-cell lung cancer (NSCLC) is the leading cause of cancer death worldwide, generating an enormous economic and social impact that has not stopped growing in recent years. Cancer treatment for this neoplasm usually includes surgery, chemotherapy, molecular targeted treatments, and ionizing radiation. The prognosis in terms of overall survival (OS) and the disparate therapeutic responses among patients can be explained, to a great extent, by the existence of widely heterogeneous molecular profiles. The main objective of this study was to identify prognostic and predictive gene signatures of response to cancer treatment involving radiotherapy, which could help in making therapeutic decisions in patients with NSCLC. To achieve this, we took as a reference the differential gene expression pattern among commercial cell lines, differentiated by their response profile to ionizing radiation (radiosensitive versus radioresistant lines), and extrapolated these results to a cohort of 107 patients with NSCLC who had received radiotherapy (among other therapies). We obtained a six-gene signature (APOBEC3B, GOLM1, FAM117A, KCNQ1OT1, PCDHB2, and USP43) with the ability to predict overall survival and progression-free survival (PFS), which could translate into a prediction of the response to the cancer treatment received. Patients who had an unfavorable prognostic signature had a median OS of 24.13 months versus 71.47 months for those with a favorable signature, and the median PFS was 12.65 months versus 47.11 months, respectively. We also carried out a univariate analysis of multiple clinical and pathological variables and a bivariate analysis by Cox regression without any factors that substantially modified the HR value of the proposed gene signature.

Comparison of Feature Selection Methods for Predicting RT-Induced Toxicity.

This work addresses a scoping review of Feature Selection (FS) methods applied to a Lung Cancer dataset to elucidate parameters' relevance when predicting radiotherapy (RT) induced toxicity. Subsetting-based and Ranking-based FS methods were implemented along with 4 advanced classifiers to predict the onset of RT-induced acute esophagitis, cough, pneumonitis and dyspnea. Their prediction performance was measured in terms of the AUC for each model to find the best FS.

High casein kinase 1 epsilon levels are correlated with better prognosis in subsets of patients with breast cancer.

Reliable biological markers that predict breast cancer (BC) outcomes after multidisciplinary therapy have not been fully elucidated. We investigated the association between casein kinase 1 epsilon (CK1ε) and the risk of recurrence in patients with BC. Using 168 available tumor samples from patients with BC treated with surgery +/- chemo(radio)therapy, we scored the CK1ε expression as high (≥ 1.5) or low (<1.5) using an immunohistochemical method. Kaplan-Meier analysis was performed to assess the risk of relapse, and Cox proportional hazards analyses were utilized to evaluate the effect of CK1ε expression on this risk. The median age at diagnosis was 60 years (range 35-96). A total of 58% of the patients underwent breast conservation surgery, while 42% underwent mastectomy. Adjuvant chemotherapy and radiation therapy were administered in 101 (60%) and 137 cases (82%), respectively. Relapse was observed in 24 patients (14%). Multivariate analysis found high expression of CK1ε to be associated with a statistically significant higher disease-free survival (DFS) in BC patients with wild-type p53 (Hazard ratio [HR] = 0.33; 95% CI, 0.12-0.91; P = 0.018) or poor histological differentiation ([HR] = 0.34; 95% CI, 0.12-0.94; P = 0.039) or in those without adjuvant chemotherapy ([HR] = 0.11; 95% CI, 0.01-0.97; P = 0.006). Our data indicate that CK1ε expression is associated with DFS in BC patients with wild-type p53 or poor histological differentiation or in those without adjuvant chemotherapy and thus may serve as a predictor of recurrence in these subsets of patients.