Cancer Survival Analysis Research Articles

Evaluation of antioxidant network proteins as novel prognostic biomarkers for head and neck cancer patients.

Recurrence rates for head and neck squamous cell carcinoma (HNSCC) approach 50% at 5years. Current staging fails to identify patients with a worse prognosis who might benefit from intensified treatment, which warrants improved prognostic biomarkers. The purpose of this retrospective case study is to identify potential prognostic biomarkers in patients with HNSCC including APE1 (DNA repair/redox gene regulator), NRF2 and PPARGC1A (redox gene regulators), SOD3 and DCN (antioxidant proteins). Differential protein expression between benign, carcinoma in situ (CIS), and invasive HNSCC tissue specimens from 77 patients was assessed using immunohistochemistry. Protein expression was analyzed with multivariate, pair-wise, and Kaplan-Meier survival analyses to identify potential prognostic biomarkers. Utilizing The Cancer Genome Atlas's transcriptome database, pair-wise and survival analysis was performed to identify potential prognostic biomarkers. APE1, NRF2, PPARGC1A, SOD3, and DCN expression in HNSCC in relation to, lymph node invasion, and patient survival were examined. Elevated APE1 protein expression in CIS corresponded with reduced survival (p=0.0243). Increased APE1 gene expression in stage T4a HNSCC was associated with reduced patient survival (p<0.015). Increased PPARGC1A in invasive tumor correlated with reduced survival (p=0.0281). Patients with lymph node invasion at diagnosis had significantly increased APE1 protein in the primary sites (p<0.05). Patients with poorly differentiated invasive tumors had reduced PPARGC1A in CIS proximal to the invasive tumor and had elevated DCN and SOD3 in proximal benign tissue (p<0.05). The expression of APE1, DCN, and SOD3 is a potential prognostic signature that identifies patients with worsened survival.

Quantitative assessment of cardiovascular autonomic impairment in cancer survivors: a single center case series

BackgroundCardiovascular autonomic dysfunction in cancer survivors is poorly understood.ObjectivesTo better characterize the clinical characteristics and types of autonomic dysfunction in this population.MethodsA retrospective analysis of cancer survivors within an academic cardio-oncology program referred for suspected autonomic dysfunction was performed. Autonomic reflex testing of adrenergic, cardiovagal, and sudomotor function was done. Autonomic impairment was graded on severity based on the Composite Autonomic Severity Score system. Patients with pre-existing autonomic dysfunction prior to their cancer diagnosis were excluded.ResultsOf approximately 282 total patients in the UCLA Cardio-Oncology program, 24 were referred for suspected autonomic dysfunction and met the inclusion criteria. 22 had autonomic impairment on autonomic reflex testing. Eight patients were female, and the mean age at time of autonomic testing was 51.3 years. The average duration from cancer diagnosis to autonomic testing was 10.3 years. The reasons for referral included dizziness, tachycardia, palpitations, and syncope. The majority of patients (75%) had hematologic disorders. The most common chemotherapies administered were vinca alkaloids (54.2%), alkylating agents (66.7%), and anthracyclines (54.2%). Most patients received radiation to the thorax (66.7%) and neck (53.3%). Eleven patients had mild autonomic impairment, 7 had moderate, and 4 had severe autonomic impairment. Dysfunction was commonly present in the sympathetic and parasympathetic branches, but most pronounced in the sympathetic system. The majority of patients were diagnosed with orthostatic hypotension (50%), inappropriate sinus tachycardia (20.8%), and postural orthostatic tachycardia syndrome (12.5%) and had subjective improvement with treatment.ConclusionCardiovascular autonomic dysfunction occurs in cancer survivors, and commonly affects both the sympathetic and parasympathetic systems. Symptom recognition in patients should prompt autonomic testing and treatment where appropriate.

Semiparametric estimation of the cure fraction in population-based cancer survival analysis.

With rapid development in medical research, the treatment of diseases including cancer has progressed dramatically and those survivors may die from causes other than the one under study, especially among elderly patients. Motivated by the Surveillance, Epidemiology, and End Results (SEER) female breast cancer study, background mortality is incorporated into the mixture cure proportional hazards (MCPH) model to improve the cure fraction estimation in population-based cancer studies. Here, that patients are "cured" is defined as when the mortality rate of the individuals in diseased group returns to the same level as that expected in the general population, where the population level mortality is presented by the mortality table of the United States. The semiparametric estimation method based on the EM algorithm for the MCPH model with background mortality (MCPH+BM) is further developed and validated via comprehensive simulation studies. Real data analysis shows that the proposed semiparametric MCPH+BM model may provide more accurate estimation in population-level cancer study.

Survival analysis in gastric cancer: a multi-center study among Iranian patients

BackgroundGastric cancer (GC) has been considered as the 5th most common type of cancer and the third leading cause of cancer-associated death worldwide. The aim of this historical cohort study was to evaluate the survival predictors for all patients with GC using the Cox proportional hazards, extended Cox, and gamma-frailty models.MethodsThis historical cohort study was performed according to documents of 1695 individuals having GC referred to three medical centers in Iran from 2001 to 2018. First, most significant prognostic risk factors on survival were selected, Cox proportional hazards, extended Cox, gamma-frailty models were applied to evaluate the effects of the risk factors, and then these models were compared with the Akaike information criterion.ResultsThe age of patients, body mass index (BMI), tumor size, type of treatment and grade of the tumor increased the hazard rate (HR) of GC patients in both the Cox and frailty models (P < 0.05). Also, the size of the tumor and BMI were considered as time-varying variables in the extended Cox model. Moreover, the frailty model showed that there is at least an unknown factor, genetic or environmental factors, in the model that is not measured (P < 0.05).ConclusionsSome prognostic factors, including age, tumor size, the grade of the tumor, type of treatment and BMI, were regarded as indispensable predictors in patients of GC. Frailty model revealed that there are unknown or latent factors, genetic and environmental factors, resulting in the biased estimates of the regression coefficients.

Construction and analysis of a lncRNA-miRNA-mRNA network based on competitive endogenous RNA reveal functional lncRNAs in oral cancer

BackgroundA growing evidence suggests that long non-coding RNAs (lncRNAs) can function as a microRNA (miRNA) sponge in various diseases including oral cancer. However, the pathophysiological function of lncRNAs remains unclear.MethodsBased on the competitive endogenous RNA (ceRNA) theory, we constructed a lncRNA-miRNA-mRNA network in oral cancer with the human expression profiles GSE74530 from the Gene Expression Omnibus (GEO) database. We used topological analysis to determine the hub lncRNAs in the regulatory ceRNA network. Then, function enrichment analysis was performed using the clusterProfiler R package. Clinical information was downloaded from The Cancer Genome Atlas (TCGA) database and survival analysis was performed with Kaplan-Meier analysis.ResultsA total of 238 potential co-dysregulated competing triples were obtained in the lncRNA-associated ceRNA network in oral cancer, which consisted of 10 lncRNA nodes, 41 miRNA nodes and 122 mRNA nodes. Additionally, we found lncRNA HCG22 exhibiting superior potential as a diagnostic and prognostic marker of oral cancer.ConclusionsOur findings provide novel insights to understand the ceRNA regulation in oral cancer and identify a novel lncRNA as a potential molecular biomarker.

Abstract C023: Cervical cancer survival analysis based on screening practices and the socioeconomic position index in Puerto Rico

Abstract Background: Cervical cancer ranks as the 7th most diagnosed cancer among women in Puerto Rico (PR). When compared to all the states and territories of the United States of America (USA), PR has the highest rate of cervical cancer. Cervical cancer is preventable through screening. While the Healthy People’s 2020 goal is a screening rate of 93%, PR falls short at 77%. There is extensive research on how women with screen-detected cancers have better survival than those whose cancer is detected symptomatically. Cancer stage is the primary prognostic factor affecting survival. Yet, stage at diagnosis is a complex interaction among many factors, including social and economic inequities. Evaluating how socioeconomic status and screening behavior affect survival will provide insight into how disparities affect health outcomes. Objective: The purpose of this study is to evaluate the effect of screening behavior and socioeconomic position on the survivorship of cervical cancer patients diagnosed in PR during the period of 2011-2014. Methods: The Puerto Rico Central Cancer Registry (PRCCR) was used to identify all incident cases of primary cervical cancer in women 21 years of age or older, determine the occurrence of the event of interest, and obtain study variables. The database acquired through the PRCCR was linked with health insurance data claims. Through this linkage, cervical cancer screening behavior was documented for at least 3 years before patients’ diagnosis. From 937 patients with primary cervical cancer diagnosed during this period, 44% (n=408) were included for this analysis as they had complete information on insurance coverage during the past 3 years (continuous enrollment) and had a microscopically confirmed diagnosis. These patients were followed up to 2017 in order to measure 3-year survival. The Kaplan-Meier Method was used to describe 3-year survival. Also, the Cox Model was used to estimate the risk of death from cervical cancer depending on screening behavior, while controlling for potential confounders. Results: Over half of the patients (52.7%) were insured by Medicaid, 41.2% resided in the area of lowest socioeconomic position (SEP), and only 72.1% were screened 3 years before their diagnosis. Although there weren’t any observable differences in survival curves by SEP, 3-year survival was 77% (95% CI 71.3-81.0) among screened patients and 60% (95% CI 50.1-84.6) among unscreened patients. The overall 25th percentile survival time was about 31 months. The 25th percentile survival time was higher for screened patients (42 months) when compared to unscreened patients (14 months). After controlling for age, stage, histology, and tumor stage, screened patients had lower risk of death when compared to unscreened patients (HR= 0.53, 95% CI 0.4-0.8). Conclusions: Screening behavior is an important predictor in cervical cancer survival. Study findings highlight the relevance to continue increasing compliance with screening guidelines in PR in order to positively impact patient survival. Citation Format: Vanessa Gómez-Vargas, Israel Almodóvar-Rivera, Karen J Ortiz-Ortiz, Carlos R Torres-Cintrón, Ana P Ortiz-Martínez. Cervical cancer survival analysis based on screening practices and the socioeconomic position index in Puerto Rico [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr C023.

Cardiac events and breast cancer survival analysis in caucasians and minorities: A retrospective study in central Illinois.

e13622 Background: Breast cancer is the most prevalent malignancy affecting the female population in the United States and poses a significant mortality risk. In central Illinois, our objective was to analyze cardiac events (described as a new diagnosis of heart failure, heart failure hospitalizations, and acute coronary syndrome) in patients diagnosed with breast cancer and overall survival. Methods: A retrospective analysis completed in February 2020 identified patients (n = 1618) based on ICD diagnosis codes with a primary diagnosis of breast cancer. The inclusion criteria comprised of confirmed pathology report with diagnosis of breast cancer between January 2014 and December 2017. A cohort of 478 patients was identified as having met the inclusion criteria. Baseline characteristics and qualitative information were obtained from retrospective chart review. Continuous variables were described with measures of central tendency (mean, median) and dispersion (range, standard deviation). Categorical variables were summarized as frequencies and percentages. Furthermore, comparison between categorical variables were analyzed using Chi-Square test. Survival curves were estimated using kaplan-meier methodology and analyzed with a log rank test. Qualitative variables were analyzed using chi-square test and reported as hazard ratio using SAS software. All significance was assumed at the p &lt; 0.05 level. Results: Of the 478 patients that were included as part of the analysis, 425 (89.6%) were of caucasian background and 49 (10.34%) were minorities (Asian, African American, and Hispanic). In regards to heart failure, our data demonstrated increased heart failure hospitalizations in minorities (7.89%) compared to caucasians (1.22%) which was a significant association (p = 0.026). In the overall survival among caucasian and minority groups there was no statistically significant difference (hazard ratio 0.939 with a 95% CI between 0.218-4.048), however it showed slight trend suggesting improved survival in the caucasian population. Conclusions: Our data illustrates that minorities with breast cancer experienced a higher incidence of heart failure when compared to the caucasian population. This supports closer observation at cardiac risk factors of such patients and medical optimization in an effort to prevent cardiac complications. While we did not see a significant difference in overall survival, further studies may be warranted in such population.

Expressions and significances of the angiotensin-converting enzyme 2 gene, the receptor of SARS-CoV-2 for COVID-19.

The ACE2 gene is a receptor of SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) for COVID-19 (coronavirus disease 2019). To analyze the expression profiles and clinical significances for this gene in humans, RNA-seq data representing 27 different tissues were analyzed using NCBI; total RNA was extracted from different tissues of mouse and semi-quantitative reverse transcriptional-polymerase chain reaction (Q-RT-PCR) was carried out. Immunohistochemistry expression profiles in normal tissues and cancer tissues and TCGA survival analysis in renal and liver cancer were conducted. ACE2 was highly conserved in different species. In normal tissues, ACE2 expression distributions were organ-specific, mainly in the kidney, male testis and female breast, and cardiovascular and gastrointestinal systems. High level of expression in testis, cardiovascular and gastrointestinal system indicated that SARS-CoV-2 might not only attack the lungs, but also affect other organs, particularly the testes, thus it may severely damage male sexual development for younger male and lead to infertility in an adult male, if he contracted COVID-19. On the other side, high expression of ACE2 was correlated with increased survival rate in renal and liver cancer, indicating that ACE2 is a prognostic marker in both renal cancer and liver cancers. Thus, the ACE2 is a functional receptor for SARS-CoV-2 and has a potential anti-tumor role in cancer. Taken together, this study may not only provide potential clues for further medical pathogenesis of COVID-19 and male fertility, but also indicate the clinical significance of the role of the ACE2 gene in cancer.

A population-based analysis of breast cancer incidence and survival by subtype in Ontario women.

Breast cancer (bca) is the type of cancer most frequently diagnosed among women in Canada. Breast cancer is categorized into various molecular subtypes by the expression of estrogen receptor (er), progesterone receptor (pgr), and her2 (human epidermal growth factor receptor 2). Currently, Canada has no national cancer registry with epidemiology data by subtype. Thus, we conducted a study to determine incidence, survival, and clinicopathologic characteristics by bca subtype [triple negative breast cancer (tnbc); her2+; and hormone receptor-positive (hr+), her2-] in Canadian women newly diagnosed with bca. Female patients diagnosed between 1 April 2012 and 31 March 2016 (fiscal 2012-2015) were identified in the Ontario Cancer Registry, and individual patient data were linked to data in provincial health administrative databases. Descriptive statistics and Kaplan-Meier curves were generated. In this cohort, 3277 women (9.5%) had tnbc, 4902 (14.3%) had her2+ bca, and 22,247 (64.8%) had hr+, her2-breast cancer. The annual incidence was 15 per 100,000 for the tnbc group, 21-23 per 100,000 for the her2+ group, and 97-105 per 100,000 for the hr+, her2- group. The lowest median overall survival (mos) of 8.9 months was observed in women with clinical stage iv tnbc. In comparison, the mos was 37.3 months in those with her2+ disease and 35.2 months in those with and hr+, her2- metastatic bca. In the present study, the most recent and largest administrative database analysis of a Canadian population to date, we observed a subtype distribution consistent with previously reported data, together with comparable annual incidence and overall survival patterns.

Application of Machine Learning for Prediction of Lung Cancer using Omics Data

Cancer is one of the deadly diseases across many countries. However, cancer can be cured, if detected at an early stage. Researchers are working on healthcare for early detection and prevention of cancer. Medical data has reached its utmost potential by providing researchers with huge data sets collected from all over the globe. In the present scenario, Machine Learning has been widely used in the area of cancer diagnosis and prognosis. Survival analysis may help in the prediction of the early onset of disease, relapse, re-occurrence of diseases and biomarker identification. Applications of machine learning and data mining methods in medical field are currently the most widespread in cancer detection and survival analysis. In this survey, different ways to detect and predict lung cancer using latest Machine learning algorithms combined with data mining has been analyzed. Comparative study of various machine learning techniques and technologies has been done over different types of data such as clinical data, omics data, image data etc.

A deep learning-based framework for lung cancer survival analysis with biomarker interpretation

BackgroundLung cancer is the leading cause of cancer-related deaths in both men and women in the United States, and it has a much lower five-year survival rate than many other cancers. Accurate survival analysis is urgently needed for better disease diagnosis and treatment management.ResultsIn this work, we propose a survival analysis system that takes advantage of recently emerging deep learning techniques. The proposed system consists of three major components. 1) The first component is an end-to-end cellular feature learning module using a deep neural network with global average pooling. The learned cellular representations encode high-level biologically relevant information without requiring individual cell segmentation, which is aggregated into patient-level feature vectors by using a locality-constrained linear coding (LLC)-based bag of words (BoW) encoding algorithm. 2) The second component is a Cox proportional hazards model with an elastic net penalty for robust feature selection and survival analysis. 3) The third commponent is a biomarker interpretation module that can help localize the image regions that contribute to the survival model’s decision. Extensive experiments show that the proposed survival model has excellent predictive power for a public (i.e., The Cancer Genome Atlas) lung cancer dataset in terms of two commonly used metrics: log-rank test (p-value) of the Kaplan-Meier estimate and concordance index (c-index).ConclusionsIn this work, we have proposed a segmentation-free survival analysis system that takes advantage of the recently emerging deep learning framework and well-studied survival analysis methods such as the Cox proportional hazards model. In addition, we provide an approach to visualize the discovered biomarkers, which can serve as concrete evidence supporting the survival model’s decision.

Applications of competing risks analysis in public health

In medical and public health research, survival statistics are of particular interest as they can reflect patient prognoses and improvements in health care systems. However, measures of survival differ in their use and interpretation depending on how they deal with competing causes of death. Cause-specific survival estimates survival function based on the event of interest while treating other events as censored, thereby representing the “net” impact of cancer on survival. On the other hand, cumulative incidence function and subdistribution hazard consider the number of subjects experiencing competing risks in their formulations, thereby providing measures for investigating the patients’ actual prognoses. In this paper, we review competing risks survival models used in public health study. We introduce the concept of competing risks methods and compare these with traditional net approaches (e.g. relative and cause-specific). We demonstrate how competing risks analysis can be used in population-based cancer survival analysis utilizing the Surveillance, Epidemiology, and End Result (SEER) cancer registry data. As the scope of public health study has extended beyond prognosis and risk prediction, competing risks analysis has been applied in such studies as well. We further discuss the uptake of the competing risks approach in personalized and precision medicine. Various methods and applications used in risk predicting prognostic models with competing risks are reviewed, aiming to provide effective analytical tools for researchers who plan to implement competing risks models on public health studies.

Comparison of pathway and gene-level models for cancer prognosis prediction

BackgroundCancer prognosis prediction is valuable for patients and clinicians because it allows them to appropriately manage care. A promising direction for improving the performance and interpretation of expression-based predictive models involves the aggregation of gene-level data into biological pathways. While many studies have used pathway-level predictors for cancer survival analysis, a comprehensive comparison of pathway-level and gene-level prognostic models has not been performed. To address this gap, we characterized the performance of penalized Cox proportional hazard models built using either pathway- or gene-level predictors for the cancers profiled in The Cancer Genome Atlas (TCGA) and pathways from the Molecular Signatures Database (MSigDB).ResultsWhen analyzing TCGA data, we found that pathway-level models are more parsimonious, more robust, more computationally efficient and easier to interpret than gene-level models with similar predictive performance. For example, both pathway-level and gene-level models have an average Cox concordance index of ~ 0.85 for the TCGA glioma cohort, however, the gene-level model has twice as many predictors on average, the predictor composition is less stable across cross-validation folds and estimation takes 40 times as long as compared to the pathway-level model. When the complex correlation structure of the data is broken by permutation, the pathway-level model has greater predictive performance while still retaining superior interpretative power, robustness, parsimony and computational efficiency relative to the gene-level models. For example, the average concordance index of the pathway-level model increases to 0.88 while the gene-level model falls to 0.56 for the TCGA glioma cohort using survival times simulated from uncorrelated gene expression data.ConclusionThe results of this study show that when the correlations among gene expression values are low, pathway-level analyses can yield better predictive performance, greater interpretative power, more robust models and less computational cost relative to a gene-level model. When correlations among genes are high, a pathway-level analysis provides equivalent predictive power compared to a gene-level analysis while retaining the advantages of interpretability, robustness and computational efficiency.

CATA: a comprehensive chromatin accessibility database for cancer.

Accessible chromatin refers to the active regions of a chromosome that are bound by many transcription factors (TFs). Changes in chromatin accessibility play a critical role in tumorigenesis. With the emergence of novel methods like Assay for Transposase-accessible Chromatin Sequencing, a sequencing method that maps chromatin-accessible regions (CARs) and enables the computational analysis of TF binding at chromatin-accessible sites, the regulatory landscape in cancer can be dissected. Herein, we developed a comprehensive cancer chromatin accessibility database named CATA, which aims to provide available resources of cancer CARs and to annotate their potential roles in the regulation of genes in a cancer type-specific manner. In this version, CATA stores 2 991 163 CARs from 23 cancer types, binding information of 1398 TFs within the CARs, and provides multiple annotations about these regions, including common single nucleotide polymorphisms (SNPs), risk SNPs, copy number variation, somatic mutations, motif changes, expression quantitative trait loci, methylation and CRISPR/Cas9 target loci. Moreover, CATA supports cancer survival analysis of the CAR-associated genes and provides detailed clinical information of the tumor samples. Database URL: CATA is available at http://www.xiejjlab.bio/cata/.

Cancer Survival Analysis using RNA Sequencing and Clinical Data

Abstract Survival outcomes were assessed in cancer patients from whom cancer tissue was analyzed with Illumina Hi-Seq ribonucleic acid (RNA) sequencing (RNA-Seq) (accessible on National Cancer Institute Genomic Data Commons (GDC) and The Cancer Genome Atlas (TCGA)). Cancer-relevant genes with the most significant correlations with the clinical outcome of overall survival were assessed in Kaplan Meier survival analysis plots. In addition, clinical variables as well as the interaction of clinical variables and cancer relevant genes were assessed in survival analysis. Results show that TP53, BRCA1, NBN, MADIL1, and EP300 were significant predictors of overall survival for prostate cancer patients. While these genes and clinical variables (Gleason Score group and biochemical recurrence) were significant predictors of overall survival when assessed separately, the combination of gene levels along with Gleason score groups provided the most predictive power for overall survival. In this study, cancer-relevant genes predicted survival outcomes, although various genes may interact with genes currently known to contribute to cancer. These findings indicate that multiple cancer types should be assessed together to determine which genes are relevant for cancer in general and for specific cancer types. Future studies will assess all RNA sequencing results available on the Genomic Data Commons, including those not yet associated with cancer. These findings have implications for assessing gene-gene interactions and gene-environment interactions prostate cancer as well as for other types of cancer.

Estimation of Population Based Colorectal Cancer Survival Analysis Using Cox Proportional Hazards Model

Colorectal cancer (CRC) is a tumour of the colon and rectum. Most cases of CRC are sporadic; meaning there are no known hereditary (genetic) components, and it develops slowly over several years through adenomatous polyps. Changes in bowel habits, blood in the stool, and anaemia are cardinal symptoms and sings of CRC. In later stages, fatigue, anorexia, weight loss, pain, jaundice, and other signs and symptoms of locally advanced and metastatic disease occur. The aim of this study is to estimate the population based colorectal cancer survival analysis using cox Proportional Hazards model, in order to fits colorectal cancer data in population-based research. This research was a five-year retrospective study on data from a record of colorectal cancer patients that received treatments from 2013 to 2017 in Radiotherapy Department of Usmanu Danfodiyo University Teaching Hospital, Sokoto, being it one of the cancer registries in Nigeria. 9 covariates were selected to fit colorectal cancer data using Cox Regression Models. The 5-year median survival was found to be 121 days. From the results, it was concluded that the predictor variables could significantly predict the survival of colorectal cancer patients using Cox proportional model. Also the results show that the data met Cox Proportional Hazards Assumptions.

GPSM2 Serves as an Independent Prognostic Biomarker for Liver Cancer Survival.

Background and Objective:Liver cancer is a malignancy with a poor prognosis. G protein signaling modulator 2 is mainly related to cell division and cell cycle regulation. In this review, the relationship between G protein signaling modulator 2 and clinical characteristics of patients with liver cancer has been explored, especially with respect to its prognostic value.Methods:G protein signaling modulator 2 messenger RNA expression and clinicopathological characteristics of patients with liver cancer were obtained from The Cancer Genome Atlas. The expression level of G protein signaling modulator 2 RNA-Seq was validated by using Gene Expression Omnibus. Chi-square test was performed to evaluate the relationship between G protein signaling modulator 2 expression and clinical characteristics. The threshold value of G protein signaling modulator 2 in the diagnosis of liver cancer was evaluated by a receiver–operating characteristic curve. Cox regression analysis and Kaplan-Meier curves were performed to evaluate the relationship between G protein signaling modulator 2 and liver cancer prognosis, which included overall and residual-free survival, and explored the prognostic value of G protein signaling modulator 2. Liver cancer survival analyses were validated by using the data of G protein signaling modulator 2 RNA-Seq from the International Cancer Genome Consortium.Results:The expression level of G protein signaling modulator 2 messenger RNA was remarkably higher in liver cancer than that in healthy tissues (P < 2.2 × e−16), which was also validated by data from the GSE14520 database. In addition, high G protein signaling modulator 2 expression significantly correlated with histological grade (P = .020), vital status (P < .001), clinical (P = .001), and T stage (P = .001). The receiver–operating characteristic curves showed G protein signaling modulator 2 to be an advantageous diagnostic molecule for liver cancer (area under curve = 0.893). Furthermore, the results of Cox analysis and Kaplan-Meier curves suggested that the upregulation of G protein signaling modulator 2 expression is linked to poor prognosis and G protein signaling modulator 2 messenger RNA could be an independent predictor for liver cancer, which was validated by data from the International Cancer Genome Consortium database.Conclusions:G protein signaling modulator 2 messenger RNA was overexpressed in liver cancer, and G protein signaling modulator 2 is an independent prognostic factor. G protein signaling modulator 2 is expected to be a treatment target for cancer.

Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data

BackgroundUnderstanding the complex biological mechanisms of cancer patient survival using genomic and clinical data is vital, not only to develop new treatments for patients, but also to improve survival prediction. However, highly nonlinear and high-dimension, low-sample size (HDLSS) data cause computational challenges to applying conventional survival analysis.ResultsWe propose a novel biologically interpretable pathway-based sparse deep neural network, named Cox-PASNet, which integrates high-dimensional gene expression data and clinical data on a simple neural network architecture for survival analysis. Cox-PASNet is biologically interpretable where nodes in the neural network correspond to biological genes and pathways, while capturing the nonlinear and hierarchical effects of biological pathways associated with cancer patient survival. We also propose a heuristic optimization solution to train Cox-PASNet with HDLSS data. Cox-PASNet was intensively evaluated by comparing the predictive performance of current state-of-the-art methods on glioblastoma multiforme (GBM) and ovarian serous cystadenocarcinoma (OV) cancer. In the experiments, Cox-PASNet showed out-performance, compared to the benchmarking methods. Moreover, the neural network architecture of Cox-PASNet was biologically interpreted, and several significant prognostic factors of genes and biological pathways were identified.ConclusionsCox-PASNet models biological mechanisms in the neural network by incorporating biological pathway databases and sparse coding. The neural network of Cox-PASNet can identify nonlinear and hierarchical associations of genomic and clinical data to cancer patient survival. The open-source code of Cox-PASNet in PyTorch implemented for training, evaluation, and model interpretation is available at: https://github.com/DataX-JieHao/Cox-PASNet.

Distribution of Cancer cases and survival analysis results: A retrospective cohort study

Abstract Background Cancer was the second leading cause of death in Turkey. The aim of this study is to examine the changes in the incidence of cancer cases and survival analysis in a region. Methods This retrospective cohort study was carried out on data of 977 people had been diagnosed with cancer between the years 2013-2015 in Bingol province located in the east of Turkey. All cancer cases are included in the study. The data of the patients were obtained from the Ministry of Health Cancer Registry System and International Disease Classification Oncology (ICD-O) codes published by World Health Organization were used. Patients were evaluated topographically (tissue originating from the lesion) in 41 categories. Kruskal Wallis, Mann Whitney U test, Chi-square test and Cox regression analysis were used to analyze the data. Results The mean age of the patients was 58.73 ± 16.80 (min: 1-max: 107). It was observed that 28.7% of the patients died during the follow-up period and 71.3% of the patients were alive. 30% of the cancer cases digestive system, 11% respiratory system, 23% female breast- genitalia, 15% blood-lymph-thyroid, 11.8% muscle-bone, 15,4% other cancer species. In men, lung, stomach, colon, prostate; breast, thyroid and skin cancers are more common in women respectively. The incidence rate for all age groups was 97 per hundred thousand in 2013, 161 per hundred thousand in 2015. The increase trend did not differ by gender. The mean follow-up period of the patients who died was 388 days and 1541 days for the living patients. The risk of death is increased significantly in patients who cannot be operated (RR = 2.779, HR = 3.048). In patients not receiving chemotherapy, the risk of death is reduced significantly (RR = 0.656, HR = 0.340). Conclusions The incidence of cancer increases in the study area. Survival times vary according to treatment efficacy. Broad-based cohort studies evaluating the effects of treatment types on survival may be recommended. Key messages Cancer in the study area shows a significant increase from year to year. The average 5-year survival period of cancer patients was found is 71% as a good result.

Patterns of Relapse and Survival Analysis of Advanced Epithelial Ovarian Cancers Operated in a Tertiary Cancer Centre

Advanced epithelial ovarian cancers have a dismal prognosis with a 5-year survival rate of 29.2%. A relapse rate in excess of 75% is noted in patients who achieve remission following neoadjuvant chemotherapy and cytoreductive surgery. Our study aims to find out the progression-free survival and overall survival of advanced epithelial ovarian cancers in South Indian population. Demographic data, neoadjuvant treatment and surgery details, recurrence, and survival status of patients operated for advanced ovarian cancer in our department from year 2015 till 2018 were retrospectively collected through medical records. Statistical analyses were performed using SPSS version 25. Our study included 111 patients with the majority staged as IIIC (82%). 107 patients received neoadjuvant chemotherapy. Six patients were lost to follow-up. The median pre- and post-chemotherapy CA 125 level was 694 IU/L and 33.5 IU/L, respectively. Optimal cytoreduction was achieved in 87.6% of patients. At a median follow-up of 24 months, median progression-free survival was 30 months (95% CI: 26.5–34.5) in the optimally cytoreduced versus 13 months (95% CI: 10.3–15.4) in the suboptimally cytoreduced patients (P0.005). The median overall survival was 36 months (95% CI: 32–39) in optimally cytoreduced versus 19 months (95% CI: 17.6–20.3) in suboptimally cytoreduced patients (P0.001). Patients achieving optimal cytoreduction after neoadjuvant chemotherapy have significantly lower recurrence rates, higher progression-free survival, and overall survival rates than those suboptimally cytoreduced. Results from our centre are equivalent to apex centres in India.