Reference-free transcriptome signatures for prostate cancer prognosis

A Novel Nomogram for Identifying the Patients at Risk for Rapid Progression of Advanced Hormone-Sensitive Prostate Cancer

Purpose

The goal of this study was to assess the prognostic impact of the lower urinary tract symptoms (LUTS) in advanced prostate cancer (PCa) patients before progression to castration-resistant prostate cancer (CRPC).

Methods

A retrospective analysis of the follow-up data for 152 CRPC patients was performed. Severe LUTS symptom was defined as an International Prostate Symptoms Score (IPSS) ≥20 at baseline. Cox regression analysis was conducted to assess CRPC prognostic factors. Nomogram model was created and assessed using the concordance index (C-index), calibration curves, receiver operating characteristic (ROC) curves, and decision curve analyses (DCA).

Results

The median CRPC free survival of patients with severe LUTS was 20.5 months, significantly longer than that (7.5 months) of less symptomatic patients. Furthermore, severe LUTS, the hemoglobin, albumin, lymphocyte, and platelet (HALP) score, and Gleason sum were determined to be independent prognostic markers and combined to establish a nomogram, which performed well in the customized prediction of CRPC progression at 6th, 12th, 18th and 24th month. The C-index (0.794 and 0.816 for the training and validation cohorts, respectively), calibration curve, and ROC curve all validated the prediction accuracy. DCA curve showed that it could be effective in helping doctors make judgments. The Nomogram-related risk score separated the patients into two groups with notable progression differences.

Conclusion

Severe LUTS was significantly associated with decreased risk for rapid progression to CRPC. The developed nomogram could help identify patients who are at a high risk of rapid CRPC progression and provide tailored follow-up and therapeutic advice.

Gene expression analysis reveals a 5-gene signature for progression-free survival in prostate cancer

Prostate cancer (PCa) is the second most common male cancer worldwide, but effective biomarkers for the presence or progression risk of disease are currently elusive. In a series of nine matched histologically confirmed PCa and benign samples, we carried out an integrated transcriptome-wide gene expression analysis, including differential gene expression analysis and weighted gene co-expression network analysis (WGCNA), which identified a set of potential gene markers highly associated with tumour status (malignant vs. benign). We then used these genes to establish a minimal progression-free survival (PFS)-associated gene signature (GS) (PCBP1, PABPN1, PTPRF, DANCR, and MYC) using least absolute shrinkage and selection operator (LASSO) and stepwise multivariate Cox regression analyses from The Cancer Genome Atlas prostate adenocarcinoma (TCGA-PRAD) dataset. Our signature was able to predict PFS over 1, 3, and 5 years in TCGA-PRAD dataset, with area under the curve (AUC) of 0.64–0.78, and our signature remained as a prognostic factor independent of age, Gleason score, and pathological T and N stages. A nomogram combining the signature and Gleason score demonstrated improved predictive capability for PFS (AUC: 0.71–0.85) and was superior to the Cambridge Prognostic Group (CPG) model alone and some conventionally used clinicopathological factors in predicting PFS. In conclusion, we have identified and validated a novel five-gene signature and established a nomogram that effectively predicted PFS in patients with PCa. Findings may improve current prognosis tools for PFS and contribute to clinical decision-making in PCa treatment.

Replicability in cancer omics data analysis: measures and empirical explorations

In biomedical research, the replicability of findings across studies is highly desired. In this study, we focus on cancer omics data, for which the examination of replicability has been mostly focused on important omics variables identified in different studies. In published literature, although there have been extensive attention and ad hoc discussions, there is insufficient quantitative research looking into replicability measures and their properties. The goal of this study is to fill this important knowledge gap. In particular, we consider three sensible replicability measures, for which we examine distributional properties and develop a way of making inference. Applying them to three The Cancer Genome Atlas (TCGA) datasets reveals in general low replicability and significant across-data variations. To further comprehend such findings, we resort to simulation, which confirms the validity of the findings with the TCGA data and further informs the dependence of replicability on signal level (or equivalently sample size). Overall, this study can advance our understanding of replicability for cancer omics and other studies that have identification as a key goal.

The contribution of uncharted RNA sequences to tumor identity in lung adenocarcinoma

The identity of cancer cells is defined by the interplay between genetic, epigenetic transcriptional and post-transcriptional variation. A lot of this variation is present in RNA-seq data and can be captured at once using reference-free, k-mer analysis. An important issue with k-mer analysis, however, is the difficulty of distinguishing signal from noise. Here, we use two independent lung adenocarcinoma datasets to identify all reproducible events at the k-mer level, in a tumor versus normal setting. We find reproducible events in many different locations (introns, intergenic, repeats) and forms (spliced, polyadenylated, chimeric etc.). We systematically analyze events that are ignored in conventional transcriptomics and assess their value as biomarkers and for tumor classification, survival prediction, neoantigen prediction and correlation with the immune microenvironment. We find that unannotated lincRNAs, novel splice variants, endogenous HERV, Line1 and Alu repeats and bacterial RNAs each contribute to different, important aspects of tumor identity. We argue that differential RNA-seq analysis of tumor/normal sample collections would benefit from this type k-mer analysis to cast a wider net on important cancer-related events. The code is available at https://github.com/Transipedia/dekupl-lung-cancer-inter-cohort.

Identification and validation of an immune-related gene pairs signature for three urologic cancers

Reliable biomarkers are needed to recognize urologic cancer patients at high risk for recurrence. In this study, we built a novel immune-related gene pairs signature to simultaneously predict recurrence for three urologic cancers. We gathered 14 publicly available gene expression profiles including bladder, prostate and kidney cancer. A total of 2,700 samples were classified into the training set (n = 1,622) and validation set (n = 1,078). The 25 immune-related gene pairs signature consisting of 41 unique genes was developed by the least absolute shrinkage and selection operator regression analysis and Cox regression model. The signature stratified patients into high- and low-risk groups with significantly different relapse-free survival in the meta-training set and its subpopulations, and was an independent prognostic factor of urologic cancers. This signature showed a robust ability in the meta-validation and multiple independent validation cohorts. Immune and inflammatory response, chemotaxis and cytokine activity were enriched with genes relevant to the signature. A significantly higher infiltration level of M1 macrophages was found in the high-risk group versus the low-risk group. In conclusion, our signature is a promising prognostic biomarker for predicting relapse-free survival in patients with urologic cancer.

A Personalized Genomics Approach of the Prostate Cancer

Decades of research identified genomic similarities among prostate cancer patients and proposed general solutions for diagnostic and treatments. However, each human is a dynamic unique with never repeatable transcriptomic topology and no gene therapy is good for everybody. Therefore, we propose the Genomic Fabric Paradigm (GFP) as a personalized alternative to the biomarkers approach. Here, GFP is applied to three (one primary—“A”, and two secondary—“B” & “C”) cancer nodules and the surrounding normal tissue (“N”) from a surgically removed prostate tumor. GFP proved for the first time that, in addition to the expression levels, cancer alters also the cellular control of the gene expression fluctuations and remodels their networking. Substantial differences among the profiled regions were found in the pathways of P53-signaling, apoptosis, prostate cancer, block of differentiation, evading apoptosis, immortality, insensitivity to anti-growth signals, proliferation, resistance to chemotherapy, and sustained angiogenesis. ENTPD2, AP5M1 BAIAP2L1, and TOR1A were identified as the master regulators of the “A”, “B”, “C”, and “N” regions, and potential consequences of ENTPD2 manipulation were analyzed. The study shows that GFP can fully characterize the transcriptomic complexity of a heterogeneous prostate tumor and identify the most influential genes in each cancer nodule.