Feature selection with the R package MXM

A robust clustering strategy for stratification unveils unique patient subgroups in acutely decompensated cirrhosis

BACKGROUND

Patient heterogeneity poses significant challenges for managing individuals and designing clinical trials, especially in complex diseases. Existing classifications rely on outcome-predicting scores, potentially overlooking crucial elements contributing to heterogeneity without necessarily impacting prognosis.

METHODS

To address patient heterogeneity, we developed ClustALL, a computational pipeline that simultaneously faces diverse clinical data challenges like mixed types, missing values, and collinearity. ClustALL enables the unsupervised identification of patient stratifications while filtering for stratifications that are robust against minor variations in the population (population-based) and against limited adjustments in the algorithm's parameters (parameter-based).

RESULTS

Applied to a European cohort of patients with acutely decompensated cirrhosis (n = 766), ClustALL identified five robust stratifications, using only data at hospital admission. All stratifications included markers of impaired liver function and number of organ dysfunction or failure, and most included precipitating events. When focusing on one of these stratifications, patients were categorized into three clusters characterized by typical clinical features; notably, the 3-cluster stratification showed a prognostic value. Re-assessment of patient stratification during follow-up delineated patients' outcomes, with further improvement of the prognostic value of the stratification. We validated these findings in an independent prospective multicentre cohort of patients from Latin America (n = 580).

CONCLUSIONS

By applying ClustALL to patients with acutely decompensated cirrhosis, we identified three patient clusters. Following these clusters over time offers insights that could guide future clinical trial design. ClustALL is a novel and robust stratification method capable of addressing the multiple challenges of patient stratification in most complex diseases.

Comprehensive methylome sequencing reveals prognostic epigenetic biomarkers for prostate cancer mortality

BACKGROUND

Prostate cancer is a clinically heterogeneous disease with a subset of patients rapidly progressing to lethal-metastatic prostate cancer. Current clinicopathological measures are imperfect predictors of disease progression. Epigenetic changes are amongst the earliest molecular changes in tumourigenesis. To find new prognostic biomarkers to enable earlier intervention and improved outcomes, we performed methylome sequencing of DNA from patients with localised prostate cancer and long-term clinical follow-up.

METHODS

We used whole-genome bisulphite sequencing (WGBS) to comprehensively map and compare DNA methylation of radical prostatectomy tissue between patients with lethal disease (n = 7) and non-lethal (n = 8) disease (median follow-up 19.5 years). Validation of differentially methylated regions (DMRs) was performed in an independent cohort (n = 185, median follow-up 15 years) using targeted multiplex bisulphite sequencing of candidate regions. Survival was assessed via univariable and multivariable analyses including clinicopathological measures (log-rank and Cox regression models).

RESULTS

WGBS data analysis identified cancer-specific methylation patterns including CpG island hypermethylation, and hypomethylation of repetitive elements, with increasing disease risk. We identified 1420 DMRs associated with prostate cancer-specific mortality (PCSM), which showed enrichment for gene sets downregulated in prostate cancer and de novo methylated in cancer. Through comparison with public prostate cancer datasets, we refined the DMRs to develop an 18-gene prognostic panel. Applying this panel to an independent cohort, we found significant associations between PCSM and hypermethylation at EPHB3, PARP6, TBX1, MARCH6 and a regulatory element within CACNA2D4. Strikingly in a multivariable model, inclusion of CACNA2D4 methylation was a better predictor of PCSM versus grade alone (Harrell's C-index: 0.779 vs. 0.684).

CONCLUSIONS

Our study provides detailed methylome maps of non-lethal and lethal prostate cancer and identifies novel genic regions that distinguish these patient groups. Inclusion of our DNA methylation biomarkers with existing clinicopathological measures improves prognostic models of prostate cancer mortality, and holds promise for clinical application.

Current Achievements and Applications of Transcriptomics in Personalized Cancer Medicine

Over the last decades, transcriptome profiling emerged as one of the most powerful approaches in oncology, providing prognostic and predictive utility for cancer management. The development of novel technologies, such as revolutionary next-generation sequencing, enables the identification of cancer biomarkers, gene signatures, and their aberrant expression affecting oncogenesis, as well as the discovery of molecular targets for anticancer therapies. Transcriptomics contribute to a change in the holistic understanding of cancer, from histopathological and organic to molecular classifications, opening a more personalized perspective for tumor diagnostics and therapy. The further advancement on transcriptome profiling may allow standardization and cost reduction of its analysis, which will be the next step for transcriptomics to become a canon of contemporary cancer medicine.

Applicability of an Automated Model and Parameter Selection in the Prediction of Screening-Level PTSD in Danish Soldiers Following Deployment: Development Study of Transferable Predictive Models Using Automated Machine Learning

BACKGROUND

Posttraumatic stress disorder (PTSD) is a relatively common consequence of deployment to war zones. Early postdeployment screening with the aim of identifying those at risk for PTSD in the years following deployment will help deliver interventions to those in need but have so far proved unsuccessful.

OBJECTIVE

This study aimed to test the applicability of automated model selection and the ability of automated machine learning prediction models to transfer across cohorts and predict screening-level PTSD 2.5 years and 6.5 years after deployment.

METHODS

Automated machine learning was applied to data routinely collected 6-8 months after return from deployment from 3 different cohorts of Danish soldiers deployed to Afghanistan in 2009 (cohort 1, N=287 or N=261 depending on the timing of the outcome assessment), 2010 (cohort 2, N=352), and 2013 (cohort 3, N=232).

RESULTS

Models transferred well between cohorts. For screening-level PTSD 2.5 and 6.5 years after deployment, random forest models provided the highest accuracy as measured by area under the receiver operating characteristic curve (AUC): 2.5 years, AUC=0.77, 95% CI 0.71-0.83; 6.5 years, AUC=0.78, 95% CI 0.73-0.83. Linear models performed equally well. Military rank, hyperarousal symptoms, and total level of PTSD symptoms were highly predictive.

CONCLUSIONS

Automated machine learning provided validated models that can be readily implemented in future deployment cohorts in the Danish Defense with the aim of targeting postdeployment support interventions to those at highest risk for developing PTSD, provided the cohorts are deployed on similar missions.