Individual risk prediction: Comparing random forests with Cox proportional-hazards model by a simulation study

With big data becoming widely available in healthcare, machine learning algorithms such as random forest (RF) that ignores time-to-event information and random survival forest (RSF) that handles right-censored data are used for individual risk prediction alternatively to the Cox proportional hazards (Cox-PH) model. We aimed to systematically compare RF and RSF with Cox-PH. RSF with three split criteria [log-rank (RSF-LR), log-rank score (RSF-LRS), maximally selected rank statistics (RSF-MSR)]; RF, Cox-PH, and Cox-PH with splines (Cox-S) were evaluated through a simulation study based on real data. One hundred eighty scenarios were investigated assuming different associations between the predictors and the outcome (linear/linear and interactions/nonlinear/nonlinear and interactions), training sample sizes (500/1000/5000), censoring rates (50%/75%/93%), hazard functions (increasing/decreasing/constant), and number of predictors (seven, 15 including noise variables). Methods' performance was evaluated with time-dependent area under curve and integrated Brier score. In all scenarios, RF had the worst performance. In scenarios with a low number of events (⩽70), Cox-PH was at least noninferior to RSF, whereas under linearity assumption it outperformed RSF. Under the presence of interactions, RSF performed better than Cox-PH as the number of events increased whereas Cox-S reached at least similar performance with RSF under nonlinear effects. RSF-LRS performed slightly worse than RSF-LR and RSF-MSR when including noise variables and interaction effects. When applied to real data, models incorporating survival time performed better. Although RSF algorithms are a promising alternative to conventional Cox-PH as data complexity increases, they require a higher number of events for training. In time-to-event analysis, algorithms that consider survival time should be used.     

Bias-reduced estimators of conditional odds ratios in matched case-control studies with unmatched confounding

We study bias-reduced estimators of exponentially transformed parameters in general linear models (GLMs) and show how they can be used to obtain bias-reduced conditional (or unconditional) odds ratios in matched case-control studies. Two options are considered and compared: the explicit approach and the implicit approach. The implicit approach is based on the modified score function where bias-reduced estimates are obtained by using iterative procedures to solve the modified score equations. The explicit approach is shown to be a one-step approximation of this iterative procedure. To apply these approaches for the conditional analysis of matched case-control studies, with potentially unmatched confounding and with several exposures, we utilize the relation between the conditional likelihood and the likelihood of the unconditional logit binomial GLM for matched pairs and Cox partial likelihood for matched sets with appropriately setup data. The properties of the estimators are evaluated by using a large Monte Carlo simulation study and an illustration of a real dataset is shown. Researchers reporting the results on the exponentiated scale should use bias-reduced estimators since otherwise the effects can be under or overestimated, where the magnitude of the bias is especially large in studies with smaller sample sizes.     

Some new results on Cox–Czanner divergence and their applications in survival studies

In the present communication, we propose a quantile-based measure for the divergence between two survival functions. This can also be used in a dynamic way where the divergence between survival functions varies with time. Several new properties of the proposed measure are investigated with suitable examples. The behavior of the measure for various reliability models is also investigated. A real data analysis is employed to compare the relative efficacy of two treatment groups using the proposed divergence measure.     

[PSI+] turns 50

abstract

The year 2015 sees the fiftieth anniversary of the publication of a research paper that underpins much of our understanding of fungal prion biology, namely “ψ, a cytoplasmic suppressor of super-suppressor in yeast” by Brian Cox. Here we show how our understanding of the molecular nature of the [PSI⁺] determinant evolved from an ‘occult’ determinant to a transmissible amyloid form of a translation termination factor. We also consider the impact studies on [PSI] have had – and continue to have - on prion research. To demonstrate this, leading investigators in the yeast prion field who have made extensive use of the [PSI⁺] trait in their research, provide their own commentaries on the discovery and significance of [PSI].     

A prognostic eight-gene expression signature for patients with breast cancer receiving adjuvant chemotherapy

Breast cancer is a popularly diagnosed malignant tumor. Genomic profiling studies suggest that breast cancer is a disease with heterogeneity. Chemotherapy is one of the chief means to treat breast cancer, while its responses and clinical outcomes vary largely due to the conventional clinicopathological factors and inherent chemosensitivity of breast cancer. Using the least absolute shrinkage and selection operator (LASSO) Cox regression model, our study established a multi-mRNA-based signature model and constructed a relative nomogram in predicting distant-recurrence-free survival for patients receiving surgery and following chemotherapy. We constructed a signature of eight mRNAs (IPCEF1, SYNDIG1, TIGIT, SPESP1, C2CD4A, CLCA2, RLN2, and CCL19) with the LASSO model, which was employed to separate subjects into groups with high- and low-risk scores. Obvious differences of distant-recurrence-free survival were found between these two groups. This eight-mRNA-based signature was independently associated with the prognosis and had better prognostic value than classical clinicopathologic factors according to multivariate Cox regression results. Receiver operating characteristic results demonstrated excellent performance in diagnosing 3-year distant-recurrence by the eight-mRNA signature. A nomogram that combined both the eight-mRNA-based signature and clinicopathological risk factors was constructed. Comparing with an ideal model, the nomograms worked well both in the training and validation sets. Through the results that the eight-mRNA signature effectively classified patients into low- and high-risk of distant recurrence, we concluded that this eight-mRNA-based signature played a promising predictive role in prognosis and could be clinically applied in breast cancer patients receiving adjuvant chemotherapy.     

Five genes as a novel signature for predicting the prognosis of patients with laryngeal cancer

In this study, we purpose to investigate a novel five-gene signature for predicting the prognosis of patients with laryngeal cancer. The laryngeal cancer datasets were obtained from The Cancer Genome Atlas (TCGA). Both univariate and multivariate Cox regression analysis was applied to screening for prognostic differential expressed genes (DEGs), and a novel gene signature was obtained. The performance of this Cox regression model was tested by receiver operating characteristic (ROC) curves and area under the curve (AUC). Further survival analysis for each of the five genes was carried out through the Kaplan-Meier curve and Log-rank test. Totally, 622 DEGs were screened from the TCGA datasets in this study. We construct a five-gene signature through Cox survival analysis. Patients were divided into low- and high-risk groups depending on the median risk score, and a significant difference of the 5-year overall survival was found between these two groups (P < .05). ROC curves verified that this five-gene signature had good performance to predict the prognosis of laryngeal cancer (AUC = 0.862, P < .05). In conclusion, the five-gene signature consist of EMP1, HOXB9, DPY19L2P1, MMP1, and KLHDC7B might be applied as an independent prognosis predictor of laryngeal cancer.