Background/Objectives: KRAS mutations are among the most prevalent oncogenic drivers in non-small cell lung cancer (NSCLC), with their impact on survival influenced by co-mutations. SMARCA4 mutations are increasingly associated with poor prognosis and can be classified as class 1 or class 2 mutations. This study evaluates the prognostic implications of KRAS and SMARCA4 mutations, including their co-mutations and their impact on NSCLC patients by utilizing real-world evidence. Methods: A retrospective analysis was conducted using the AACR GENIE Biopharma Collaborative (BPC) NSCLC 2.0 dataset. NSCLC patients with KRAS mutations, SMARCA4 mutations, or KRAS/SMARCA4 co-mutations were identified. Survival outcomes were assessed using univariate and multivariate Cox proportional hazards models, incorporating key clinical variables such as sex, race, smoking history, and stage. Results: Among 659 NSCLC patients with KRAS or SMARCA4 mutations analyzed, KRAS mutations were the most prevalent (79%, n = 518). SMARCA4 mutations were identified in 14% of cases (n = 95) across two classes. Six percent (n = 41) with class 1 mutations and 8% (n = 54) with class 2. Neither SMARCA4 class was associated with worse survival outcomes compared to KRAS-mutated patients (p = 0.438 & 0.720). Patients harboring KRAS/SMARCA4 class 1 co-mutations (3%, n = 18) had significantly worse overall survival compared to those with KRAS mutations alone (hazard ratio [HR] = 3.23, p < 0.001). In contrast, KRAS/SMARCA4 class 2 co-mutations (4%, n = 28) did not significantly impact survival compared to KRAS-mutated patients (HR = 1.34, p = 0.205). Conclusions: KRAS/SMARCA4 class 1 co-mutations are associated with significantly worse overall survival compared to KRAS-mutated NSCLC patients. Our multivariate analysis demonstrates the critical need to incorporate routine next-generation sequencing (NGS) testing in managing NSCLC patients at the time of metastatic diagnosis, with particular emphasis on identifying SMARCA4 mutation class as a potential prognostic biomarker in those with KRAS co-mutations.

Marginal likelihood plays a central role in Bayesian model comparison and hypothesis testing, but its computation is often challenging in practice. This article reviews recent Monte Carlo methods that rely on the availability of Markov chain Monte Carlo (MCMC) samples from the posterior and prior distributions along with the corresponding unnormalized kernels that can be evaluated numerically. Within this scope, we summarize the strengths, limitations, differences, and connections of different methods. Two in‐depth applications are presented to illustrate their relative performance. This article is categorized under: Statistical and Graphical Methods of Data Analysis > Monte Carlo Methods Statistical Models > Model Selection Statistical Models > Bayesian Models

The improvement of mortality projection is a pivotal topic in the diverse branches related to insurance, demography, and public policy. Motivated by the thread of Lee-Carter related models, we propose a Bayesian model to estimate and predict mortality rates for multi-population. This new model features in information borrowing among populations and properly reflecting variations of data. It also provides a solution to a long-time overlooked problem: model selection for dependence structures of population-specific time parameters. By introducing a Dirac spike function, simultaneous model selection and estimation for population-specific time effects can be achieved without much extra computation cost. We use the Japanese mortality data from Human Mortality Database to illustrate the desirable properties of our model.