Abstract:ABSTRACT It is difficult to characterize complex variations of biological processes, often longitudinally measured using biomarkers that yield noisy data. While joint modeling with a longitudinal submodel for the biomarker measurements and a survival submodel for assessing the hazard of events can alleviate measurement error issues, the continuous longitudinal submodel often uses random intercepts and slopes to estimate both between- and within-patient heterogeneity in biomarker trajectories. To overcome longitudinal submodel challenges, we replace random slopes with scaled integrated fractional Brownian motion (IFBM). As a more generalized version of integrated Brownian motion, IFBM reasonably depicts noisily measured biological processes. From this longitudinal IFBM model, we derive novel target functions to monitor the risk of rapid disease progression as real-time predictive probabilities. Predicted biomarker values from the IFBM submodel are used as inputs in a Cox submodel to estimate event hazard. This two-stage approach to fit the submodels is performed via Bayesian posterior computation and inference. We use the proposed approach to predict dynamic lung disease progression and mortality in women with a rare disease called lymphangioleiomyomatosis who were followed in a national patient registry. We compare our approach to those using integrated Ornstein-Uhlenbeck or conventional random intercepts-and-slopes terms for the longitudinal submodel. In the comparative analysis, the IFBM model consistently demonstrated superior predictive performance.

What problem does this paper attempt to address?

### Problems the paper attempts to solve This paper aims to solve the difficult problem of characterizing changes in complex biological processes, especially when these processes are longitudinally measured by biomarkers and the data are often noisy. Although the joint - modeling method can correct errors in biomarker measurements through the longitudinal sub - model and evaluate the risk of events through the survival sub - model, traditional continuous longitudinal sub - models usually use random intercepts and slopes to estimate heterogeneity between patients and within - patient heterogeneity. This method has limitations in describing complex fluctuations in biomarker trajectories. To solve this problem, the authors propose a new two - stage Bayesian modeling method that uses Integrated Fractional Brownian Motion (IFBM) to replace the traditional random slope. IFBM is a more general random process and can more reasonably depict noisy biological processes. Through this longitudinal IFBM model, the authors derive a new objective function for real - time monitoring of the risk of rapid disease progression and use the predicted biomarker values as input to estimate the event risk in the Cox sub - model. Specifically, the objectives of this study include: 1. **Improve the longitudinal model**: Use IFBM instead of the traditional random slope to better capture complex fluctuations in biomarker trajectories. 2. **Real - time prediction**: Calculate the risk of dynamic lung function decline through the objective function to provide real - time prediction. 3. **Survival analysis**: Combine the prediction results of the longitudinal model and use the Cox proportional - hazards model to evaluate the risk of death or lung transplantation. The authors apply this method to predict the risk of dynamic lung function decline and death in female patients with lymphangioleiomyomatosis (LAM) and compare it with traditional methods. The results show that the IFBM model has significant advantages in prediction performance.

Bayesian two-stage modeling of longitudinal and time-to-event data with an integrated fractional Brownian motion covariance structure

A Bayesian joint longitudinal-survival model with a latent stochastic process for intensive longitudinal data

Bayesian Estimation of Two-Part Joint Models for a Longitudinal Semicontinuous Biomarker and a Terminal Event with R-INLA: Interests for Cancer Clinical Trial Evaluation

A joint model for (un)bounded longitudinal markers, competing risks, and recurrent events using patient registry data

Joint Models for Time-to-Event Data and Longitudinal Biomarkers of High Dimension

Bayesian Latent-Class Mixed-Effect Hybrid Models for Dyadic Longitudinal Data with Non-Ignorable Dropouts.

Bayesian semiparametric joint modeling of a count outcome and inconveniently timed longitudinal predictors

Joint modeling of zero-inflated longitudinal measurements and time-to-event outcomes with applications to dynamic prediction

Bayesian Analysis of Longitudinal Dyadic Data with Informative Missing Data Using a Dyadic Shared-Parameter Model

Bayesian blockwise inference for joint models of longitudinal and multistate data with application to longitudinal multimorbidity analysis

Joint Dynamic Models and Statistical Inference for Recurrent Competing Risks, Longitudinal Marker, and Health Status

Jointly Modelling Longitudinal and Survival Data

Joint Modeling of Geometric Features of Longitudinal Process and Discrete Survival Time Measured on Nested Timescales: An Application to Fecundity Studies

A Flexible Joint Model for Multiple Longitudinal Biomarkers and A Time-to-Event Outcome: with Applications to Dynamic Prediction Using Highly Correlated Biomarkers

Bayesian Joint Modeling of Multivariate Longitudinal and Survival Data with an Application to Diabetes Study.

A Bayesian quantile joint modeling of multivariate longitudinal and time-to-event data

A Bayesian Approach to Modeling Variance of Intensive Longitudinal Biomarker Data as a Predictor of Health Outcomes

Semiparametric multivariate joint model for skewed-longitudinal and survival data: A Bayesian approach

Integrating Latent Classes in the Bayesian Shared Parameter Joint Model of Longitudinal and Survival Outcomes

Bayesian Semiparametric Joint Model of Multivariate Longitudinal and Survival Data with Dependent Censoring.

Robust Bayesian cumulative probit linear mixed models for longitudinal ordinal data