Dataset Information

Time-to-event prediction using survival analysis methods for Alzheimer's disease progression.

ABSTRACT:

Introduction

Many research studies have well investigated Alzheimer's disease (AD) detection and progression. However, the continuous-time survival prediction of AD is not yet fully explored to support medical practitioners with predictive analytics. In this study, we develop a survival analysis approach to examine interactions between patients' inherent temporal and medical patterns and predict the probability of the AD next stage progression during a time period. The likelihood of reaching the following AD stage is unique to a patient, helping the medical practitioner analyze the patient's condition and provide personalized treatment recommendations ahead of time.

Methodologies

We simulate the disease progression based on patient profiles using non-linear survival methods-non-linear Cox proportional hazard model (Cox-PH) and neural multi-task logistic regression (N-MTLR). In addition, we evaluate the concordance index (C-index) and Integrated Brier Score (IBS) to describe the evolution to the next stage of AD. For personalized forecasting of disease, we also developed deep neural network models using the dataset provided by the National Alzheimer's Coordinating Center with their multiple-visit details between 2005 and 2017.

Results

The experiment results show that our N-MTLR based survival models outperform the CoxPH models, the best of which gives Concordance-Index of 0.79 and IBS of 0.09. We obtained 50 critical features out of 92 by applying recursive feature elimination and random forest techniques on the clinical data; the top ones include normal cognition and behavior, criteria for dementia, community affairs, etc. Our study demonstrates that selecting critical features can improve the effectiveness of probabilities at each time interval.

Conclusions

The proposed deep learning-based survival method and model can be used by medical practitioners to predict the patients' AD shift efficiently and recommend personalized treatment to mitigate or postpone the effects of AD. More generally, our proposed survival analysis approach for predicting disease stage shift can be used for other progressive diseases such as cancer, Huntington's disease, and scleroderma, just to mention a few, using the corresponding clinical data.

SUBMITTER: Sharma R

PROVIDER: S-EPMC8719343 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundThe clinical meaningfulness of the effects of recently approved disease-modifying treatments (DMT) in Alzheimer's disease is under debate. Available evidence is limited to short-term effects on clinical rating scales which may be difficult to interpret and have limited intrinsic meaning to patients. The main value of DMTs accrues over the long term as they are expected to cause a delay or slowing of disease progression. While awaiting such evidence, the translation of short-term effects to time delays or slowing of progression could offer a powerful and readily interpretable representation of clinical outcomes.MethodsWe simulated disease progression trajectories representing two arms, active and placebo, of a hypothetical clinical trial of a DMT. The placebo arm was simulated based on estimated mean trajectories of clinical dementia rating scale-sum of boxes (CDR-SB) recordings from amyloid-positive subjects with mild cognitive impairment (MCI) from Alzheimer's Disease Neuroimaging Initiative (ADNI). The active arm was simulated to show an average slowing of disease progression versus placebo of 20% at each visit. The treatment effects in the simulated trials were estimated with a progression model for repeated measures (PMRM) and a mixed model for repeated measures (MMRM) for comparison. For PMRM, the treatment effect is expressed in units of time (e.g., days) and for MMRM in units of the outcome (e.g., CDR-SB points). PMRM results were implemented in a health economics Markov model extrapolating disease progression and death over 15 years.ResultsThe PMRM model estimated a 19% delay in disease progression at 18 months and 20% (~ 7 months delay) at 36 months, while the MMRM model estimated a 25% reduction in CDR-SB (~ 0.5 points) at 36 months. The PMRM model had slightly greater power compared to MMRM. The health economic model based on the estimated time delay suggested an increase in life expectancy (10 months) without extending time in severe stages of disease.ConclusionPMRM methods can be used to estimate treatment effects in terms of slowing of progression which translates to time metrics that can be readily interpreted and appreciated as meaningful outcomes for patients, care partners, and health care practitioners.

Project description:Background Breast cancer (BC) survival prediction can be a helpful tool for identifying important factors selecting the effective treatment reducing mortality rates. This study aims to predict the time-related survival probability of BC patients in different molecular subtypes over 30 years of follow-up. Materials and methods This study retrospectively analyzed 3580 patients diagnosed with invasive breast cancer (BC) from 1991 to 2021 in the Cancer Research Center of Shahid Beheshti University of Medical Science. The dataset contained 18 predictor variables and two dependent variables, which referred to the survival status of patients and the time patients survived from diagnosis. Feature importance was performed using the random forest algorithm to identify significant prognostic factors. Time-to-event deep-learning-based models, including Nnet-survival, DeepHit, DeepSurve, NMLTR and Cox-time, were developed using a grid search approach with all variables initially and then with only the most important variables selected from feature importance. The performance metrics used to determine the best-performing model were C-index and IBS. Additionally, the dataset was clustered based on molecular receptor status (i.e., luminal A, luminal B, HER2-enriched, and triple-negative), and the best-performing prediction model was used to estimate survival probability for each molecular subtype. Results The random forest method identified tumor state, age at diagnosis, and lymph node status as the best subset of variables for predicting breast cancer (BC) survival probabilities. All models yielded very close performance, with Nnet-survival (C-index=0.77, IBS=0.13) slightly higher using all 18 variables or the three most important variables. The results showed that the Luminal A had the highest predicted BC survival probabilities, while triple-negative and HER2-enriched had the lowest predicted survival probabilities over time. Additionally, the luminal B subtype followed a similar trend as luminal A for the first five years, after which the predicted survival probability decreased steadily in 10- and 15-year intervals. Conclusion This study provides valuable insight into the survival probability of patients based on their molecular receptor status, particularly for HER2-positive patients. This information can be used by healthcare providers to make informed decisions regarding the appropriateness of medical interventions for high-risk patients. Future clinical trials should further explore the response of different molecular subtypes to treatment in order to optimize the efficacy of breast cancer treatments.

Project description:Event-time or continuous-time statistical approaches have been applied throughout the biostatistical literature and have led to numerous scientific advances. However, these techniques have traditionally relied on knowing failure times. This has limited application of these analyses, particularly, within the ecological field where fates of marked animals may be unknown. To address these limitations, we developed an integrated approach within a Bayesian framework to estimate hazard rates in the face of unknown fates. We combine failure/survival times from individuals whose fates are known and times of which are interval-censored with information from those whose fates are unknown, and model the process of detecting animals with unknown fates. This provides the foundation for our integrated model and permits necessary parameter estimation. We provide the Bayesian model, its derivation, and use simulation techniques to investigate the properties and performance of our approach under several scenarios. Lastly, we apply our estimation technique using a piece-wise constant hazard function to investigate the effects of year, age, chick size and sex, sex of the tending adult, and nesting habitat on mortality hazard rates of the endangered mountain plover (Charadrius montanus) chicks. Traditional models were inappropriate for this analysis because fates of some individual chicks were unknown due to failed radio transmitters. Simulations revealed biases of posterior mean estimates were minimal (? 4.95%), and posterior distributions behaved as expected with RMSE of the estimates decreasing as sample sizes, detection probability, and survival increased. We determined mortality hazard rates for plover chicks were highest at <5 days old and were lower for chicks with larger birth weights and/or whose nest was within agricultural habitats. Based on its performance, our approach greatly expands the range of problems for which event-time analyses can be used by eliminating the need for having completely known fate data.

Project description:BackgroundTime-to-event data is frequently reported in both clinical and preclinical research spheres. Systematic review and meta-analysis is a tool that can help to identify pitfalls in preclinical research conduct and reporting that can help to improve translational efficacy. However, pooling of studies using hazard ratios (HRs) is cumbersome especially in preclinical meta-analyses including large numbers of small studies. Median survival is a much simpler metric although because of some limitations, which may not apply to preclinical data, it is generally not used in survival meta-analysis. We aimed to appraise its performance when compared with hazard ratio-based meta-analysis when pooling large numbers of small, imprecise studies.MethodsWe simulated a survival dataset with features representative of a typical preclinical survival meta-analysis, including with influence of a treatment and a number of covariates. We calculated individual patient data-based hazard ratios and median survival ratios (MSRs), comparing the summary statistics directly and their performance at random-effects meta-analysis. Finally, we compared their sensitivity to detect associations between treatment and influential covariates at meta-regression.ResultsThere was an imperfect correlation between MSR and HR, although the opposing direction of treatment effects between summary statistics appeared not to be a major issue. Precision was more conservative for HR than MSR, meaning that estimates of heterogeneity were lower. There was a slight sensitivity advantage for MSR at meta-analysis and meta-regression, although power was low in all circumstances.ConclusionsWe believe we have validated MSR as a summary statistic for use in a meta-analysis of small, imprecise experimental survival studies-helping to increase confidence and efficiency in future reviews in this area. While assessment of study precision and therefore weighting is less reliable, MSR appears to perform favourably during meta-analysis. Sensitivity of meta-regression was low for this set of parameters, so pooling of treatments to increase sample size may be required to ensure confidence in preclinical survival meta-regressions.