Objective To explore a Bayesian logistic regression model for step－in dosing regimens(eBLRM),which considers the cumulative toxicity probability across different dosing cycles to identify the maximum tolerated schedule(MTS).Methods The Bayesian logistic regression model(BLRM)was extended to obtain a posterior estimate for the cumulative toxicity probability of the last cycle based on accumulated patient data,enabling exploration of dose sequences.Results The performance of eBLRM was evaluated by comparison with the existing methods.Simulation results indicated that eBLRM performed better or equivalent in the proportion of the correct selection of MTS and patients assigned to real MTS under low－toxicity scenarios.In the case of high－toxicity scenarios,eBLRM had a higher proportion of early trial termination due to safety,resulting in slightly inferior performance compared to the existing method.Conclusion The eBLRM method demonstrates relatively good performance,providing a simple and comprehensible dose exploration approach for step－in dosing regimens.

Objective To explore a Bayesian logistic regression model for step－in dosing regimens(eBLRM),which considers the cumulative toxicity probability across different dosing cycles to identify the maximum tolerated schedule(MTS).Methods The Bayesian logistic regression model(BLRM)was extended to obtain a posterior estimate for the cumulative toxicity probability of the last cycle based on accumulated patient data,enabling exploration of dose sequences.Results The performance of eBLRM was evaluated by comparison with the existing methods.Simulation results indicated that eBLRM performed better or equivalent in the proportion of the correct selection of MTS and patients assigned to real MTS under low－toxicity scenarios.In the case of high－toxicity scenarios,eBLRM had a higher proportion of early trial termination due to safety,resulting in slightly inferior performance compared to the existing method.Conclusion The eBLRM method demonstrates relatively good performance,providing a simple and comprehensible dose exploration approach for step－in dosing regimens.

Objective To explore the statistical performance and applicable conditions of Bayesian additive regression tree(BART)for estimating average treatment effect in observational studies.Methods The difference of estimates between BART and multivariate regression,propensity score matching,and inverse probability weighting through simulations and actual epidemiological data was compared.Results The results of these simulations showed that under the linear assumption,the performance of BART was close to that of the commonly used methods;when the relationship among variables in the data was complex and non-linear,BART performed markedly better than the others.When the ignorability assumption was not satisfied and there was unobserved confounding,all methods performed worse,but BART was still significantly better than the others and relatively robust.In the actual epidemiological data,this method was used to estimate the average treatment effect of smoking cessation on weight change.Conclusion In most observational studies,outcomes are influenced by multiple factors,making it difficult for researchers to properly specify relationships between variables.It is difficult to identify all these variables or determine the relationship between them.In terms of model fitting and result accuracy,BART is worth recommending.

AIM: To investigate the application of two-stage estimation (TSE) on adjustment for treatment switch in oncology trials. METHODS: The theory and implementation of TSE method was described, and was applied to adjust the data from a two-arm randomized controlled trial of anti-tumor drugs. The changes of survival curves and hazard ratio of two groups after adjustment for cross-over were evaluated. In addition, the results of two-stage estimation and rank preserving structural failure time model (RPSFT) were compared. RESULTS: After adjustment for cross-over using TSE methods, the results showed that the median survival time of control group was shorter than the original one, and the hazard ratio was lower than the observed value. Moreover, TSE method showed similar results to rank preserving structural failure time model. CONCLUSION: The TSE method is relatively simple to use, reliable and has a good practice property in cross-over analysis of oncology trials. At the same time, it is necessary to pay attention to its application scopes.