arXiv subjects

Zhongheng Cai

Publications and source records attributed to Zhongheng Cai.

4 recordsLinked to original sources

arXiv2026

Interim Monitoring as an Information-Time Alignment Problem: The WCR Framework for Time-to-Event Trials

Interim monitoring in time-to-event trials must balance inferential maturity with operationally meaningful timing. Event-driven designs align analyses with event accumulation but can produce substantial and unpredictable calendar delays, whereas enrollment-driven designs provide predictable timing but may rely on immature follow-up. We propose the Window-Cohort with Calibrated Follow-Up Requirement (WCR) framework, which directly parameterizes follow-up maturity through a locked cohort size and a post-lock follow-up requirement. The interim analysis is conducted after the prespecified cohort has accrued the calibrated minimum follow-up, while enrollment may continue and later patients are reserved for the final analysis. The framework distinguishes restricted follow-up for landmark survival estimands from unrestricted follow-up for proportional hazards estimands, thereby linking the effective information horizon to the estimand. Design parameters and decision thresholds are jointly calibrated through constrained optimization to control type I error and power while balancing calendar time, interim maturity, and decision-lag burden. Simulation studies motivated by a rare pediatric oncology trial show that WCR attains target operating characteristics under the calibration model and offers more stable and interpretable interim timing than conventional event-driven and enrollment-driven approaches. The methodology is implemented in the open-source R package WCRBayesDesign, available on CRAN. WCR reframes interim monitoring as an information-time alignment problem and provides a practical design strategy for single-arm trials with sparse events, slow accrual, and long-horizon endpoints.

stat.ME↗

arXiv2025

A Bayesian Optimal Phase II Design for Randomized Immunotherapy Trials with Delayed Treatment Effects

Immunotherapy has transformed cancer treatment, yet its delayed therapeutic effects often lead to non-proportional hazards, rendering many conventional phase II designs underpowered and prone to type I error inflation. To address this issue, we propose a novel Bayesian Optimal Phase II design (DTE-BOP2) that explicitly models the uncertainty in the separation timing of treatment effect. The treatment separation timepoint (denoted by S) is endowed with a truncated-Gamma prior, whose parameters can be elicited from experts or inferred from historical data, with default settings available when prior knowledge is scarce. Built upon the BOP2 framework (Zhou et al. 2017, 2020), our design retains operational simplicity while incorporating type I error control and maintaining the power. Extensive simulations demonstrate that DTE-BOP2 uniformly controls type I error at the nominal level across a wide range of treatment effect separation timepoint S. We further observe that the power decreases monotonically as S increases. Importantly, we find that the power is primarily driven by the relative magnitude of treatment benefit before and after the separation time, i.e., the ratio of medians, rather than their absolute values. Compared to the original BOP2, the piecewise weighted log-rank, and the conventional log-rank tests, DTE-BOP2 achieves higher power with smaller sample sizes while preserving type I error robustness across plausible delay scenarios. An open-source R package, DTEBOP2 (CRAN), with detailed vignettes, enables investigators to implement the design and analyse phase-II trials exhibiting delayed treatment effects.

stat.ME↗

arXiv2025

A Novel Bayesian Extrapolation Design for Assessing Equivalence in Exposure-Response Curves between Pediatric and Adult Populations

Development of effective treatments in pediatric population poses unique scientific and ethical challenges in addition to the small population. In this regard, both the U.S. and E.U. regulations suggest a complementary strategy, pediatric extrapolation, based on assessing the relevance of existing information in the adult population to the pediatric population. The pediatric extrapolation approach often relies on data extrapolation from adults, contingent upon evidence of similar disease progression, pharmacology and clinical response to treatment between adult and children. Similarity evaluation in pharmacology is usually characterized through the exposure-response relationship. Current methodologies for comparing exposure-response (E-R) curves between these groups are inadequate, typically focusing on isolated data points rather than the entire curve spectrum (Zhang et al., 2021). To overcome this limitation, we introduce an innovative Bayesian approach for a comprehensive evaluation of E-R curve similarities between adult and pediatric populations. This method encompasses the entire curve, employing logistic regression for binary endpoints. We have developed an algorithm to determine sample size and key design parameters, such as the Bayesian posterior probability threshold, and utilize the maximum curve distance as a measure of similarity. Integrating Bayesian and frequentist principles, our approach involves developing a method to simulate datasets under both null and alternative hypotheses, allowing for type I error and type II error control. Simulation studies and sensitivity analyses demonstrate that our method maintains a stable performance with type I error and type II error control.

stat.AP↗

arXiv2024

Bayesian Hierarchical Model for Synthesizing Registry and Survey Data on Female Breast Cancer Prevalence

In public health, it is critical for policymakers to assess the relationship between the disease prevalence and associated risk factors or clinical characteristics, facilitating effective resources allocation. However, for diseases like female breast cancer (FBC), reliable prevalence data at specific geographical levels, such as the county-level, are limited because the gold standard data typically come from long-term cancer registries, which do not necessarily collect needed risk factors. In addition, it remains unclear whether fitting each model separately or jointly results in better estimation. In this paper, we identify two data sources to produce reliable county-level prevalence estimates in Missouri, USA: the population-based Missouri Cancer Registry (MCR) and the survey-based Missouri County-Level Study (CLS). We propose a two-stage Bayesian model to synthesize these sources, accounting for their differences in the methodological design, case definitions, and collected information. The first stage involves estimating the county-level FBC prevalence using the raking method for CLS data and the counting method for MCR data, calibrating the differences in the methodological design and case definition. The second stage includes synthesizing two sources with different sets of covariates using a Bayesian generalized linear mixed model with Zeller-Siow prior for the coefficients. Our data analyses demonstrate that using both data sources have better results than at least one data source, and including a data source membership matters when there exist systematic differences in these sources. Finally, we translate results into policy making and discuss methodological differences for data synthesis of registry and survey data.

stat.AP↗