Participants who are randomized to treatment but have no post-baseline data pose a unique challenge. These participants need to be included to preserve randomization. Because there is no information about the outcome or the intercurrent event(s) that led to missing data, an estimand of interest, and the focus of this study, is a hypothetical strategy to estimate what would have been observed if participants had not discontinued. Various imputation-based and likelihood-based analyses were compared in simulated and real clinical trial data. Models that used baseline as a covariate or constrained baseline values to be equal yielded similar results and had greater power than an unconstrained analysis that fit baseline as a response. Assigning change to the first post-baseline visit as 0 and applying an analysis with baseline as a covariate controlled Type I error at the nominal level and had power equal to or greater than other methods. Treatment contrasts were not biased when the reason for missing all post-baseline data was random or treatment related. Within-group bias occurred with outcome-related missingness of all post-baseline data, but the bias was nearly equal in the two arms, leading to unbiased treatment contrasts. Bias occurred when missing all post-baseline data was related to treatment and outcome. Given the idiosyncratic nature of clinical trials, no universally best analytic approach exists for dealing with participants that have a baseline but no post-baseline data. Analysts can choose among the methods to tailor an approach to the situation at hand.

Ongoing dose-escalation trials present unique challenges in assessing safety, all with the goal to establish the maximum tolerated dose (MTD) and/or the recommended phase II dose (RP2D). Due to the extended duration of trials, it is common to declare doses as safe even if the entire adaptive and iterative dose escalation process has not been completed. This practice is often applied to aid trial management decisions, such as starting region-specific dose-escalation studies, initiating combination therapies, or allowing intra-patient dose escalation in previous cohorts, to enhance efficiency and accelerate the pace of drug development. Whereas regulatory agencies often require a safety margin for the starting dose for regional or combination dose-escalation studies, there exists a misperception within the research community that once an escalation assessment team has cleared a cohort for escalation, the associated dose level is inherently safe. Given that dose escalation algorithms permit repeated de-escalations, previously cleared doses may not be truly safe, necessitating a quantification of this risk and identifying an adequate safety margin. Through rigorous simulation studies, we identified criteria that must be met to confidently declare dose levels as safe within the Bayesian optimal interval (BOIN) design. Our comparative analysis of safety criteria shows that evaluating a minimum of three subjects at the dose level above the one to be declared safe is essential for accurate safety assessments and protecting patients. More aggressive dose-escalation paradigms would risk exposing patients to doses that are not ultimately declared safe.

The estimand framework proposes different strategies to address intercurrent events. The treatment policy strategy seems to be the most favoured as it is closely aligned with the pre-addendum intention-to-treat principle. All data for all patients should ideally be collected; however, in reality patients may withdraw from a study leading to missing data. This needs to be dealt with as part of the estimation. A common intercurrent event we focus on is treatment discontinuation. Several areas of research have been conducted exploring models to estimate the estimand when intercurrent events are handled using a treatment policy strategy; however, the research is limited for binary endpoints. We explore different retrieved dropout models, where post-intercurrent event, the observed data can be used to multiply impute the missing post-intercurrent event data. We compare our proposed models to a simple imputation model that makes no distinction between the pre- and post-intercurrent event data, and assess varying statistical properties through a simulation study. We then provide an example of how retrieved dropout models were used in practice for Phase 3 clinical trials in rheumatoid arthritis. From the models explored, we conclude that a simple retrieved dropout model including an indicator for whether or not the intercurrent event occurred is the most pragmatic choice. However, at least 50% of observed post-intercurrent event data is required for these models to work well. Therefore, the suitability of implementing this model in practice will depend on the amount of observed post-intercurrent event data available and missing data.

The introduction of checkpoint inhibitors in immuno-oncology has raised questions about the suitability of the log-rank test as the default primary analysis method in confirmatory studies, particularly when survival curves exhibit non-proportional hazards. The log-rank test, while effective in controlling false positive rates, may lose power in scenarios where survival curves remain similar for extended periods before diverging. To address this, various weighted versions of the log-rank test have been proposed, including the "MaxCombo" test, which combines multiple weighted log-rank statistics to enhance power across a range of alternative hypotheses. Despite its potential, the MaxCombo test has seen limited adoption, possibly owing to its proneness to produce counterintuitive results insituations where the hazard functions on the two arms cross. In response, the modestly weighted log-rank test was developed to provide a balanced approach, giving greater weight to later event times while avoiding undue influence from early detrimental effects. However, this test also faces limitations, particularly if the possibility of early separation of survival curves cannot be ruled out a priori. We propose a novel test statistic that integrates the strengths of the standard log-rank test, the modestly weighted log-rank test, and the MaxCombo test. By considering the maximum of the standard log-rank statistic and a modestly weighted log-rank statistic, the new test aims to maintain power under delayed effect scenarios while minimizing power loss relative to the log-rank test in worst-case scenarios. Simulation studies and a case study demonstrate the efficiency and robustness of this approach, highlighting its potential as a robust alternative for primary analysis in immuno-oncology trials.

Motivated by the analysis of data from a clinical trial on patients with early breast cancer, we propose in this paper a new joint model that uses a Tobit partly linear mixed model for longitudinal measurements which are bounded in an interval and have a nonlinear relationship with the observation times and a semiparametric mixture cure model that incorporates a B-spline baseline hazard for survival times with cure proportion. A procedure is developed for estimating parameters in the proposed model using the partial likelihood and Laplace approximation. Additionally, a method of random weighting is proposed to compute the variances of the parameter estimators. The performance of the proposed model and the inference procedures is evaluated through simulation studies and data from the clinical trial that motivated this study.

ABSTRACTDuring a vaccine development program, if the assay used to measure immunological endpoints is changed, ideally, a bridging study is performed to establish the relationship between results obtained with the new and previous assay. However, this is not always feasible, and when bridging study data are absent, this can limit the ability to use historical study information to strengthen evidence generated in the clinical program. We present a case study on GSK's quadrivalent meningococcal vaccine (MenACWY‐CRM), where the immunogenicity assay was changed over time. A large amount of study data was collected in randomized controlled clinical trials, providing a valuable source of information to support vaccine development, but the introduction of the new assay complicated the comparison of antibody responses across studies. Several causal inference techniques, developed for the analysis of non‐randomized studies, can be used to estimate the assay bridging effect and, as observed in our case study, address the presence of confounding factors resulting from pooling group data from different sources. Cutting‐edge propensity score‐based methods were evaluated, highlighting their advantages and limitations. Within the family of propensity score weighting methods, the widely used inverse probability weighting was compared to the novel overlap weighting technique. The latter was shown to resolve the problem of extreme weights in a situation where there was poor overlap in covariate distribution between two groups. Automated selection of specific methods should be approached with caution, carefully considering the different estimands targeted by different methods.

ABSTRACTOverdispersion, a common issue in clustered multinomial data, can lead to biased standard errors and compromised statistical inference if not adequately addressed. This study describes a comprehensive procedure for constructing multiple comparisons of interest and applying multiplicity adjustments in the analysis of clustered, potentially overdispersed multinomial data. We investigate four quasi‐likelihood estimators and the Dirichlet‐multinomial model to account for overdispersion. Through a simulation study, we evaluate the performance of these methods under various scenarios, focusing on family‐wise error rate, statistical power and coverage probability. Our findings indicate that the Afroz quasi‐likelihood estimator is recommended when strict error control is required, whereas the Dirichlet‐multinomial model is preferable when high statistical power is desired, albeit with a slightly increased tolerance for false positives. Additionally, we address the challenge of zero‐count categories within groups, demonstrating that incorporating pseudo‐observations can effectively mitigate associated estimation difficulties. Practical applications to real datasets from toxicology and flow cytometry underscore the robustness and practical utility of these methods.

ABSTRACTThe conduct of dose‐finding trials can be specifically challenging in small populations, for example, in pediatric settings. Recently, research has shown that Bayesian borrowing from adult trials combined with appropriately robust prior distributions enables the conduct of pediatric dose‐finding trials with very small sample sizes. However, the appropriate degree of borrowing remains a subjective choice, relying on default methods or expert opinion. This paper proposes an approach to empirically determine the degree of borrowing based on a meta‐analysis of the similarity between population‐specific dose‐toxicity curves of other biologically similar compounds. Focusing on the pediatric use case, the approach may be applicable to any dose‐finding trial with information borrowing from another population. With the ExNex and a hierarchical model, two popular statistical modeling approaches are applied. The estimated degree of similarity is then translated into the statistical model for the dose‐finding algorithm using either variance inflation or robust mixture prior distributions. The performance of each combination of statistical model approaches is investigated in a simulation study. The results with mixture priors are promising for the application of the proposed methods, especially with many (20) compounds, while variance inflation models require additional fine‐tuning and seem to be less robust. With fewer (3 or 7) compounds, our proposed methods are either in line with robust priors that ignore the data from other compounds or are slightly better. We further provide a case study analyzing real dose‐finding data from 6 compounds with our models, demonstrating applicability in real‐world situations. For clinical trials teams, the decision for or against the proposed approach might be connected to the efforts in terms of time and cost to receive the external data.