Bayesian design of clinical trials using the scale transformed power prior

There are many Bayesian design methods allowing for the incorporation of historical data for sample size determination (SSD) in situations where the outcome in the historical data is the same as the outcome of a new study. However, there is a dearth of methods supporting the incorporation of data from a previously completed clinical trial that investigated the same or similar treatment as the new trial but had a primary outcome that is different. We propose a simulation-based Bayesian SSD framework using the partial-borrowing scale transformed power prior (straPP). The partial-borrowing straPP is developed by applying a novel scale transformation to a traditional power prior on the parameters from the historical data model to make the information better align with the new data model. The scale transformation is based on the assumption that the standardized parameters (i.e., parameters multiplied by the square roots of their respective Fisher information matrices) are equal. To illustrate the method, we present results from simulation studies that use real data from a previously completed clinical trial to design a new clinical trial with a primary time-to-event endpoint.

Adaptively leverage multiple real-world data sources for treatment effect estimation based on similarity

The incorporation of real-world data (RWD) into medical product development and evaluation has exhibited consistent growth. However, there is no universally adopted method of how much information to borrow from external data. This paper proposes a study design methodology called Tree-based Monte Carlo (TMC) that dynamically integrates patients from various RWD sources to calculate the treatment effect based on the similarity between clinical trial and RWD. Initially, a propensity score is developed to gauge the resemblance between clinical trial data and each real-world dataset. Utilizing this similarity metric, we construct a hierarchical clustering tree that delineates varying degrees of similarity between each RWD source and the clinical trial data. Ultimately, a Gaussian process methodology is employed across this hierarchical clustering framework to synthesize the projected treatment effects of the external group. Simulation result shows that our clustering tree could successfully identify similarity. Data sources exhibiting greater similarity with clinical trial are accorded higher weights in treatment estimation process, while less congruent sources receive comparatively lower emphasis. Compared with another Bayesian method, meta-analytic predictive prior (MAP), our proposed method's estimator is closer to the true value and has smaller bias.

A generalized calibrated Bayesian hierarchical modeling approach to basket trials with multiple endpoints

A basket trial simultaneously evaluates a treatment in multiple cancer subtypes, offering an effective way to accelerate drug development in multiple indications. Many basket trials are designed and monitored based on a single efficacy endpoint, primarily the tumor response. For molecular targeted or immunotherapy agents, however, a single efficacy endpoint cannot adequately characterize the treatment effect. It is increasingly important to use more complex endpoints to comprehensively assess the risk-benefit profile of such targeted therapies. We extend the calibrated Bayesian hierarchical modeling approach to monitor phase II basket trials with multiple endpoints. We propose two generalizations, one based on the latent variable approach and the other based on the multinomial-normal hierarchical model, to accommodate different types of endpoints and dependence assumptions regarding information sharing. We introduce shrinkage parameters as functions of statistics measuring homogeneity among subgroups and propose a general calibration approach to determine the functional forms. Theoretical properties of the generalized hierarchical models are investigated. Simulation studies demonstrate that the monitoring procedure based on the generalized approach yields desirable operating characteristics.

Adaptive Bayesian information borrowing methods for finding and optimizing subgroup-specific doses

In precision oncology, integrating multiple cancer patient subgroups into a single master protocol allows for the simultaneous assessment of treatment effects in these subgroups and promotes the sharing of information between them, ultimately reducing sample sizes and costs and enhancing scientific validity. However, the safety and efficacy of these therapies may vary across different subgroups, resulting in heterogeneous outcomes. Therefore, identifying subgroup-specific optimal doses in early-phase clinical trials is crucial for the development of future trials. In this article, we review various innovative Bayesian information-borrowing strategies that aim to determine and optimize subgroup-specific doses. Specifically, we discuss Bayesian hierarchical modeling, Bayesian clustering, Bayesian model averaging or selection, pairwise borrowing, and other relevant approaches. By employing these Bayesian information-borrowing methods, investigators can gain a better understanding of the intricate relationships between dose, toxicity, and efficacy in each subgroup. This increased understanding significantly improves the chances of identifying an optimal dose tailored to each specific subgroup. Furthermore, we present several practical recommendations to guide the design of future early-phase oncology trials involving multiple subgroups when using the Bayesian information-borrowing methods.

Generalized triple outcome decision-making in basket trials

Making the go/no-go decision is critical in Phase II (or Ib) clinical trials. The conventional decision-making framework based on a binary hypothesis testing has been gradually replaced by the TODeM (Triple Outcome Decision-Making) which has three zones of outcomes: go, no-go, and consider. The TODeM provides more flexibility in decision-making with considering both of statistical significance and clinical relevance. However, Bayesian methods (e.g. EXNEX, MUCE, etc.) for the information borrowing are still based on the binary decision-making framework. We propose a new decision-making process G-TODeM (Generalized Triple Outcome Decision-Making) to apply those Bayesian methods with information borrowing across different cohorts to the TODeM framework. Essentially, the information borrowed from other cohorts can shrink the consider zone of the inference cohort.

Basket trials in oncology: a systematic review of practices and methods, comparative analysis of innovative methods, and an appraisal of a missed opportunity

Background

Basket trials are increasingly used in oncology drug development for early signal detection, accelerated tumor-agnostic approvals, and prioritization of promising tumor types in selected patients with the same mutation or biomarker. Participants are grouped into so-called baskets according to tumor type, allowing investigators to identify tumors with promising responses to treatment for further study. However, it remains a question as to whether and how much the adoption of basket trial designs in oncology have translated into patient benefits, increased pace and scale of clinical development, and de-risking of downstream confirmatory trials.

Methods

Innovation in basket trial design and analysis includes methods that borrow information across tumor types to increase the quality of statistical inference within each tumor type. We build on the existing systematic reviews of basket trials in oncology to discuss the current practices and landscape. We conceptually illustrate recent innovative methods for basket trials, with application to actual data from recently completed basket trials. We explore and discuss the extent to which innovative basket trials can be used to de-risk future trials through their ability to aid prioritization of promising tumor types for subsequent clinical development.

Results

We found increasing adoption of basket trial design in oncology, but largely in the design of single-arm phase II trials with a very low adoption of innovative statistical methods. Furthermore, the current practice of basket trial design, which does not consider its impact on the clinical development plan, may lead to a missed opportunity in improving the probability of success of a future trial. Gating phase II with a phase Ib basket trial reduced the size of phase II trials, and losses in the probability of success as a result of not using innovative methods may not be recoverable by running a larger phase II trial.

Conclusion

Innovative basket trial methods can reduce the size of early phase clinical trials, with sustained improvement in the probability of success of the clinical development plan. We need to do more as a community to improve the adoption of these methods.

A comparison of Bayesian information borrowing methods in basket trials and a novel proposal of modified exchangeability-nonexchangeability method

Recent innovation in trial design to improve study efficiency has led to the development of basket trials in which a single therapeutic treatment is tested on several patient populations, each of which forms a basket. In a common setting, patients across all baskets share a genetic marker and as such, an assumption can be made that all patients may have a homogeneous response to treatments. Bayesian information borrowing procedures utilize this assumption to draw on information regarding the response in one basket when estimating the response rate in others. This can improve power and precision of estimates particularly in the presence of small sample sizes, however, can come at a cost of biased estimates and an inflation of error rates, bringing into question validity of trial conclusions. We review and compare the performance of several Bayesian borrowing methods, namely: the Bayesian hierarchical model (BHM), calibrated Bayesian hierarchical model (CBHM), exchangeability-nonexchangeability (EXNEX) model and a Bayesian model averaging procedure. A generalization of the CBHM is made to account for unequal sample sizes across baskets. We also propose a modification of the EXNEX model that allows for better control of a type I error. The proposed method uses a data-driven approach to account for the homogeneity of the response data, measured through Hellinger distances. Through an extensive simulation study motivated by a real basket trial, for both equal and unequal sample sizes across baskets, we show that in the presence of a basket with a heterogeneous response, unlike the other methods discussed, this model can control type I error rates to a nominal level whilst yielding improved power.

A Bayesian adaptive design for dual-agent phase I-II oncology trials integrating efficacy data across stages

Combination of several anticancer treatments has typically been presumed to have enhanced drug activity. Motivated by a real clinical trial, this paper considers phase I-II dose finding designs for dual-agent combinations, where one main objective is to characterize both the toxicity and efficacy profiles. We propose a two-stage Bayesian adaptive design that accommodates a change of patient population in-between. In stage I, we estimate a maximum tolerated dose combination using the escalation with overdose control (EWOC) principle. This is followed by a stage II, conducted in a new yet relevant patient population, to find the most efficacious dose combination. We implement a robust Bayesian hierarchical random-effects model to allow sharing of information on the efficacy across stages, assuming that the related parameters are either exchangeable or nonexchangeable. Under the assumption of exchangeability, a random-effects distribution is specified for the main effects parameters to capture uncertainty about the between-stage differences. The inclusion of nonexchangeability assumption further enables that the stage-specific efficacy parameters have their own priors. The proposed methodology is assessed with an extensive simulation study. Our results suggest a general improvement of the operating characteristics for the efficacy assessment, under a conservative assumption about the exchangeability of the parameters a priori.

Synthesizing secondary data into survival analysis to improve estimation efficiency

The accelerated failure time (AFT) model and Cox proportional hazards (PH) model are broadly used for survival endpoints of primary interest. However, the estimation efficiency from those models can be further enhanced by incorporating the information from secondary outcomes that are increasingly available and highly correlated with primary outcomes. Those secondary outcomes could be longitudinal laboratory measures collected from doctor visits or cross-sectional disease-relevant variables, which are believed to contain extra information related to primary survival endpoints to a certain extent. In this paper, we develop a two-stage estimation framework to combine a survival model with a secondary model that contains secondary outcomes, named as the empirical-likelihood-based weighting (ELW), which comprises two weighting schemes accommodated to the AFT model (ELW-AFT) and the Cox PH model (ELW-Cox), respectively. This innovative framework is flexibly adaptive to secondary outcomes with complex data features, and it leads to more efficient parameter estimation in the survival model even if the secondary model is misspecified. Extensive simulation studies showcase more efficiency gain from ELW compared to conventional approaches, and an application in the Atherosclerosis Risk in Communities study also demonstrates the superiority of ELW by successfully detecting risk factors at the time of hospitalization for acute myocardial infarction.

The scale transformed power prior for use with historical data from a different outcome model

We develop the scale transformed power prior for settings where historical and current data involve different data types, such as binary and continuous data. This situation arises often in clinical trials, for example, when historical data involve binary responses and the current data involve some other type of continuous or discrete outcome. The power prior, proposed by Ibrahim and Chen, does not address the issue of different data types. Herein, we develop a new type of power prior, which we call the scale transformed power prior (straPP). The straPP is constructed by transforming the power prior for the historical data by rescaling the parameter using a function of the Fisher information matrices for the historical and current data models, thereby shifting the scale of the parameter vector from that of the historical to that of the current data. Examples are presented to motivate the need for such a transformation, and simulation studies are presented to illustrate the performance advantages of the straPP over the power prior and other informative and noninformative priors. A real dataset from a clinical trial undertaken to study a novel transitional care model for stroke survivors is used to illustrate the methodology.

Bayesian hierarchical models for adaptive basket trial designs

Basket trials evaluate a single drug targeting a single genetic variant in multiple cancer cohorts. Empirical findings suggest that treatment efficacy across baskets may be heterogeneous. Most modern basket trial designs use Bayesian methods. These methods require the prior specification of at least one parameter that permits information sharing across baskets. In this study, we provide recommendations for selecting a prior for scale parameters for adaptive basket trials by using Bayesian hierarchical modeling. Heterogeneity among baskets attracts much attention in basket trial research, and substantial heterogeneity challenges the basic assumption of exchangeability of Bayesian hierarchical approach. Thus, we also allowed each stratum-specific parameter to be exchangeable or nonexchangeable with similar strata by using data observed in an interim analysis. Through a simulation study, we evaluated the overall performance of our design based on statistical power and type I error rates. Our research contributes to the understanding of the properties of Bayesian basket trial designs.

A structured framework for adaptively incorporating external evidence in sequentially monitored clinical trials

We present a Bayesian framework for sequential monitoring that allows for use of external data, and that can be applied in a wide range of clinical trial applications. The basis for this framework is the idea that, in many cases, specification of priors used for sequential monitoring and the stopping criteria can be semi-algorithmic byproducts of the trial hypotheses and relevant external data, simplifying the process of prior elicitation. Monitoring priors are defined using the family of generalized normal distributions, which comprise a flexible class of priors, naturally allowing one to construct a prior that is peaked or flat about the parameter values thought to be most likely. External data are incorporated into the monitoring process through mixing an a priori skeptical prior with an enthusiastic prior using a weight that can be fixed or adaptively estimated. In particular, we introduce an adaptive monitoring prior for efficacy evaluation that dynamically weighs skeptical and enthusiastic prior components based on the degree to which observed data are consistent with an enthusiastic perspective. The proposed approach allows for prospective and pre-specified use of external data in the monitoring procedure. We illustrate the method for both single-arm and two-arm randomized controlled trials. For the latter case, we also include a retrospective analysis of actual trial data using the proposed adaptive sequential monitoring procedure. Both examples are motivated by completed pediatric trials, and the designs incorporate information from adult trials to varying degrees. Preposterior analysis and frequentist operating characteristics of each trial design are discussed.

Monotonicity conditions for avoiding counterintuitive decisions in basket trials

In a basket trial, a new treatment is tested in different subgroups, called the baskets. In oncology, the baskets usually comprise patients with different primary tumor sites but a common biomarker. Most basket trials are uncontrolled phase II trials and investigate a binary endpoint such as tumor response. To combine the data of baskets that show a similar response to the treatment, many basket trial designs use Bayesian borrowing methods. This increases the power compared to a basketwise analysis. However, it can lead to posterior probabilities that are not monotonically increasing in the number of responses. We show that, as a consequence, two types of counterintuitive decisions can arise-one that occurs within a single trial and one that occurs when the results are compared between different trials. We propose two monotonicity conditions for the inference in basket trials. Using a design recently proposed by Fujikawa and colleagues, we investigate the case of a single-stage basket trial with equal sample sizes in all baskets and show that, as the number of baskets increases, these conditions are violated for a wide range of different borrowing strengths. We show that in the investigated scenarios pruning baskets can help to ensure that the monotonicity conditions hold and investigate how this affects type I error rate and power.

Improving main analysis by borrowing information from auxiliary data

In many clinical and observational studies, auxiliary data from the same subjects, such as repeated measurements or surrogate variables, will be collected in addition to the data of main interest. Not directly related to the main study, these auxiliary data in practice are rarely incorporated into the main analysis, though they may carry extra information that can help improve the estimation in the main analysis. Under the setting where part of or all subjects have auxiliary data available, we propose an effective weighting approach to borrow the auxiliary information by building a working model for the auxiliary data, where improvement of estimation precision over the main analysis is guaranteed regardless of the specification of the working model. An information index is also constructed to assess how well the selected working model works to improve the main analysis. Both theoretical and numerical studies show the excellent and robust performance of the proposed method in comparison to estimation without using the auxiliary data. Finally, we utilize the Atherosclerosis Risk in Communities study for illustration.

PA-CRM: A continuous reassessment method for pediatric phase I oncology trials with concurrent adult trials

Pediatric phase I trials are usually carried out after the adult trial testing the same agent has started, but not completed yet. As the pediatric trial progresses, in light of the accrued interim data from the concurrent adult trial, the pediatric protocol often is amended to modify the original pediatric dose escalation design. In practice, this is done frequently in an ad hoc way, interrupting patient accrual and slowing down the trial. We developed a pediatric-continuous reassessment method (PA-CRM) to streamline this process, providing a more efficient and rigorous method to find the maximum tolerated dose for pediatric phase I oncology trials. We use a discounted joint likelihood of the adult and pediatric data, with a discount parameter controlling information borrowing between pediatric and adult trials. According to the interim adult and pediatric data, the discount parameter is adaptively updated using the Bayesian model averaging method. Numerical study shows that the PA-CRM improves the efficiency and accuracy of the pediatric trial and is robust to various model assumptions.

Bayesian hierarchical classification and information sharing for clinical trials with subgroups and binary outcomes

Bayesian hierarchical models have been applied in clinical trials to allow for information sharing across subgroups. Traditional Bayesian hierarchical models do not have subgroup classifications; thus, information is shared across all subgroups. When the difference between subgroups is large, it suggests that the subgroups belong to different clusters. In that case, placing all subgroups in one pool and borrowing information across all subgroups can result in substantial bias for the subgroups with strong borrowing, or a lack of efficiency gain with weak borrowing. To resolve this difficulty, we propose a hierarchical Bayesian classification and information sharing (BaCIS) model for the design of multigroup phase II clinical trials with binary outcomes. We introduce subgroup classification into the hierarchical model. Subgroups are classified into two clusters on the basis of their outcomes mimicking the hypothesis testing framework. Subsequently, information sharing takes place within subgroups in the same cluster, rather than across all subgroups. This method can be applied to the design and analysis of multigroup clinical trials with binary outcomes. Compared to the traditional hierarchical models, better operating characteristics are obtained with the BaCIS model under various scenarios.