Bayesian hierarchical modeling based on multisource exchangeability

Leveraging information from secondary endpoints to enhance dynamic borrowing across subpopulations

Randomized trials seek efficient treatment effect estimation within target populations, yet scientific interest often also centers on subpopulations. Although there are typically too few subjects within each subpopulation to efficiently estimate these subpopulation treatment effects, one can gain precision by borrowing strength across subpopulations, as is the case in a basket trial. While dynamic borrowing has been proposed as an efficient approach to estimating subpopulation treatment effects on primary endpoints, additional efficiency could be gained by leveraging the information found in secondary endpoints. We propose a multisource exchangeability model (MEM) that incorporates secondary endpoints to more efficiently assess subpopulation exchangeability. Across simulation studies, our proposed model almost uniformly reduces the mean squared error when compared to the standard MEM that only considers data from the primary endpoint by gaining efficiency when subpopulations respond similarly to the treatment and reducing the magnitude of bias when the subpopulations are heterogeneous. We illustrate our model's feasibility using data from a recently completed trial of very low nicotine content cigarettes to estimate the effect on abstinence from smoking within three priority subpopulations. Our proposed model led to increases in the effective sample size two to four times greater than under the standard MEM.

Propensity score weighted multi-source exchangeability models for incorporating external control data in randomized clinical trials

Among clinical trialists, there has been a growing interest in using external data to improve decision-making and accelerate drug development in randomized clinical trials (RCTs). Here we propose a novel approach that combines the propensity score weighting (PW) and the multi-source exchangeability modelling (MEM) approaches to augment the control arm of a RCT in the rare disease setting. First, propensity score weighting is used to construct weighted external controls that have similar observed pre-treatment characteristics as the current trial population. Next, the MEM approach evaluates the similarity in outcome distributions between the weighted external controls and the concurrent control arm. The amount of external data we borrow is determined by the similarities in pretreatment characteristics and outcome distributions. The proposed approach can be applied to binary, continuous and count data. We evaluate the performance of the proposed PW-MEM method and several competing approaches based on simulation and re-sampling studies. Our results show that the PW-MEM approach improves the precision of treatment effect estimates while reducing the biases associated with borrowing data from external sources.

Multidisciplinary considerations for implementing Bayesian borrowing in basket trials

Drug development has historically relied on phase I-III clinical trials including participants sharing the same disease. However, drug development has evolved as the discovery of mechanistic drivers of disease demonstrated that the same therapeutic target may provide benefits across different diseases. A basket trial condenses evaluation of one therapy among multiple related diseases into a single trial and presents an opportunity to borrow information across them rather than viewing each in isolation. Borrowing is a statistical tool but requires a foundation of clinical and therapeutic mechanistic justification. We review the Bayesian borrowing approach, including its assumptions, and provide a framework for how this approach can be evaluated for successful use in a basket trial for drug development.

Discussion on "LEAP: the latent exchangeability prior for borrowing information from historical data" by Ethan M. Alt, Xiuya Chang, Xun Jiang, Qing Liu, May Mo, H. Amy Xia, and Joseph G. Ibrahim

In the following discussion, we describe the various assumptions of exchangeability that have been made in the context of Bayesian borrowing and related models. In this context, we are able to highlight the difficulty of dynamic Bayesian borrowing under the assumption of individuals in the historical data being exchangeable with the current data and thus the strengths and novel features of the latent exchangeability prior. As borrowing methods are popular within clinical trials to augment the control arm, some potential challenges are identified with the application of the approach in this setting.

Semi-supervised mixture multi-source exchangeability model for leveraging real-world data in clinical trials

The traditional trial paradigm is often criticized as being slow, inefficient, and costly. Statistical approaches that leverage external trial data have emerged to make trials more efficient by augmenting the sample size. However, these approaches assume that external data are from previously conducted trials, leaving a rich source of untapped real-world data (RWD) that cannot yet be effectively leveraged. We propose a semi-supervised mixture (SS-MIX) multisource exchangeability model (MEM); a flexible, two-step Bayesian approach for incorporating RWD into randomized controlled trial analyses. The first step is a SS-MIX model on a modified propensity score and the second step is a MEM. The first step targets a representative subgroup of individuals from the trial population and the second step avoids borrowing when there are substantial differences in outcomes among the trial sample and the representative observational sample. When comparing the proposed approach to competing borrowing approaches in a simulation study, we find that our approach borrows efficiently when the trial and RWD are consistent, while mitigating bias when the trial and external data differ on either measured or unmeasured covariates. We illustrate the proposed approach with an application to a randomized controlled trial investigating intravenous hyperimmune immunoglobulin in hospitalized patients with influenza, while leveraging data from an external observational study to supplement a subgroup analysis by influenza subtype.

Discussion on "LEAP: the latent exchangeability prior for borrowing information from historical data" by Ethan M. Alt, Xiuya Chang, Xun Jiang, Qing Liu, May Mo, H. Amy Xia, and Joseph G. Ibrahim

This discussion provides commentary on the paper by Ethan M. Alt, Xiuya Chang, Xun Jiang, Qing Liu, May Mo, H. Amy Xia, and Joseph G. Ibrahim entitled "LEAP: the latent exchangeability prior for borrowing information from historical data". The authors propose a novel method to bridge the incorporation of supplemental information into a study while also identifying potentially exchangeable subgroups to better facilitate information sharing. In this discussion, we highlight the potential relationship with other Bayesian model averaging approaches, such as multisource exchangeability modeling, and provide a brief numeric case study to illustrate how the concepts behind latent exchangeability prior may also improve the performance of existing methods. The results provided by Alt et al. are exciting, and we believe that the method represents a meaningful approach to more efficient information sharing.

Biomarker-driven basket trial designs: origins and new methodological developments

Due to increased use of gene sequencing techniques, understanding of cancer on a molecular level has evolved, in terms of both diagnosis and evaluation in response to initial therapies. In parallel, clinical trials meant to evaluate molecularly-driven interventions through assessment of both treatment effects and putative predictive biomarker effects are being employed to advance the goals of precision medicine. Basket trials investigate one or more biomarker-targeted therapies across multiple cancer types in a tumor location agnostic fashion. The review article offers an overview of the traditional forms of such designs, the practical challenges facing each type of design, and then review novel adaptations proposed in the last few years, categorized into Bayesian and Classical Frequentist perspectives. The review article concludes by summarizing potential advantages and limitations of the new trial design solutions.

A Bayesian platform trial design with hybrid control based on multisource exchangeability modelling

Enrolling patients to the standard of care (SOC) arm in randomized clinical trials, especially for rare diseases, can be very challenging due to the lack of resources, restricted patient population availability, and ethical considerations. As the therapeutic effect for the SOC is often well documented in historical trials, we propose a Bayesian platform trial design with hybrid control based on the multisource exchangeability modelling (MEM) framework to harness historical control data. The MEM approach provides a computationally efficient method to formally evaluate the exchangeability of study outcomes between different data sources and allows us to make better informed data borrowing decisions based on the exchangeability between historical and concurrent data. We conduct extensive simulation studies to evaluate the proposed hybrid design. We demonstrate the proposed design leads to significant sample size reduction for the internal control arm and borrows more information compared to competing Bayesian approaches when historical and internal data are compatible.

Optimal timing for an accelerated interim futility analysis incorporating real world data

BACKGROUND

Randomized controlled trials include interim monitoring guidelines to stop early for safety, efficacy, or futility. Futility monitoring facilitates re-allocation of limited resources. However, conventional methods for interim futility monitoring require a trial to accrue nearly half of the outcome data to make a reliable early stopping decision, limiting its benefit. As early stopping for futility will not inflate type-I error, these analyses are an appealing venue for incorporating external data to improve efficiency.

METHODS

We propose a Bayesian approach to futility monitoring leveraging real world data using Semi-Supervised MIXture Multi-source Exchangeability Models, which accounts for both measured and unmeasured differences between data sources. We implement futility monitoring using predictive probabilities and investigate the optimal timing with respect to the expected sample size under the null hypothesis. Because we only incorporate external data during the interim futility analysis the proposed design is not limited by type-I error inflation.

RESULTS

When the external and trial data are exchangeable, the proposed method provides a roughly 70 person reduction in expected sample size under the null. Under scenarios where exchangeability does not hold, our approach still provides a 10-20 person reduction in expected sample size under the null with about 80% power.

CONCLUSIONS

External data borrowing in interim futility monitoring is a promising venue to improve trial efficiency without type-I error inflation. Approaches that are acceptable to regulatory authorities and leverage the complementary strengths of real world and trial data are vital to more efficiently allocate limited resources amongst clinical trials.

Optimal ensemble construction for multistudy prediction with applications to mortality estimation

It is increasingly common to encounter prediction tasks in the biomedical sciences for which multiple datasets are available for model training. Common approaches such as pooling datasets before model fitting can produce poor out-of-study prediction performance when datasets are heterogeneous. Theoretical and applied work has shown multistudy ensembling to be a viable alternative that leverages the variability across datasets in a manner that promotes model generalizability. Multistudy ensembling uses a two-stage stacking strategy which fits study-specific models and estimates ensemble weights separately. This approach ignores, however, the ensemble properties at the model-fitting stage, potentially resulting in performance losses. Motivated by challenges in the estimation of COVID-attributable mortality, we propose optimal ensemble construction, an approach to multistudy stacking whereby we jointly estimate ensemble weights and parameters associated with study-specific models. We prove that limiting cases of our approach yield existing methods such as multistudy stacking and pooling datasets before model fitting. We propose an efficient block coordinate descent algorithm to optimize the loss function. We use our method to perform multicountry COVID-19 baseline mortality prediction. We show that when little data is available for a country before the onset of the pandemic, leveraging data from other countries can substantially improve prediction accuracy. We further compare and characterize the method's performance in data-driven simulations and other numerical experiments. Our method remains competitive with or outperforms multistudy stacking and other earlier methods in the COVID-19 data application and in a range of simulation settings.

Incorporating Efficacy Data from Initial Trials Into Subsequent Evaluations: Application to Vaccines Against Respiratory Syncytial Virus

BACKGROUND

When a randomized controlled trial fails to demonstrate statistically significant efficacy against the primary endpoint, a potentially costly new trial would need to be conducted to receive licensure. Incorporating data from previous trials might allow for more efficient follow-up trials to demonstrate efficacy, speeding the availability of effective vaccines.

METHODS

Based on the outcomes from a failed trial of a maternal vaccine against respiratory syncytial virus (RSV), we simulated data for a new Bayesian group-sequential trial. We analyzed the data either ignoring data from the previous trial (i.e., weakly informative prior distributions) or using prior distributions incorporating the historical data into the analysis. We evaluated scenarios where efficacy in the new trial was the same, greater than, or less than that in the original trial. For each scenario, we evaluated the statistical power and type I error rate for estimating the vaccine effect following interim analyses.

RESULTS

When we used a stringent threshold to control the type I error rate, analyses incorporating historical data had a small advantage over trials that did not. If control of type I error is less important (e.g., in a postlicensure evaluation), the incorporation of historical data can provide a substantial boost in efficiency.

CONCLUSIONS

Due to the need to control the type I error rate in trials used to license a vaccine, incorporating historical data provides little additional benefit in terms of stopping the trial early. However, these statistical approaches could be promising in evaluations that use real-world evidence following licensure.

Bayesian Methods for Information Borrowing in Basket Trials: An Overview

Basket trials allow simultaneous evaluation of a single therapy across multiple cancer types or subtypes of the same cancer. Since the same treatment is tested across all baskets, it may be desirable to borrow information across them to improve the statistical precision and power in estimating and detecting the treatment effects in different baskets. We review recent developments in Bayesian methods for the design and analysis of basket trials, focusing on the mechanism of information borrowing. We explain the common components of these methods, such as a prior model for the treatment effects that embodies an assumption of exchangeability. We also discuss the distinct features of these methods that lead to different degrees of borrowing. Through simulation studies, we demonstrate the impact of information borrowing on the operating characteristics of these methods and discuss its broader implications for drug development. Examples of basket trials are presented in both phase I and phase II settings.

Decoupling power and type I error rate considerations when incorporating historical control data using a test-then-pool approach

To accelerate a randomized controlled trial, historical control data may be used after ensuring little heterogeneity between the historical and current trials. The test-then-pool approach is a simple frequentist borrowing method that assesses the similarity between historical and current control data using a two-sided test. A limitation of the conventional test-then-pool method is the inability to control the type I error rate and power for the primary hypothesis separately and flexibly for heterogeneity between trials. This is because the two-sided test focuses on the absolute value of the mean difference between the historical and current controls. In this paper, we propose a new test-then-pool method that splits the two-sided hypothesis of the conventional method into two one-sided hypotheses. Testing each one-sided hypothesis with different significance levels allows for the separate control of the type I error rate and power for heterogeneity between trials. We also propose a significance-level selection approach based on the maximum type I error rate and the minimum power. The proposed method prevented a decrease in power even when there was heterogeneity between trials while controlling type I error at a maximum tolerable type I error rate larger than the targeted type I error rate. The application of depression trial data and hypothetical trial data further supported the usefulness of the proposed method.

MIDAS-2: an enhanced Bayesian platform design for immunotherapy combinations with subgroup efficacy exploration

Although immunotherapy combinations have revolutionised cancer treatment, the rapid screening of effective and optimal therapies from large numbers of candidate combinations, as well as exploring subgroup efficacy, remains challenging. This necessitates innovative, integrated, and efficient trial designs. In this study, we extend the MIDAS design to include subgroup exploration and propose an enhanced Bayesian information borrowing platform design called MIDAS-2. MIDAS-2 enables quick and continuous screening of promising combination strategies and exploration of their subgroup effects within a unified platform design framework. We use a regression model to characterize the efficacy pattern in subgroups. Information borrowing is applied through Bayesian hierarchical modelling to improve trial efficiency considering the limited sample size in subgroups. Time trend calibration is also employed to avoid potential baseline drifts. Simulation results demonstrate that MIDAS-2 yields high probabilities for identifying the effective drug combinations as well as promising subgroups, facilitating appropriate selection of the best treatments for each subgroup. The proposed design is robust against small time trend drifts, and the type I error is successfully controlled after calibration when a large drift is expected. Overall, MIDAS-2 provides an adaptive drug screening and subgroup exploring framework to accelerate immunotherapy development in an efficient, accurate, and integrated fashion.

SAM: Self-adapting mixture prior to dynamically borrow information from historical data in clinical trials

Mixture priors provide an intuitive way to incorporate historical data while accounting for potential prior-data conflict by combining an informative prior with a noninformative prior. However, prespecifying the mixing weight for each component remains a crucial challenge. Ideally, the mixing weight should reflect the degree of prior-data conflict, which is often unknown beforehand, posing a significant obstacle to the application and acceptance of mixture priors. To address this challenge, we introduce self-adapting mixture (SAM) priors that determine the mixing weight using likelihood ratio test statistics or Bayes factors. SAM priors are data-driven and self-adapting, favoring the informative (noninformative) prior component when there is little (substantial) evidence of prior-data conflict. Consequently, SAM priors achieve dynamic information borrowing. We demonstrate that SAM priors exhibit desirable properties in both finite and large samples and achieve information-borrowing consistency. Moreover, SAM priors are easy to compute, data-driven, and calibration-free, mitigating the risk of data dredging. Numerical studies show that SAM priors outperform existing methods in adopting prior-data conflicts effectively. We developed R package "SAMprior" and web application that are freely available at CRAN and www.trialdesign.org to facilitate the use of SAM priors.

Frequentist analysis of basket trials with one-sample Mantel-Haenszel procedures

Recent substantial advances of molecular targeted oncology drug development is requiring new paradigms for early-phase clinical trial methodologies to enable us to evaluate efficacy of several subtypes simultaneously and efficiently. The concept of the basket trial is getting of much attention to realize this requirement borrowing information across subtypes, which are called baskets. Bayesian approach is a natural approach to this end and indeed the majority of the existing proposals relies on it. On the other hand, it required complicated modeling and may not necessarily control the type 1 error probabilities at the nominal level. In this article, we develop a purely frequentist approach for basket trials based on one-sample Mantel-Haenszel procedure relying on a very simple idea for borrowing information under the common treatment effect assumption over baskets. We show that the proposed Mantel-Haenszel estimator for the treatment effect is consistent under two limiting models of the large strata and sparse data limiting models (dually consistent) and propose dually consistent variance estimators. The proposed estimators are interpretable even if the common treatment effect assumptions are violated. Then, we can design basket trials in a confirmatory matter. We also propose an information criterion approach to identify effective subclasses of baskets.

Bayesian approach for design and analysis of medical device trials in the era of modern clinical studies

Medical device technology develops rapidly, and the life cycle of a medical device is much shorter than drugs. It is necessary to evaluate the safety and effectiveness of a medical device in a timely manner to keep up with technology flux. Bayesian methods provides an efficient approach to addressing this challenge. In this article, we review the characteristics of the Bayesian approach and some Bayesian designs that were commonly used in medical device regulatory setting, including Bayesian adaptive design, Bayesian diagnostic design, Bayesian multiregional design, and Bayesian label expansion study. We illustrate these designs with medical devices approved by the US Food and Drug Administration (FDA). We also review several innovative Bayesian information borrowing methods, and briefly discuss the challenges and future directions of the Bayesian application in medical device trials. Our objective is to promote the use of the Bayesian approach to accelerate the development of innovative medical devices and their accessibility to patients for effective disease diagnoses and treatments.

Dynamic use of historical controls in clinical trials for rare disease research: A re-evaluation of the MILES trial

BACKGROUND

Randomized controlled trials offer the best design for eliminating bias in estimating treatment effects but can be slow and costly in rare disease research. Additionally, an equal randomization approach may not be optimal in studies in which prior evidence of superiority of one or more treatments exist. Supplementing prospectively enrolled, concurrent controls with historical controls can reduce recruitment requirements and provide patients a higher likelihood of enrolling in a new and possibly superior treatment arm. Appropriate methods need to be employed to ensure comparability of concurrent and historical controls to minimize bias and variability in the treatment effect estimates and reduce the chances of drawing incorrect conclusions regarding treatment benefit.

METHODS

MILES was a phase III placebo-controlled trial employing 1:1 randomization that led to US Food and Drug Administration approval of sirolimus for treating patients with lymphangioleiomyomatosis. We re-analyzed the MILES trial data to learn whether substituting concurrent controls with controls from a historical registry could have accelerated subject enrollment while leading to similar study conclusions. We used propensity score matching to identify exchangeable historical controls from a registry balancing the baseline characteristics of the two control groups. This allowed more new patients to be assigned to the sirolimus arm. We used trial data and simulations to estimate key outcomes under an array of alternative designs.

RESULTS

Borrowing information from historical controls would have allowed the trial to enroll fewer concurrent controls while leading to the same conclusion reached in the trial. Simulations showed similar statistical performance for borrowing as for the actual trial design without producing type I error inflation and preserving power for the same study size when concurrent and historical controls are comparable.

CONCLUSION

Substituting concurrent controls with propensity score-matched historical controls can allow more prospectively enrolled patients to be assigned to the active treatment and enable the trial to be conducted with smaller overall sample size, while maintaining covariate balance and study power and minimizing bias in response estimation. This approach does not fully eliminate the concern that introducing non-randomized historical controls in a trial may lead to bias in estimating treatment effects, and should be carefully considered on a case-by-case basis. Borrowing historical controls is best suited when conducting randomized controlled trials with conventional designs is challenging, as in rare disease research. High-quality data on covariates and outcomes must be available for candidate historical controls to ensure the validity of these designs. Additional precautions are needed to maintain blinding of the treatment assignment and to ensure comparability in the assessment of treatment safety.MILES ClinicalTrials.gov Number: NCT00414648.

Recent innovations in adaptive trial designs: A review of design opportunities in translational research

Clinical trials are constantly evolving in the context of increasingly complex research questions and potentially limited resources. In this review article, we discuss the emergence of "adaptive" clinical trials that allow for the preplanned modification of an ongoing clinical trial based on the accumulating evidence with application across translational research. These modifications may include terminating a trial before completion due to futility or efficacy, re-estimating the needed sample size to ensure adequate power, enriching the target population enrolled in the study, selecting across multiple treatment arms, revising allocation ratios used for randomization, or selecting the most appropriate endpoint. Emerging topics related to borrowing information from historic or supplemental data sources, sequential multiple assignment randomized trials (SMART), master protocol and seamless designs, and phase I dose-finding studies are also presented. Each design element includes a brief overview with an accompanying case study to illustrate the design method in practice. We close with brief discussions relating to the statistical considerations for these contemporary designs.

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.

Data Integration in Causal Inference

Integrating data from multiple heterogeneous sources has become increasingly popular to achieve a large sample size and diverse study population. This paper reviews development in causal inference methods that combines multiple datasets collected by potentially different designs from potentially heterogeneous populations. We summarize recent advances on combining randomized clinical trial with external information from observational studies or historical controls, combining samples when no single sample has all relevant variables with application to two-sample Mendelian randomization, distributed data setting under privacy concerns for comparative effectiveness and safety research using real-world data, Bayesian causal inference, and causal discovery methods.

Integrating non-concurrent controls in the analyses of late-entry experimental arms in multi-arm trials with a shared control group in the presence of parameter drift

BACKGROUND

Under a master protocol, open platform trials allow new experimental treatments to enter an existing clinical trial. Whether late-entry experimental treatments should be compared to all available or concurrently randomized controls is not well established. Using all available data can increase power and precision; however, drift in population parameters can yield biased estimates and impact type I error rate.

METHODS

We explored the application of methods developed to incorporate historical controls in two-arm trials to the analysis of a late-entry arm in a simulated open platform trial under varying scenarios of parameter drift. Methods explored include test-then-pool, fixed power prior, dynamic power prior, and multi-source exchangeability model approaches.

RESULTS/CONCLUSIONS

Simulated trial results confirm that in the presence of no drift, naively pooling all controls increases power and produces more precise, unbiased estimates when compared to using concurrent controls only. However, under drift, pooling can result in type I error rate inflation or deflation and biased estimates. In the presence of parameter drift, methods that partially borrow non-concurrent data, either through a static weighting mechanism or through methods that allow the heterogeneity between non-concurrent and concurrent data to determine the degree of borrowing, are superior to naively pooling the data. However, compared to using concurrent controls only, these approaches cannot guarantee type I error control or unbiased estimates. Thus, concurrent controls should be used as comparators in confirmatory studies.

Bayesian and frequentist approaches to sequential monitoring for futility in oncology basket trials: A comparison of Simon's two-stage design and Bayesian predictive probability monitoring with information sharing across baskets

This article discusses and compares statistical designs of basket trial, from both frequentist and Bayesian perspectives. Baskets trials are used in oncology to study interventions that are developed to target a specific feature (often genetic alteration or immune phenotype) that is observed across multiple tissue types and/or tumor histologies. Patient heterogeneity has become pivotal to the development of non-cytotoxic treatment strategies. Treatment targets are often rare and exist among several histologies, making prospective clinical inquiry challenging for individual tumor types. More generally, basket trials are a type of master protocol often used for label expansion. Master protocol is used to refer to designs that accommodates multiple targets, multiple treatments, or both within one overarching protocol. For the purpose of making sequential decisions about treatment futility, Simon's two-stage design is often embedded within master protocols. In basket trials, this frequentist design is often applied to independent evaluations of tumor histologies and/or indications. In the tumor agnostic setting, rarer indications may fail to reach the sample size needed for even the first evaluation for futility. With recent innovations in Bayesian methods, it is possible to evaluate for futility with smaller sample sizes, even for rarer indications. Novel Bayesian methodology for a sequential basket trial design based on predictive probability is introduced. The Bayesian predictive probability designs allow interim analyses with any desired frequency, including continual assessments after each patient observed. The sequential design is compared with and without Bayesian methods for sharing information among a collection of discrete, and potentially non-exchangeable tumor types. Bayesian designs are compared with Simon's two-stage minimax design.

Bayesian local exchangeability design for phase II basket trials

We propose an information borrowing strategy for the design and monitoring of phase II basket trials based on the local multisource exchangeability assumption between baskets (disease types). In our proposed local-MEM framework, information borrowing is only allowed to occur locally, that is, among baskets with similar response rate and the amount of information borrowing is determined by the level of similarity in response rate, whereas baskets not considered similar are not allowed to share information. We construct a two-stage design for phase II basket trials using the proposed strategy. The proposed method is compared to competing Bayesian methods and Simon's two-stage design in a variety of simulation scenarios. We demonstrate the proposed method is able to maintain the family-wise type I error rate at a reasonable level and has desirable basket-wise power compared to Simon's two-stage design. In addition, our method is computationally efficient compared to existing Bayesian methods in that the posterior profiles of interest can be derived explicitly without the need for sampling algorithms. R scripts to implement the proposed method are available at https://github.com/yilinyl/Bayesian-localMEM.

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.

Comparative Study of Bayesian Information Borrowing Methods in Oncology Clinical Trials

With deeper insight into precision medicine, more innovative oncology trial designs have been proposed to contribute to the characteristics of novel antitumor drugs. Bayesian information borrowing is an indispensable part of these designs, which shows great advantages in improving the efficiency of clinical trials. Bayesian methods provide an effective framework when incorporating information. However, the key point lies in how to choose an appropriate method for complex oncology clinical trials.

A group-sequential randomized trial design utilizing supplemental trial data

Definitive clinical trials are resource intensive, often requiring a large number of participants over several years. One approach to improve the efficiency of clinical trials is to incorporate historical information into the primary trial analysis. This approach has tremendous potential in the areas of pediatric or rare disease trials, where achieving reasonable power is difficult. In this article, we introduce a novel Bayesian group-sequential trial design based on Multisource Exchangeability Models, which allows for dynamic borrowing of historical information at the interim analyses. Our approach achieves synergy between group sequential and adaptive borrowing methodology to attain improved power and reduced sample size. We explore the frequentist operating characteristics of our design through simulation and compare our method to a traditional group-sequential design. Our method achieves earlier stopping of the primary study while increasing power under the alternative hypothesis but has a potential for type I error inflation under some null scenarios. We discuss the issues of decision boundary determination, power and sample size calculations, and the issue of information accrual. We present our method for a continuous and binary outcome, as well as in a linear regression setting.

Bayesian basket trial design with false-discovery rate control

BACKGROUND

Recent advances in developing "tumor agnostic" oncology therapies have identified molecular targets that define patient subpopulations in a manner that supersedes conventional criteria for cancer classification. These successes have produced effective targeted therapies that are administered to patients regardless of their tumor histology. Trials have evolved as well with master protocol designs. By blending translational and clinical science, basket trials in particular are well-suited to investigate and develop targeted therapies among multiple cancer histologies. However, basket trials intrinsically involve more complex design decisions, including issues of multiple testing across baskets, and guidance for investigators is needed.

METHODS

The sensitivity of the multisource exchangeability model to prior specification under differing degrees of response heterogeneity is explored through simulation. Then, a multisource exchangeability model design that incorporates control of the false-discovery rate is presented and a simulation study compares the operating characteristics to a design where the family-wise error rate is controlled and to the frequentist approach of treating the baskets as independent. Simulations are based on the original design of a real-world clinical trial, the SUMMIT trial, which investigated Neratinib treatment for a variety of solid tumors. The methods studied here are specific to single-arm phase II trials with binary outcomes.

RESULTS

Values of prior probability of exchangeability in the multisource exchangeability model between 0.1 and 0.3 provide the best trade-offs between gain in precision and bias, especially when per-basket sample size is below 30. Application of these calibration results to a re-analysis of the SUMMIT trial showed that the breast basket exceeded the null response rate with posterior probability of 0.999 while having low posterior probability of exchangeability with all other baskets. Simulations based on the design of the SUMMIT trial revealed that there is meaningful improvement in power even in baskets with small sample size when the false-discovery rate is controlled as opposed to the family-wise error rate. For example, when only the breast basket was active, with a sample size of 25, the power was 0.76 when the false-discovery rate was controlled at 0.05 but only 0.56 when the family-wise error rate was controlled at 0.05, indicating that impractical sample sizes for the phase II setting would be needed to achieve acceptable power while controlling the family-wise error rate in this setting of a trial with 10 baskets.

CONCLUSION

Selection of the prior exchangeability probability based on calibration and incorporation of false-discovery rate control result in multisource exchangeability model designs with high power to detect promising treatments in the context of phase II basket trials.

A flexible Bayesian framework for individualized inference via adaptive borrowing

The explosion in high-resolution data capture technologies in health has increased interest in making inferences about individual-level parameters. While technology may provide substantial data on a single individual, how best to use multisource population data to improve individualized inference remains an open research question. One possible approach, the multisource exchangeability model (MEM), is a Bayesian method for integrating data from supplementary sources into the analysis of a primary source. MEM was originally developed to improve inference for a single study by asymmetrically borrowing information from a set of similar previous studies and was further developed to apply a more computationally intensive symmetric borrowing in the context of basket trial; however, even for asymmetric borrowing, its computational burden grows exponentially with the number of supplementary sources, making it unsuitable for applications where hundreds or thousands of supplementary sources (i.e., individuals) could contribute to inference on a given individual. In this article, we propose the data-driven MEM (dMEM), a two-stage approach that includes both source selection and clustering to enable the inclusion of an arbitrary number of sources to contribute to individualized inference in a computationally tractable and data-efficient way. We illustrate the application of dMEM to individual-level human behavior and mental well-being data collected via smartphones, where our approach increases individual-level estimation precision by 84% compared with a standard no-borrowing method and outperforms recently proposed competing methods in 80% of individuals.

Incorporating external data into the analysis of clinical trials via Bayesian additive regression trees

Most clinical trials involve the comparison of a new treatment to a control arm (eg, the standard of care) and the estimation of a treatment effect. External data, including historical clinical trial data and real-world observational data, are commonly available for the control arm. With proper statistical adjustments, borrowing information from external data can potentially reduce the mean squared errors of treatment effect estimates and increase the power of detecting a meaningful treatment effect. In this article, we propose to use Bayesian additive regression trees (BART) for incorporating external data into the analysis of clinical trials, with a specific goal of estimating the conditional or population average treatment effect. BART naturally adjusts for patient-level covariates and captures potentially heterogeneous treatment effects across different data sources, achieving flexible borrowing. Simulation studies demonstrate that BART maintains desirable and robust performance across a variety of scenarios and compares favorably to alternatives. We illustrate the proposed method with an acupuncture trial and a colorectal cancer trial.

BRIDGING RANDOMIZED CONTROLLED TRIALS AND SINGLE-ARM TRIALS USING COMMENSURATE PRIORS IN ARM-BASED NETWORK META-ANALYSIS

Network meta-analysis (NMA) is a powerful tool to compare multiple treatments directly and indirectly by combining and contrasting multiple independent clinical trials. Because many NMAs collect only a few eligible randomized controlled trials (RCTs), there is an urgent need to synthesize different sources of information, e.g., from both RCTs and single-arm trials. However, single-arm trials and RCTs may have different populations and quality, so that assuming they are exchangeable may be inappropriate. This article presents a novel method using a commensurate prior on variance (CPV) to borrow variance (rather than mean) information from single-arm trials in an arm-based (AB) Bayesian NMA. We illustrate the advantages of this CPV method by reanalyzing an NMA of immune checkpoint inhibitors in cancer patients. Comprehensive simulations investigate the impact on statistical inference of including single-arm trials. The simulation results show that the CPV method provides efficient and robust estimation even when the two sources of information are moderately inconsistent.

Project 2025: Proposals for the Continued Success of Drug Development in Acute Myeloid Leukemia

The Food and Drug Administration Oncology Center of Excellence initiated Project 2025 to develop five-year goals in specific areas of oncology drug development. This meeting, in October 2020, brought together a panel of regulators and academic experts in acute myeloid leukemia (AML) to discuss opportunities to maximize the success that has recently occurred in AML drug development. The panel discussed challenges and opportunities in clinical trial design and novel endpoints, and outlined key considerations for drug development to facilitate continued growth in the field.

Calibrated dynamic borrowing using capping priors

Multisource exchangeability models (MEMs), a BayeTsian approach for dynamically integrating information from multiple clinical trials, are a promising approach for gaining efficiency in randomized controlled trials. When the supplementary trials are considerably larger than the primary trial, care must be taken when integrating supplementary data to avoid overwhelming the primary trial. In this paper, we propose "capping priors," which controls the extent of dynamic borrowing by placing an a priori cap on the effective supplemental sample size. We demonstrate the behavior of this technique via simulation, and apply our method to four randomized trials of very low nicotine content cigarettes.

Dynamic borrowing in the presence of treatment effect heterogeneity

A number of statistical approaches have been proposed for incorporating supplemental information in randomized clinical trials. Existing methods often compare the marginal treatment effects to evaluate the degree of consistency between sources. Dissimilar marginal treatment effects would either lead to increased bias or down-weighting of the supplemental data. This represents a limitation in the presence of treatment effect heterogeneity, in which case the marginal treatment effect may differ between the sources solely due to differences between the study populations. We introduce the concept of covariate-adjusted exchangeability, in which differences in the marginal treatment effect can be explained by differences in the distributions of the effect modifiers. The potential outcomes framework is used to conceptualize covariate-adjusted and marginal exchangeability. We utilize a linear model and the existing multisource exchangeability models framework to facilitate borrowing when marginal treatment effects are dissimilar but covariate-adjusted exchangeability holds. We investigate the operating characteristics of our method using simulations. We also illustrate our method using data from two clinical trials of very low nicotine content cigarettes. Our method has the ability to incorporate supplemental information in a wider variety of situations than when only marginal exchangeability is considered.

Robust Adaptive Incorporation of Historical Control Data in a Randomized Trial of External Cooling to Treat Septic Shock

This paper proposes randomized controlled clinical trial design to evaluate external cooling as a means to control fever and thereby reduce mortality in patients with septic shock. The trial will include concurrent external cooling and control arms while adaptively incorporating historical control arm data. Bayesian group sequential monitoring will be done using a posterior comparative test based on the 60-day survival distribution in each concurrent arm. Posterior inference will follow from a Bayesian discrete time survival model that facilitates adaptive incorporation of the historical control data through an innovative regression framework with a multivariate spike-and-slab prior distribution on the historical bias parameters. For each interim test, the amount of information borrowed from the historical control data will be determined adaptively in a manner that reflects the degree of agreement between historical and concurrent control arm data. Guidance is provided for selecting Bayesian posterior probability group-sequential monitoring boundaries. Simulation results elucidating how the proposed method borrows strength from the historical control data are reported. In the absence of historical control arm bias, the proposed design controls the type I error rate and provides substantially larger power than reasonable comparators, whereas in the presence bias of varying magnitude, type I error rate inflation is curbed.

Borrowing from supplemental sources to estimate causal effects from a primary data source

The increasing multiplicity of data sources offers exciting possibilities in estimating the effects of a treatment, intervention, or exposure, particularly if observational and experimental sources could be used simultaneously. Borrowing between sources can potentially result in more efficient estimators, but it must be done in a principled manner to mitigate increased bias and Type I error. Furthermore, when the effect of treatment is confounded, as in observational sources or in clinical trials with noncompliance, causal effect estimators are needed to simultaneously adjust for confounding and to estimate effects across data sources. We consider the problem of estimating causal effects from a primary source and borrowing from any number of supplemental sources. We propose using regression-based estimators that borrow based on assuming exchangeability of the regression coefficients and parameters between data sources. Borrowing is accomplished with multisource exchangeability models and Bayesian model averaging. We show via simulation that a Bayesian linear model and Bayesian additive regression trees both have desirable properties and borrow under appropriate circumstances. We apply the estimators to recently completed trials of very low nicotine content cigarettes investigating their impact on smoking behavior.

Optimal Bayesian hierarchical model to accelerate the development of tissue-agnostic drugs and basket trials

Tissue-agnostic trials and basket trials enroll patients based on their genetic biomarkers, not tumor type, in an attempt to determine if a new drug can successfully treat disease conditions based on biomarkers. The Bayesian hierarchical model (BHM) provides an attractive approach to design phase II tissue-agnostic trials by allowing information borrowing across multiple disease types. In this article, we elucidate two intrinsic and inevitable issues that may limit the use of BHM to tissue-agnostic trials: sensitivity to the prior specification of the shrinkage parameter and the competing "interest" among disease types in increasing power and controlling type I error. To address these issues, we propose the optimal BHM (OBHM) and clustered OBHM (COBHM) approaches. In these approach, we first specify a flexible utility function to quantify the tradeoff between type I error and power across disease types based on the study objectives, and then we select the prior of the shrinkage parameter to optimize the utility function of clinical and regulatory interest. COBHM further utilizes a simple Bayesian rule to cluster tumor types into sensitive and insensitive subgroups to achieve more accurate information borrowing. Simulation study shows that the OBHM and especially COBHM have desirable operating characteristics, outperforming some existing methods. COBHM effectively balances power and type I error, addresses the sensitivity of the prior selection, and reduces the "unwarranted" subjectivity in the prior selection. It provides a systematic, rigorous way to apply BHM and solve the common problem of blindingly using a non-informative inverse-gamma prior (with a large variance) or priors arbitrarily chosen that may lead to problematic statistical properties.

Moving Beyond 3+3: The Future of Clinical Trial Design

Misgivings have been raised about the operating characteristics of the canonical 3+3 dose-escalation phase I clinical trial design. Yet, the traditional 3+3 design is still the most commonly used. Although it has been implied that adhering to this design is due to a stubborn reluctance to adopt change despite other designs performing better in hypothetical computer-generated simulation models, the continued adherence to 3+3 dose-escalation phase I strategies is more likely because these designs perform the best in the real world, pinpointing the correct dose and important side effects with an acceptable degree of precision. Beyond statistical simulations, there are little data to refute the supposed shortcomings ascribed to the 3+3 method. Even so, to address the unique nuances of gene- and immune-targeted compounds, a variety of inventive phase 1 trial designs have been suggested. Strategies for developing these therapies have launched first-in-human studies devised to acquire a breadth of patient data that far exceed the size of a typical phase I design and blur the distinction between dose selection and efficacy evaluation. Recent phase I trials of promising cancer therapies assessed objective tumor response and durability at various doses and schedules as well as incorporated multiple expansion cohorts spanning a variety of histology or biomarker-defined tumor subtypes, sometimes resulting in U.S. Food and Drug Administration approval after phase I. This article reviews recent innovations in phase I design from the perspective of multiple stakeholders and provides recommendations for future trials.

Categories, components, and techniques in a modular construction of basket trials for application and further research

Basket trials have become a virulent topic in medical and statistical research during the last decade. The core idea of them is to treat patients, who express the same genetic predisposition-either personally or their disease-with the same treatment irrespective of the location of the disease. The location of the disease defines each basket and the pathway of the treatment uses the common genetic predisposition among the baskets. This opens the opportunity to share information among baskets, which can consequently increase the information of the basket-wise response with respect to the investigated treatment. This further allows dynamic decisions regarding futility and efficacy of individual baskets during the ongoing trial. Several statistical designs have been proposed on how a basket trial can be conducted and this has left an unclear situation with many options. The different designs propose different mathematical and statistical techniques, different decision rules, and also different trial purposes. This paper presents a broad overview of existing designs, categorizes them, and elaborates their similarities and differences. A uniform and consistent notation facilitates the first contact, introduction, and understanding of the statistical methodologies and techniques used in basket trials. Finally, this paper presents a modular approach for the construction of basket trials in applied medical science and forms a base for further research of basket trial designs and their techniques.

Hierarchical Bayesian clustering design of multiple biomarker subgroups (HCOMBS)

Given the Food and Drug Administration's (FDA's) acceptance of master protocol designs in recent guidance documents, the oncology field is rapidly moving to address the paradigm shift to molecular subtype focused studies. Identifying new "marker-based" treatments requires new methodologies to address the growing demand to conduct clinical trials in smaller molecular subpopulations, identify effective treatment and marker interactions, and control for false positives. We introduce our methodology, Hierarchical Bayesian Clustering Design of Multiple Biomarker Subgroups (HCOMBS), a two-stage umbrella Phase II design with effect size clustering and information borrowing across multiple biomarker-treatment pairs. HCOMBS was designed to reduce required sample size, differentiate between varying effect sizes, and control for operating characteristics in the multi-arm setting. When compared to independently applied Simon's Optimal two-stage design, we showed through simulations that HCOMBS required less participants per treatment arm with a well-controlled family-wise error rate and desirable marginal power. Additionally, HCOMBS features a statistical approach that simultaneously conducts clustering and hypothesis testing in one step. We also applied the proposed design on the alliance brain metastases umbrella trial.

Statistical design considerations for trials that study multiple indications

Breakthroughs in cancer biology have defined new research programs emphasizing the development of therapies that target specific pathways in tumor cells. Innovations in clinical trial design have followed with master protocols defined by inclusive eligibility criteria and evaluations of multiple therapies and/or histologies. Consequently, characterization of subpopulation heterogeneity has become central to the formulation and selection of a study design. However, this transition to master protocols has led to challenges in identifying the optimal trial design and proper calibration of hyperparameters. We often evaluate a range of null and alternative scenarios; however, there has been little guidance on how to synthesize the potentially disparate recommendations for what may be optimal. This may lead to the selection of suboptimal designs and statistical methods that do not fully accommodate the subpopulation heterogeneity. This article proposes novel optimization criteria for calibrating and evaluating candidate statistical designs of master protocols in the presence of the potential for treatment effect heterogeneity among enrolled patient subpopulations. The framework is applied to demonstrate the statistical properties of conventional study designs when treatments offer heterogeneous benefit as well as identify optimal designs devised to monitor the potential for heterogeneity among patients with differing clinical indications using Bayesian modeling.

Histology-agnostic drug development - considering issues beyond the tissue

With advances in tumour biology and immunology that continue to refine our understanding of cancer, therapies are now being developed to treat cancers on the basis of specific molecular alterations and markers of immune phenotypes that transcend specific tumour histologies. With the landmark approvals of pembrolizumab for the treatment of patients whose tumours have high microsatellite instability and larotrectinib and entrectinib for those harbouring NTRK fusions, a regulatory pathway has been created to facilitate the approval of histology-agnostic indications. Negative results presented in the past few years, however, highlight the intrinsic complexities faced by drug developers pursuing histology-agnostic therapeutic agents. When patient selection and statistical analysis involve multiple potentially heterogeneous histologies, guidance is needed to navigate the challenges posed by trial design. Additionally, as new therapeutic agents are tested and post-approval data become available, the regulatory framework for acting on these data requires further optimization. In this Review, we summarize the development and testing of approved histology-agnostic therapeutic agents and present data on other agents currently under development. Finally, we discuss the challenges intrinsic to histology-agnostic drug development in oncology, including biological, regulatory, design and statistical considerations.

Iterated multisource exchangeability models for individualized inference with an application to mobile sensor data

Researchers are increasingly interested in using sensor technology to collect accurate activity information and make individualized inference about treatments, exposures, and policies. How to optimally combine population data with data from an individual remains an open question. Multisource exchangeability models (MEMs) are a Bayesian approach for increasing precision by combining potentially heterogeneous supplemental data sources into analysis of a primary source. MEMs are a potentially powerful tool for individualized inference but can integrate only a few sources; their model space grows exponentially, making them intractable for high-dimensional applications. We propose iterated MEMs (iMEMs), which identify a subset of the most exchangeable sources prior to fitting a MEM model. iMEM complexity scales linearly with the number of sources, and iMEMs greatly increase precision while maintaining desirable asymptotic and small sample properties. We apply iMEMs to individual-level behavior and emotion data from a smartphone app and show that they achieve individualized inference with up to 99% efficiency gain relative to standard analyses that do not borrow information.

Bayesian hierarchical meta-analytic methods for modeling surrogate relationships that vary across treatment classes using aggregate data

Surrogate endpoints play an important role in drug development when they can be used to measure treatment effect early compared to the final clinical outcome and to predict clinical benefit or harm. Such endpoints are assessed for their predictive value of clinical benefit by investigating the surrogate relationship between treatment effects on the surrogate and final outcomes using meta‐analytic methods. When surrogate relationships vary across treatment classes, such validation may fail due to limited data within each treatment class. In this paper, two alternative Bayesian meta‐analytic methods are introduced which allow for borrowing of information from other treatment classes when exploring the surrogacy in a particular class. The first approach extends a standard model for the evaluation of surrogate endpoints to a hierarchical meta‐analysis model assuming full exchangeability of surrogate relationships across all the treatment classes, thus facilitating borrowing of information across the classes. The second method is able to relax this assumption by allowing for partial exchangeability of surrogate relationships across treatment classes to avoid excessive borrowing of information from distinctly different classes. We carried out a simulation study to assess the proposed methods in nine data scenarios and compared them with subgroup analysis using the standard model within each treatment class. We also applied the methods to an illustrative example in colorectal cancer which led to obtaining the parameters describing the surrogate relationships with higher precision.

Bayesian adaptive design for concurrent trials involving biologically related diseases

We develop a Bayesian design method for a clinical program where an investigational product is to be studied concurrently in a set of clinical trials involving related diseases with the goal of demonstrating superiority to a control in each. The approach borrows information on treatment effectiveness using correlated mixture priors using an analysis procedure that is closely related Bayesian model averaging. Mixture priors are constructed by eliciting conjugate priors based on pessimistic and enthusiastic predictions for the data to be observed for each disease and then by eliciting mixture weights for all possible configurations of the pessimistic and enthusiastic priors across the diseases to be studied. The proposed approach provides a robust framework for information borrowing in settings where the diseases may have endpoints based on different data types. We show via simulation that operating characteristics based on the proposed design framework are favorable compared to those based on information borrowing designs using the Bayesian hierarchical model which is poorly suited for information borrowing when there are different data types underpinning the endpoints across which information is to be borrowed.

Basket Designs: Statistical Considerations for Oncology Trials

Progress in the areas of genomics, disease pathways, and drug discovery has advanced into clinical and translational cancer research. The latest innovations in clinical trials have followed with master protocols, which are defined by inclusive eligibility criteria and devised to interrogate multiple therapies for a given tumor histology and/or multiple histologies for a given therapy under one protocol. The use of master protocols for oncology has become more common with the desire to improve the efficiency of clinical research and accelerate overall drug development. Basket trials have been devised to ascertain the extent to which a treatment strategy offers benefit to various patient subpopulations defined by a common molecular target. Conventionally conducted within the phase II setting, basket designs have become popular as drug developers seek to effectively evaluate and identify preliminary efficacy signals among clinical indications identified as promising in preclinical study. This article reviews basket trial designs in oncology settings and discusses several issues that arise with their design and analysis.

Design and Evaluation of an External Control Arm Using Prior Clinical Trials and Real-World Data

PURPOSE

We discuss designs and interpretable metrics of bias and statistical efficiency of "externally controlled" trials (ECT) and compare ECT performance to randomized and single-arm designs.

EXPERIMENTAL DESIGN

We specify an ECT design that leverages information from real-world data (RWD) and prior clinical trials to reduce bias associated with interstudy variations of the enrolled populations. We then used a collection of clinical studies in glioblastoma (GBM) and RWD from patients treated with the current standard of care to evaluate ECTs. Validation is based on a "leave one out" scheme, with iterative selection of a single-arm from one of the studies, for which we estimate treatment effects using the remaining studies as external control. This produces interpretable and robust estimates on ECT bias and type I errors.

RESULTS

We developed a model-free approach to evaluate ECTs based on collections of clinical trials and RWD. For GBM, we verified that inflated false positive error rates of standard single-arm trials can be considerably reduced (up to 30%) by using external control data.

CONCLUSIONS

The use of ECT designs in GBM, with adjustments for the clinical profiles of the enrolled patients, should be preferred to single-arm studies with fixed efficacy thresholds extracted from published results on the current standard of care.

Seamless Designs: Current Practice and Considerations for Early-Phase Drug Development in Oncology

Traditionally, drug development has evaluated dose, safety, activity, and comparative benefit in a sequence of phases using trial designs and endpoints specifically devised for each phase. Innovations in drug development seek to consolidate the phases and rapidly expand accrual with "seamless" trial designs. Although consolidation and rapid accrual may yield efficiencies, widespread use of seamless first-in-human (FiH) trials without careful consideration of objectives, statistical analysis plans, or trial oversight raises concerns. A working group formed by the National Cancer Institute convened to consider and discuss opportunities and challenges for such trials as well as encourage responsible use of these designs. We reviewed all abstracts presented at American Society of Clinical Oncology annual meetings from 2010 to 2017 for FiH trials enrolling at least 100 patients. We identified 1786 early-phase trials enrolling 57 559 adult patients. Fifty-one of the trials (2.9%) investigated 50 investigational new drugs, were seamless, and accounted for 14.6% of the total patients. The seamless trials included a median of 3 (range = 1-13) expansion cohorts. The overall risk of clinically significant treatment-related adverse events (grade 3-4) was 49.1% (range = 0.0-100%), and seven studies reported at least one toxic death. Rapid expansion of FiH trials may lead to earlier drug approval and corresponding widespread patient access to active therapeutics. Nevertheless, seamless designs must adhere to established ethical, scientific, and statistical standards. Protocols should include prospectively planned analyses of efficacy in disease- or biomarker-defined cohorts of sufficient rigor to support accelerated approval.