Time-to-event model-assisted designs for dose-finding trials with delayed toxicity

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.

A Bayesian Dynamic Model-Based Adaptive Design for Oncology Dose Optimization in Phase I/II Clinical Trials

With the development of targeted therapy, immunotherapy, and antibody-drug conjugates (ADCs), there is growing concern over the "more is better" paradigm developed decades ago for chemotherapy, prompting the US Food and Drug Administration (FDA) to initiate Project Optimus to reform dose optimization and selection in oncology drug development. For early-phase oncology trials, given the high variability from sparse data and the rigidity of parametric model specifications, we use Bayesian dynamic models to borrow information across doses with only vague order constraints. Our proposed adaptive design simultaneously incorporates toxicity and efficacy outcomes to select the optimal dose (OD) in Phase I/II clinical trials, utilizing Bayesian model averaging to address the uncertainty of dose-response relationships and enhance the robustness of the design. Additionally, we extend the proposed design to handle delayed toxicity and efficacy outcomes. We conduct extensive simulation studies to evaluate the operating characteristics of the proposed method under various practical scenarios. The results demonstrate that the proposed designs have desirable operating characteristics. A trial example is presented to demonstrate the practical implementation of the proposed designs.

Bayesian Methods for Quality Tolerance Limit (QTL) Monitoring

In alignment with the ICH guideline for Good Clinical Practice [ICH E6(R2)], quality tolerance limit (QTL) monitoring has become a standard component of risk-based monitoring of clinical trials by sponsor companies. Parameters that are candidates for QTL monitoring are critical to participant safety and quality of trial results. Breaching the QTL of a given parameter could indicate systematic issues with the trial that could impact participant safety or compromise the reliability of trial results. Methods for QTL monitoring should detect potential QTL breaches as early as possible while limiting the rate of false alarms. Early detection allows for the implementation of remedial actions that can prevent a QTL breach at the end of the trial. We demonstrate that statistically based methods that account for the expected value and variability of the data generating process outperform simple methods based on fixed thresholds with respect to important operating characteristics. We also propose a Bayesian method for QTL monitoring and an extension that allows for the incorporation of partial information, demonstrating its potential to outperform frequentist methods originating from the statistical process control literature.

T3 + 3: 3 + 3 Design With Delayed Outcomes

Delayed outcome is common in phase I oncology clinical trials. It causes logistic difficulty, wastes resources, and prolongs the trial duration. This article investigates this issue and proposes the time-to-event 3 + 3 (T3 + 3) design, which utilizes the actual follow-up time for at-risk patients with pending toxicity outcomes. The T3 + 3 design allows continuous accrual without unnecessary trial suspension and is costless and implementable with pretabulated dose decision rules. Besides, the T3 + 3 design uses the isotonic regression to estimate the toxicity rates across dose levels and therefore can accommodate for any targeted toxicity rate for maximum tolerated dose (MTD). It dramatically facilitates the trial preparation and conduct without intensive computation and statistical consultation. The extension to other algorithm-based phase I dose-finding designs (e.g., i3 + 3 design) is also studied. Comprehensive computer simulation studies are conducted to investigate the performance of the T3 + 3 design under various dose-toxicity scenarios. The results confirm that the T3 + 3 design substantially shortens the trial duration compared with the conventional 3 + 3 design and yields much higher accuracy in MTD identification than the rolling six design. In summary, the T3 + 3 design addresses the delayed outcome issue while keeping the desirable features of the 3 + 3 design, such as simplicity, transparency, and costless implementation. It has great potential to accelerate early-phase drug development.

PKBOIN-12: A Bayesian Optimal Interval Phase I/II Design Incorporating Pharmacokinetics Outcomes to Find the Optimal Biological Dose

Immunotherapies and targeted therapies have gained popularity due to their promising therapeutic effects across multiple treatment areas. The focus of early phase dose-finding clinical trials has shifted from finding the maximum tolerated dose (MTD) to identifying the optimal biological dose (OBD), which aims to balance the toxicity and efficacy outcomes, thus optimizing the risk-benefit trade-off. These trials often collect multiple pharmacokinetics (PK) outcomes to assess drug exposure, which has shown correlations with toxicity and efficacy outcomes but has not been utilized in the current dose-finding designs for OBD selection. Moreover, PK outcomes are usually available within days after initial treatment, much faster than toxicity and efficacy outcomes. To bridge this gap, we introduce the innovative model-assisted PKBOIN-12 design, which enhances BOIN12 by integrating PK information into both the dose-finding algorithm and the final OBD determination process. We further extend PKBOIN-12 to TITE-PKBOIN-12 to address the challenges of late-onset toxicity and efficacy outcomes. Simulation results demonstrate that PKBOIN-12 more effectively identifies the OBD and allocates a greater number of patients to it than BOIN12. Additionally, PKBOIN-12 decreases the probability of selecting inefficacious doses as the OBD by excluding those with low drug exposure. Comprehensive simulation studies and sensitivity analysis confirm the robustness of both PKBOIN-12 and TITE-PKBOIN-12 in various scenarios.

Phase I clinical trial designs in oncology: A systematic literature review from 2020 to 2022

Background

Phase I clinical trials aim to find the highest dose of a novel drug that may be administrated safely without having serious adverse effects. Model-based designs have recently become popular in dose-finding procedures. Our objective is to provide an overview of phase I clinical trials in oncology.

Methods

A retrospective analysis of phase I clinical trials in oncology was performed by using the PubMed database between January 1, 2020, and December 31, 2022. We extracted all papers with the inclusion of trials in oncology and kept only those in which dose escalation or/ and dose expansion were conducted. We also compared the study parameters, design parameters, and patient parameters between industry-sponsored studies and academia-sponsored research.

Result

Among the 1450 papers retrieved, 256 trials described phase I clinical trials in oncology. Overall, 71.1% of trials were done with a single study cohort, 56.64% of trials collected a group of at least 20 study volunteers, 55.1% were sponsored by industry, and 99.2% of trials had less than 10 patients who experienced DLTs.The traditional 3 + 3 (73.85%) was still the most prevailing method for the dose-escalation approach. More than 50% of the trials did not reach MTDs. Industry-sponsored study enrolled more patients in dose-escalation trials with benefits of continental cooperation. Compared to previous findings, the usage of model-based design increased to about 10%, and the percentage of traditional 3 + 3 design decreased to 74%.

Conclusions

Phase I traditional 3 + 3 designs perform well, but there is still room for development in novel model-based dose-escalation designs in clinical practice.

MC-Keyboard: A Practical Phase I Trial Design for Targeted Therapies and Immunotherapies Integrating Multiple-Grade Toxicities

In targeted therapies and immunotherapies, the occurrence of low-grade (e.g., grade 1-2) toxicities (LGT) is common, while dose-limiting toxicities (DLT) are relatively rare. As a result, conventional phase I trial designs, solely based on DLTs and disregarding milder toxicities, are problematic when evaluating these novel therapies. Methods: To address this issue, we propose a novel phase I design called a multiple-constraint keyboard (MC-Keyboard) that integrates multiple toxicity constraints, accounting for both DLT and LGT, for precise dose escalation and de-escalation, and identification of the maximum tolerated dose (MTD). As a model-assisted design, an important feature of MC-Keyboard is that its dose-escalation or de-escalation rule can be pretabulated and incorporated into the trial protocol before the initiation of the trial, greatly simplifying its implementation. Results: The simulation study showed that the MC-Keyboard had high accuracy in identifying the MTD and is safer than some existing designs. Conclusion: The MC-Keyboard provides a novel, simple, and safe approach to assessing safety and identifying the MTD for targeted therapies and immunotherapies.

Dose optimization for cancer treatments with considerations for late-onset toxicities

Given that novel anticancer therapies have different toxicity profiles and mechanisms of action, it is important to reconsider the current approaches for dose selection. In an effort to move away from considering the maximum tolerated dose as the optimal dose, the Food and Drug Administration Project Optimus points to the need of incorporating long-term toxicity evaluation, given that many of these novel agents lead to late-onset or cumulative toxicities and there are no guidelines on how to handle them. Numerous methods have been proposed to handle late-onset toxicities in dose-finding clinical trials. A summary and comparison of these methods are provided. Moreover, using PI3K inhibitors as a case study, we show how late-onset toxicity can be integrated into the dose-optimization strategy using current available approaches. We illustrate a re-design of this trial to compare the approach to those that only consider early toxicity outcomes and disregard late-onset toxicities. We also provide proposals going forward for dose optimization in early development of novel anticancer agents with considerations for late-onset toxicities.

A Bayesian quasi-likelihood design for identifying the minimum effective dose and maximum utility dose in dose-ranging studies

Most existing dose-ranging study designs focus on assessing the dose-efficacy relationship and identifying the minimum effective dose. There is an increasing interest in optimizing the dose based on the benefit-risk tradeoff. We propose a Bayesian quasi-likelihood dose-ranging design that jointly considers safety and efficacy to simultaneously identify the minimum effective dose and the maximum utility dose to optimize the benefit-risk tradeoff. The binary toxicity endpoint is modeled using a beta-binomial model. The efficacy endpoint is modeled using the quasi-likelihood approach to accommodate various types of data (e.g. binary, ordinal or continuous) without imposing any parametric assumptions on the dose-response curve. Our design utilizes a utility function as a measure of benefit-risk tradeoff and adaptively assign patients to doses based on the doses' likelihood of being the minimum effective dose and maximum utility dose. The design takes a group-sequential approach. At each interim, the doses that are deemed overly toxic or futile are dropped. At the end of the trial, we use posterior probability criteria to assess the strength of the dose-response relationship for establishing the proof-of-concept. If the proof-of-concept is established, we identify the minimum effective dose and maximum utility dose. Our simulation study shows that compared with some existing designs, the Bayesian quasi-likelihood dose-ranging design is robust and yields competitive performance in establishing proof-of-concept and selecting the minimum effective dose. Moreover, it includes an additional feature for further maximum utility dose selection.

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.

Application of multi-armed bandits to dose-finding clinical designs

Multi-armed bandits are very simple and powerful methods to determine actions to maximize a reward in a limited number of trials. An early phase in dose-finding clinical trials needs to identify the maximum tolerated dose among multiple doses by repeating the dose-assignment. We consider applying the superior selection performance of multi-armed bandits to dose-finding clinical designs. Among the multi-armed bandits, we first consider the use of Thompson sampling which determines actions based on random samples from a posterior distribution. In the small sample size, as shown in dose-finding trials, because the tails of posterior distribution are heavier and random samples are too much variability, we also consider an application of regularized Thompson sampling and greedy algorithm. The greedy algorithm determines a dose based on a posterior mean. In addition, we also propose a method to determine a dose based on a posterior mode. We evaluate the performance of our proposed designs for nine scenarios via simulation studies.

A Bayesian optimal interval design for dose optimization with a randomization scheme based on pharmacokinetics outcomes in oncology

The primary objective of an oncology dose-finding trial for novel therapies, such as molecularly targeted agents and immune-oncology therapies, is to identify the optimal dose (OD) that is tolerable and therapeutically beneficial for subjects in subsequent clinical trials. Pharmacokinetic (PK) information is considered an appropriate indicator for evaluating the level of drug intervention in humans from a pharmacological perspective. Several novel anticancer agents have been shown to have significant exposure-efficacy relationships, and some PK information has been considered an important predictor of efficacy. This paper proposes a Bayesian optimal interval design for dose optimization with a randomization scheme based on PK outcomes in oncology. A simulation study shows that the proposed design has advantages compared to the other designs in the percentage of correct OD selection and the average number of patients allocated to OD in various realistic settings.

TITE-gBOIN-ET: Time-to-event generalized Bayesian optimal interval design to accelerate dose-finding accounting for ordinal graded efficacy and toxicity outcomes

One of the primary objectives of an oncology dose-finding trial for novel therapies, such as molecular-targeted agents and immune-oncology therapies, is to identify an optimal dose (OD) that is tolerable and therapeutically beneficial for subjects in subsequent clinical trials. These new therapeutic agents appear more likely to induce multiple low or moderate-grade toxicities than dose-limiting toxicities. Besides, for efficacy, evaluating the overall response and long-term stable disease in solid tumors and considering the difference between complete remission and partial remission in lymphoma are preferable. It is also essential to accelerate early-stage trials to shorten the entire period of drug development. However, it is often challenging to make real-time adaptive decisions due to late-onset outcomes, fast accrual rates, and differences in outcome evaluation periods for efficacy and toxicity. To solve the issues, we propose a time-to-event generalized Bayesian optimal interval design to accelerate dose finding, accounting for efficacy and toxicity grades. The new design named "TITE-gBOIN-ET" design is model-assisted and straightforward to implement in actual oncology dose-finding trials. Simulation studies show that the TITE-gBOIN-ET design significantly shortens the trial duration compared with the designs without sequential enrollment while having comparable or higher performance in the percentage of correct OD selection and the average number of patients allocated to the ODs across various realistic settings.

Time-to-event calibration-free odds design: A robust efficient design for phase I trials with late-onset outcomes

Compared with most of the existing phase I designs, the recently proposed calibration-free odds (CFO) design has been demonstrated to be robust, model-free, and easy to use in practice. However, the original CFO design cannot handle late-onset toxicities, which have been commonly encountered in phase I oncology dose-finding trials with targeted agents or immunotherapies. To account for late-onset outcomes, we extend the CFO design to its time-to-event (TITE) version, which inherits the calibration-free and model-free properties. One salient feature of CFO-type designs is to adopt game theory by competing three doses at a time, including the current dose and the two neighboring doses, while interval-based designs only use the data at the current dose and is thus less efficient. We conduct comprehensive numerical studies for the TITE-CFO design under both fixed and randomly generated scenarios. TITE-CFO shows robust and efficient performances compared with interval-based and model-based counterparts. As a conclusion, the TITE-CFO design provides robust, efficient, and easy-to-use alternatives for phase I trials when the toxicity outcome is late-onset.

A comparative study of adaptive trial designs for dose optimization

Recently, the US Food and Drug Administration Oncology Center of Excellence initiated Project Optimus to reform the dose optimization and dose selection paradigm in oncology drug development. The agency pointed out that the current paradigm for dose selection-based on the maximum tolerated dose (MTD)-is not sufficient for molecularly targeted therapies and immunotherapies, for which efficacy may not increase after the dose reaches a certain level. In these cases, it is more appropriate to identify the optimal biological dose (OBD) that optimizes the risk-benefit tradeoff of the drug. Project Optimus has spurred tremendous interest and urgent need for guidance on designing dose optimization trials. In this article, we review several representative dose optimization designs, including model-based and model-assisted designs, and compare their operating characteristics based on 10,000 randomly generated scenarios with various dose-toxicity and dose-efficacy curves and some fixed representative scenarios. The results show that, compared with model-based designs, model-assisted methods have advantages of easy-to-implement, robustness, and high accuracy to identify OBD. Some guidance is provided to help biostatisticians and clinicians to choose appropriate dose optimization methods in practice.

Comparison Between Simultaneous and Sequential Utilization of Safety and Efficacy for Optimal Dose Determination in Bayesian Model-Assisted Designs

It has become quite common in recent early oncology trials to include both the dose-finding and the dose-expansion parts within the same study. This shift can be viewed as a seamless way of conducting the trials to obtain information on safety and efficacy hence identifying an optimal dose (OD) rather than just the maximum tolerated dose (MTD). One approach is to conduct a dose-finding part based solely on toxicity outcomes, followed by a dose expansion part to evaluate efficacy outcomes. Another approach employs only the dose-finding part, where the dose-finding decisions are made utilizing both the efficacy and toxicity outcomes of those enrolled patients. In this paper, we compared the two approaches through simulation studies under various realistic settings. The percentage of correct ODs selection, the average number of patients allocated to the ODs, and the average trial duration are reported in choosing the appropriate designs for their early-stage dose-finding trials, including expansion cohorts.

Modified isotonic regression based phase I/II clinical trial design identifying optimal biological dose

Conventional phase I/II clinical trial designs often use complicated parametric models to characterize the dose-response relationships and conduct the trials. However, the parametric models are hard to justify in practice, and the misspecification of parametric models can lead to substantially undesirable performances in phase I/II trials. Moreover, it is difficult for the physicians conducting phase I/II trials to clinically interpret the parameters of these complicated models, and such significant learning costs impede the translation of novel statistical designs into practical trial implementation. To solve these issues, we propose a transparent and efficient phase I/II clinical trial design, referred to as the modified isotonic regression-based design (mISO), to identify the optimal biological doses for molecularly targeted agents and immunotherapy. The mISO design makes no parametric model assumptions on the dose-response relationship and yields desirable performances under any clinically meaningful dose-response curves. The concise, clinically interpretable dose-response models and dose-finding algorithm make the proposed designs highly translational from the statistical community to the clinical community. We further extend the mISO design and develop the mISO-B design to handle the delayed outcomes. Our comprehensive simulation studies show that the mISO and mISO-B designs are highly efficient in optimal biological dose selection and patients allocation and outperform many existing phase I/II clinical trial designs. We also provide a trial example to illustrate the practical implementation of the proposed designs. The software for simulation and trial implementation are available for free download.

Data-dependent early completion of dose-finding trials for drug-combination

PURPOSE

Model-assisted designs for drug combination trials have been proposed as novel designs with simple and superior performance. However, model-assisted designs have the disadvantage that the sample size must be set in advance, and trials cannot be completed until the number of patients treated reaches the pre-set sample size. Model-assisted designs have a stopping rule that can be used to terminate the trial if the number of patients treated exceeds the predetermined number, there is no statistical basis for the predetermined number. Here, I propose two methods for data-dependent early completion of dose-finding trials for drug combination: (1) an early completion method based on dose retainment probability, and (2) an early completion method in which the dose retainment probability is adjusted by a bivariate isotonic regression.

METHODS

Early completion is determined when the dose retainment probability using both trial data and the number of remaining patients is high. Early completion of a virtual trial was demonstrated. The performances of the early completion methods were evaluated by simulation studies with 12 scenarios.

RESULTS

The simulation studies showed that the percentage of early completion was an average of approximately 70%, and the number of patients treated was 25% less than the planned sample size. The percentage of correct maximum tolerated dose combination selection for the early completion methods was similar to that of non-early completion methods with an average difference of approximately 3%.

CONCLUSION

The performance of the proposed early completion methods was similar to that of the non-early completion methods. Furthermore, the number of patients for determining early completion before the trial starts was determined and a program code for calculating the dose retainment probability was proposed.

Isotonic design for phase I cancer clinical trials with late-onset toxicities

This article addresses the problem of identifying the maximum tolerated dose (MTD) in Phase I dose-finding clinical trials with late-onset toxicities. The main design challenge is how best to adaptively allocate study participants to tolerable doses when the evaluation window for the toxicity endpoint is long relative to the accrual rate of new participants. We propose a new design framework based on order-restricted statistical inference that addresses this challenge in sequential dose assignments. We illustrate the proposed method on real data from a Phase I trial of bortezomib in lymphoma patients and apply it to a Phase I trial of radiotherapy in prostate cancer patients. We conduct extensive simulation studies to compare our design's operating characteristics to existing published methods. Overall, our proposed design demonstrates good performance relative to existing methods in allocating participants at and around the MTD during the study and accurately recommending the MTD at the study conclusion.

Dose finding studies for therapies with late-onset toxicities: A comparison study of designs

An objective of phase I dose-finding trials is to find the maximum tolerated dose; the dose with a particular risk of toxicity. Frequently, this risk is assessed across the first cycle of therapy. However, in oncology, a course of treatment frequently consists of multiple cycles of therapy. In many cases, the overall risk of toxicity for a given treatment is not fully encapsulated by observations from the first cycle, and hence it is advantageous to include toxicity outcomes from later cycles in phase I trials. Extending the follow up period in a trial naturally extends the total length of the trial which is undesirable. We present a comparison of eight methods that incorporate late onset toxicities while not extensively extending the trial length. We conduct simulation studies over a number of scenarios and in two settings; the first setting with minimal stopping rules and the second setting with a full set of standard stopping rules expected in such a dose finding study. We find that the model-based approaches in general outperform the model-assisted approaches, with an interval censored approach and a modified version of the time-to-event continual reassessment method giving the most promising overall performance in terms of correct selections and trial length. Further recommendations are made for the implementation of such methods.

Surv-CRM-12: A Bayesian phase I/II survival CRM for right-censored toxicity endpoints with competing disease progression

The growing interest in new classes of anti-cancer agents, such as molecularly-targeted therapies and immunotherapies with modes of action different from those of cytotoxic chemotherapies, has changed the dose-finding paradigm. In this setting, the observation of late-onset toxicity endpoints may be precluded by treatment and trial discontinuation due to disease progression, defining a competing event to toxicity. Trial designs where dose-finding is modeled in the framework of a survival competing risks model appear particularly well-suited. We aim to provide a phase I/II dose-finding design that allows dose-limiting toxicity (DLT) outcomes to be delayed or unobserved due to competing progression within the possibly long observation window. The proposed design named the Survival-continual reassessment method-12, uses survival models for right-censored DLT and progression endpoints. In this competing risks framework, cause-specific hazards for DLT and progression-free of DLT were considered, with model parameters estimated using Bayesian inference. It aims to identify the optimal dose (OD), by minimizing the cumulative incidence of disease progression, given an acceptable toxicity threshold. In a simulation study, design operating characteristics were evaluated and compared to the TITE-BOIN-ET design and a nonparametric benchmark approach. The performance of the proposed method was consistent with the complexity of scenarios as assessed by the nonparametric benchmark. We found that the proposed design presents satisfying operating characteristics in selecting the OD and safety.

A Road Map for Designing Phase I Clinical Trials of Radiotherapy-Novel Agent Combinations

Radiotherapy has proven efficacy in a wide range of cancers. There is growing interest in evaluating radiotherapy-novel agent combinations and a drive to initiate this earlier in the clinical development of the novel agent, where the scientific rationale and preclinical evidence for a radiotherapy combination approach are high. Optimal design, delivery, and interpretation of studies are essential. In particular, the design of phase I studies to determine safety and dosing is critical to an efficient development strategy. There is significant interest in early-phase research among scientific and clinical communities over recent years, at a time when the scrutiny of the trial methodology has significantly increased. To enhance trial design, optimize safety, and promote efficient trial conduct, this position paper reviews the current phase I trial design landscape. Key design characteristics extracted from 37 methodology papers were used to define a road map and a design selection process for phase I radiotherapy-novel agent trials. Design selection is based on single- or dual-therapy dose escalation, dose-limiting toxicity categorization, maximum tolerated dose determination, subgroup evaluation, software availability, and design performance. Fifteen of the 37 designs were identified as being immediately accessible and relevant to radiotherapy-novel agent phase I trials. Applied examples of using the road map are presented. Developing these studies is intensive, highlighting the need for funding and statistical input early in the trial development to ensure appropriate design and implementation from the outset. The application of this road map will improve the design of phase I radiotherapy-novel agent combination trials, enabling a more efficient development pathway.

Borrowing historical information to improve phase I clinical trials using meta-analytic-predictive priors

Multiple phase I clinical trials may be performed to determine specific maximum tolerated doses (MTD) for specific races or cancer types. In these situations, borrowing historical information has potential to improve the accuracy of estimating toxicity rate and increase the probability of correctly targeting MTD. To utilize historical information in phase I clinical trials, we proposed using the Meta-Analytic-Predictive (MAP) priors to automatically estimate the heterogeneity between historical trials and give a relatively reasonable amount of borrowed information. We then applied MAP priors in some famous phase I trial designs, such as the continual reassessment method (CRM), Keyboard design and Bayesian optimal interval design (BOIN), to accomplish the process of dose finding. A clinical trial example and extended simulation studies show that our proposed methods have robust and efficient statistical performance, compared with those designs which do not consider borrowing information.

TITE-BOIN12: A Bayesian phase I/II trial design to find the optimal biological dose with late-onset toxicity and efficacy

In the era of immunotherapies and targeted therapies, the focus of early phase clinical trials has shifted from finding the maximum tolerated dose to identifying the optimal biological dose (OBD), which maximizes the toxicity-efficacy trade-off. One major impediment to using adaptive designs to find OBD is that efficacy or/and toxicity are often late-onset, hampering the designs' real-time decision rules for treating new patients. To address this issue, we propose the model-assisted TITE-BOIN12 design to find OBD with late-onset toxicity and efficacy. As an extension of the BOIN12 design, the TITE-BOIN12 design also uses utility to quantify the toxicity-efficacy trade-off. We consider two approaches, Bayesian data augmentation and an approximated likelihood method, to enable real-time decision making when some patients' toxicity and efficacy outcomes are pending. Extensive simulations show that, compared to some existing designs, TITE-BOIN12 significantly shortens the trial duration while having comparable or higher accuracy to identify OBD and a lower risk of overdosing patients. To facilitate the use of the TITE-BOIN12 design, we develop a user-friendly software freely available at http://www.trialdesign.org.

Adaptive design for identifying maximum tolerated dose early to accelerate dose-finding trial

Purpose

The early identification of maximum tolerated dose (MTD) in phase I trial leads to faster progression to a phase II trial or an expansion cohort to confirm efficacy.

Methods

We propose a novel adaptive design for identifying MTD early to accelerate dose-finding trials. The early identification of MTD is determined adaptively by dose-retainment probability using a trial data via Bayesian analysis. We applied the early identification design to an actual trial. A simulation study evaluates the performance of the early identification design.

Results

In the actual study, we confirmed the MTD could be early identified and the study period was shortened. In the simulation study, the percentage of the correct MTD selection in the early identification Keyboard and early identification Bayesian optimal interval (BOIN) designs was almost same from the non-early identification version. The early identification Keyboard and BOIN designs reduced the study duration by about 50% from the model-assisted designs. In addition, the early identification Keyboard and BOIN designs reduced the study duration by about 20% from time-to-event model-assisted designs.

Conclusion

We proposed the early identification of MTD maintaining the accuracy to be able to short the study period.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12874-022-01584-y.

SCI: A Bayesian adaptive phase I/II dose-finding design accounting for semi-competing risks outcomes for immunotherapy trials

An immunotherapy trial often uses the phase I/II design to identify the optimal biological dose, which monitors the efficacy and toxicity outcomes simultaneously in a single trial. The progression‐free survival rate is often used as the efficacy outcome in phase I/II immunotherapy trials. As a result, patients developing disease progression in phase I/II immunotherapy trials are generally seriously ill and are often treated off the trial for ethical consideration. Consequently, the happening of disease progression will terminate the toxicity event but not vice versa, so the issue of the semi‐competing risks arises. Moreover, this issue can become more intractable with the late‐onset outcomes, which happens when a relatively long follow‐up time is required to ascertain progression‐free survival. This paper proposes a novel Bayesian adaptive phase I/II design accounting for semi‐competing risks outcomes for immunotherapy trials, referred to as the dose‐finding design accounting for semi‐competing risks outcomes for immunotherapy trials (SCI) design. To tackle the issue of the semi‐competing risks in the presence of late‐onset outcomes, we re‐construct the likelihood function based on each patient's actual follow‐up time and develop a data augmentation method to efficiently draw posterior samples from a series of Beta‐binomial distributions. We propose a concise curve‐free dose‐finding algorithm to adaptively identify the optimal biological dose using accumulated data without making any parametric dose–response assumptions. Numerical studies show that the proposed SCI design yields good operating characteristics in dose selection, patient allocation, and trial duration.

Early Completion of Model-Assisted Designs for Dose-Finding Trials

PURPOSE

We propose a novel early completion method for phase I dose-finding trials using model-assisted designs. The trials can halt when a maximum tolerated dose (MTD) is estimated with sufficient accuracy. Early completion can reduce the average number of patients treated relative to the planned number, thereby allowing the trial to proceed to enrolling an expansion cohort for efficacy and enabling the trial to reach the next phase faster.

METHODS

Early completion is conducted on the basis of a dose-retainment probability using dose-assignment decisions. We evaluated early the completion for two actual trials. In addition, we performed a computer simulation to confirm the percentage of correctly selected MTDs, the early completion percentage, and the average number of patients treated.

RESULTS

In the evaluation of the two actual trials, we confirmed that the trials completed early. In the simulation results, we confirmed that the percentages of correct MTD selection were maintained relative to the original model-assisted designs. The early completion percentages ranged from 50% to 90%, and the number of patients treated reduced from 20%-60% relative to the planned number of patients.

CONCLUSION

We conclude that the early completion method can be applied unproblematically to the model-assisted design of phase I dose-finding trials.

An adaptive gBOIN design with shrinkage boundaries for phase I dose-finding trials

Background

With the emergence of molecularly targeted agents and immunotherapies, the landscape of phase I trials in oncology has been changed. Though these new therapeutic agents are very likely induce multiple low- or moderate-grade toxicities instead of DLT, most of the existing phase I trial designs account for the binary toxicity outcomes. Motivated by a pediatric phase I trial of solid tumor with a continuous outcome, we propose an adaptive generalized Bayesian optimal interval design with shrinkage boundaries, gBOINS, which can account for continuous, toxicity grades endpoints and regard the conventional binary endpoint as a special case.

Result

The proposed gBOINS design enjoys convergence properties, e.g., the induced interval shrinks to the toxicity target and the recommended dose converges to the true maximum tolerated dose with increased sample size.

Conclusion

The proposed gBOINS design is transparent and simple to implement. We show that the gBOINS design has the desirable finite property of coherence and large-sample property of consistency. Numerical studies show that the proposed gBOINS design yields good performance and is comparable with or superior to the competing design.

TITE-gBOIN: Time-to-event Bayesian optimal interval design to accelerate dose-finding accounting for toxicity grades

The new therapeutic agents, such as molecular targeted agents and immuno-oncology therapies, appear more likely to induce multiple toxicities at different grades than dose-limiting toxicities defined in traditional dose-finding trials. In addition, it is often challenging to make adaptive decisions on dose escalation and de-escalation on time because of the fast accrual rate and/or the late-onset toxicity outcomes, causing the potential suspension of the enrollment and the delay of the trials. To address these issues, we propose a time-to-event Bayesian optimal interval design to accelerate the dose-finding process utilizing toxicity grades based on both cumulative and pending toxicity outcomes. The proposed design, named "TITE-gBOIN" design, is a nonparametric and model-assisted design and has the virtues of robustness, simplicity and straightforward to implement in actual oncology dose-finding trials. A simulation study shows that the TITE-gBOIN design has a higher probability of selecting the MTDs correctly and allocating more patients to the MTDs across various realistic settings while reducing the trial duration significantly, therefore can accelerate early-stage dose-finding trials.

Probability-of-decision interval 3+3 (POD-i3+3) design for phase I dose finding trials with late-onset toxicity

Late-onset toxicities often occur in phase I trials investigating novel immunotherapy and molecular targeted therapies. For trials with cohort based designs (such as modified toxicity probability interval, Bayesian optimal interval, and i3+3), patients are often turned away since the current cohort are still being followed without definite dose-limiting toxicities, which results in prolonged trial duration and waste of patient resources. In this paper, we incorporate a probability-of-decision framework into the i3+3 design and allow real-time dosing inference when the next patient becomes available. Both follow-up time for the pending patients and time to dose-limiting toxicities for the observed patients are used in calculating the posterior probability of each possible dosing decision. An intensive simulation study is conducted to evaluate the operating characteristics of the newly proposed probability-of-decision-i3+3 design under various dosing scenarios and patient accrual settings. Results show that the probability-of-decision-i3+3 design achieves comparable safety and reliability performances but much shorter trial duration compared to the complete designs.

Bayesian adaptive model selection design for optimal biological dose finding in phase I/II clinical trials

Identification of the optimal dose presents a major challenge in drug development with molecularly targeted agents, immunotherapy, as well as chimeric antigen receptor T-cell treatments. By casting dose finding as a Bayesian model selection problem, we propose an adaptive design by simultaneously incorporating the toxicity and efficacy outcomes to select the optimal biological dose (OBD) in phase I/II clinical trials. Without imposing any parametric assumption or shape constraint on the underlying dose-response curves, we specify curve-free models for both the toxicity and efficacy endpoints to determine the OBD. By integrating the observed data across all dose levels, the proposed design is coherent in dose assignment and thus greatly enhances efficiency and accuracy in pinning down the right dose. Not only does our design possess a completely new yet flexible dose-finding framework, but it also has satisfactory and robust performance as demonstrated by extensive simulation studies. In addition, we show that our design enjoys desirable coherence properties, while most of existing phase I/II designs do not. We further extend the design to accommodate late-onset outcomes which are common in immunotherapy. The proposed design is exemplified with a phase I/II clinical trial in chronic lymphocytic leukemia.

A Bayesian hierarchical monitoring design for phase II cancer clinical trials: Incorporating information on response duration into monitoring rules

We propose a Bayesian hierarchical monitoring design for single-arm phase II clinical trials of cancer treatments that incorporates the information on the duration of response (DOR) into the monitoring rules. To screen a new treatment by evaluating its preliminary therapeutic effect, futility monitoring rules are commonly used in phase II clinical trials to make "go/no-go" decisions timely and efficiently. These futility monitoring rules are usually focused on a single outcome (eg, response rate), although a single outcome may not adequately determine the efficacy of the experimental treatment. For example, targeted agents with a long response duration but a similar response rate may be worth further evaluation in cancer research. To address this issue, we propose Bayesian hierarchical futility monitoring rules to consider both the response rate and duration. The first level of monitoring evaluates whether the response rate provides evidence that the experimental treatment is worthy of further evaluation. If the evidence from the response rate does not support continuing the trial, the second level monitoring rule, which is based on the DOR, will be triggered. If both stopping rules are satisfied, the trial will be stopped for futility. We conducted simulation studies to evaluate the operating characteristics of the proposed monitoring rules and compared them to those of standard method. We illustrated the proposed design with a single-arm phase II cancer clinical trial to assess the safety and efficacy of combined treatment of nivolumab and azacitidine in patients with relapsed/refractory acute myeloid leukemia. The proposed design avoids an aggressive early termination for futility when the experimental treatment substantially prolongs the DOR but fails to improve the response rate.

Response adaptive designs for Phase II trials with binary endpoint based on context-dependent information measures

In many rare disease Phase II clinical trials, two objectives are of interest to an investigator: maximising the statistical power and maximising the number of patients responding to the treatment. These two objectives are competing, therefore, clinical trial designs offering a balance between them are needed. Recently, it was argued that response-adaptive designs such as families of multi-arm bandit (MAB) methods could provide the means for achieving this balance. Furthermore, response-adaptive designs based on a concept of context-dependent (weighted) information criteria were recently proposed with a focus on Shannon's differential entropy. The information-theoretic designs based on the weighted Renyi, Tsallis and Fisher informations are also proposed. Due to built-in parameters of these novel designs, the balance between the statistical power and the number of patients that respond to the treatment can be tuned explicitly. The asymptotic properties of these measures are studied in order to construct intuitive criteria for arm selection. A comprehensive simulation study shows that using the exact criteria over asymptotic ones or using information measures with more parameters, namely Renyi and Tsallis entropies, brings no sufficient gain in terms of the power or proportion of patients allocated to superior treatments. The proposed designs based on information-theoretical criteria are compared to several alternative approaches. For example, via tuning of the built-in parameter, one can find designs with power comparable to the fixed equal randomisation's but a greater number of patients responded in the trials.

Rolling continual reassessment method with overdose control: An efficient and safe dose escalation design

In phase 1 dose escalation studies, dose limiting toxicities (DLTs) are defined as adverse events of concern occurring during a predefined time window after first dosing of patients. Standard dose escalation designs, such as the continual reassessment method (CRM), only utilize this binary DLT information. Thus, late-onset DLTs are usually not accounted for when CRM guiding the dose escalation and finally defining the maximum tolerated dose (MTD) of the drug, which brings safety concerns for patients. Previously, several extensions of CRMs, such as the time-to-event CRM (TITE-CRM), fractional CRM (fCRM) and the data augmented CRM (DA-CRM), have been proposed to handle this issue without prolonging trial duration. However, among the model-based designs, none of the designs have explicitly controlled the risk of overdosing as in the escalation with overdose control (EWOC) design. Here we propose a novel dose escalation with overdose control design using a two-parameter logistic regression model for the probability of DLT depending on the dose and a piecewise exponential model for the time to DLT distribution, which we call rolling-CRM design. A comprehensive simulation study has been conducted to compare the performance of the rolling-CRM design with other dose escalation designs. Of note, the trial duration is significantly shorter compared to traditional CRM designs. The proposed design also retains overdose control characteristics, but might require a larger sample size compared to traditional CRM designs.

A novel framework of Bayesian optimal interval design for phase I trials with late-onset toxicities

As molecularly targeted agents (MTAs) and immunotherapies have widely demonstrated delayed toxicity profile after multiple treatment cycles, the traditional phase I dose-finding designs may not be appropriate anymore because they just account for the acute toxicities occurring in the early period of treatment. When the dose-limiting toxicity (DLT) assessment window is prolonged to account for late-onset DLTs, it will cause logistic issues if the enrollment is suspended until all the DLT information is collected. We propose a novel framework to estimate the toxicity probability in the scenarios where some patients' DLT information are not complete and then implement the Bayesian optimal interval (BOIN) design to make decisions on dose escalation/de-escalation. Our proposed approach maintains BOIN's transparency by simply comparing the estimated toxicity probability with the escalation/de-escalation boundaries to decide the next dose level. The numerical studies show that our proposed framework can achieve comparable operating characteristics as other dose-finding designs considering late-onset DLTs, thus providing an attractive option of phase I dose-finding clinical trials for MTAs and immunotherapies.

SPA: Single patient acceleration in oncology dose-escalation trials

Efficient identification of the optimal dose and dosing scheme is one of the most critical and challenging tasks in early-phase oncology trials. The results are far-reaching because advancing a sub-optimal dose to late-stage development may not only jeopardize patients' safety or fail to deliver desired efficacy, but also be costly to sponsors as refined doses must be evaluated further before seeking regulatory approval. A good dose-escalation design is anticipated to yield high accuracy of selecting the correct dose while using fewer patients and keeping the trial duration short. Recently, treating a single patient at each lower dose level until certain events are triggered to switch to larger cohorts has gained much popularity. We name this approach "Single Patient Acceleration" (SPA), which is essentially a variant of the Accelerated Titration Design (ATD) by Simon et al. [25]. Although literature on novel dose-escalation methods is abundant in the past decade, there is a surprisingly lack of research on evaluating the ATD/SPA framework. In this article, we conduct comprehensive simulations to evaluate the performance of dose-escalation designs with or without SPA, and show that SPA improves design efficiency with similar or better accuracy to those without the "single patient" component under certain circumstances (e.g., slow initial enrollment, or the true maximum tolerated dose is at higher candidate dose levels). Potential safety concerns as a cost of efficiency improvement are also investigated in a quantitative manner to illustrate a comprehensive benefit-risk profile of SPA. Practical considerations and recommendations in using SPA are also discussed.

uTPI: A utility-based toxicity probability interval design for phase I/II dose-finding trials

Unlike chemotherapy, the maximum tolerated dose (MTD) of molecularly targeted agents and immunotherapy may not pose significant clinical benefit over the lower doses. By simultaneously considering both toxicity and efficacy endpoints, phase I/II trials can identify a more clinically meaningful dose for subsequent phase II trials than traditional toxicity-based phase I trials in terms of risk-benefit tradeoff. To strengthen and simplify the current practice of phase I/II trials, we propose a utility-based toxicity probability interval (uTPI) design for finding the optimal biological dose, based on a numerical utility that provides a clinically meaningful, one-dimensional summary representation of the patient's bivariate toxicity and efficacy outcome. The uTPI design does not rely on any parametric specification of the dose-response relationship, and it directly models the dose desirability through a quasi binomial likelihood. Toxicity probability intervals are used to screen out overly toxic dose levels, and then the dose escalation/de-escalation decisions are made adaptively by comparing the posterior desirability distributions of the adjacent levels of the current dose. The uTPI design is flexible in accommodating various dose desirability formulations, while only requiring minimum design parameters. It has a clear decision structure such that a dose-assignment decision table can be calculated before the trial starts and can be used throughout the trial, which simplifies the practical implementation of the design. Extensive simulation studies demonstrate that the proposed uTPI design yields desirable as well as robust performance under various scenarios.

Fractional design: An alternative paradigm for late-onset toxicities in oncology dose-finding studies

Late-onset (LO) toxicities often arise in the new era of phase I oncology dose-finding trials with targeted agents or immunotherapies. The current LO toxicities modelling is often formulated in a weighted likelihood framework, where the time-to-event continual reassessment method (TITE-CRM) is commonly used. The TITE-CRM uses the patient exposure time as a weight for the censored observation, while there is large uncertainty on which weight function to be used. As an alternative, the fractional scheme formulates an efficient and robust paradigm to address LO toxicity issues in dose finding. We review the fractional continual reassessment method (fCRM) and compare its operating characteristics with those of the TITE-CRM as well as other competitive designs via extensive simulation studies based on both the fixed and randomly generated scenarios. The fCRM is shown to possess desirable operating characteristics in identifying the maximum tolerated dose (MTD) and deliver competitive performances in comparison with other designs. It provides an alternative efficient and robust paradigm for interpreting and addressing LO toxicities in the new era of phase I dose-finding trials in precision oncology. A real trial example is used to illustrate the practical use of the fCRM design.

Practical considerations for the implementation of adaptive designs for oncology Phase I dose-finding trials

The traditional 3 + 3 design continues to be commonly used for Phase I dose-finding oncology trials, despite increasing criticisms and development of innovative methods. Unfortunately, it is a challenge to convince principal investigators to use novel designs. The goal of this paper is to persuade researchers to break away from 3 + 3 design and provide potential solutions to better designs and implementation strategy. We reviewed the statistical methods for adaptive Phase I designs. The barriers among all the major components of the implementation team have been emphasized and potential solutions have been discussed. Institutional support to the principal investigators and statistician, as well as to other team members is essential to design and implement adaptive trials in academic medical institutions.

A new pragmatic design for dose escalation in phase 1 clinical trials using an adaptive continual reassessment method

BACKGROUND

A key challenge in phase I trials is maintaining rapid escalation in order to avoid exposing too many patients to sub-therapeutic doses, while preserving safety by limiting the frequency of toxic events. Traditional rule-based designs require temporarily stopping recruitment whilst waiting to see whether enrolled patients develop toxicity. This can be both inefficient and introduces logistic challenges to recruitment in the clinic. We describe a novel two-stage dose assignment procedure designed for a phase I clinical trial (STARPAC), where a good estimation of prior was possible.

METHODS

The STARPAC design uses rule-based design until the first patient has a dose limiting toxicity (DLT) and then switches to a modified CRM, with rules to handle patient recruitment during follow-up of earlier patients. STARPAC design is compared via simulations with the TITE-CRM and 3 + 3 methods in various toxicity estimate (T1-5), rate of recruitment (R1-2), and DLT events timing (DT1-4), scenarios using several metrics: accuracy of maximum tolerated dose (MTD), numbers of DLTs, number of patients enrolled and those missed; duration of trial; and proportion of patients treated at the therapeutic dose or MTD.

RESULTS

The simulations suggest that STARPAC design performed well in MTD estimation and in treating patients at the highest possible therapeutic levels. STARPAC and TITE-CRM were comparable in the number of patients required and DLTs incurred. The 3 + 3 design often had fewer patients and DLTs although this is due to its low escalation rate leading to poor MTD estimation. For the numbers of declined patients and MTD estimation 3 + 3 is uniformly worse, with STARPAC being better in those metrics for high toxicity scenarios and TITE-CRM better with low toxicity. In situations including doses with toxicities both above and below 30%, the STARPAC design outperformed TITE-CRM with respect to every metric.

CONCLUSION

When considering doses with toxicities both above and below the target of 30% toxicities, the two-stage STARPAC dose escalation design provides a more efficient phase I trial design than either the traditional 3 + 3 or the TITE-CRM design. Trialists should model various designs via simulation to adopt the most efficient design for their clinical scenario.

TRIAL REGISTRATION

Clinical Trials NCT03307148 (11 October 2017).