Covariate adjusted meta-analytic predictive (CA-MAP) prior for historical borrowing using patient-level data.

Utilization of historical data is increasingly common for gaining efficiency in the drug development and decision-making processes. The underlying issue of between-trial heterogeneity in clinical trials is a barrier in making these methods standard practice in the pharmaceutical industry. Common methods for historical borrowing discount the borrowed information based on the similarity between outcomes in the historical and current data. However, individual clinical trials and their outcomes are intrinsically heterogenous due to differences in study design, patient characteristics, and changes in standard of care. Additionally, differences in covariate distributions can produce inconsistencies in clinical outcome data between historical and current data when there may be a consistent covariate effect. In such scenario, borrowing historical data is still advantageous even though the population level outcome summaries are different. In this paper, we propose a covariate adjusted meta-analytic-predictive (CA-MAP) prior for historical control borrowing. A MAP prior is assigned to each covariate effect, allowing the amount of borrowing to be determined by the consistency of the covariate effects across the current and historical data. This approach integrates between-trial heterogeneity with covariate level heterogeneity to tune the amount of information borrowed. Our method is unique as it directly models the covariate effects instead of using the covariates to select a similar population to borrow from. In summary, our proposed patient-level extension of the MAP prior allows for the amount of historical control borrowing to depend on the similarity of covariate effects rather than similarity in clinical outcomes.

Evolution of Phase II Oncology Trial Design: from Single Arm to Master Protocol.

The recent development of novel anticancer treatments with diverse mechanisms of action has accelerated the detection of treatment candidates tremendously. The rapidly changing drug development landscapes and the high failure rates in Phase III trials both underscore the importance of more efficient and robust phase II designs. The goals of phase II oncology studies are to explore the preliminary efficacy and toxicity of the investigational product and to inform future drug development strategies such as go/no-go decisions for phase III development, or dose/indication selection. These complex purposes of phase II oncology designs call for efficient, flexible, and easy-to-implement clinical trial designs. Therefore, innovative adaptive study designs with the potential of improving the efficiency of the study, protecting patients, and improving the quality of information gained from trials have been commonly used in Phase II oncology studies. Although the value of adaptive clinical trial methods in early phase drug development is generally well accepted, there is no comprehensive review and guidance on adaptive design methods and their best practice for phase II oncology trials. In this paper, we review the recent development and evolution of phase II oncology design, including frequentist multistage design, Bayesian continuous monitoring, master protocol design, and innovative design methods for randomized phase II studies. The practical considerations and the implementation of these complex design methods are also discussed.

Characterization and management of adverse events observed with mobocertinib (TAK-788) treatment for EGFR exon 20 insertion-positive non-small cell lung cancer.

BACKGROUND: Mobocertinib has demonstrated durable clinical benefit in platinum-pretreated patients (PPP) with epidermal growth factor receptor exon 20 insertion-positive non-small cell lung cancer (NSCLC). RESEARCH DESIGN AND METHODS: Pooled safety analysis of two studies included patients with NSCLC (N = 257) treated with the recommended phase 2 dose (RP2D) of mobocertinib (160 mg once daily). We report overall safety (treatment-emergent adverse events [TEAEs]) in the RP2D population; characterization of GI and skin-related events in 114 PPP from a phase 1/2 study (NCT02716116); and clinical activity in PPP with and without dose reductions due to TEAEs. RESULTS: In the RP2D population (N = 257), the most common TEAEs were diarrhea (93%), nausea (47%), rash (38%), and vomiting (37%). In PPP (N = 114), median times to diarrhea onset and resolution were 5 and 2 days, respectively. Median times to onset and resolution of skin-related events were 9 and 78 days, respectively. Among PPP with (n = 29) or without (n = 85) dose reductions due to TEAEs, overall response rates were 21% and 31% and median durations of response were 5.7 and 17.5 months, respectively. CONCLUSIONS: GI and skin-related events are common with mobocertinib; minimizing dose reductions with proactive management may improve clinical outcomes. TRIAL REGISTRATION: NCT02716116; NCT03807778.

Mobocertinib is a treatment for patients with a certain type of lung cancer. We analyzed the safety of mobocertinib in 257 patients with lung cancer. The most common side effects with mobocertinib were diarrhea, nausea, vomiting, and skin rash. In 114 patients with lung cancer who were treated in the past with chemotherapy that included platinum-based drugs, diarrhea started after about 5 days of mobocertinib treatment and went away in about 2 days. Skin-related side effects started after about 9 days and went away in about 2.5 months. One-fifth of patients who had to receive a smaller amount of mobocertinib because of side effects responded to treatment compared with one-third of patients who received the recommended mobocertinib amount. Managing side effects quickly can better help patients with lung cancer who are treated with mobocertinib.

Bayesian Additive Regression Trees (BART) with covariate adjusted borrowing in subgroup analyses.

It is crucial in clinical trials to investigate treatment effect consistency across subgroups defined by patient baseline characteristics. However, there may be treatment effect variability across subgroups due to small subgroup sample size. Various Bayesian models have been proposed to incorporate this variability when borrowing information across subgroups. These models rely on the underlying assumption that patients with similar characteristics will have similar outcomes to the same treatment. Patient populations within each subgroup must subjectively be deemed similar enough Pocock (1976) to borrow response information across subgroups. We propose utilizing the machine learning method of Bayesian Additive Regression Trees (BART) to provide a method for subgroup borrowing that does not rely on an underlying assumption of homogeneity between subgroups. BART is a data-driven approach that utilizes patient-level observations. The amount of borrowing between subgroups automatically adjusts as BART learns the covariate-response relationships. Modeling patient-level data rather than treating the subgroup as a single unit minimizes assumptions regarding homogeneity across subgroups. We illustrate the use of BART in this context by comparing performance from existing subgroup borrowing methods in a simulation study and a case study in non-small cell lung cancer. The application of BART in the context of subgroup analyses alleviates the need to subjectively choose how much information to borrow based on subgroup similarity. Having the amount of borrowing be analytically determined and controlled for based on the similarity of individual patient-level characteristics allows for more objective decision making in the drug development process with many other applications including basket trials.

Role of stem cell transplant in CD30+ PTCL following frontline brentuximab vedotin plus CHP or CHOP in ECHELON-2.

Peripheral T-cell lymphomas (PTCLs) are a heterogeneous group of aggressive non-Hodgkin lymphomas, the majority of which have high relapse rates following standard therapy. Despite use of consolidative stem cell transplant (SCT) following frontline therapy, there remains no consensus on its utility. The double-blind randomized phase 3 ECHELON-2 study (#NCT01777152; clinicaltrials.gov) demonstrated improved progression-free survival (PFS) and overall survival with frontline brentuximab vedotin plus cyclophosphamide, doxorubicin, and prednisone (A+CHP). Herein, we conducted an exploratory subgroups analysis of the impact of consolidative SCT on PFS in patients with previously untreated CD30+ PTCL (ALK- anaplastic large cell lymphoma [ALCL] and non-ALCL) who were in complete response (CR) after frontline treatment with A+CHP or cyclophosphamide, doxorubicin, vincristine, and prednisone. Median PFS follow-up was 47.57 months. The PFS hazard ratio was 0.36, equating to a 64% reduction in the risk of a PFS event in patients who underwent SCT. The median PFS in patients who underwent SCT was not reached, vs 55.66 months in patients who did not undergo SCT. PFS results favored the use of SCT in both ALK- ALCL and non-ALCL subgroups. These data support the consideration of consolidative SCT in patients with CD30+PTCL who achieve CR following treatment with A+CHP.

Treatment Outcomes and Safety of Mobocertinib in Platinum-Pretreated Patients With EGFR Exon 20 Insertion-Positive Metastatic Non-Small Cell Lung Cancer: A Phase 1/2 Open-label Nonrandomized Clinical Trial.

IMPORTANCE: Metastatic non-small cell lung cancer (mNSCLC) with EGFR exon 20 insertion (EGFRex20ins) mutations is associated with a poor prognosis. Mobocertinib is an oral tyrosine kinase inhibitor designed to selectively target EGFRex20ins mutations. OBJECTIVE: To evaluate treatment outcomes and safety of mobocertinib in patients with previously treated EGFRex20ins-positive mNSCLC. DESIGN, SETTING, AND PARTICIPANTS: This 3-part, open-label, phase 1/2 nonrandomized clinical trial with dose-escalation/dose-expansion cohorts (28 sites in the US) and a single-arm extension cohort (EXCLAIM; 40 sites in Asia, Europe, and North America) was conducted between June 2016 and November 2020 (data cutoff date). The primary analysis populations were the platinum-pretreated patients (PPP) cohort and the EXCLAIM cohort. The PPP cohort included 114 patients with platinum-pretreated EGFRex20ins-positive mNSCLC who received mobocertinib 160 mg once daily from the dose-escalation (n = 6), dose-expansion (n = 22), and EXCLAIM (n = 86) cohorts. The EXCLAIM cohort included 96 patients with previously treated EGFRex20ins-positive mNSCLC (10 were not platinum pretreated and thus were excluded from the PPP cohort). INTERVENTIONS: Mobocertinib 160 mg once daily. MAIN OUTCOMES AND MEASURES: The primary end point of the PPP and EXCLAIM cohorts was confirmed objective response rate (ORR) assessed by independent review committee (IRC). Secondary end points included confirmed ORR by investigator, duration of response, progression-free survival, overall survival, and safety. RESULTS: Among the PPP (n = 114) and EXCLAIM (n = 96) cohorts, the median (range) age was 60 (27-84) and 59 (27-80) years, respectively; most patients were women (75 [66%] and 62 [65%], respectively) and of Asian race (68 [60%] and 66 [69%], respectively). At data cutoff, median follow-up was 14.2 months in the PPP cohort (median 2 prior anticancer regimens; 40 [35%] had baseline brain metastases), with confirmed ORR of 28% (95% CI, 20%-37%) by IRC assessment and 35% (95% CI, 26%-45%) by investigator assessment; median duration of response by IRC assessment was 17.5 months (95% CI, 7.4-20.3). Median progression-free survival by IRC assessment was 7.3 months (95% CI, 5.5-9.2). Median overall survival was 24.0 months (95% CI, 14.6-28.8). In the EXCLAIM cohort, median follow-up was 13.0 months, with confirmed ORR by IRC assessment of 25% (95% CI, 17%-35%) and by investigator assessment of 32% (95% CI, 23%-43%). The most common treatment-related adverse events were diarrhea and rash. CONCLUSIONS AND RELEVANCE: In this open-label, phase 1/2 nonrandomized clinical trial, mobocertinib was associated with clinically meaningful benefit in patients with previously treated EGFRex20ins-positive mNSCLC, with a manageable safety profile. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT02716116.

Randomized phase 3 ALCANZA study of brentuximab vedotin vs physician's choice in cutaneous T-cell lymphoma: final data.

The primary analysis of the phase 3 ALCANZA trial showed significantly improved objective responses lasting ≥4 months (ORR4; primary endpoint) and progression-free survival (PFS) with brentuximab vedotin vs physician's choice (methotrexate or bexarotene) in CD30-expressing mycosis fungoides (MF) or primary cutaneous anaplastic large-cell lymphoma (C-ALCL). Cutaneous T-cell lymphomas often cause pruritus and pain; brentuximab vedotin improved skin symptom burden with no negative effects on quality of life. We report final data from ALCANZA (median follow-up, 45.9 months). Adults with previously treated CD30-expressing MF/C-ALCL were randomly assigned to brentuximab vedotin (n = 64) or physician's choice (n = 64). Final data demonstrated improved responses per independent review facility with brentuximab vedotin vs physician's choice: ORR4; 54.7% vs 12.5% (P < .001); complete response, 17.2% vs 1.6% (P = .002). Median PFS with brentuximab vedotin vs physician's choice was 16.7 months vs 3.5 months (P < .001). Median time to the next treatment was significantly longer with brentuximab vedotin than with physician's choice (14.2 vs 5.6 months; hazard ratio, 0.27; 95% confidence interval, 0.17-0.42; P < .001). Of 44 patients in the brentuximab vedotin arm who experienced any-grade peripheral neuropathy, (grade 3, n = 6; grade 4, n = 0), 86% (38 of 44) had complete resolution (26 of 44) or improvement to grades 1 and 2 (12 of 44). Peripheral neuropathy was ongoing in 18 patients (all grades 1-2). These final analyses confirm improved, clinically meaningful, durable responses and longer PFS with brentuximab vedotin vs physician's choice in CD30-expressing MF or C-ALCL. This trial was registered at https://www.clinicaltrials.gov as #NCT01578499.

Propensity-score-based meta-analytic predictive prior for incorporating real-world and historical data.

As the availability of real-world data sources (eg, EHRs, claims data, registries) and historical data has rapidly surged in recent years, there is an increasing interest and need from investigators and health authorities to leverage all available information to reduce patient burden and accelerate both drug development and regulatory decision making. Bayesian meta-analytic approaches are a popular historical borrowing method that has been developed to leverage such data using robust hierarchical models. The model structure accounts for various degrees of between-trial heterogeneity, resulting in adaptively discounting the external information in the case of data conflict. In this article, we propose to integrate the propensity score method and Bayesian meta-analytic-predictive (MAP) prior to leverage external real-world and historical data. The propensity score methodology is applied to select a subset of patients from external data that are similar to those in the current study with regards to key baseline covariates and to stratify the selected patients together with those in the current study into more homogeneous strata. The MAP prior approach is used to obtain stratum-specific MAP prior and derive the overall propensity score integrated meta-analytic predictive (PS-MAP) prior. Additionally, we allow for tuning the prior effective sample size for the proposed PS-MAP prior, which quantifies the amount of information borrowed from external data. We evaluate the performance of the proposed PS-MAP prior by comparing it to the existing propensity score-integrated power prior approach in a simulation study and illustrate its implementation with an example of a single-arm phase II trial.

Bayesian semiparametric meta-analytic-predictive prior for historical control borrowing in clinical trials.

When designing a clinical trial, borrowing historical control information can provide a more efficient approach by reducing the necessary control arm sample size while still yielding increased power. Several Bayesian methods for incorporating historical information via a prior distribution have been proposed, for example, (modified) power prior, (robust) meta-analytic predictive prior. When utilizing historical control borrowing, the prior parameter(s) must be specified to determine the magnitude of borrowing before the current data are observed. Thus, a flexible prior is needed in case of heterogeneity between historic trials or prior data conflict with the current trial. To incorporate the ability to selectively borrow historic information, we propose a Bayesian semiparametric meta-analytic-predictive prior. Using a Dirichlet process mixture prior allows for relaxation of parametric assumptions, and lets the model adaptively learn the relationship between the historic and current control data. Additionally, we generalize a method for estimating the prior effective sample size (ESS) for the proposed prior. This gives an intuitive quantification of the amount of information borrowed from historical trials, and aids in tuning the prior to the specific task at hand. We illustrate the effectiveness of the proposed methodology by comparing performance between existing methods in an extensive simulation study and a phase II proof-of-concept trial in ankylosing spondylitis. In summary, our proposed robustification of the meta-analytic-predictive prior alleviates the need for prespecifying the amount of borrowing, providing a more flexible and robust method to integrate historical data from multiple study sources in the design and analysis of clinical trials.

Response to brentuximab vedotin versus physician's choice by CD30 expression and large cell transformation status in patients with mycosis fungoides: An ALCANZA sub-analysis.

INTRODUCTION: Mycosis fungoides (MF), the most common type of cutaneous T-cell lymphoma, can lead to disfiguring lesions, debilitating pruritus and frequent skin infections. This study assessed response to brentuximab vedotin in patients with MF in the phase III ALCANZA study. METHODS: Baseline CD30 levels and large-cell transformation (LCT) status were centrally reviewed in patients with previously-treated CD30-positive MF using ≥2 skin biopsies obtained at screening; eligible patients required ≥1 biopsy with ≥10% CD30 expression. Patients were categorised as CD30min < 10% (≥1 biopsy with <10% CD30 expression), or CD30min ≥ 10% (all biopsies with ≥10% CD30 expression) and baseline LCT present or absent. Efficacy analyses were the proportion of patients with objective response lasting ≥4 months (ORR4) and progression-free survival (PFS). RESULTS: Clinical activity with brentuximab vedotin was observed across all CD30 expression levels in patients with ≥1 biopsy showing ≥10% CD30 expression. Superior ORR4 was observed with brentuximab vedotin versus physician's choice in patients: with CD30min < 10% (40.9% versus 9.5%), with CD30min ≥ 10% (57.1% versus 10.3%), with LCT (64.7% versus 17.6%) and without LCT (38.7% versus 6.5%). Brentuximab vedotin improved median PFS versus physician's choice in patients: with CD30min < 10% (16.7 versus 2.3 months), with CD30min ≥ 10% (15.5 versus 3.9 months), with LCT (15.5 versus 2.8 months) and without LCT (16.1 versus 3.5 months). Safety profiles were generally comparable across subgroups. CONCLUSION: These exploratory analyses demonstrated that brentuximab vedotin improved rates of ORR4 and PFS versus physician's choice in patients with CD30-positive MF and ≥1 biopsy showing ≥10% CD30 expression, regardless of LCT status. CLINICAL TRIAL REGISTRATION: Clinicaltrials.gov, NCT01578499.

Activity and Safety of Mobocertinib (TAK-788) in Previously Treated Non-Small Cell Lung Cancer with EGFR Exon 20 Insertion Mutations from a Phase I/II Trial.

Mobocertinib, an oral epidermal growth factor receptor (EGFR) inhibitor targeting EGFR gene mutations, including exon 20 insertions (EGFRex20ins), in non-small cell lung cancer, was evaluated in a phase I/II dose-escalation/expansion trial (ClinicalTrials.gov NCT02716116). Dose escalation identified 160 mg/d as the recommended phase 2 dose and maximum tolerated dose. Among 136 patients treated with 160 mg/d, the most common any-grade treatment-related adverse events (TRAE; >25%) were diarrhea (83%), nausea (43%), rash (33%), and vomiting (26%), with diarrhea (21%) the only grade ≥3 TRAE >5%. Among 28 EGFRex20ins patients treated at 160 mg/d, the investigator-assessed confirmed response rate was 43% (12/28; 95% confidence interval, 24%-63%) with median duration of response of 14 months (5.0-not reached) and median progression-free survival of 7.3 months (4.4-15.6). Mobocertinib demonstrated antitumor activity in patients with diverse EGFRex20ins variants with a safety profile consistent with other EGFR inhibitors. SIGNIFICANCE: No oral EGFR-targeted therapies are currently approved for patients with EGFRex20ins NSCLC. Mobocertinib demonstrated antitumor activity with manageable toxicity in patients with advanced EGFRex20ins NSCLC in this study, supporting additional development of mobocertinib in this patient population.See related commentary by Pacheco, p. 1617.This article is highlighted in the In This Issue feature, p. 1601.

Bayesian analysis of survival data with missing censoring indicators.

In some large clinical studies, it may be impractical to perform the physical examination to every subject at his/her last monitoring time in order to diagnose the occurrence of the event of interest. This gives rise to survival data with missing censoring indicators where the probability of missing may depend on time of last monitoring and some covariates. We present a fully Bayesian semi-parametric method for such survival data to estimate regression parameters of the proportional hazards model of Cox. Theoretical investigation and simulation studies show that our method performs better than competing methods. We apply the proposed method to analyze the survival data with missing censoring indicators from the Orofacial Pain: Prospective Evaluation and Risk Assessment study.

Flexible Bayesian subgroup analysis in early and confirmatory trials.

Subgroup analysis is one of the most important issues in clinical trials. In confirmatory trials, it is critical to investigate consistency of the treatment effect across subgroups, which could potentially result in incorrect scientific conclusion or regulatory decision. There are many challenges and methodological complications of interpreting subgroup results beyond the regulatory setting. For the early phase or proof of concept trials, particularly in basket trials, it is also important to have reliable estimation of subgroup treatment effect in order to guide the next phase go/no-go decision making when large biases can be introduced due to small sample size or random variability. In this paper, we review several recent methods that have been proposed for subgroup analysis in the Bayesian framework to correct for bias. We present simulation results from applying various novel Bayesian hierarchical models for subgroup analysis to a phase II basket trial. For different scenarios considered, we compare the average total sample size, and frequentist-like operating characteristics of power and familywise type I error rate. We compare the precision of the model estimates of the treatment effect by assessing average relative bias and the width of the 95% credible interval for the bias. We also demonstrate flexible Bayesian hierarchical models in a case study of a phase III oncology trial for subgroup treatment effect estimation to help with regulatory decision making. Finally, we conclude our findings in the discussion section and give recommendations on how these methods could be implemented in confirmatory and early phase clinical trials.

Patient-reported quality of life in patients with relapsed/refractory cutaneous T-cell lymphoma: Results from the randomised phase III ALCANZA study.

BACKGROUND: Brentuximab vedotin was approved for adult patients with CD30-expressing cutaneous T-cell lymphoma treated with prior systemic therapy based on improved response rates and progression-free survival with brentuximab vedotin (1.8 mg/kg once every 3 weeks; ≤16 cycles) versus physician's choice (methotrexate/bexarotene; ≤48 weeks) in the phase III ALCANZA study. Quality of life (QoL) in ALCANZA patients was also examined. METHODS: QoL measures in ALCANZA were based on the Skindex-29, Functional Assessment of Cancer Therapy-General (FACT-G) and European QoL 5-dimension (EQ-5D) questionnaires. RESULTS: Mean maximum reduction from the baseline Skindex-29 symptom domain score (key secondary end-point) was greater with brentuximab vedotin than physician's choice (-27.96 versus -8.62); the difference, -18.9 (95% confidence interval -26.6, -11.2; adjusted p < 0.001), exceeded the study-defined minimally important difference (9.0-12.3). Mean changes from baseline to end-of-treatment visit total FACT-G scores were similar with brentuximab vedotin and physician's choice (0.15 versus -2.29). EQ-5D changes were also comparable between arms. Among brentuximab vedotin-treated patients with peripheral neuropathy (PN), mean maximum reduction in Skindex-29 symptom domain was -35.54 versus -11.11 in patients without PN. PN had no meaningful effect on FACT-G and EQ-5D QoL scores. CONCLUSIONS: In summary, brentuximab vedotin produced superior reductions in symptom burden compared with physician's choice, without adversely impacting QoL. QoL was unaffected by the presence of PN in brentuximab vedotin-treated patients. CLINICAL TRIAL REGISTRATION: NCT01578499.

Bayesian Semi-parametric Design (BSD) for adaptive dose-finding with multiple strata.

In the era of precision medicine, it is of increasing interest to consider multiple strata (e.g. indications, regions, or subgroups) within a single oncology dose-finding study when identifying the maximum tolerated dose (MTD). We propose two Bayesian semi-parametric designs (BSD) for dose-finding with multiple strata to allow for both adaptively dosing patients based on various toxicity profiles and efficient identification of the MTD for each stratum. We develop non-parametric priors based on the Dirichlet process to allow for a flexible prior distribution and negate the need for a pre-specified exchangeability parameter. The two BSD models are built under different prior beliefs of strata heterogeneity and allow for appropriate borrowing of information across similar strata. Simulation studies are performed to evaluate the BSD model performance by comparing it with existing methods, including the fully stratified, exchangeability, and exchangeability-non-exchangeability models. In general, our BSD models outperform the competing methods in correctly identifying the MTD for different strata and necessitate a smaller sample size to determine the MTD. The BSD models are robust to various heterogeneity assumptions and can be easily extended to other binary and time to event endpoints.

Comparison of multi-arm multi-stage design and adaptive randomization in platform clinical trials.

Platforms trials are clinical trials that allow for concurrent evaluations of multiple treatments, thus allowing for more efficient and ethical studies compared to traditional two-arm trials. Conventional group-sequential multi-arm multi-stage (MAMS) designs use pre-specified stopping boundaries and treatment selection rules to determine if experimental treatments should be dropped. Flexible MAMS designs allow for interim modifications to the design plan without compromising error rates. Bayesian response adaptive randomization (BRAR) designs increase patient allocation to treatment arms that are performing well during the course of the trial. In this paper, we compare these two major methods and their extensions under several scenarios in the platform trials setting. Results show that BRAR and flexible MAMS designs have comparable power and type 1 error rate under varying simulated scenarios, allowing for addition of flexible treatment selection. BRAR outperforms flexible MAMS when there is a single effective treatment. Flexible MAMS designs are more efficient compared to BRAR when there are no effective treatments. BRAR performance increases as the probability of a treatment arm being dropped increases.