Futility Stopping Research Articles

An automated platform trial framework for A/B testing

This paper proposes a platform trial for conducting A/B tests with multiple arms and interim monitoring to investigate the impact of several factors on the expected sample size and probability of early stopping. We examined the performance of three stopping boundaries: O’Brien Fleming (OBF) stopping for either futility or difference (both), Pocock stopping for futility only, and fixed sample size design. We simulated twelve scenarios of different orders of arms based on various effect sizes, as well as considered 1 or 3 interim looks. The overall findings are summarizing in a flowchart to provide intuitive guidance for the design of the platform based on the simulation. We found that it is better to use OBF stopping for both if there is any effective variant and the trial is sufficiently powered to detect the expected effect size. If the study is underpowered to detect a difference, we recommend fixed sample size design to gather as much information as possible, however if the expected sample size is important to minimize, we recommend using Pocock boundaries with futility monitoring. Our results aimed at helping high-tech companies conduct their own studies without requiring extensive knowledge of clinical trial design and statistical methodology.

Adaptive Clinical Trials in Stroke.

Designing a clinical trial to evaluate the efficacy of an intervention is often complicated by uncertainty over aspects of the study population, potential treatment effects, most relevant outcomes, dropouts, and other factors. However, once participants begin to be enrolled and partial trial data become available, this level of uncertainty is reduced. Adaptive clinical trials are designed to take advantage of the accumulating data during the conduct of a trial to make changes according to prespecified decision rules to increase the likelihood of success or statistical efficiency. Common adaptive rules address early stopping for benefit or futility, sample size reestimation, adding or dropping treatment arms or altering randomization ratios, and changing the eligibility criteria to focus on responder patient subgroups. Adaptive clinical trials are gaining popularity for clinical stroke research. We provide an overview of the methods, practical considerations, challenges and limitations, and potential future role of adaptive clinical trials in advancing knowledge and practice in stroke.

TRLS-03. A PHASE 1/2 STUDY ASSESSING SAFETY AND PHARMACOKINETICS OF CEMIPLIMAB IN PEDIATRIC PATIENTS WITH RELAPSED OR REFRACTORY SOLID OR CENTRAL-NERVOUS-SYSTEM TUMORS AND SAFETY AND EFFICACY OF CEMIPLIMAB IN COMBINATION WITH RADIOTHERAPY IN PATIENTS WITH NEWLY DIAGNOSED DIFFUSE INTRINSIC PONTINE GLIOMA, NEWLY DIAGNOSED HIGH-GRADE GLIOMA, OR RECURRENT HIGH-GRADE GLIOMA

Abstract BACKGROUND We investigated, for the first time, combination PD-1 inhibition (cemiplimab) and radiation therapy (RT) followed by cemiplimab monotherapy in children with solid tumors, diffuse intrinsic pontine glioma (DIPG) and high-grade glioma (HGG). METHODS We conducted a multicenter study through the Pediatric Neuro-Oncology Consortium. Phase 1 established the safety/pharmacokinetics of cemiplimab monotherapy in children (<18 years) with recurrent solid and central-nervous-system (CNS) tumors. Phase 2 evaluated cemiplimab/RT in children and young adults (3-25 years) with newly diagnosed (nd) DIPG, ndHGG or recurrent HGG (rHGG). Patients with ndDIPG and ndHGG were randomized to hypofractionated RT (hfRT) or conventionally fractionated RT (cRT). Primary endpoints included overall survival at 12 months (OS12) for ndDIPG and rHGG and progression-free survival at 12 months (PFS12) for ndHGG. Correlates included quality-of-life, circulating tumor DNA (ctDNA), immunogenicity, and PD-1 pathway components. RESULTS 25 patients (solid tumors, n=8; CNS tumors, n=17) were treated in phase 1; serum drug concentrations were comparable to adults. 32 patients were treated in phase 2. OS12 was 33.3% for ndDIPG treated with cemiplimab/hfRT (n=6), 0.0% for ndDIPG with cemiplimab/cRT (n=5), and 31.3% for rHGG (n=9). PFS12 was not estimable for ndHGG treated with cemiplimab/hfRT (n=5) and 20.0% for ndHGG treated with cemiplimab/cRT (n=7). 20 patients (35.1%) had grade 3/4 treatment-related AEs (TRAEs) and 13 (22.8%) had serious TRAEs. The most frequent grade 3/4 TRAE was decreased lymphocyte count (n=4 [7.0%]). 12 patients (21.1%) demonstrated treatment-emergent pseudo-progression, 2 (3.5%) grade 3/4. The most frequent serious TRAE was pseudo-progression (n=3 [5.3%]). CONCLUSIONS Cemiplimab/RT did not demonstrate an improvement in OS (ndDIPG and rHGG) or PFS (ndHGG), leading to early stopping for futility. However, the combination was tolerable, demonstrated similar outcomes across RT schedules, and showed similar exposure in serum compared to adults. Biologic and clinical correlate analyses are underway.

The Bayesian Group-Sequential Predictive Evidence Value Design for Phase II Clinical Trials with Binary Endpoints

AbstractIn clinical research, the initial efficacy of a new agent is typically assessed in a phase IIA study. Bayesian group-sequential designs are often based on predictive probability of trial success. In this paper, the novel Bayesian group-sequential predictive evidence value design is introduced, and we prove that the predictive probability approach is a special case of it. A comparison with Simon’s two-stage and competing Bayesian designs based on phase IIA cancer trials is provided. Results show that the novel design can improve operating characteristics such as the false-positive rate, probability of early stopping for futility and expected sample size of the trial. Given these advantages, the predictive evidence value design constitutes an important addition to the biostatistician’s toolbelt when planning a phase IIA trial the Bayesian way, in particular, when small sample sizes and a large probability for early termination under the null hypothesis are desired.

Beta spending function based on conditional power in group sequential design.

Conditional power (CP) serves as a widely utilized approach for futility monitoring in group sequential designs. However, adopting the CP methods may lead to inadequate control of the type II error rate at the desired level. In this study, we introduce a flexible beta spending function tailored to regulate the type II error rate while employing CP based on a predetermined standardized effect size for futility monitoring (a so-called CP-beta spending function). This function delineates the expenditure of type II error rate across the entirety of the trial. Unlike other existing beta spending functions, the CP-beta spending function seamlessly incorporates beta spending concept into the CP framework, facilitating precise stagewise control of the type II error rate during futility monitoring. In addition, the stopping boundaries derived from the CP-beta spending function can be calculated via integration akin to other traditional beta spending function methods. Furthermore, the proposed CP-beta spending function accommodates various thresholds on the CP-scale at different stages of the trial, ensuring its adaptability across different information time scenarios. These attributes render the CP-beta spending function competitive among other forms of beta spending functions, making it applicable to any trials in group sequential designs with straightforward implementation. Both simulation study and example from an acute ischemic stroke trial demonstrate that the proposed method accurately captures expected power, even when the initially determined sample size does not consider futility stopping, and exhibits a good performance in maintaining overall type I error rates for evidentfutility.

Optimal futility stopping boundaries for binary endpoints

BackgroundGroup sequential designs incorporating the option to stop for futility at the time point of an interim analysis can save time and resources. Thereby, the choice of the futility boundary importantly impacts the design’s resulting performance characteristics, including the power and probability to correctly or wrongly stop for futility. Several authors contributed to the topic of selecting good futility boundaries. For binary endpoints, Simon’s designs (Control Clin Trials 10:1–10, 1989) are commonly used two-stage designs for single-arm phase II studies incorporating futility stopping. However, Simon’s optimal design frequently yields an undesirably high probability of falsely declaring futility after the first stage, and in Simon’s minimax design often a high proportion of the planned sample size is already evaluated at the interim analysis leaving only limited benefit in case of an early stop.MethodsThis work focuses on the optimality criteria introduced by Schüler et al. (BMC Med Res Methodol 17:119, 2017) and extends their approach to binary endpoints in single-arm phase II studies. An algorithm for deriving optimized futility boundaries is introduced, and the performance of study designs implementing this concept of optimal futility boundaries is compared to the common Simon’s minimax and optimal designs, as well as modified versions of these designs by Kim et al. (Oncotarget 10:4255–61, 2019).ResultsThe introduced optimized futility boundaries aim to maximize the probability of correctly stopping for futility in case of small or opposite effects while also setting constraints on the time point of the interim analysis, the power loss, and the probability of stopping the study wrongly, i.e. stopping the study even though the treatment effect shows promise. Overall, the operating characteristics, such as maximum sample size and expected sample size, are comparable to those of the classical and modified Simon’s designs and sometimes better. Unlike Simon’s designs, which have binding stopping rules, the optimized futility boundaries proposed here are not adjusted to exhaust the full targeted nominal significance level and are thus still valid for non-binding applications.ConclusionsThe choice of the futility boundary and the time point of the interim analysis have a major impact on the properties of the study design. Therefore, they should be thoroughly investigated at the planning stage. The introduced method of selecting optimal futility boundaries provides a more flexible alternative to Simon’s designs with non-binding stopping rules. The probability of wrongly stopping for futility is minimized and the optimized futility boundaries don’t exhibit the unfavorable properties of an undesirably high probability of falsely declaring futility or a high proportion of the planned sample evaluated at the interim time point.

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

BackgroundRandomized 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. MethodsWe 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. ResultsWhen 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. ConclusionsExternal 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.

Iadademstat and Gilteritinib for the Treatment of FLT3-Mutated Relapsed/Refractory Acute Myeloid Leukemia: The Frida Study

Background: Frontline and relapsed /refractory (R/R) acute myeloid leukemia (AML) patients carrying fms-related tyrosine kinase 3 [FLT3]-mutations (FLT3 mut+) benefit from FLT3 inhibitors (FLT3i) that have emerged as active therapeutic agents for this subset of AML encompassing up to 30-40% of all AML cases. The FLT3i gilteritinib, as monotherapy for R/R AML pts, has improved outcomes but the duration of remission achieved is transient and often brief. The Phase 3 randomized ADMIRAL trial of gilteritinib demonstrated a 20% CR rate and 2.8 months event-free-survival (EFS) (Perl, et al., NEJM 2019). The development of new targeted agents which synergize with FLT3 inhibitors, and ideally also inhibit the emergence of resistance, is therefore a major medical need in AML. Iadademstat (iada/ORY-1001) is a specific, oral, potent, covalent inhibitor of the epigenetic Lysine-Specific Demethylase 1 (LSD1/KDMA1) enzyme. Epigenetic dysregulation is a hallmark of AML. Up to 70% of recurring mutations in AML patients target epigenetic regulators of gene expression, resulting in myeloid differentiation blockade and enhanced leukemic stem cell renewal, underscoring the potential of epigenetic therapies to change the natural history of the disease. Preclinically, iada produces striking synergy with FLT3is, particularly gilteritinib in FLT3 wild-type and FLT3 mut+ AML cells and in derived cell lines resistant to venetoclax, azacitidine and other FLT3is (Sacilotto et al., 2022 Eur J. of Cancer 174S1). The novel MOAs generating this synergy in combinations include activation of a pro-differentiating epigenetic transcriptional program with simultaneous suppression of MYC-driven target genes (Yashar, et al. 2023, Mol. Cancer Res). The recently completed Phase 2 study of iada in combination with azacitidine (ALICE) produced a high rate of composite remission, including complete remission (CR) and CR with incomplete count recovery (CRi) (mostly MRD negative), as well as durable responses, in treatment naïve, unfit AML patients (pts), without exacerbating the toxicity profile of azacitidine (Salamero et al., 2022, Blood S1). The FRIDA study (NCT05546580) aims to establish the safety, tolerability, and the recommended phase 2 dose (RP2D) of the combination of iada plus gilteritinib in FLT3 mut+ R/R AML. Methods: Adult pts with body weight ≥50 Kg and ECOG 0-2, with FLT3 mut+ R/R AML, after no more than 2 prior lines of therapy, are enrolling. Certain patients with prior exposure to FLT3is including gilteritinib may be eligible. Up to 18 pts, in a 3+3 escalation phase, will receive iada at doses of 75 to 150 ug, orally, in 5 days ON - 2 days OFF schedule, with continuous gilteritinib, at 120 mg/day orally. In the expansion phase, up to 14 pts at the selected safe and pharmacologically active dose/s (determined based on all available data from escalation pts including PK, target engagement (TE), safety, tolerability, and emerging activity) will be enrolled. Primary endpoints of the study are safety and RP2D determination, based on the same 5 criteria outlined above. Bayesian posterior probability efficacy monitoring will be performed periodically for each dose cohort in the expansion phase. Bayesian efficacy futility and early stopping boundary will be applied during the monitoring. Posterior probability criterion (Prob (CR>0.3) ≥ 0.60) at the end of the study will warrant additional development. The safety of the combination treatment will also be continuously evaluated during expansion following a Bayesian design stopping rule.Secondary endpoints include overall survival, EFS, CR, CR/CR with partial hematologic recovery (CRh), overall response rates, time to response, duration of response, and transfusion rate. Exploratory endpoints include measurable residual disease and gene mutational analysis. At the time of the submission, FRIDA is enrolling a second dose level cohort in escalation phase and plans to have the 15 sites open to accrual in the US by the end of 2023. Additional sites will be added for a subsequent randomized controlled double-blinded FRIDA 2 study to assess the efficacy of the iada and gilteritinib combination in R/R FLT3 mut+ AML.

Temsirolimus (T) in patients (pts) with solid tumors with PIK3CA mutation (mut): Results from the Targeted Agent and Profiling Utilization Registry (TAPUR) study.

3117 Background: TAPUR is a phase II basket study evaluating antitumor activity of commercially available targeted agents in pts with advanced cancers with genomic alterations. Results in cohorts of pts with breast cancer (BC) and other solid tumors with PIK3CA mut treated with T are reported. Methods: Eligible pts had BC or other solid tumors, measurable disease, ECOG performance status (PS) 0-2, adequate organ function, and no standard treatment (tx) options. Genomic testing was performed in CLIA-certified, CAP-accredited site selected labs. After antihistamine pre-tx, 25 mg of T was infused over 30-60 minutes once weekly until disease progression. Primary endpoint was disease control (DC), defined as complete or partial (PR) response, or stable disease of at least 16+ weeks (wks) duration (SD16+) per RECIST v1.1. For the BC cohort, Simon’s optimal 2-stage design with null DC rate of 15% vs. 35% (power=0.85, α=0.10) was used with stage 1 (n=10) stopping for futility if < 2/10 pts had DC. Low accruing histology-specific cohorts with PIK3CA and T tx were collapsed into 1 histology-pooled (HP) cohort. For the HP cohort, the hypothesized null DC rate of 15% was evaluated by a 1-sided exact binomial test with α=0.10. Secondary endpoints were progression-free survival (PFS), overall survival (OS), and safety. Results: 12 pts with PIK3CA mut with BC and 29 pts with PIK3CA mut in other solid tumors (across 9 tumor types) were enrolled. 2 pts (1 in each cohort) were found to be ineligible after enrolling and were not included in efficacy analyses. Demographics and outcomes for each cohort are shown. At the end of stage 1 in the BC cohort, 1 PR was observed for DC and OR rates of 9%; the cohort was closed for futility (p=0.83). For the HP cohort, 3 PR and 5 SD16+ were observed for DC rate of 29% (p=0.049) and OR rate of 11%; the null hypothesis was rejected. Cancer types in pts with OR or SD16+ included cervical, ovarian and head/neck; most common muts were H1047R/L (3), E545K (2) and E542K (2). 1 pt with ovarian and H1047R has ongoing PR at 86 wks. 11/41 pts had ≥1 tx-related grade 3-4 adverse or serious adverse event, including anemia, headache, hyperglycemia, hypertension, hypertriglyceridemia, mucositis oral, lymphocyte, neutrophil or platelet count decrease, pneumonitis, and sepsis. Conclusions: Although T does not appear to have antitumor activity in pts with BC with PIK3CA mut, it does show antitumor activity in pts with other solid tumors with PIK3CA mut and warrants further study. Clinical trial information: NCT02693535 . [Table: see text]

Iadademstat in combination with gilteritinib for patients with mutated FLT3 relapsed/refractory acute myeloid leukemia.

TPS7075 Background: Epigenetic dysregulation is a hallmark of acute myeloid leukemia (AML) and is caused by recurrent translocations and/or mutations in chromatin regulators and transcription factors, resulting in myeloid differentiation blockade and leukemic stem cell renewal. Accordingly, about 70% of recurring mutations in AML patients (pts) target regulators of gene expression, underscoring the potential of epigenetic therapies to change the disease natural history. Iadademstat (iada/ORY-1001) is a specific, oral, potent, covalent inhibitor of the epigenetic Lysine-Specific Demethylase 1 (LSD1/KDMA1) enzyme. In preclinical and clinical studies iada decreased leukemic stem cell survival and induced macrophage/monocytic differentiation of blasts. ALICE, a Ph2 study of iada in combination with azacitidine (aza), showed high complete remission (CR/CRi) rates and durable responses in treatment naïve, unfit AML pts without exacerbating the toxicity profile of aza (Salamero et al., Oral at ASH 2022). Despite improvements in AML therapy, relapsed and refractory (R/R) cases are frequent and contribute to the death of more than 50% of pts, particularly in those subpopulations with higher risk genetics, the largest of which (up to 30-40% AML pts) harbors fms-like tyrosine kinase 3 mutations (FLT3mut+). Use of the FLT3 inhibitor (FLT3i) gilteritinib as monotherapy for R/R pts resulted in improved outcomes but the duration of remission achieved is transient and often brief (CR rate: 20%; EFS: 2.8 months) per the ADMIRAL Ph3 study (Perl, et al., NEJM 2019). Preclinically, iada has marked synergy with FLT3i, particularly gilteritinib, in FLT3 wild-type and FLT3 mut+ AML cells and in derived cell lines resistant to venetoclax, azacitidine and FLT3is. Methods: Adult pts with ECOG 0-2 and FLT3 mut+ R/R AML, after 1 or 2 prior lines of therapy, will be enrolled. In escalation (3+3), up to 18 pts will receive iada at 75 to 150 ug, orally, in 5 days ON-2 days OFF schedule, with gilteritinib SoC. Up to approximate 14 pts will be expanded from pharmacologically active dose/s (per Project OPTIMUS). Primary endpoints of the study are safety and RP2D determination. Bayesian posterior probability efficacy monitoring will be performed periodically for each dose cohort. Bayesian efficacy futility and early stopping boundary will be applied. Posterior probability criterion (Prob (CR > 0.3) ≥ 0.60) at the end of the study will warrant further research. Safety in expansion will be evaluated continuously with a Bayesian stopping rule. Secondary endpoints include OS, EFS, CR, CR/CRh, ORR, TTR, DoR, and transfusion rate. Exploratory endpoints include PK, PD, MRD and gene mutational analysis. The study will run in 15 US sites. Additional sites will be added for a subsequent Phase 2 randomized controlled double-blinded FRIDA 2 study to assess the efficacy of this combination in R/R FLT3 mut+ AML. Clinical trial information: NCT05546580 .

Epistemic uncertainty in Bayesian predictive probabilities

ABSTRACT Bayesian predictive probabilities have become a ubiquitous tool for design and monitoring of clinical trials. The typical procedure is to average predictive probabilities over the prior or posterior distributions. In this paper, we highlight the limitations of relying solely on averaging, and propose the reporting of intervals or quantiles for the predictive probabilities. These intervals formalize the intuition that uncertainty decreases with more information. We present four different applications (Phase 1 dose escalation, early stopping for futility, sample size re-estimation, and assurance/probability of success) to demonstrate the practicality and generality of the proposed approach.

Completion and publication of clinical trials in a cooperative group: a cohort study of trials of the Swiss Group for Clinical Cancer Research (SAKK)

BackgroundPremature trial discontinuation and non-publication of trial results are still major issues negatively affecting reliable evidence generation.ObjectivesTo investigate trial completion and publication rate of cancer trials conducted within the Swiss...

Abstract PD11-03: PD11-03 Comparison of a mono Atezolizumab window followed by Atezolizumab and chemotherapy with Atezolizumab and chemotherapy in triple negative breast cancer – an interim analysis of the adaptive randomized neoadjuvant trial NeoMono

Abstract Introduction: Improvement of systemic treatment of TNBC represents an unmet medical need. Targeted therapy of regulatory immune pathways has become an important option in the treatment of many malignant diseases including breast cancer. Neodjuvant trials combining chemotherapy and checkpoint inhibitors (KEYNOTE-522 and IMpassion031) have demonstrated a meaningful benefit regarding pathological complete remission (pCR) for the addition of PD-1- or PD-L1-inhibitors to chemotherapy in patients with TNBC. In the KEYNOTE-522 trial, the addition of an immune checkpoint inhibitor (ICI) to neoadjuvant chemotherapy also had a beneficial impact on event-free survival even in patients who did not achieve a pCR. Of note, in the neoadjuvant GeparNuevo trial only those patients with TNBC who received a 2-week checkpoint inhibitor monotherapy window before the start of neoadjuvant chemotherapy in combination with checkpoint inhibition, achieved a significant pCR benefit from the addition of the PD-1 inhibitor Durvalumab to neoadjuvant chemotherapy alone. Methods: NeoMono is a phase 2 randomized multicenter trial recruiting male and female patients with primary TNBC (defined as ER/PR < 10% and HER2 negative). Neoadjuvant treatment in Arm A and B consists of Atezolizumab 1200 mg every 3 weeks in addition to neoadjuvant chemotherapy (i.e., 12 x Carboplatin and Paclitaxel q1w followed by Epirubicin and Cyclophosphamide q3w). Combination therapy in arm A is preceded by an Atezolizumab monotherapy window (i.e., 840 mg Atezolizumab once two weeks prior to initiation of combination therapy). Study goals are to compare the efficacy of neoadjuvant chemotherapy with Atezolizumab with and without an Atezolizumab two-week monotherapy window (primary endpoint: pCR) and the identification of biomarkers predicting (early) response to or resistance against Atezolizumab. The extensive translational program of the neoMono trial aims at identifying these biomarkers on tumor and patient level through analysis of sequential tissue and liquid biopsies. The NeoMono statistical design adapts the idea of a proof-of-concept trial and uses Bayesian posterior and predictive probabilities for inference about the primary hypothesis. Up to four planned efficacy interim analyses provide decision points for early stopping for success or futility. The expected maximum number of patients to be recruited is 458. Results: The predefined number of 50 patients in each arm being evaluable for the primary endpoint pCR has been reached and the results of the first planned interim analysis will be presented at the meeting. Conclusion: The addition of an ICI to state of the art neoadjuvant chemotherapy has recently been established as a new standard of care in TNBC. NeoMono has the potential to answer the question if the beneficial effect of the ICI can be increased by a chemotherapy free ICI monotherapy window prior to a combination with neoadjuvant chemotherapy. Citation Format: Hans-Christian Kolberg, Johannes Schumacher, Ramona Erber, Michael Braun, Bernhard Heinrich, Oliver Hoffmann, Peter A. Fasching, Georg Kunz, Michael P. Lux, Christian Schem, Eva-Maria Grischke, Mustafa Deryal, Kristina Lübbe, Arndt Hartmann, Sabine Kasimir-Bauer, Cornelia Kolberg-Liedtke. PD11-03 Comparison of a mono Atezolizumab window followed by Atezolizumab and chemotherapy with Atezolizumab and chemotherapy in triple negative breast cancer – an interim analysis of the adaptive randomized neoadjuvant trial NeoMono [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD11-03.

Over-accrual in Bayesian adaptive trials with continuous futility stopping.

We explore frequentist operating characteristics of a Bayesian adaptive design that allows continuous early stopping for futility. In particular, we focus on the power versus sample size relationship when more patients are accrued than originally planned. We consider the case of a phase II single-arm study and a Bayesian phase II outcome-adaptive randomization design. For the former, analytical calculations are possible; for the latter, simulations are conducted. Results for both cases show a decrease in power with an increasing sample size. It appears that this effect is due to the increasing cumulative probability of incorrectly stopping for futility. The increase in cumulative probability of incorrectly stopping for futility is related to the continuous nature of the early stopping, which increases the number of interim analyses with accrual. The issue can be addressed by, for instance, delaying the start of testing for futility, reducing the number of futility tests to be performed or by setting stricter criteria for concluding futility.

Challenges in Implementing Futility Schemes, with Reference to Aducanumab.

Stopping an ongoing clinical trial based on an interim analysis that shows poor outcomes, often referred to as a judgment of "futility", is a familiar feature in current clinical trials practice. Interim data can be misleading, and the implications of prematurely terminating a trial that should not stop are severe. It is thus critical that designs allowing futility stopping be planned and implemented carefully and cautiously. A recent Phase III development program for aducanumab in Alzheimer's disease was halted based on a pre-defined futility guideline, yet based upon updated data and closer examination, the terminated studies became the basis for a regulatory submission. Not surprisingly, this situation generated much controversy and discussion. It provides a good basis for illustrating important principles governing the planning and implementation of futility schemes.

Abstract TP11: Acute Stroke Randomized Clinical Trials: What Factors Help Trials Achieve Their Intended Sample Size?

Background: We performed a systematic review of acute stroke trials to determine features associated with premature termination of trial enrollment. Methods: MEDLINE was searched for stroke RCTs between 1/1/2013-6/11/2020. Included studies were Phase 2 or 3, had a preplanned sample size ≥100, had time to treatment within 24 hours of onset for TIA, ischemic stroke, or ICH, and were published in a major clinical journal (BMJ, Circulation, Stroke, Neurology, JAMA, JAMA Neurology, Lancet, Lancet Neurology, NEJM). Studies were deemed prematurely terminated if authors reported early termination. Study variables and inclusion criteria are listed in Table 1. Complex trials required specialized training and expertise, high degree of monitoring, or procedural intervention. Univariate Chi-square tests/Fisher’s exact test were used to determine characteristics associated with early termination, and multivariable lasso logistic regression was used to select the most important predictors. Results: Of 1,219 studies assessed, 69 trials met eligibility criteria for inclusion (Table 1); 34 (49%) were prematurely terminated. In univariate analysis, variables associated with premature termination were complexity of treatment, industry involvement, exclusion of older patients, and presence of a futility stopping rule. In lasso regression, complex treatment trials (OR 3.77, 95% CI 1.30-10.97, versus moderate or simple trials combined), presence of a futility rule (OR 3.56, 95% CI 1.02-12.40 versus no rule), and geographic region (North America OR 1.97, 95% CI 0.57-6.81 versus all other regions combined) were selected as important factors. Conclusion: Half of acute stroke trials were terminated prematurely. Broadening inclusion/exclusion criteria and simplifying the complexity of trials may decrease the likelihood of unplanned premature termination, whereas planned futility analyses may appropriately terminate trials early, saving money and resources.

A gated group sequential design for seamless Phase II/III trial with subpopulation selection

BackgroundDue to the high cost and high failure rate of Phase III trials where a classical group sequential design (GSD) is usually used, seamless Phase II/III designs are more and more popular to improve trial efficiency. A potential attraction of Phase II/III design is to allow a randomized proof-of-concept stage prior to committing to the full cost of a Phase III trial. Population selection during the trial allows a trial to adapt and focus investment where it is most likely to provide patient benefit. Previous methods have been developed for this problem when there is a single primary endpoint and two possible populations.MethodsTo find the population that potentially benefits with one or two primary endpoints (e.g., progression free survival (PFS), overall survival (OS)), we propose a gated group sequential design for a seamless Phase II/III trial design with adaptive population selection.ResultsThe investigated design controls the familywise error rate and allows multiple interim analyses to enable early stopping for efficacy or futility. Simulations and an illustrative example suggest that the proposed gated group sequential design has more power and requires less time and resources compared to the group sequential design and adaptive design.ConclusionsCombining the group sequential design and adaptive design, the gated group sequential design has more power and higher efficiency while controlling for the familywise error rate. It has the potential to save drug development cost and more quickly fulfill unmet medical needs.

Optimal predictive probability designs for randomized biomarker-guided oncology trials.

Efforts to develop biomarker-targeted anti-cancer therapies have progressed rapidly in recent years. With efforts to expedite regulatory reviews of promising therapies, several targeted cancer therapies have been granted accelerated approval on the basis of evidence acquired in single-arm phase II clinical trials. And yet, in the absence of randomization, patient prognosis for progression-free survival and overall survival may not have been studied under standard of care chemotherapies for emerging biomarker subpopulations prior to the submission of an accelerated approval application. Historical control rates used to design and evaluate emerging targeted therapies often arise as population averages, lacking specificity to the targeted genetic or immunophenotypic profile. Thus, historical trial results are inherently limited for inferring the potential "comparative efficacy" of novel targeted therapies. Consequently, randomization may be unavoidable in this setting. Innovations in design methodology are needed, however, to enable efficient implementation of randomized trials for agents that target biomarker subpopulations. This article proposes three randomized designs for early phase biomarker-guided oncology clinical trials. Each design utilizes the optimal efficiency predictive probability method to monitor multiple biomarker subpopulations for futility. Only designs with type I error between 0.05 and 0.1 and power of at least 0.8 were considered when selecting an optimal efficiency design from among the candidate designs formed by different combinations of posterior and predictive threshold. A simulation study motivated by the results reported in a recent clinical trial studying atezolizumab treatment in patients with locally advanced or metastatic urothelial carcinoma is used to evaluate the operating characteristics of the various designs. Out of a maximum of 300 total patients, we find that the enrichment design has an average total sample size under the null of 101.0 and a total average sample size under the alternative of 218.0, as compared to 144.8 and 213.8 under the null and alternative, respectively, for the stratified control arm design. The pooled control arm design enrolled a total of 113.2 patients under the null and 159.6 under the alternative, out of a maximum of 200. These average sample sizes that are 23-48% smaller under the alternative and 47-64% smaller under the null, as compared to the realized sample size of 310 patients in the phase II study of atezolizumab. Our findings suggest that potentially smaller phase II trials to those used in practice can be designed using randomization and futility stopping to efficiently obtain more information about both the treatment and control groups prior to phase III study.

The Frida Study: An Option for Mutated FLT3 Relapsed/Refractory Acute Myeloid Leukemia Patients with a Novel Iadademstat and Gilteritinib Combination Therapy

Background: Epigenetic dysregulation is a hallmark of acute myeloid leukemia (AML) and is caused by recurrent translocations and/or mutations in chromatin regulators and transcription factors, resulting in myeloid differentiation blockade and leukemic stem cell renewal. Accordingly, about 70% of recurring mutations in AML patients target regulators of gene expression, underscoring the potential of epigenetic therapies to change the natural history of the disease. Iadademstat (iada/ORY-1001) is a specific, oral, potent, covalent inhibitor of the epigenetic Lysine-Specific Demethylase 1 (LSD1/KDMA1) enzyme. Preclinical models demonstrate that exposure to iada decreases leukemic stem cell survival and preclinical and clinical studies shows iada's effect on inducing macrophage/monocytic differentiation of blast cells. Furthermore, a Phase 2 ongoing study of iada in combination with azacitidine (ALICE) shows high complete remission (CR)/ CR with incomplete count recovery (CRi) and durable responses in treatment naïve, unfit AML patients (pts) without exacerbating the toxicity profile of azacitidine (Salamero et al., Abstract P586 EHA 2022). Despite improvements in AML therapy, relapses remain a common challenge, contributing to the death of more than 50% of pts, particularly in those subpopulations with higher risk genetics, the largest of which (up to 30-40% of AML pts) harbor mutations in the fms-like tyrosine kinase 3 (FLT3) gene. Use of the FLT3 inhibitor (FLT3i) gilteritinib as monotherapy for relapsed/refractory (R/R) pts has resulted in improved outcomes but the duration of remission achieved is transient and often brief. The Phase 3 randomized ADMIRAL trial of gilteritinib revealed a CR rate of 20%; and an event-free-survival (EFS) of 2.8 months (Perl, et al., NEJM 2019). Preclinically, iada produces marked synergy with FLT3i, including gilteritinib, in FLT3 wild-type and mutated (FLT3 mut+) AML cells and in derived cell lines resistant to venetoclax, azacitidine and other FLT3is. The novel MOAs generating this synergy of the combination include activation of a pro-differentiative epigenetic and transcriptional program with simultaneous suppression of the activity of MYC target genes (Yashar, et al. 2022, https://doi.org/10.1101/2022.01.17.476469). The FRIDA clinical trial is designed to establish the safety, tolerability of the combination of iada plus gilteritinib and to identify the recommended phase 2 dose (RP2D) in FLT3 mut+ R/R AML specifically according to FDA's Project Optimus. Methods: Adult pts with body weight ≥50 Kg and ECOG 0-2, with FLT3 mut+ R/R AML, after 1 or 2 prior lines of therapy, will be enrolled. Up to 18 pts, in a 3+3 escalation phase, will receive iada at doses of 75 to 150 ug, orally, in 5 days ON - 2 days OFF schedule, with continuous gilteritinib, at 120 mg/day orally. In the expansion phase, up to 14 pts at the selected safe and pharmacologically active dose/s (determined based on all available data from escalation pts including PK, target engagement (TE), safety, tolerability, and emerging activity) will be enrolled. TE will be assessed with a proprietary validated chemo-probe assay for LSD1 occupancy in PBMCs. Primary endpoints of the study are safety and RP2D determination, based on the same 5 criteria outlined above. Bayesian posterior probability efficacy monitoring will be performed periodically for each dose cohort starting from the first 6 additional pts in the expansion phase (total of 12). Bayesian efficacy futility and early stopping boundary will be applied during the monitoring. Posterior probability criterion (Prob (θ>0.3) ≥ 0.60) at the end of the study will warrant additional development. The safety of the combination treatment will be also continuously evaluated during expansion following a Bayesian design stopping rule. Secondary endpoints include overall survival, EFS, CR, CR/CR with partial hematologic recovery (CRh), overall response rates, time to response, duration of response, and transfusion rate. Exploratory endpoints include measurable residual disease and gene mutational analysis. Pts who achieve a response can undergo hematopoietic stem cell transplant. The study plans to have 15 sites enrolling in the US. Additional sites will be added for a subsequent randomized controlled double-blinded FRIDA 2 study to assess the efficacy of the iada and gilteritinib combination in R/R FLT3 mut+ AML.

A Phase I Open Label Study of Fostamatinib, a SYK Inhibitor, in Patients with Lower-Risk Myelodysplastic Syndromes and Chronic Myelomonocytic Leukemia

INTRODUCTION: Ineffective hematopoiesis in myelodysplastic syndromes (MDS) and chronic myelomonocytic leukemia (CMML) have been associated to upregulated innate immune and inflammatory signals. Spleen tyrosine kinase (SYK) is a non-receptor tyrosine kinase that is highly expressed in hematopoietic cells and mediates signaling through Toll-like receptors and pattern recognition receptors leading to activation of the ERK, AKT, NF-kB and p38 signaling cascades. Given prior data suggesting that inhibition of these signals might be associated with improved hematopoiesis in MDS we hypothesized SYK inhibition might have therapeutic efficacy in lower-risk MDS and CMML. METHODS: To study this we designed an open label phase I clinical trial of fostamatinib, a SYK inhibitor, for patients with lower-risk MDS or CMML (defined as those with a score <3.5 by the revised International Prognostic Scoring System [IPSS-R]) and no response to prior erythroid stimulating agents or hypomethylating agents (HMAs). Patients with MDS with del(5q) must have received prior therapy with lenalidomide and those with MDS with ring sideroblasts must have received luspatercept. Patients had to be transfusion dependent, defined as prior RBC or platelet transfusions within the past 28 days, or have hemoglobin <10g/dL or platelets <75x109/L. Patients with profound neutropenia (ANC <0.5x109/L) where excluded. The study includes an intra-patient dose escalation design by which patients start therapy at a dose of fostamatinib of 100mg po bid on days 1-28 of a 28-day cycle. If no response to therapy, by IWG 2006 response criteria, is observed after 8 weeks of therapy dose will be escalated to 150mg po bid on days 1-28 of a 28-day cycle and then, in the absence of response after an additional 8 weeks of therapy, to 200mg po bid on days 1-28 of a 28 day cycle. Patients achieving no response after the second dose escalation will be discontinued from study (Figure 1a). The primary endpoint of the study is to assess the safety and tolerability of fostamatinib. Secondary objectives included overall response rates, transfusion independence and survival outcomes. The study includes futility and safety stopping rules. This study is registered at clinicaltrials.gov (NCT05030675). RESULTS: To date, a total of 5 patients have been treated on study as of July 2022 including 3 patients with MDS and ring sideroblasts. Patient characteristics are detailed in Figure 1b. Median age is 71 years (range 59-82). Median number of prior therapies was 5 (range 4-6) including one patient with a prior allogeneic stem-cell transplant. Four (80%) patients had received prior therapy with HMA. Two patients were transfusion dependent of RBC at the time of enrollment with one patient who was eligible based on the presence of thrombocytopenia. At the current data cut off, three patients developed treatment emergent adverse events the main including grade 1-2 diarrhea or constipation. One patient developed COVID-19 infection during the course of therapy and had mild symptomatology. Fostamatinib was not discontinued in this patient. No episodes of expected on-target uncontrolled hypertension or transaminitis have been observed so far. To date, only 4 patients have completed at least 2 cycles of therapy (8 weeks) and are eligible for response, and only 2 patients have dose escalated beyond 100mg po bid: 1 currently at the 150mg po bid dose level, and 1 at the maximum dose of 200mg po bid. With a median follow up of only 2.7 months (95% CI 0-5.6), no responses have been observed. All 5 patients remain on study: 3 (60%) at the 100mg po bid dose, 1 (20%) at 150mg po bid and 1 (20%) at 200mg po bid. CONCLUSIONS: Preliminary data suggests that treatment with fostamatinib at doses of up to 200mg po bid can be safely administered in patients with lower-risk MDS who have failed available therapies. Enrollment is ongoing to determine the potential efficacy of doses of up to 200mg bid. Additional correlative studies to determine predictors of response are underway. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal