Combining Multiple Comparisons and Modeling Techniques in Dose-Response Studies

Efficacy and safety of eliapixant in endometriosis-associated pelvic pain: the randomized, placebo-controlled phase 2b SCHUMANN study

BACKGROUND

The SCHUMANN study evaluated the efficacy and safety of the selective P2 × 3 antagonist eliapixant in patients with endometriosis-associated pelvic pain (EAPP).

METHODS

SCHUMANN was a randomized, placebo- and active comparator-controlled, double-blind to placebo and open-label to comparator, parallel-group, multicenter, dose-finding phase 2b study. The participants were women with surgically diagnosed endometriosis who fulfilled defined EAPP criteria. Participants were randomized 1:1:1:1 to twice daily (BID) 25 mg, 75 mg, or 150 mg oral eliapixant or a placebo for 12 weeks. An exploratory once-daily elagolix 150 mg treatment group was also included. The primary endpoint was the absolute change in mean worst EAPP from baseline to the end of intervention (EOI).

RESULTS

Overall, 215 participants were randomized for treatment (44 to eliapixant 25 mg, 44 to eliapixant 75 mg, 43 to eliapixant 150 mg, 43 to a placebo, and 41 to elagolix 150 mg). For safety reasons, the study was terminated early; both treatment and enrollment stopped immediately, producing less than 50% of the planned number of completers. The study found no significant differences in EAPP reduction from baseline between groups and no significant dose-response model. The elagolix 150 mg group showed better pain reduction than any of the other groups. No new safety signals were observed, relative to the previously known safety profile of eliapixant, which was generally well tolerated. However, one case of moderate and probably drug-induced liver injury in a participant receiving eliapixant 150 mg BID supported the association between eliapixant and a potential increase in liver function values, defined before the start of the phase 2 program.

CONCLUSIONS

This study did not meet its primary objective as no statistically significant or clinically relevant differences in changes of mean worst EAPP from baseline were observed between treatment groups. The single observed case of moderate, probably drug-induced liver injury was the second case in the eliapixant phase 2 program conducted in the following indications: refractory or unexplained chronic cough, diabetic neuropathic pain, overactive bladder, and EAPP. Due to this, the benefit-risk ratio for the study was no longer considered to be positive.

CLINICAL TRIAL REGISTRATION

ClinicalTrials.gov identifier NCT04614246; registered November 3, 2020.

Statistical and practical considerations in planning and conduct of dose-optimization trials

The U.S. Food and Drug Administration launched Project Optimus with the aim of shifting the paradigm of dose-finding and selection toward identifying the optimal biological dose that offers the best balance between benefit and risk, rather than the maximum tolerated dose. However, achieving dose optimization is a challenging task that involves a variety of factors and is considerably more complicated than identifying the maximum tolerated dose, both in terms of design and implementation. This article provides a comprehensive review of various design strategies for dose-optimization trials, including phase 1/2 and 2/3 designs, and highlights their respective advantages and disadvantages. In addition, practical considerations for selecting an appropriate design and planning and executing the trial are discussed. The article also presents freely available software tools that can be utilized for designing and implementing dose-optimization trials. The approaches and their implementation are illustrated through real-world examples.

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.

No benefit of soluble fiber on liver function

Background: To conduct a systematic review and dose-response meta-analysis of current findings from randomized controlled trials (RCTs) on the effect of soluble fiber supplementation on liver function in both healthy individuals and people with specific health conditions, PubMed, Scopus, and ISI Web of Science were systematically searched for relevant RCTs published prior to April 2022. Methods: We estimated the change in liver function parameters for each 5 g/d increment in soluble fiber in each trial and then calculated the mean difference (MD) and 95%CI. A total of 25 RCTs with 27 treatment arms (1744 subjects; 884 cases, 860 controls) were included. Results: A total of 25 RCTs with 27 treatment arms were included. The intervention duration of the included studies ranged from 3 to 52 weeks and the dose of soluble fiber supplementation varied from 0.0025 to 40 g/d. Soluble fiber supplementation could not significantly affect serum alanine transaminase (MD: -0.02 U/L, 95% CI: -1.06 to 1.01), aspartate transaminase (MD: -0.34 U/L, 95% CI: -0.84 to 0.15), alkaline phosphatase (MD: 0.29 U/L, -0.14 to 0.71), gamma-glutamyl transferase (MD: 0.12 U/L; 95% CI: -0.81 to 1.05), serum bilirubin (MD: 0.42μmol/L, 95% CI: -0.08 to 0.93) and albumin (MD: 0.64 g/dl, 95% CI: -0.42 to 1.70) levels. Conclusions: Findings from this study did not support the beneficial effects of soluble fiber supplementation on liver function biomarkers. There is a need for long-term high-quality interventions to examine the effects of different types and doses of soluble fibers on liver function as primary outcome.

Statistical considerations in model-based dose finding for binary responses under model uncertainty

The statistical methodology for model-based dose finding under model uncertainty has attracted increasing attention in recent years. While the underlying principles are simple and easy to understand, developing and implementing an efficient approach for binary responses can be a formidable task in practice. Motivated by the statistical challenges encountered in a phase II dose finding study, we explore several key design and analysis issues related to the hybrid testing-modeling approaches for binary responses. The issues include candidate model selection and specifications, optimal design and efficient sample size allocations, and, notably, the methods for dose-response testing and estimation. Specifically, we consider a class of generalized linear models suited for the candidate set and establish D-optimal designs for these models. Additionally, we propose using permutation-based tests for dose-response testing to avoid asymptotic normality assumptions typically required for contrast-based tests. We perform trial simulations to enhance our understanding of these issues.

A class of computational methods to reduce selection bias when designing Phase 3 clinical trials

When designing confirmatory Phase 3 studies, one usually evaluates one or more efficacious and safe treatment option(s) based on data from previous studies. However, several retrospective research articles reported the phenomenon of "diminished treatment effect in Phase 3" based on many case studies. Even under basic assumptions, it was shown that the commonly used estimator could substantially overestimate the efficacy of selected group(s). As alternatives, we propose a class of computational methods to reduce estimation bias and mean squared error with a broader scope of multiple treatment groups and flexibility to accommodate summary results by group as input. Based on simulation studies and a real data example, we provide practical implementation guidance for this class of methods under different scenarios. For more complicated problems, our framework can serve as a starting point with additional layers built in. Proposed methods can also be widely applied to other selection problems.

Bayesian model averaging of longitudinal dose-response models

Selecting a safe and clinically beneficial dose can be difficult in drug development. Dose justification often relies on dose-response modeling where parametric assumptions are made in advance which may not adequately fit the data. This is especially problematic in longitudinal dose-response models, where additional parametric assumptions must be made. This paper proposes a class of longitudinal dose-response models to be used in the Bayesian model averaging paradigm which improve trial operating characteristics while maintaining flexibility a priori. A new longitudinal model for non-monotonic longitudinal profiles is proposed. The benefits and trade-offs of the proposed approach are demonstrated through a case study and simulation.

The effects of coenzyme Q10 supplementation on biomarkers of exercise-induced muscle damage, physical performance, and oxidative stress: A GRADE-assessed systematic review and dose-response meta-analysis of randomized controlled trials

PURPOSE

This study aims to elucidate the dose-dependent effect of coenzyme Q10 supplementation (CoQ10) on exercise-induced muscle damage (EIMD), physical performance, and oxidative stress in adults.

METHODS

A systematic search was conducted through PubMed, Scopus, and ISI Web of Science databases up to August 2023, focusing on randomized control trials (RCTs) that investigated the effects of CoQ10 supplementation on EIMD recovery, physical performance and oxidative stress mitigation in adults. The weighted mean difference (WMD) and 95 % confidence interval (95 %CI) were estimated using the random-effects model.

RESULTS

The meta-analysis incorporated 28 RCTs, encompassing 830 subjects. CoQ10 supplementation significantly decreased creatine kinase (CK) (WMD: -50.64 IU/L; 95 %CI: -74.75, -26.53, P < 0.001), lactate dehydrogenase (LDH) (WMD: -52.10 IU/L; 95 %CI: -74.01, -30.19, P < 0.001), myoglobin (Mb) (WMD: -21.77 ng/ml; 95 %CI: -32.59, -10.94, P < 0.001), and Malondialdehyde (MDA) (WMD: -0.73 μmol/l; 95 %CI: -1.26, -0.20, P = 0.007) levels. No significant alteration in total antioxidant capacity was observed post-CoQ10 treatment. Each 100 mg/day increase in CoQ10 supplementation was correlated with a significant reduction in CK (MD: -23.07 IU/L, 95 %CI: -34.27, -11.86), LDH (WMD: -27.21 IU/L, 95 %CI: -28.23, -14.32), Mb (MD: -7.09 ng/ml; 95 %CI: -11.35, -2.83) and MDA (WMD: -0.17 μmol/l, 95 %CI: -0.29, -0.05) serum levels. Using SMD analysis, "very large" effects on LDH and "moderate" effects on CK and MDA were noted, albeit nonsignificant for other outcomes.

CONCLUSION

CoQ10 supplementation may be effective in reducing biomarkers of EIMD and oxidative stress in adults. Nevertheless, given the preponderance of studies conducted in Asia, the generalizability of these findings warrants caution. Further RCTs, particularly in non-Asian populations with large sample sizes and extended supplementation durations, are essential to substantiate these observations.

Upadacitinib for the treatment of moderate-to-severe Crohn's disease

Despite an increasing number of therapies for Crohn's disease (CD), half of patients do not respond to initial treatment or lose response over time, highlighting the need for novel therapies. Inhibition of Janus kinases (JAKs) has emerged as an important therapeutic target for CD. Upadacitinib is an orally administered selective JAK1 inhibitor, which is effective for the induction and maintenance of remission in moderately-to-severely active CD, including in patients with prior failure of biological therapy. Nonselective JAK inhibition has been associated with thromboembolic disease, cardiovascular events and malignancy in patients older than 50 years with rheumatoid arthritis and pre-existing cardiovascular risk factors, which should be considered upon prescription. Upadacitinib is the first and currently only oral advanced therapy for CD.

Combination MCP-Mod for two-drug combination dose-ranging studies

Combination therapies with multiple mechanisms of action can offer improved efficacy and/or safety profiles when compared to a single therapy with one mechanism of action. Consequently, the number of combination therapy studies have increased multi-fold, both in oncology and non-oncology indications. However, identifying the optimal doses of each drug in a combination therapy can require a large sample size and prolong study timelines, especially when full factorial designs are used. In this paper, we extend the MCP-Mod design of Bretz, Pinheiro, and Branson to a three-dimensional space to model the dose-response surface of a two-drug combination under the framework of Combination (Comb) MCP-Mod. The resulting model yields a set of dosages for each drug in the combination that elicits the target response so that an optimal dose for the combination can be selected for pivotal studies. We construct three-dimensional dose-response models for the combination and formulate the contrast test statistic to select the best model, which can then be used to select the optimal dose. Guidance to calculate power and sample size calculations are provided to assist study design. Simulation studies show that Comb MCP-Mod performs as well as the conventional multiple comparisons approach in controlling the family-wise error rate at the desired alpha level. However, Comb MCP-Mod is more powerful than the classical multiple comparisons approach in detecting dose-response relationships when treatment is non-null. The probability of correctly identifying the underlying dose-response relationship is generally higher when using Comb MCP-Mod than when using the multiple comparisons approach.

Results of a randomized double-blind placebo-controlled Phase 2 study propose iptacopan as an alternative complement pathway inhibitor for IgA nephropathy

Targeting the alternative complement pathway is an attractive therapeutic strategy given its role in the pathogenesis of immunoglobulin A nephropathy (IgAN). Iptacopan (LNP023) is an oral, proximal alternative complement inhibitor that specifically binds to Factor B. Our randomized, double-blind, parallel-group adaptive Phase 2 study (NCT03373461) enrolled patients with biopsy-confirmed IgAN (within previous three years) with estimated glomerular filtration rates of 30 mL/min/1.73 m2 and over and urine protein 0.75 g/24 hours and over on stable doses of renin angiotensin system inhibitors. Patients were randomized to four iptacopan doses (10, 50, 100, or 200 mg bid) or placebo for either a three-month (Part 1; 46 patients) or a six-month (Part 2; 66 patients) treatment period. The primary analysis evaluated the dose-response relationship of iptacopan versus placebo on 24-hour urine protein-to-creatinine ratio (UPCR) at three months. Other efficacy, safety and biomarker parameters were assessed. Baseline characteristics were generally well-balanced across treatment arms. There was a statistically significant dose-response effect, with 23% reduction in UPCR achieved with iptacopan 200 mg bid (80% confidence interval 8-34%) at three months. UPCR decreased further through six months in iptacopan 100 and 200 mg arms (from a mean of 1.3 g/g at baseline to 0.8 g/g at six months in the 200 mg arm). A sustained reduction in complement biomarker levels including plasma Bb, serum Wieslab, and urinary C5b-9 was observed. Iptacopan was well-tolerated, with no reports of deaths, treatment-related serious adverse events or bacterial infections, and led to strong inhibition of alternative complement pathway activity and persistent proteinuria reduction in patients with IgAN. Thus, our findings support further evaluation of iptacopan in the ongoing Phase 3 trial (APPLAUSE-IgAN; NCT04578834).

Interim decision making in seamless trial designs: An application in an adaptive dose-finding study in a rare kidney disease

Adaptive seamless trial designs, combining the learning and confirming cycles of drug development in a single trial, have gained popularity in recent years. Adaptations may include dose selection, sample size re-estimation and enrichment of the study population. Despite methodological advances and recognition of the potential efficiency gains such designs offer, their implementation, including how to enable efficient decision making on the adaptations in interim analyzes, remains a key challenge in their adoption. This manuscript uses a case study of an adaptive seamless proof-of-concept (Phase 2a)/dose-finding (Phase 2b) to showcase potential adaptive features that can be implemented in trial designs at earlier development stages and the role of simulations in assessing the design operating characteristics and specifying the decision rules for the adaptations. It further outlines the elements needed to support successful interim analysis decision making on the adaptations while safeguarding study integrity, including the role of different stakeholders, interactive simulation-based tools to facilitate decision making and operational aspects requiring preplanning. The benefits of the adaptive Phase 2a/2b design chosen compared to following the traditional two separate studies (2a and 2b) paradigm are discussed. With careful planning and appreciation of their complexity and components needed for their implementation, seamless adaptive designs have the potential to yield significant savings both in terms of time and resources.

Sample size calculation for multi-arm parallel design with restricted mean survival time

With the recent advances in oncology treatment, restricted mean survival time (RMST) is increasingly being used to replace the routine approach based on hazard ratios in randomized controlled trials for time-to-event outcomes. While RMST has been widely applied in single-arm and two-arm designs, challenges still exist in comparing RMST in multi-arm trials with three or more groups. In particular, it is unclear in the literature how to compare more than one intervention simultaneously or perform multiple testing based on RMST, and sample size determination is a major obstacle to its penetration to practice. In this paper, we propose a novel method of designing multi-arm clinical trials with right-censored survival endpoint based on RMST that can be applied in both phase II/III settings using a global χ 2 test as well as a modeling-based multiple comparison procedure. The framework provides a closed-form sample size formula built upon a multi-arm global test and a sample size determination procedure based on multiple-comparison in the phase II dose-finding study. The proposed method enjoys strong robustness and flexibility as it requires less a priori set-up than conventional work, and obtains a smaller sample size while achieving the target power. In the assessment of sample size, we also incorporate practical considerations, including the presence of non-proportional hazards and staggered patient entry. We evaluate the validity of our method through simulation studies under various scenarios. Finally, we demonstrate the accuracy and stability of our method by implementing it in the design of two real clinical trial examples.

Safety and efficacy of factor XIa inhibition with milvexian for secondary stroke prevention (AXIOMATIC-SSP): a phase 2, international, randomised, double-blind, placebo-controlled, dose-finding trial

BACKGROUND

People with factor XI deficiency have lower rates of ischaemic stroke than the general population and infrequent spontaneous bleeding, suggesting that factor XI has a more important role in thrombosis than in haemostasis. Milvexian, an oral small-molecule inhibitor of activated factor XI, added to standard antiplatelet therapy, might reduce the risk of non-cardioembolic ischaemic stroke without increasing the risk of bleeding. We aimed to estimate the dose-response of milvexian for recurrent ischaemic cerebral events and major bleeding in patients with recent ischaemic stroke or transient ischaemic attack (TIA).

METHODS

AXIOMATIC-SSP was a phase 2, randomised, double-blind, placebo-controlled, dose-finding trial done at 367 hospitals in 27 countries. Eligible participants aged 40 years or older, with acute (<48 h) ischaemic stroke or high-risk TIA, were randomly assigned by a web-based interactive response system in a 1:1:1:1:1:2 ratio to receive one of five doses of milvexian (25 mg once daily, 25 mg twice daily, 50 mg twice daily, 100 mg twice daily, or 200 mg twice daily) or matching placebo twice daily for 90 days. All participants received clopidogrel 75 mg daily for the first 21 days and aspirin 100 mg daily for the first 90 days. Investigators, site staff, and participants were masked to treatment assignment. The primary efficacy endpoint was the composite of ischaemic stroke or incident covert brain infarct on MRI at 90 days, assessed in all participants allocated to treatment who completed a follow-up MRI brain scan, and the primary analysis assessed the dose-response relationship with Multiple Comparison Procedure-Modelling (MCP-MOD). The main safety outcome was major bleeding at 90 days, assessed in all participants who received at least one dose of the study drug. This trial is registered with ClinicalTrials.gov (NCT03766581) and the EU Clinical Trials Register (2017-005029-19).

FINDINGS

Between Jan 27, 2019, and Dec 24, 2021, 2366 participants were randomly allocated to placebo (n=691); milvexian 25 mg once daily (n=328); or twice-daily doses of milvexian 25 mg (n=318), 50 mg (n=328), 100 mg (n=310), or 200 mg (n=351). The median age of participants was 71 (IQR 62-77) years and 859 (36%) were female. At 90 days, the estimates of the percentage of participants with either symptomatic ischaemic stroke or covert brain infarcts were 16·8 (90·2% CI 14·5-19·1) for placebo, 16·7 (14·8-18·6) for 25 mg milvexian once daily, 16·6 (14·8-18·3) for 25 mg twice daily, 15·6 (13·9-17·5) for 50 mg twice daily, 15·4 (13·4-17·6) for 100 mg twice daily, and 15·3 (12·8-19·7) for 200 mg twice daily. No significant dose-response was observed among the five milvexian doses for the primary composite efficacy outcome. Model-based estimates of the relative risk with milvexian compared with placebo were 0·99 (90·2% CI 0·91-1·05) for 25 mg once daily, 0·99 (0·87-1·11) for 25 mg twice daily, 0·93 (0·78-1·11) for 50 mg twice daily, 0·92 (0·75-1·13) for 100 mg twice daily, and 0·91 (0·72-1·26) for 200 mg twice daily. No apparent dose-response was observed for major bleeding (four [1%] of 682 participants with placebo, two [1%] of 325 with milvexian 25 mg once daily, two [1%] of 313 with 25 mg twice daily, five [2%] of 325 with 50 mg twice daily, five [2%] of 306 with 100 mg twice daily, and five [1%] of 344 with 200 mg twice daily). Five treatment-emergent deaths occurred, four of which were considered unrelated to the study drug by the investigator.

INTERPRETATION

Factor XIa inhibition with milvexian, added to dual antiplatelet therapy, did not substantially reduce the composite outcome of symptomatic ischaemic stroke or covert brain infarction and did not meaningfully increase the risk of major bleeding. Findings from our study have informed the design of a phase 3 trial of milvexian for the prevention of ischaemic stroke in patients with acute ischaemic stroke or TIA.

FUNDING

Bristol Myers Squibb and Janssen Research & Development.

How to analyze continuous and discrete repeated measures in small-sample cross-over trials?

To optimize the use of data from a small number of subjects in rare disease trials, an at first sight advantageous design is the repeated measures cross-over design. However, it is unclear how these within-treatment period and within-subject clustered data are best analyzed in small-sample trials. In a real-data simulation study based upon a recent epidermolysis bullosa simplex trial using this design, we compare non-parametric marginal models, generalized pairwise comparison models, GEE-type models and parametric model averaging for both repeated binary and count data. The recommendation of which methodology to use in rare disease trials with a repeated measures cross-over design depends on the type of outcome and the number of time points the treatment has an effect on. The non-parametric marginal model testing the treatment-time-interaction effect is suitable for detecting between group differences in the shapes of the longitudinal profiles. For binary outcomes with the treatment effect on a single time point, the parametric model averaging method is recommended, while in the other cases the unmatched generalized pairwise comparison methodology is recommended. Both provide an easily interpretable effect size measure, and do not require exclusion of periods or subjects due to incompleteness.

DROID: dose-ranging approach to optimizing dose in oncology drug development

In the era of targeted therapy, there has been increasing concern about the development of oncology drugs based on the "more is better" paradigm, developed decades ago for chemotherapy. Recently, the US Food and Drug Administration (FDA) initiated Project Optimus to reform the dose optimization and dose selection paradigm in oncology drug development. To accommodate this paradigm shifting, we propose a dose-ranging approach to optimizing dose (DROID) for oncology trials with targeted drugs. DROID leverages the well-established dose-ranging study framework, which has been routinely used to develop non-oncology drugs for decades, and bridges it with established oncology dose-finding designs to optimize the dose of oncology drugs. DROID consists of two seamlessly connected stages. In the first stage, patients are sequentially enrolled and adaptively assigned to investigational doses to establish the therapeutic dose range (TDR), defined as the range of doses with acceptable toxicity and efficacy profiles, and the recommended phase 2 dose set (RP2S). In the second stage, patients are randomized to the doses in RP2S to assess the dose-response relationship and identify the optimal dose. The simulation study shows that DROID substantially outperforms the conventional approach, providing a new paradigm to efficiently optimize the dose of targeted oncology drugs. DROID aligns with the approach of a randomized, parallel dose-response trial design recommended by the FDA in the Guidance on Optimizing the Dosage of Human Prescription Drugs and Biological Products for the Treatment of Oncologic Diseases.

Calculation of Phase 2 dose-finding study sample size for reliable Phase 3 dose selection

Sample sizes of Phase 2 dose-finding studies, usually determined based on a power requirement to detect a significant dose-response relationship, will generally not provide adequate precision for Phase 3 target dose selection. We propose to calculate the sample size of a dose-finding study based on the probability of successfully identifying the target dose within an acceptable range (e.g., 80%-120% of the target) using the multiple comparison and modeling procedure (MCP-Mod). With the proposed approach, different design options for the Phase 2 dose-finding study can also be compared. Due to inherent uncertainty around an assumed true dose-response relationship, sensitivity analyses to assess the robustness of the sample size calculations to deviations from modeling assumptions are recommended. Planning for a hypothetical Phase 2 dose-finding study is used to illustrate the main points. Codes for the proposed approach is available at https://github.com/happysundae/posMCPMod.

Designs for the simultaneous inference of concentration-response curves

BACKGROUND

An important problem in toxicology in the context of gene expression data is the simultaneous inference of a large number of concentration-response relationships. The quality of the inference substantially depends on the choice of design of the experiments, in particular, on the set of different concentrations, at which observations are taken for the different genes under consideration. As this set has to be the same for all genes, the efficient planning of such experiments is very challenging. We address this problem by determining efficient designs for the simultaneous inference of a large number of concentration-response models. For that purpose, we both construct a D-optimality criterion for simultaneous inference and a K-means procedure which clusters the support points of the locally D-optimal designs of the individual models.

RESULTS

We show that a planning of experiments that addresses the simultaneous inference of a large number of concentration-response relationships yields a substantially more accurate statistical analysis. In particular, we compare the performance of the constructed designs to the ones of other commonly used designs in terms of D-efficiencies and in terms of the quality of the resulting model fits using a real data example dealing with valproic acid. For the quality comparison we perform an extensive simulation study.

CONCLUSIONS

The design maximizing the D-optimality criterion for simultaneous inference improves the inference of the different concentration-response relationships substantially. The design based on the K-means procedure also performs well, whereas a log-equidistant design, which was also included in the analysis, performs poorly in terms of the quality of the simultaneous inference. Based on our findings, the D-optimal design for simultaneous inference should be used for upcoming analyses dealing with high-dimensional gene expression data.

Effect of soluble fiber on blood pressure in adults: a systematic review and dose-response meta-analysis of randomized controlled trials

OBJECTIVES

Pervious epidemiologic evidence indicates that soluble fiber is protective against hypertention: however, randomized controlled trials (RCTs) have presented varying results. In the present study, we aimed to conduct a systematic review and dose-response meta-analysis to summarize published RCTs which assess the effect of soluble fiber supplementation on systolic blood pressure (SBP) and diastolic blood pressure (DBP).

METHODS

Scopus, PubMed, and ISI Web of Sciences were searched to identify relevant studies up to Aug 2022. We estimated the change in blood pressure for each 5 g/d increment in soluble fiber supplementation in each trial and then calculated the weighted mean difference (WMD) and 95%CI using a random-effects model. We estimated dose-dependent effects using a dose-response meta-analysis of differences in means. The risk of bias for study was assessed using the Cochrane tool. Publication bias was evaluated via funnel plot and Begg's test and Egger's test.

RESULTS

Eighty-three eligible studies with total sample size of 5,985 participants were included in the meta-analysis. Soluble fiber supplementation significantly decreased SBP (WMD: -1.36 mmHg, 95% CI: -2.13 to -0.60, P < 0.001; I2 = 47.1%, P < 0.001) and DBP (WMD: -0.72 mmHg, 95% CI: -1.26 to -0.18, P = 0.009; I2 = 45.4%, P < 0.001). Each 5 g/d increment in soluble fiber supplementation had a significant reduction in SBP (WMD: -0.54 mmHg; 95%CI: -0.86, -0.22, P = 0.001; I2 = 52.2, Phet < 0.001) and DBP (WMD: -0.28 mmHg; 95%CI: -0.49, -0.80, P = 0.007; I2 = 43.1%, Phet < 0.001). The levels of SBP decreased proportionally with the increase in soluble fiber supplementation up to 20 g/d (MD20g/d: -1.79 mmHg, 95%CI: -2.86, -0.71).

CONCLUSION

Current evidence indicated the beneficial effect of soluble fiber supplementation on blood pressure. Our findings suggest that soluble fiber supplementation could contribute to the management of hypertension and the reduction of cardiovascular disease risk.

Guidance for statistical design and analysis of toxicological dose-response experiments, based on a comprehensive literature review

The analysis of dose-response, concentration-response, and time-response relationships is a central component of toxicological research. A major decision with respect to the statistical analysis is whether to consider only the actually measured concentrations or to assume an underlying (parametric) model that allows extrapolation. Recent research suggests the application of modelling approaches for various types of toxicological assays. However, there is a discrepancy between the state of the art in statistical methodological research and published analyses in the toxicological literature. The extent of this gap is quantified in this work using an extensive literature review that considered all dose-response analyses published in three major toxicological journals in 2021. The aspects of the review include biological considerations (type of assay and of exposure), statistical design considerations (number of measured conditions, design, and sample sizes), and statistical analysis considerations (display, analysis goal, statistical testing or modelling method, and alert concentration). Based on the results of this review and the critical assessment of three selected issues in the context of statistical research, concrete guidance for planning, execution, and analysis of dose-response studies from a statistical viewpoint is proposed.

Spesolimab Efficacy and Safety in Patients with Moderate-to-Severe Palmoplantar Pustulosis: A Multicentre, Double-Blind, Randomised, Placebo-Controlled, Phase IIb, Dose-Finding Study

INTRODUCTION

We evaluated the anti-interleukin-36 receptor antibody spesolimab in patients with moderate-to-severe palmoplantar pustulosis (PPP).

METHODS

This phase IIb trial comprised a loading dose period to week (W) 4, then maintenance dosing to W52. Patients were randomised 2:1:1:1:2 to subcutaneous spesolimab 3000 mg to W4 then 600 mg every 4 weeks (q4w), spesolimab 3000 mg to W4 then 300 mg q4w, spesolimab 1500 mg to W4 then 600 mg q4w, spesolimab 1500 mg to W4, 300 mg q4w to W16 then 300 mg every 8 weeks (q8w), or placebo switching to spesolimab 600 mg q4w at W16. The primary efficacy endpoint was percentage change from baseline in Palmoplantar Pustular Area and Severity Index (PPP ASI) at W16. Secondary endpoints included a Palmoplantar Pustular Physician's Global Assessment (PPP PGA) score of 0/1. Safety (including adverse events [AEs], local tolerability) was assessed.

RESULTS

152 patients were treated. The primary endpoint was not met; mean differences for spesolimab versus placebo ranged from - 14.6% (95% confidence interval [CI]: - 31.5%, 2.2%) to - 5.3% (95% CI: - 19.1%, 8.6%); none reached significance. At W16, 23 (21.1%) and two (4.7%) patients in the combined spesolimab and placebo groups, respectively, achieved PPP PGA 0/1 (mean difference 16.4%; 95% CI: 3.8%, 25.7%), increasing to 59 (54.1%; combined spesolimab) and 12 (27.9%; placebo switch to spesolimab) patients at W52. Non-Asian patients had significant improvements in the primary endpoint (mean difference - 17.7%; nominal P = 0.0394) and PPP PGA 0/1 at W16 with spesolimab versus placebo. Rates of AEs and AE-related discontinuations were similar for spesolimab and placebo. Local tolerability events and injection-site reactions were more frequent with spesolimab than placebo.

CONCLUSION

The primary objective to demonstrate a non-flat dose-response relationship and proof-of-concept was not achieved; improvements with spesolimab occurred in secondary endpoints and in non-Asian patients, indicating potential modest benefits. Spesolimab was generally well tolerated (ClinicalTrials.gov NCT04015518).

Bayesian hierarchical model for dose-finding trial incorporating historical data

The Multiple Comparison Procedure and Modelling (MCPMod) approach has been shown to be a powerful statistical technique that can significantly improve the design and analysis of dose-finding studies under model uncertainty. Due to its frequentist nature, however, it is difficult to incorporate information into MCPMod from historical trials on the same drug. BMCPMod, a recently introduced Bayesian version of MCPMod, is designed to take into account historical information on the placebo dose group. We introduce a Bayesian hierarchical framework capable of incorporating historical information on an arbitrary number of dose groups, including both placebo and active ones, taking into account the relationship between responses of these dose groups. Our approach can also model both prognostic and predictive between-trial heterogeneity and is particularly useful in situations where the effect sizes of two trials are different. Our goal is to reduce the necessary sample size in the dose-finding trial while maintaining its target power.

Seamless phase II/III design: a useful strategy to reduce the sample size for dose optimization

BACKGROUND

The traditional more-is-better dose selection paradigm, originally developed for cytotoxic chemotherapeutics, can be problematic when applied to the development of novel molecularly targeted agents. Recognizing this issue, the US Food and Drug Administration initiated Project Optimus to reform the dose optimization and selection paradigm in oncology drug development, emphasizing the need for greater attention to benefit-risk considerations.

METHODS

We identify different types of phase II/III dose-optimization designs, classified according to trial objectives and endpoint types. Through computer simulations, we examine their operating characteristics and discuss the relevant statistical and design considerations for effective dose optimization.

RESULTS

Phase II/III dose-optimization designs are capable of controlling family-wise type I error rates and achieving appropriate statistical power with substantially smaller sample sizes than the conventional approach while also reducing the number of patients who experience toxicity. Depending on the design and scenario, the sample size savings range from 16.6% to 27.3%, with a mean savings of 22.1%.

CONCLUSIONS

Phase II/III dose-optimization designs offer an efficient way to reduce sample sizes for dose optimization and accelerate the development of targeted agents. However, because of interim dose selection, the phase II/III dose-optimization design presents logistical and operational challenges and requires careful planning and implementation to ensure trial integrity.

Identifying alert concentrations using a model-based bootstrap approach

The determination of alert concentrations, where a pre-specified threshold of the response variable is exceeded, is an important goal of concentration-response studies. The traditional approach is based on investigating the measured concentrations and attaining statistical significance of the alert concentration by using a multiple t-test procedure. In this paper, we propose a new model-based method to identify alert concentrations, based on fitting a concentration-response curve and constructing a simultaneous confidence band for the difference of the response of a concentration compared to the control. In order to obtain these confidence bands, we use a bootstrap approach which can be applied to any functional form of the concentration-response curve. This particularly offers the possibility to investigate also those situations where the concentration-response relationship is not monotone and, moreover, to detect alerts at concentrations which were not measured during the study, providing a highly flexible framework for the problem at hand.

Improved inference for MCP-Mod approach using time-to-event endpoints with small sample sizes

The Multiple Comparison Procedures with Modeling Techniques (MCP-Mod) framework has been recently approved by the U.S. Food, Administration, and European Medicines Agency as fit-for-purpose for phase II studies. Nonetheless, this approach relies on the asymptotic properties of Maximum Likelihood (ML) estimators, which might not be reasonable for small sample sizes. In this paper, we derived improved ML estimators and correction for their covariance matrices in the censored Weibull regression model based on the corrective and preventive approaches. We performed two simulation studies to evaluate ML and improved ML estimators with their covariance matrices in (i) a regression framework (ii) the Multiple Comparison Procedures with Modeling Techniques framework. We have shown that improved ML estimators are less biased than ML estimators yielding Wald-type statistics that controls type I error without loss of power in both frameworks. Therefore, we recommend the use of improved ML estimators in the MCP-Mod approach to control type I error at nominal value for sample sizes ranging from 5 to 25 subjects per dose.

A weighted quantile sum regression with penalized weights and two indices

Background

New statistical methodologies were developed in the last decade to face the challenges of estimating the effects of exposure to multiple chemicals. Weighted Quantile Sum (WQS) regression is a recent statistical method that allows estimating a mixture effect associated with a specific health effect and identifying the components that characterize the mixture effect.

Objectives

In this study, we propose an extension of WQS regression that estimates two mixture effects of chemicals on a health outcome in the same model through the inclusion of two indices, one in the positive direction and one in the negative direction, with the introduction of a penalization term.

Methods

To evaluate the performance of this new model we performed both a simulation study and a real case study where we assessed the effects of nutrients on obesity among adults using the National Health and Nutrition Examination Survey (NHANES) data.

Results

The method showed good performance in estimating both the regression parameter and the weights associated with the single elements when the penalized term was set equal to the magnitude of the Akaike information criterion of the unpenalized WQS regression. The two indices further helped to give a better estimate of the parameters [Positive direction Median Error (PME): 0.022; Negative direction Median Error (NME): -0.044] compared to the standard WQS without the penalization term (PME: -0.227; NME: 0.215). In the case study, WQS with two indices was able to find a significant effect of nutrients on obesity in both directions identifying sodium and magnesium as the main actors in the positive and negative association, respectively.

Discussion

Through this work, we introduced an extension of WQS regression that improved the accuracy of the parameter estimates when considering a mixture of elements that can have both a protective and a harmful effect on the outcome; and the advantage of adding a penalization term when estimating the weights.

Robust and Adaptive Two-stage Designs in Nonlinear Mixed Effect Models

To get informative studies for nonlinear mixed effect models (NLMEM), design optimization can be performed based on Fisher Information Matrix (FIM) using the D-criterion. Its computation requires knowledge about models and parameters, which are often prior guesses. Thus, adaptive designs composed of several stages may be used. Robust approach can also be used to account for various candidate models. In the estimation step of a given stage, model selection (MS) or model averaging (MA) can be performed. In this work we propose a new two-stage adaptive design strategy, based on the robust expected FIM and MA over several candidate models. The methodology is applied to a clinical trial simulation in ophthalmology to optimize doses and time measurements. A set of dose-response candidate models is defined, and one-stage designs are compared to two-stage 50/50 designs (i.e., each stage performed with half of the available subjects), using either local optimal design or robust design, and performing analysis with one model, MS or MA. Performing a two-stage design with MS at the interim analysis can correct the choice of a wrong model for designing the first stage. Overall, starting from a robust design (1- or 2-stage) is valuable and leads to reasonable bias and precision. The proposed robust adaptive design strategy is a new tool to design longitudinal studies that could be used in different therapeutic areas.

Efficacy and Safety of Eliapixant in Refractory Chronic Cough: The Randomized, Placebo-Controlled Phase 2b PAGANINI Study

INTRODUCTION

The PAGANINI study evaluated the efficacy and safety of the selective P2X3 antagonist eliapixant in patients with refractory chronic cough (RCC).

METHODS

PAGANINI was a randomized, double-blind, parallel-group, placebo-controlled, multicenter, dose-finding, phase 2b study. Adults with RCC lasting ≥ 12 months and cough severity ≥ 40 mm on a visual analog scale at screening were enrolled. Participants were randomized 1:1:1:1 to twice-daily 25 mg, 75 mg, or 150 mg oral eliapixant or placebo for 12 weeks. The primary endpoint was change from baseline in 24-h cough count after 12 weeks of intervention.

RESULTS

Overall, 310 participants were randomized to twice-daily eliapixant 25 mg (n = 75), 75 mg (n = 78), 150 mg (n = 80), or placebo (n = 77). A statistically significant dose-response signal with eliapixant was detected for the primary endpoint (all dose-response models, adjusted p < 0.1; one-sided). Adverse events (AEs) were reported in 39 (51%) participants with placebo and 43-51 (57-65%) participants receiving eliapixant. The most common AE was dysgeusia, occurring in 1% (n = 1) of the placebo group and 1-16% (n = 1-13) of the eliapixant groups in a dose-related manner. One case of a moderate drug-induced liver injury occurred in a participant receiving 150 mg twice-daily eliapixant.

CONCLUSION

Eliapixant demonstrated efficacy and a favorable taste tolerability profile in RCC. However, a drug-induced liver injury contributed to intensified liver monitoring in clinical trials with eliapixant and discontinuation of the entire development program in all indications by Bayer AG.

TRIAL REGISTRATION

ClinicalTrials.gov identifier NCT04562155; registered September 18, 2020.

Sample size re-estimation in Phase 2 dose-finding: Conditional power versus Bayesian predictive power

Unblinded sample size re-estimation (SSR) is often planned in a clinical trial when there is large uncertainty about the true treatment effect. For Proof-of Concept (PoC) in a Phase II dose finding study, contrast test can be adopted to leverage information from all treatment groups. In this article, we propose two-stage SSR designs using frequentist conditional power (CP) and Bayesian predictive power (PP) for both single and multiple contrast tests. The Bayesian SSR can be implemented under a wide range of prior settings to incorporate different prior knowledge. Taking the adaptivity into account, all type I errors of final analysis in this paper are rigorously protected. Simulation studies are carried out to demonstrate the advantages of unblinded SSR in multi-arm trials.

A Bayesian design for finding optimal biological dose with mixed types of responses of toxicity and efficacy

For molecularly targeted therapy and immunotherapy, the targeted dose in the early phase clinical trial has been shifted from the maximum tolerated dose for the cytotoxic drug to the optimal biological dose where both toxicity and efficacy are considered. In this paper, we consider the situation that the responses of toxicity and efficacy are mixed in binary and continuous types, respectively, where the continuous endpoint bears more magnitude information than the binary endpoint after dichotomization. We propose combining two model-based designs to sequentially identify the most efficacious and tolerably safe dose. The employed designs both take the dose level information into account to achieve high estimation efficiency. We demonstrate the superiority of the proposed method to some existing methods by simulation.

Soluble Fiber Supplementation and Serum Lipid Profile: A Systematic Review and Dose-Response Meta-Analysis of Randomized Controlled Trials

To present a comprehensive synthesis of the effect of soluble fiber supplementation on blood lipid parameters in adults, a systematic search was undertaken in PubMed, Scopus, and ISI Web of Science of relevant articles published before November 2021. Randomized controlled trials (RCTs) evaluating the effects of soluble fibers on blood lipids in adults were included. We estimated the change in blood lipids for each 5 g/d increment in soluble fiber supplementation in each trial and then calculated the mean difference (MD) and 95% CI using a random-effects model. We estimated dose-dependent effects using a dose-response meta-analysis of differences in means. The risk of bias and certainty of the evidence was evaluated using the Cochrane risk of bias tool and the Grading Recommendations Assessment, Development, and Evaluation methodology, respectively. A total of 181 RCTs with 220 treatment arms (14,505 participants: 7348 cases and 7157 controls) were included. There was a significant reduction in LDL cholesterol (MD: -8.28 mg/dL, 95% CI: -11.38, -5.18), total cholesterol (TC) (MD: -10.82 mg/dL, 95% CI: -12.98, -8.67), TGs (MD: -5.55 mg/dL, 95% CI: -10.31, -0.79), and apolipoprotein B (Apo-B) (MD: -44.99 mg/L, 95% CI: -62.87, -27.12) after soluble fiber supplementation in the overall analysis. Each 5 g/d increase in soluble fiber supplementation had a significant reduction in TC (MD: -6.11 mg/dL, 95% CI: -7.61, -4.61) and LDL cholesterol (MD: -5.57 mg/dl, 95% CI: -7.44, -3.69). In a large meta-analysis of RCTs, results suggest that soluble fiber supplementation could contribute to the management of dyslipidemia and the reduction of cardiovascular disease risk.

Venglustat, a Novel Glucosylceramide Synthase Inhibitor, in Patients at Risk of Rapidly Progressing ADPKD: Primary Results of a Double-Blind, Placebo-Controlled, Phase 2/3 Randomized Clinical Trial

RATIONALE & OBJECTIVE

Autosomal dominant polycystic kidney disease (ADPKD) is characterized by the formation of multiple kidney cysts that leads to growth in total kidney volume (TKV) and progression to kidney failure. Venglustat is a glucosylceramide synthase inhibitor that has been shown to inhibit cyst growth and reduce kidney failure in preclinical models of ADPKD.

STUDY DESIGN

STAGED-PKD was a 2-stage, multicenter, double-blind, randomized, placebo-controlled phase 2/3 study in adults with ADPKD at risk of rapidly progressive disease, who were selected based on Mayo Clinic imaging classification of ADPKD class 1C, 1D, or 1E and an estimated glomerular filtration rate (eGFR) of 30-89.9mL/min/1.73m².

SETTING & PARTICIPANTS

Enrollment included 236 and 242 patients in stages 1 and 2, respectively.

INTERVENTIONS

In trial stage 1, the patients were randomized 1:1:1 to venglustat, 8mg; venglustat, 15mg; or placebo. In stage 2, the patients were randomized 1:1 to venglustat, 15mg (highest dose identified as safe and well tolerated in stage 1), or placebo.

OUTCOMES

Primary end points were rate of change in TKV over 18 months in stage 1 and eGFR slope over 24 months in stage 2. Secondary end points were eGFR slope over 18 months (stage 1), rate of change in TKV (stage 2), and safety/tolerability, pain, and fatigue (stages 1 and 2).

RESULTS

A prespecified interim futility analysis showed that venglustat treatment had no effect on the annualized rate of change in TKV over 18 months (stage 1) and had a faster rate of decline in eGFR slope over 24 months (stage 2). Due to this lack of efficacy, the study was terminated early.

LIMITATIONS

The short follow-up period after the end of treatment and limited generalizability of the findings.

CONCLUSIONS

In patients with rapidly progressing ADPKD, treatment with venglustat at either 8mg or 15mg showed no change in the rate of change in TKV and a faster rate of eGFR decline in STAGED-PKD despite a dose-dependent decrease in plasma glucosylceramide levels.

FUNDING

This study was funded by Sanofi.

TRIAL REGISTRATION

Registered at ClinicalTrials.gov with study number NCT03523728.

Remibrutinib, a novel BTK inhibitor, demonstrates promising efficacy and safety in chronic spontaneous urticaria

BACKGROUND

Chronic spontaneous urticaria (CSU) is inadequately controlled in many patients and greatly affects quality of life. Remibrutinib, a highly selective, oral, novel covalent Bruton tyrosine kinase inhibitor, might be effective in CSU.

OBJECTIVE

This first-in-patient trial aimed to evaluate the efficacy and safety of remibrutinib in CSU treatment and characterize the dose-response.

METHODS

This randomized, double-blind, placebo-controlled, phase 2b dose-finding trial evaluated remibrutinib (12 weeks) in patients inadequately controlled with second-generation H₁-antihistamines, with at least moderately active CSU, with or without prior anti-IgE treatment (NCT03926611). Patients received remibrutinib 10 mg once daily, 35 mg once daily, 100 mg once daily, 10 mg twice daily, 25 mg twice daily, 100 mg twice daily, or placebo (1:1:1:1:1:1:1 ratio). The main end points were weekly Urticaria Activity Score change from baseline at week 4 and safety.

RESULTS

Overall, 311 patients were randomized. Reduced symptom score was observed for all remibrutinib doses from week 1 until week 12, with weekly Urticaria Activity Score change from baseline at week 4: -19.1 (10 mg once daily), -19.1 (35 mg once daily), -14.7 (100 mg once daily), -16.0 (10 mg twice daily), -20.0 (25 mg twice daily), -18.1 (100 mg twice daily), and -5.4 for placebo (nominal P < .0001 for all doses vs placebo). Most adverse events were mild or moderate, with no dose-dependent pattern.

CONCLUSION

Remibrutinib was highly effective in the treatment of CSU over the entire dose range, with a rapid onset of action and a favorable safety profile.

In vitro/in silico prediction of drug induced steatosis in relation to oral doses and blood concentrations by the Nile Red assay

The accumulation of lipid droplets in hepatocytes is a key feature of drug-induced liver injury (DILI) and can be induced by a subset of hepatotoxic compounds. In the present study, we optimized and evaluated an in vitro technique based on the fluorescent dye Nile Red, further named Nile Red assay to quantify lipid droplets induced by the exposure to chemicals. The Nile Red assay and a cytotoxicity test (CTB assay) were then performed on cells exposed concentration-dependently to 60 different compounds. Of these, 31 were known to induce hepatotoxicity in humans, and 13 were reported to also cause steatosis. In order to compare in vivo relevant blood concentrations, pharmacokinetic models were established for all compounds to simulate the maximal blood concentrations (C_max) at therapeutic doses. The results showed that several hepatotoxic compounds induced an increase in lipid droplets at sub-cytotoxic concentrations. To compare how well (1) the cytotoxicity test alone, (2) the Nile Red assay alone, and (3) the combination of the cytotoxicity test and the Nile Red assay (based on the lower EC₁₀ of both assays) allow the differentiation between hepatotoxic and non-hepatotoxic compounds, a previously established performance metric, the Toxicity Separation Index (TSI) was calculated. In addition, the Toxicity Estimation Index (TEI) was calculated to determine how well blood concentrations that cause an increased DILI risk can be estimated for hepatotoxic compounds. Our findings indicate that the combination of both assays improved the TSI and TEI compared to each assay alone. In conclusion, the study demonstrates that inclusion of the Nile Red assay into in vitro test batteries may improve the prediction of DILI compounds.

Rationale and design of the AXIOMATIC-SSP phase II trial: Antithrombotic treatment with factor XIa inhibition to Optimize Management of Acute Thromboembolic events for Secondary Stroke Prevention

BACKGROUND

Individuals with ischemic stroke or transient ischemic attack (TIA) have a high early risk of ischemic stroke despite dual antiplatelet therapy. The risk of ischemic stroke, and associated disability, represents a significant unmet clinical need. Genetic variants resulting in reduced factor XI levels are associated with reduced risk for ischemic stroke but are not associated with increased intracranial bleeding. Milvexian is an oral small-molecule inhibitor of FXIa that binds activated factor XI with high affinity and selectivity and may reduce the risk of stroke when added to antiplatelet drugs without significant bleeding. We aimed to evaluate the dose-response relationship of milvexian in participants treated with dual antiplatelets.

METHODS

We began a phase II, double-blinded, randomized, placebo-controlled trial at 367 sites in 2019. Participants (N = 2366) with ischemic stroke (National Institutes of Health Stroke Scale score ≤7) or high-risk TIA (ABCD² score ≥6) were randomized to 1 of 5 doses of milvexian or placebo for 90 days. Participants also received clopidogrel 75 mg daily for the first 21 days and aspirin 100 mg for 90 days. The efficacy endpoint was the composite of ischemic stroke or incident infarct on magnetic resonance imaging. Major bleeding, defined as type 3 or 5 bleeding according to the Bleeding Academic Research Consortium, was the safety endpoint. Participant follow-up will end in 2022.

CONCLUSION

The AXIOMATIC-SSP trial will evaluate the dose-response of milvexian for ischemic stroke occurrence in participants with ischemic stroke or TIA.

Trial of Anti-BDCA2 Antibody Litifilimab for Cutaneous Lupus Erythematosus

BACKGROUND

Blood dendritic cell antigen 2 (BDCA2) is a receptor that is exclusively expressed on plasmacytoid dendritic cells, which are implicated in the pathogenesis of lupus erythematosus. Whether treatment with litifilimab, a humanized monoclonal antibody against BDCA2, would be efficacious in reducing disease activity in patients with cutaneous lupus erythematosus has not been extensively studied.

METHODS

In this phase 2 trial, we randomly assigned adults with histologically confirmed cutaneous lupus erythematosus with or without systemic manifestations in a 1:1:1:1 ratio to receive subcutaneous litifilimab (at a dose of 50, 150, or 450 mg) or placebo at weeks 0, 2, 4, 8, and 12. We used a dose-response model to assess whether there was a response across the four groups on the basis of the primary end point, which was the percent change from baseline to 16 weeks in the Cutaneous Lupus Erythematosus Disease Area and Severity Index-Activity score (CLASI-A; scores range from 0 to 70, with higher scores indicating more widespread or severe skin involvement). Safety was also assessed.

RESULTS

A total of 132 participants were enrolled; 26 were assigned to the 50-mg litifilimab group, 25 to the 150-mg litifilimab group, 48 to the 450-mg litifilimab group, and 33 to the placebo group. Mean CLASI-A scores for the groups at baseline were 15.2, 18.4, 16.5, and 16.5, respectively. The difference from placebo in the change from baseline in CLASI-A score at week 16 was -24.3 percentage points (95% confidence interval [CI] -43.7 to -4.9) in the 50-mg litifilimab group, -33.4 percentage points (95% CI, -52.7 to -14.1) in the 150-mg group, and -28.0 percentage points (95% CI, -44.6 to -11.4) in the 450-mg group. The least squares mean changes were used in the primary analysis of a best-fitting dose-response model across the three drug-dose levels and placebo, which showed a significant effect. Most of the secondary end points did not support the results of the primary analysis. Litifilimab was associated with three cases each of hypersensitivity and oral herpes infection and one case of herpes zoster infection. One case of herpes zoster meningitis occurred 4 months after the participant received the last dose of litifilimab.

CONCLUSIONS

In a phase 2 trial involving participants with cutaneous lupus erythematosus, treatment with litifilimab was superior to placebo with regard to a measure of skin disease activity over a period of 16 weeks. Larger and longer trials are needed to determine the effect and safety of litifilimab for the treatment of cutaneous lupus erythematosus. (Funded by Biogen; LILAC ClinicalTrials.gov number, NCT02847598.).

The Acute Stroke Therapy by Inhibition of Neutrophils study - Key features and impact

The Acute Stroke Therapy by Inhibition of Neutrophils (ASTIN) study, initiated in November of the year 2000, is now widely recognized as having been a landmark study in the history of clinical trials. We look at why this is the case by considering its key features and impact. These key features are: the use of Bayesian design and analysis; the use of the normal dynamic linear model; the response adaptive nature of the study; the use of real-time dosing decisions; and the use of an integrated model to predict 90-day response on the Scandinavian Stroke Scale. Our overall conclusion is that the ASTIN study's main impact came from showing the clinical trial community the feasibility of the novel design and analysis used when most of these key features were rarely used in industry trials, let alone used together in one trial in a disease area with a tremendous unmet medical need.

Simulation-based sequential design

We review some simulation-based methods to implement optimal decisions in sequential design problems as they naturally arise in clinical trial design. As a motivating example we use a stylized version of a dose-ranging design in the ASTIN trial. The approach can be characterized as constrained backward induction. The nature of the constraint is a restriction of the decisions to a set of actions that are functions of the current history only implicitly through a low-dimensional summary statistic. In addition, the action set is restricted to time-invariant policies. Time-dependence is only introduced indirectly through the change of the chosen summary statistic over time. This restriction allows computationally efficient solutions to the sequential decision problem. A further simplification is achieved by restricting optimal actions to be described by decision boundaries on the space of such summary statistics.

Confirmatory efficacy testing for individual dose-placebo comparisons using serial gatekeeping procedure in dose-finding trials with multiple comparison procedures-modeling

Dose-finding trials play a key role in the entire drug development process to determine optimal doses for regulatory approval. We address confirmatory efficacy testing for individual dose-placebo comparisons in the context of a dose-finding trial designed with multiple comparison procedures-modeling (MCP-Mod). An extension of the MCP-Mod, called closed MCP-Mod, has been proposed to carry out the MCP-Mod in conjunction with pairwise dose-placebo comparisons; however, an issue associated with the misspecification of candidate dose-response models remains. We consider another way to combine the MCP-Mod and the individual dose-placebo comparisons using serial gatekeeping procedures with fixed sequence, Holm, Hochberg, and step-down Dunnett procedure. The method controls the family-wise error rate in the strong sense and is simple enough to be implemented by existing software. Simulation studies suggested that the serial gatekeeping procedure was comparable with the closed MCP-Mod in terms of statistical power to detect the efficacy of at least one dose, and both methods were capable of pursuing the efficacy claim rather than just establishing the dose-response signal with less than a 20% increase in sample size when assuming monotonic dose-response shapes. The serial gatekeeping procedure would have advantages in the simplicity of implementation and ease of interpretation. The dose-finding trials aiming to declare the dose-response signal, as well as the efficacy of individual doses, would be worth considering as an option to accelerate the drug development program in certain situations.

A straightforward meta-analysis approach for oncology phase I dose-finding studies

Phase I early-phase clinical studies aim at investigating the safety and the underlying dose-toxicity relationship of a drug or combination. While little may still be known about the compound's properties, it is crucial to consider quantitative information available from any studies that may have been conducted previously on the same drug. A meta-analytic approach has the advantages of being able to properly account for between-study heterogeneity, and it may be readily extended to prediction or shrinkage applications. Here we propose a simple and robust two-stage approach for the estimation of maximum tolerated dose(s) utilizing penalized logistic regression and Bayesian random-effects meta-analysis methodology. Implementation is facilitated using standard R packages. The properties of the proposed methods are investigated in Monte Carlo simulations. The investigations are motivated and illustrated by two examples from oncology.

Model selection characteristics when using MCP-Mod for dose-response gene expression data

We extend the scope of application for MCP-Mod (Multiple Comparison Procedure and Modeling) to in vitro gene expression data and assess its characteristics regarding model selection for concentration gene expression curves. Precisely, we apply MCP-Mod on single genes of a high-dimensional gene expression data set, where human embryonic stem cells were exposed to eight concentration levels of the compound valproic acid (VPA). As candidate models we consider the sigmoid E max $E_{\max }$ (four-parameter log-logistic), linear, quadratic, E max $E_{\max }$ , exponential, and beta model. Through simulations we investigate the impact of omitting one or more models from the candidate model set to uncover possibly superfluous models and to evaluate the precision and recall rates of selected models. Each model is selected according to Akaike information criterion (AIC) for a considerable number of genes. For less noisy cases the popular sigmoid E max $E_{\max }$ model is frequently selected. For more noisy data, often simpler models like the linear model are selected, but mostly without relevant performance advantage compared to the second best model. Also, the commonly used standard E max $E_{\max }$ model has an unexpected low performance.

The pan-JAK inhibitor delgocitinib in a cream formulation demonstrates dose-response in chronic hand eczema in a 16-week randomised phase 2b trial

BACKGROUND

Chronic hand eczema (CHE) is a burdensome disease, and new well-documented, safe, and efficacious treatments are warranted. In a recent CHE phase 2a trial, the pan-Janus kinase (JAK) inhibitor delgocitinib in an ointment formulation was found to be efficacious and well-tolerated.

OBJECTIVES

This trial assessed the dose-response, efficacy, and safety of delgocitinib cream in CHE.

METHODS

In this double-blind, phase 2b dose-ranging trial, adults with CHE and a recent history of inadequate response or contraindication to topical corticosteroids were randomised to delgocitinib cream 1, 3, 8, 20 mg/g or vehicle treatment twice daily for 16 weeks. Primary endpoint was Investigator's Global Assessment for CHE (IGA-CHE) treatment success (0 [clear] or 1 [almost clear] with a ≥2-point improvement from baseline to Week 16). Secondary endpoints were time to IGA-CHE treatment success and changes in Hand Eczema Severity Index (HECSI); other endpoints were itch and pain NRS scores, and Patient's Global Assessment (PaGA) at Week 16.

RESULTS

258 patients were randomised 1:1:1:1:1 to delgocitinib cream 1, 3, 8, 20 mg/g or vehicle. A significant dose-response relationship was established for IGA-CHE (p<0.025). IGA-CHE treatment success at Week 16 was achieved in 21.2% (1 mg/g), 7.8% (3 mg/g), 36.5% (8 mg/g), 37.7% (20 mg/g), and 8.0% (vehicle) of patients. Delgocitinib 8 and 20 mg/g showed a treatment effect against vehicle (p<0.001). Similarly, there were improvements in HECSI, itch and pain NRS scores, and PaGA. Delgocitinib cream was well-tolerated with majority of adverse events being mild or moderate and considered unrelated to treatment. Most frequently reported adverse events were nasopharyngitis (17.3%-29.4% in delgocitinib groups vs 40% in vehicle group), eczema (5.8%-11.3% in delgocitinib groups vs 16.0% in vehicle group) and headache (3.8%-11.5% in delgocitinib groups vs 4.0% in vehicle group).

CONCLUSIONS

In this trial, delgocitinib cream showed a dose-response relationship in terms of efficacy and was well-tolerated.

Safety and efficacy of subcutaneous ianalumab (VAY736) in patients with primary Sjögren's syndrome: a randomised, double-blind, placebo-controlled, phase 2b dose-finding trial

BACKGROUND

Sjögren's syndrome is an autoimmune disease characterised by dry eyes and mouth, systemic features, and reduced quality of life. There are no disease-modifying treatments. A new biologic, ianalumab (VAY736), with two modes of suppressing B cells, has previously shown preliminary efficacy. This dose-finding trial aimed to assess the safety and efficacy of different subcutaneous doses of ianalumab in patients with moderate to severe primary Sjögren's syndrome.

METHODS

VAY736A2201 was a randomised, parallel, double-blind, placebo-controlled, phase 2b dose-finding study done in 56 centres in 19 countries. Patients aged 18-75 years with primary Sjögren's syndrome with moderate to severe disease activity (European Alliance of Associations for Rheumatology [EULAR] Sjögren's Syndrome Disease Activity Index [ESSDAI] score ≥6) and symptom severity (EULAR Sjögren's Syndrome Patient Reported Index score ≥5) were eligible. Participants were randomly assigned (1:1:1:1) to receive subcutaneous placebo or ianalumab (5 mg, 50 mg, or 300 mg) every 4 weeks for 24 weeks using a secure, online randomisation system. Randomisation was stratified by the ESSDAI score at baseline (≥10 or <10). Study personnel and patients were masked to treatment assignment. The primary outcome was the change in ESSDAI score from baseline to 24 weeks in all randomly assigned patients. Dose-related change in disease activity (ESSDAI) from baseline at week 24 was assessed by multiple comparison procedure with modelling analysis. Safety was measured in all patients who received at least one dose of study drug. This trial is registered with ClinicalTrials.gov, NCT02962895.

FINDINGS

Between June 27, 2017, and Dec 06, 2018, 293 patients were screened, 190 of whom were randomly assigned (placebo n=49, ianalumab 5 mg n=47, ianalumab 50 mg n=47, ianalumab 300 mg n=47). Statistically significant dose-responses were seen for overall disease activity (ESSDAI score) in four of the five dose-response models tested (p<0·025 in four models, p=0·060 in one model). The ESSDAI score decreased from baseline in all ianalumab groups, with the maximal ESSDAI score change from baseline observed in the ianalumab 300 mg group: placebo-adjusted least-squares mean change from baseline -1·92 points (95% CI -4·15 to 0·32; p=0·092). There were four serious adverse events in three patients considered treatment-related (pneumonia [n=1] and gastroenteritis [n=1] in the placebo group; appendicitis plus tubo-ovarian abscess in the same patient in the ianalumab 50 mg group).

INTERPRETATION

The study met its primary objective, showing a dose-related decrease in disease activity as measured by ESSDAI at week 24. Overall, ianalumab was well tolerated and safe, with no increase in infections. To our knowledge, this is the first large, randomised, controlled trial in primary Sjögren's syndrome that met its primary endpoint, and its results mean there is potential for more studies of this mechanism in the future.

FUNDING

Novartis.

Optimal adaptive allocation using deep reinforcement learning in a dose-response study

Estimation of the dose-response curve for efficacy and subsequent selection of an appropriate dose in phase II trials are important processes in drug development. Various methods have been investigated to estimate dose-response curves. Generally, these methods are used with equal allocation of subjects for simplicity; nevertheless, they may not fully optimize performance metrics because of nonoptimal allocation. Optimal allocation methods, which include adaptive allocation methods, have been proposed to overcome the limitations of equal allocation. However, they rely on asymptotics, and thus sometimes cannot efficiently optimize the performance metric with the sample size in an actual clinical trial. The purpose of this study is to construct an adaptive allocation rule that directly optimizes a performance metric, such as power, accuracy of model selection, accuracy of the estimated target dose, or mean absolute error over the estimated dose-response curve. We demonstrate that deep reinforcement learning with an appropriately defined state and reward can be used to construct such an adaptive allocation rule. The simulation study shows that the proposed method can successfully improve the performance metric to be optimized when compared with the equal allocation, D-optimal, and TD-optimal methods. In particular, when the mean absolute error was set to the metric to be optimized, it is possible to construct a rule that is superior for many metrics.

The hepatocyte export carrier inhibition assay improves the separation of hepatotoxic from non-hepatotoxic compounds

An in vitro/in silico method that determines the risk of human drug induced liver injury in relation to oral doses and blood concentrations of drugs was recently introduced. This method utilizes information on the maximal blood concentration (C_max) for a specific dose of a test compound, which can be estimated using physiologically-based pharmacokinetic modelling, and a cytotoxicity test in cultured human hepatocytes. In the present study, we analyzed if the addition of an assay that measures the inhibition of bile acid export carriers, like BSEP and/or MRP2, to the existing method improves the differentiation of hepatotoxic and non-hepatotoxic compounds. Therefore, an export assay for 5-chloromethylfluorescein diacetate (CMFDA) was established. We tested 36 compounds in a concentration-dependent manner for which the risk of hepatotoxicity for specific oral doses and the capacity to inhibit hepatocyte export carriers are known. Compared to the CTB cytotoxicity test, substantially lower EC₁₀ values were obtained using the CMFDA assay for several known BSEP and/or MRP2 inhibitors. To quantify if the addition of the CMFDA assay to our test system improves the overall separation of hepatotoxic from non-hepatotoxic compounds, the toxicity separation index (TSI) was calculated. We obtained a better TSI using the lower alert concentration from either the CMFDA or the CTB test (TSI: 0.886) compared to considering the CTB test alone (TSI: 0.775). In conclusion, the data show that integration of the CMFDA assay with an in vitro test battery improves the differentiation of hepatotoxic and non-hepatotoxic compounds in a set of compounds that includes bile acid export carrier inhibitors.

Adaptive dose-response studies to establish proof-of-concept in learning-phase clinical trials

In learning-phase clinical trials in drug development, adaptive designs can be efficient and highly informative when used appropriately. In this article, we extend the multiple comparison procedures with modeling techniques (MCP-Mod) procedure with generalized multiple contrast tests (GMCTs) to two-stage adaptive designs for establishing proof-of-concept. The results of an interim analysis of first-stage data are used to adapt the candidate dose-response models and the dosages studied in the second stage. GMCTs are used in both stages to obtain stage-wise p -values, which are then combined to determine an overall p -value. An alternative approach is also considered that combines the t -statistics across stages, employing the conditional rejection probability principle to preserve the Type I error probability. Simulation studies demonstrate that the adaptive designs are advantageous compared to the corresponding tests in a nonadaptive design if the selection of the candidate set of dose-response models is not well informed by evidence from preclinical and early-phase studies.