Estimating treatment effect in randomized trial after control to treatment crossover using external controls.

In clinical trials, it is common to design a study that permits the administration of an experimental treatment to participants in the placebo or standard of care group post primary endpoint. This is often seen in the open-label extension phase of a phase III, pivotal study of the new medicine, where the focus is on assessing long-term safety and efficacy. With the availability of external controls, proper estimation and inference of long-term treatment effect during the open-label extension phase in the absence of placebo-controlled patients are now feasible. Within the framework of causal inference, we propose several difference-in-differences (DID) type methods and a synthetic control method (SCM) for the combination of randomized controlled trials and external controls. Our realistic simulation studies demonstrate the desirable performance of the proposed estimators in a variety of practical scenarios. In particular, DID methods outperform SCM and are the recommended methods of choice. An empirical application of the methods is demonstrated through a phase III clinical trial in rare disease.

Opportunities and challenges with decentralized trials in Neuroscience.

Decentralized clinical trials (DCTs), that is, studies integrating elements of telemedicine and mobile/local healthcare providers allowing for home-based assessments, are an important concept to make studies more resilient and more patient-centric by taking into consideration participant's views and shifting trial activities to better meet the needs of trial participants. There are however, not only advantages but also challenges associated with DCTs. An area to be addressed by appropriate statistical methodology is the integration of data resulting from a possible mix of home and clinic assessments at different visits for the same variable, especially in adjusting for sources of possible systematic differences. One source of systematic bias may be how a participant perceives their disease and treatment in their home versus in a clinical setting. In this paper, we will discuss these issues with a focus on Neuroscience when participants have the choice between home and clinic assessments to illustrate how to identify systematic biases and describe appropriate approaches to maintain clinical trial scientific rigor. We will describe the benefits and challenges of DCTs in Neuroscience and then describe the relevance of home versus clinic assessments using the estimand framework. We outline several options to enable home assessments in a study. Results of simulations will be presented to help deciding between design and analysis options in a simple scenario where there might be differences in response between clinic and home assessments.

The use of external controls: To what extent can it currently be recommended?

With more and better clinical data being captured outside of clinical studies and greater data sharing of clinical studies, external controls may become a more attractive alternative to randomized clinical trials (RCTs). Both industry and regulators recognize that in situations where a randomized study cannot be performed, external controls can provide the needed contextualization to allow a better interpretation of studies without a randomized control. It is also agreed that external controls will not fully replace RCTs as the gold standard for formal proof of efficacy in drug development and the yardstick of clinical research. However, it remains unclear in which situations conclusions about efficacy and a positive benefit/risk can reliably be based on the use of an external control. This paper will provide an overview on types of external control, their applications and the different sources of bias their use may incur, and discuss potential mitigation steps. It will also give recommendations on how the use of external controls can be justified.

A multistate model for early decision-making in oncology.

The development of oncology drugs progresses through multiple phases, where after each phase, a decision is made about whether to move a molecule forward. Early phase efficacy decisions are often made on the basis of single-arm studies based on a set of rules to define whether the tumor improves ("responds"), remains stable, or progresses (response evaluation criteria in solid tumors [RECIST]). These decision rules are implicitly assuming some form of surrogacy between tumor response and long-term endpoints like progression-free survival (PFS) or overall survival (OS). With the emergence of new therapies, for which the link between RECIST tumor response and long-term endpoints is either not accessible yet, or the link is weaker than with classical chemotherapies, tumor response-based rules may not be optimal. In this paper, we explore the use of a multistate model for decision-making based on single-arm early phase trials. The multistate model allows to account for more information than the simple RECIST response status, namely, the time to get to response, the duration of response, the PFS time, and time to death. We propose to base the decision on efficacy on the OS hazard ratio (HR) comparing historical control to data from the experimental treatment, with the latter predicted from a multistate model based on early phase data with limited survival follow-up. Using two case studies, we illustrate feasibility of the estimation of such an OS HR. We argue that, in the presence of limited follow-up and small sample size, and making realistic assumptions within the multistate model, the OS prediction is acceptable and may lead to better early decisions within the development of a drug.

Best practice for analysis of shared clinical trial data.

BACKGROUND: Greater transparency, including sharing of patient-level data for further research, is an increasingly important topic for organisations who sponsor, fund and conduct clinical trials. This is a major paradigm shift with the aim of maximising the value of patient-level data from clinical trials for the benefit of future patients and society. We consider the analysis of shared clinical trial data in three broad categories: (1) reanalysis - further investigation of the efficacy and safety of the randomized intervention, (2) meta-analysis, and (3) supplemental analysis for a research question that is not directly assessing the randomized intervention. DISCUSSION: In order to support appropriate interpretation and limit the risk of misleading findings, analysis of shared clinical trial data should have a pre-specified analysis plan. However, it is not generally possible to limit bias and control multiplicity to the extent that is possible in the original trial design, conduct and analysis, and this should be acknowledged and taken into account when interpreting results. We highlight a number of areas where specific considerations arise in planning, conducting, interpreting and reporting analyses of shared clinical trial data. A key issue is that that these analyses essentially share many of the limitations of any post hoc analyses beyond the original specified analyses. The use of individual patient data in meta-analysis can provide increased precision and reduce bias. Supplemental analyses are subject to many of the same issues that arise in broader epidemiological analyses. Specific discussion topics are addressed within each of these areas. Increased provision of patient-level data from industry and academic-led clinical trials for secondary research can benefit future patients and society. Responsible data sharing, including transparency of the research objectives, analysis plans and of the results will support appropriate interpretation and help to address the risk of misleading results and avoid unfounded health scares.

Data sharing and the evolving role of statisticians.

BACKGROUND: Greater transparency and, in particular, sharing of clinical study reports and patient level data for further research is an increasingly important topic for the pharmaceutical and biotechnology industry and other organisations who sponsor and conduct clinical research as well as academic researchers and patient advocacy groups. Statisticians are ambassadors for data sharing and are central to its success. They play an integral role in data sharing discussions within their companies and also externally helping to shape policy and processes while providing input into practical solutions to aid data sharing. Data sharing is generating changes in the required profile for statisticians in the pharmaceutical and biotechnology industry, as well as academic institutions and patient advocacy groups. DISCUSSION: Successful statisticians need to possess many qualities required in today's pharmaceutical environment such as collaboration, diplomacy, written and oral skills and an ability to be responsive; they are also knowledgeable when debating strategy and analytical techniques. However, increasing data transparency will require statisticians to evolve and learn new skills and behaviours during their career which may not have been an accepted part of the traditional role. Statisticians will move from being the gate-keepers of data to be data facilitators. To adapt successfully to this new environment, the role of the statistician is likely to be broader, including defining new responsibilities that lie beyond the boundaries of the traditional role. Statisticians should understand how data transparency can benefit them and the potential strategic advantage it can bring and be fully aware of the pharmaceutical and biotechnology industry commitments to data transparency and the policies within their company or research institute in addition to focusing on reviewing requests and provisioning data. Data transparency will evolve the role of statisticians within the pharmaceutical and biotechnology industry, academia and research bodies to a level which may not have been an accepted part of their traditional role or career. In the future, skills will be required to manage challenges arising from data sharing; statisticians will need strong scientific and statistical guiding principles for reanalysis and supplementary analyses based on researchers' requests, have enhanced consultancy skills, in particular the ability to defend good statistical practice in the face of criticism and the ability to critique methods of analysis. Statisticians will also require expertise in data privacy regulations, data redaction and anonymisation and be able to assess the probability of re-identification, an ability to understand analyses conducted by researchers and recognise why such analyses may propose different results compared to the original analyses. Bringing these skills to the implementation of data sharing and interpretation of the results will help to maximise the value of shared data while guarding against misleading conclusions.

Bayesian predictive power: choice of prior and some recommendations for its use as probability of success in drug development.

Bayesian predictive power, the expectation of the power function with respect to a prior distribution for the true underlying effect size, is routinely used in drug development to quantify the probability of success of a clinical trial. Choosing the prior is crucial for the properties and interpretability of Bayesian predictive power. We review recommendations on the choice of prior for Bayesian predictive power and explore its features as a function of the prior. The density of power values induced by a given prior is derived analytically and its shape characterized. We find that for a typical clinical trial scenario, this density has a u-shape very similar, but not equal, to a ß-distribution. Alternative priors are discussed, and practical recommendations to assess the sensitivity of Bayesian predictive power to its input parameters are provided. Copyright © 2016 John Wiley & Sons, Ltd.

Comparison of design strategies for a three-arm clinical trial with time-to-event endpoint: Power, time-to-analysis, and operational aspects.

To optimize resources, randomized clinical trials with multiple arms can be an attractive option to simultaneously test various treatment regimens in pharmaceutical drug development. The motivation for this work was the successful conduct and positive final outcome of a three-arm randomized clinical trial primarily assessing whether obinutuzumab plus chlorambucil in patients with chronic lympocytic lymphoma and coexisting conditions is superior to chlorambucil alone based on a time-to-event endpoint. The inference strategy of this trial was based on a closed testing procedure. We compare this strategy to three potential alternatives to run a three-arm clinical trial with a time-to-event endpoint. The primary goal is to quantify the differences between these strategies in terms of the time it takes until the first analysis and thus potential approval of a new drug, number of required events, and power. Operational aspects of implementing the various strategies are discussed. In conclusion, using a closed testing procedure results in the shortest time to the first analysis with a minimal loss in power. Therefore, closed testing procedures should be part of the statistician's standard clinical trials toolbox when planning multiarm clinical trials.

Statistical and data reporting guidelines for the European Journal of Cardio-Thoracic Surgery and the Interactive CardioVascular and Thoracic Surgery.

As part of the peer review process for the European Journal of Cardio-Thoracic Surgery (EJCTS) and the Interactive CardioVascular and Thoracic Surgery (ICVTS), a statistician reviews any manuscript that includes a statistical analysis. To facilitate authors considering submitting a manuscript and to make it clearer about the expectations of the statistical reviewers, we present up-to-date guidelines for authors on statistical and data reporting specifically in these journals. The number of statistical methods used in the cardiothoracic literature is vast, as are the ways in which data are presented. Therefore, we narrow the scope of these guidelines to cover the most common applications submitted to the EJCTS and ICVTS, focusing in particular on those that the statistical reviewers most frequently comment on.

The use of phase 2 interim analysis to expedite drug development decisions.

PURPOSE: To expedite drug development, we propose a two-step decision-making process that utilizes interim efficacy results from a comparative phase 2 trial to determine whether to accelerate subsequent phase 3 preparations, and final analysis to ultimately determine whether to conduct phase 3 testing. METHODS: The operational characteristics of this process were evaluated by modeling simulated data of oncology trials and retrospectively analyzing data from historical comparative phase 2 trials. Progression-free survival (PFS) was used as the primary endpoint; the estimated PFS hazard ratios (HRs) of ≤0.60 at interim and of ≤0.65 at final analysis favoring the experimental arm were defined as positive results. The conditional probability of achieving a target PFS HR at final analysis, based on observed interim results, was also estimated by imputing post-interim data with and without the proportional hazard assumption. RESULTS: Simulations of phase 2 trials showed that estimated interim PFS HRs correlated with estimated final PFS HRs, with reasonably low false-positive rates for supporting phase 3 "go" decisions at interim. Using observed historical data, decisions based on interim PFS analyses also matched final phase 3 "go" and "no-go" decisions with a false-positive rate of 16.7% (2 of 12 trials) and a false-negative rate of 9.4% (3 of 32 trials). Analytical modeling accurately predicted final PFS HRs from observed interim data when accounting for appropriate underlying assumptions. CONCLUSIONS: The results support the usefulness of a two-step decision-making process that utilizes interim phase 2 results to reduce the interval between phase 2 completion and phase 3 initiation.

European Federation of Statisticians in the Pharmaceutical Industry's position on access to clinical trial data.

The European Federation of Statisticians in the Pharmaceutical Industry (EFSPI) believes access to clinical trial data should be implemented in a way that supports good research, avoids misuse of such data, lies within the scope of the original informed consent and fully protects patient confidentiality. In principle, EFSPI supports responsible data sharing. EFSPI acknowledges it is in the interest of patients that their data are handled in a strictly confidential manner to avoid misuse under all possible circumstances. It is also in the interest of the altruistic nature of patients participating in trials that such data will be used for further development of science as much as possible applying good statistical principles. This paper summarises EFSPI's position on access to clinical trial data. The position was developed during the European Medicines Agency (EMA) advisory process and before the draft EMA policy on publication and access to clinical trial data was released for consultation; however, the EFSPI's position remains unchanged following the release of the draft policy. Finally, EFSPI supports a need for further guidance to be provided on important technical aspects relating to re-analyses and additional analyses of clinical trial data, for example, multiplicity, meta-analysis, subgroup analyses and publication bias.

Perspectives on the use of adaptive designs in clinical trials. Part II. Panel discussion.

In the midst of gaining more experience in pursuing scientifically sound approaches of adaptive designs in clinical trials, a panel discussion with international representatives from industry, academia, and regulatory agencies was held at the Basel Biometric Society Spring Conference, March 12, 2010. The goal was to develop some consensus among industry, government, and academic statisticians concerning requirements and methods for adaptive designs in clinical trials. In this paper, we summarize the panelists' perspectives given at that session.

Comments on the draft guidance on "adaptive design clinical trials for drugs and biologics" of the U.S. Food and Drug Administration.

The U.S. FDA has published a draft guidance on "Adaptive Design Clinical Trials for Drugs and Biologics", which gives regulatory guidance on methodological issues in exploratory and confirmatory clinical trials planned with an adaptive design. This comment summarizes the discussion within the joint working group "Adaptive Designs and Multiple Testing Procedures" of the Austro-Swiss and German regions of the International Biometric Society held at the 90-day public comment period in spring 2010.

Value of N-terminal brain natriuretic peptide as a prognostic marker in patients with CKD: results from the CREATE study.

BACKGROUND AND OBJECTIVES: This study assessed plasma N-terminal prohormone brain natriuretic peptide (NT-proBNP) as a prognostic marker of cardiovascular risk in patients with chronic kidney disease stages 3-4 and anaemia treated with epoetin beta to two haemoglobin target ranges. DESIGN, SETTING, PARTICIPANTS & MEASUREMENTS: Of 603 patients enrolled in the Cardiovascular Risk Reduction by Early Anaemia Treatment with Epoetin Beta (CREATE) trial (baseline creatinine clearance 15-35 mL/min; haemoglobin 11.0-12.5 g/dL), 291 were included in this sub-study. Patients received subcutaneous epoetin beta either immediately after randomisation (target 13.0-15.0 g/dL; Group 1), or after their haemoglobin levels had fallen < 10.5 g/dL (target 10.5-11.5 g/dL; Group 2). Chronic heart failure New York Heart Association class III-IV was an exclusion criterion. (ClinicalTrials.gov Identifier: NCT00321919) RESULTS: Cardiovascular event rates were higher in patients with baseline NT-proBNP > 400 vs. ≤ 400 pg/mL (39 vs. 13 events; p = 0.0002). Dialysis was initiated in 68 vs. 42 patients with NT-proBNP > 400 vs. ≤ 400 pg/mL (p = 0.0003). Amongst patients with NT-proBNP > 400 pg/mL, there was no significant difference between treatment groups in risk of cardiovascular events (HR = 0.57; p = 0.08) or time to dialysis (HR = 0.65; p = 0.08). The overall interpretation of this substudy is, however, limited by its relatively small sample size which, together with low clinical event rates, result in a lack of statistical power for some analyses and should be viewed as being hypothesis-generating in nature. CONCLUSIONS: In chronic kidney disease patients with mild-to-moderate anaemia, elevated baseline plasma NT-proBNP levels are associated with a higher risk of cardiovascular events and an accelerated progression towards end-stage renal disease.

What is the impact of antithrombotic therapy and risk factors on the frequency of thrombovascular events in patients with metastatic breast cancer receiving epoetin beta?

UNLABELLED: PURPOSE, PATIENTS AND METHODS: This retrospective analysis of the BRAVE study evaluated the impact of baseline risk factors and antithrombotic therapy on the risk of thrombovascular events (TVEs) in patients receiving epoetin compared to patients not receiving epoetin. RESULTS: Baseline risk factors have a significant impact on TVE risk under epoetin therapy. More than 2 risk factors increased the risk of TVEs in patients receiving epoetin (hazard ratio [HR] 2.89, confidence interval [CI] 1.04-8.02, p value [p]=0.04). In patients on epoetin without antithrombotic therapy, the risk for TVEs was higher (HR 4.11, CI 1.37-12.4, p=0.01) compared to those who received antithrombotics (HR 1.37, CI 0.59-3.18, p=0.45). CONCLUSIONS: Our analysis has identified several risk factors which may impact the risk of TVEs under epoetin therapy. These data suggest that antithrombotic therapy may have the potential to reduce the risk of TVEs under epoetin therapy. These findings are hypothesis-generating and need to be confirmed in a prospective, randomised study.

Effect of once-weekly epoetin beta on survival in patients with metastatic breast cancer receiving anthracycline- and/or taxane-based chemotherapy: results of the Breast Cancer-Anemia and the Value of Erythropoietin (BRAVE) study.

PURPOSE: The Breast Cancer-Anemia and the Value of Erythropoietin (BRAVE) study evaluated whether epoetin beta would improve survival in patients with metastatic breast cancer (MBC). PATIENTS AND METHODS: BRAVE was an open-label, randomized, multicenter study in patients with MBC treated with anthracycline- and/or taxane-based chemotherapy. Patients (hemoglobin [Hb] < 12.9 g/dL) were randomly assigned (1:1) to epoetin beta 30,000 U subcutaneously once weekly or control for 24 weeks. The primary efficacy variable was overall survival. Secondary efficacy outcomes included progression-free survival, transfusion- and severe anemia-free survival, Hb response, safety, and quality of life (QoL). RESULTS: After 18 months of follow-up, 62 (27%) of 231 patients survived with epoetin beta therapy and 63 (27%) of 232 with control. No difference was detected in overall survival (hazard ratio [HR] = 1.07; 95% CI, 0.87 to 1.33, P = .522) or progression-free survival (HR = 1.07; 95% CI, 0.89 to 1.30, P = .448). There was a statistically significant benefit on transfusion- and severe anemia-free survival compared with control (HR = 0.59; P = .0097). Median Hb level increased with epoetin beta (11.7 g/dL at baseline to 13.3 g/dL at 24 weeks) but did not change with control (11.5 v 11.4 g/dL). Patients receiving epoetin beta experienced more thromboembolic events (TEEs) compared with controls (13% v 6%; P = .012) with no difference in serious TEEs (4% v 3%). Epoetin beta did not significantly improve QoL in this study where patients had a high baseline Hb value. CONCLUSION: In patients with MBC receiving chemotherapy and initial Hb less than 12.9 g/dL, epoetin beta increased Hb. No difference was detected in overall survival. Because of its superiority design, this study cannot, however, exclude clinically important differences in survival with absolute certainty.

Normalization of hemoglobin level in patients with chronic kidney disease and anemia.

BACKGROUND: Whether correction of anemia in patients with stage 3 or 4 chronic kidney disease improves cardiovascular outcomes is not established. METHODS: We randomly assigned 603 patients with an estimated glomerular filtration rate (GFR) of 15.0 to 35.0 ml per minute per 1.73 m2 of body-surface area and mild-to-moderate anemia (hemoglobin level, 11.0 to 12.5 g per deciliter) to a target hemoglobin value in the normal range (13.0 to 15.0 g per deciliter, group 1) or the subnormal range (10.5 to 11.5 g per deciliter, group 2). Subcutaneous erythropoietin (epoetin beta) was initiated at randomization (group 1) or only after the hemoglobin level fell below 10.5 g per deciliter (group 2). The primary end point was a composite of eight cardiovascular events; secondary end points included left ventricular mass index, quality-of-life scores, and the progression of chronic kidney disease. RESULTS: During the 3-year study, complete correction of anemia did not affect the likelihood of a first cardiovascular event (58 events in group 1 vs. 47 events in group 2; hazard ratio, 0.78; 95% confidence interval, 0.53 to 1.14; P=0.20). Left ventricular mass index remained stable in both groups. The mean estimated GFR was 24.9 ml per minute in group 1 and 24.2 ml per minute in group 2 at baseline and decreased by 3.6 and 3.1 ml per minute per year, respectively (P=0.40). Dialysis was required in more patients in group 1 than in group 2 (127 vs. 111, P=0.03). General health and physical function improved significantly (P=0.003 and P<0.001, respectively, in group 1, as compared with group 2). There was no significant difference in the combined incidence of adverse events between the two groups, but hypertensive episodes and headaches were more prevalent in group 1. CONCLUSIONS: In patients with chronic kidney disease, early complete correction of anemia does not reduce the risk of cardiovascular events. (ClinicalTrials.gov number, NCT00321919 [ClinicalTrials.gov].).