Joint modeling of survival and longitudinal non-survival data: current methods and issues. Report of the DIA Bayesian joint modeling working group.

Explicitly modeling underlying relationships between a survival endpoint and processes that generate longitudinal measured or reported outcomes potentially could improve the efficiency of clinical trials and provide greater insight into the various dimensions of the clinical effect of interventions included in the trials. Various strategies have been proposed for using longitudinal findings to elucidate intervention effects on clinical outcomes such as survival. The application of specifically Bayesian approaches for constructing models that address longitudinal and survival outcomes explicitly has been recently addressed in the literature. We review currently available methods for carrying out joint analyses, including issues of implementation and interpretation, identify software tools that can be used to carry out the necessary calculations, and review applications of the methodology.

Balancing benefit and risk of medicines: a systematic review and classification of available methodologies.

BACKGROUND: The need for formal and structured approaches for benefit-risk assessment of medicines is increasing, as is the complexity of the scientific questions addressed before making decisions on the benefit-risk balance of medicines. We systematically collected, appraised and classified available benefit-risk methodologies to facilitate and inform their future use. METHODS: A systematic review of publications identified benefit-risk assessment methodologies. Methodologies were appraised on their fundamental principles, features, graphical representations, assessability and accessibility. We created a taxonomy of methodologies to facilitate understanding and choice. RESULTS: We identified 49 methodologies, critically appraised and classified them into four categories: frameworks, metrics, estimation techniques and utility survey techniques. Eight frameworks describe qualitative steps in benefit-risk assessment and eight quantify benefit-risk balance. Nine metric indices include threshold indices to measure either benefit or risk; health indices measure quality-of-life over time; and trade-off indices integrate benefits and risks. Six estimation techniques support benefit-risk modelling and evidence synthesis. Four utility survey techniques elicit robust value preferences from relevant stakeholders to the benefit-risk decisions. CONCLUSIONS: Methodologies to help benefit-risk assessments of medicines are diverse and each is associated with different limitations and strengths. There is not a 'one-size-fits-all' method, and a combination of methods may be needed for each benefit-risk assessment. The taxonomy introduced herein may guide choice of adequate methodologies. Finally, we recommend 13 of 49 methodologies for further appraisal for use in the real-life benefit-risk assessment of medicines.

Guidance on the implementation and reporting of a drug safety Bayesian network meta-analysis.

The Drug Information Association Bayesian Scientific Working Group (BSWG) was formed in 2011 with a vision to ensure that Bayesian methods are well understood and broadly utilized for design and analysis and throughout the medical product development process, and to improve industrial, regulatory, and economic decision making. The group, composed of individuals from academia, industry, and regulatory, has as its mission to facilitate the appropriate use and contribute to the progress of Bayesian methodology. In this paper, the safety sub-team of the BSWG explores the use of Bayesian methods when applied to drug safety meta-analysis and network meta-analysis. Guidance is presented on the conduct and reporting of such analyses. We also discuss different structural model assumptions and provide discussion on prior specification. The work is illustrated through a case study involving a network meta-analysis related to the cardiovascular safety of non-steroidal anti-inflammatory drugs.

Statistical aspects in comparative benefit-risk assessment: challenges and opportunities for pharmaceutical statisticians.

Benefit-risk assessment is a fundamental element of drug development with the aim to strengthen decision making for the benefit of public health. Appropriate benefit-risk assessment can provide useful information for proactive intervention in health care settings, which could save lives, reduce litigation, improve patient safety and health care outcomes, and furthermore, lower overall health care costs. Recent development in this area presents challenges and opportunities to statisticians in the pharmaceutical industry. We review the development and examine statistical issues in comparative benefit-risk assessment. We argue that a structured benefit-risk assessment should be a multi-disciplinary effort involving experts in clinical science, safety assessment, decision science, health economics, epidemiology and statistics. Well planned and conducted analyses with clear consideration on benefit and risk are critical for appropriate benefit-risk assessment. Pharmaceutical statisticians should extend their knowledge to relevant areas such as pharmaco-epidemiology, decision analysis, modeling, and simulation to play an increasingly important role in comparative benefit-risk assessment.

Nonparametric estimation for cumulative duration of adverse events.

Analysis of adverse events (AE) for drug safety assessment presents challenges to statisticians in observational studies as well as in clinical trials since AEs are typically recurrent with varying duration and severity. Routine analyses often concentrate on the number of patients who had at least one occurrence of a specific AE or a group of AEs, or the time to occurrence of the first event. We argue that other information in AE data particularly cumulative duration of events is also important, particularly for benefit-risk assessment. We propose a nonparametric method to estimate the mean cumulative duration (MCD) based on the nonparametric cumulative mean function estimate, together with a robust estimate for the variance of the estimate, as in Lawless and Nadeau (1995). This approach can be easily used to analyze multiple, overlapped and severity weighted AE durations. This method can also be used for estimating the difference between two MCDs. Estimation in the presence of censoring due to informative dropouts and/or a terminal event is also considered. The method can be implemented in standard softwares such as SAS. We illustrate the use of the method with a numerical example. Small sample properties of this approach are examined via simulation.

A review of risk measures in pharmacoepidemiology with tips for statisticians in the pharmaceutical industry.

Pharmacoepidemiology is the study of the therapeutic effects, risk, and use of drugs in large populations, which applies epidemiological methods and reasoning. As reflected in the recent strengthening of the pharmacovigilance legislation in Europe, greater attention has been placed to epidemiological research in response to an increasing call by the public for further post-marketing studies on the safety and efficacy of drugs. Various measures of risk are used in pharmacoepidemiology to quantify the probability of experiencing an adverse outcome and capture the relative increases in risk between treated and untreated populations: cumulative incidence, incidence rate, absolute risk reduction, relative risk, odds ratio, incidence rate ratio, and time to event outcomes. We review in this paper the commonly used measures of risk in pharmacoepidemiology and provide some practical tips for the industry statistician.

Why a Bayesian approach to safety analysis in pharmacovigilance is important.

Large databases of routinely collected data are a valuable source of information for detecting potential associations between drugs and adverse events (AE). A pharmacovigilance system starts with a scan of these databases for potential signals of drug-AE associations that will subsequently be examined by experts to aid in regulatory decision-making. The signal generation process faces some key challenges: (1) an enormous volume of drug-AE combinations need to be tested (i.e. the problem of multiple testing); (2) the results are not in a format that allows the incorporation of accumulated experience and knowledge for future signal generation; and (3) the signal generation process ignores information captured from other processes in the pharmacovigilance system and does not allow feedback. Bayesian methods have been developed for signal generation in pharmacovigilance, although the full potential of these methods has not been realised. For instance, Bayesian hierarchical models will allow the incorporation of established medical and epidemiological knowledge into the priors for each drug-AE combination. Moreover, the outputs from this analysis can be incorporated into decision-making tools to help in signal validation and posterior actions to be taken by the regulators and companies. We discuss in this paper the apparent advantage of the Bayesian methods used in safety signal generation and the similarities and differences between the two widely used Bayesian methods. We will also propose the use of Bayesian hierarchical models to address the three key challenges and discuss the reasons why Bayesian methodology still have not been fully utilised in pharmacovigilance activities.

Opportunities for minimization of confounding in observational research.

Observational epidemiological studies are increasingly used in pharmaceutical research to evaluate the safety and effectiveness of medicines. Such studies can complement findings from randomized clinical trials by involving larger and more generalizable patient populations by accruing greater durations of follow-up and by representing what happens more typically in the clinical setting. However, the interpretation of exposure effects in observational studies is almost always complicated by non-random exposure allocation, which can result in confounding and potentially lead to misleading conclusions. Confounding occurs when an extraneous factor, related to both the exposure and the outcome of interest, partly or entirely explains the relationship observed between the study exposure and the outcome. Although randomization can eliminate confounding by distributing all such extraneous factors equally across the levels of a given exposure, methods for dealing with confounding in observational studies include a careful choice of study design and the possible use of advanced analytical methods. The aim of this paper is to introduce some of the approaches that can be used to help minimize the impact of confounding in observational research to the reader working in the pharmaceutical industry.

In vivo detection of macrophages in human carotid atheroma: temporal dependence of ultrasmall superparamagnetic particles of iron oxide-enhanced MRI.

UNLABELLED: Background- It has been suggested that inflammatory cells within vulnerable plaques may be visualized by superparamagnetic iron oxide particle-enhanced MRI. The purpose of this study was to determine the time course for macrophage visualization with in vivo contrast-enhanced MRI using an ultrasmall superparamagnetic iron oxide (USPIO) agent in symptomatic human carotid disease. METHODS: Eight patients scheduled for carotid endarterectomy underwent multisequence MRI of the carotid bifurcation before and 24, 36, 48, and 72 hours after Sinerem (2.6 mg/kg) infusion. RESULTS: USPIO particles accumulated in macrophages in 7 of 8 patients given Sinerem. Areas of signal intensity reduction, corresponding to USPIO/macrophage-positive histological sections, were visualized in all 7 of these patients, optimally between 24 and 36 hours, decreasing after 48 hours, but still evident up to 96 hours after infusion. CONCLUSIONS: USPIO-enhanced MRI of carotid atheroma can be used to identify macrophages in vivo. The temporal change in the resultant signal intensity reduction on MRI suggests an optimal time window for the detection of macrophages on postinfusion imaging.

Effect of dose on the frequency of major birth defects following fetal exposure to lamotrigine monotherapy in an international observational study.

Data from the International Lamotrigine Pregnancy Registry were analyzed to examine the effect of maximal first-trimester maternal dose of lamotrigine monotherapy on the risk of major birth defects (MBDs). Among 802 exposures, the frequency of MBDs was 2.7% (95% confidence interval [CI] 1.8-4.2%). The distribution of dose did not differ between infants with and those without MBDs (mean 248.3 milligrams per day [mg/day] and 278.9 mg/day, respectively, median 200 mg/day for both groups). A logistic regression analysis showed no difference in the risk of MBDs as a continuous function of dose (summary odds ratio [OR] per 100 mg increase =0.999, 95% CI 0.996-1.001). There was also no effect of dose, up to 400 mg/day, on the frequency of MBDs.