Since 1998, the US Food and Drug Administration (FDA) has been exploring new automated and rapid Bayesian data mining techniques. These techniques have been used to systematically screen the FDA's huge MedWatch database of voluntary reports of adverse drug events for possible events of concern. The data mining method currently being used is the Multi-Item Gamma Poisson Shrinker (MGPS) program that replaced the Gamma Poisson Shrinker (GPS) program we originally used with the legacy database. The MGPS algorithm, the technical aspects of which are summarised in this paper, computes signal scores for pairs, and for higher-order (e.g. triplet, quadruplet) combinations of drugs and events that are significantly more frequent than their pair-wise associations would predict. MGPS generates consistent, redundant, and replicable signals while minimising random patterns. Signals are generated without using external exposure data, adverse event background information, or medical information on adverse drug reactions. The MGPS interface streamlines multiple input-output processes that previously had been manually integrated. The system, however, cannot distinguish between already-known associations and new associations, so the reviewers must filter these events. In addition to detecting possible serious single-drug adverse event problems, MGPS is currently being evaluated to detect possible synergistic interactions between drugs (drug interactions) and adverse events (syndromes), and to detect differences among subgroups defined by gender and by age, such as paediatrics and geriatrics. In the current data, only 3.4% of all 1.2 million drug-event pairs ever reported (with frequencies > or = 1) generate signals [lower 95% confidence interval limit of the adjusted ratios of the observed counts over expected (O/E) counts (denoted EB05) of > or = 2]. The total frequency count that contributed to signals comprised 23% (2.4 million) of the total number, 10.4 million of drug-event pairs reported, greatly facilitating a more focused follow-up and evaluation. The algorithm provides an objective, systematic view of the data alerting reviewers to critically important, new safety signals. The study of signals detected by current methods, signals stored in the Center for Drug Evaluation and Research's Monitoring Adverse Reports Tracking System, and the signals regarding cerivastatin, a cholesterol-lowering drug voluntarily withdrawn from the market in August 2001, exemplify the potential of data mining to improve early signal detection. The operating characteristics of data mining in detecting early safety signals, exemplified by studying a drug recently well characterised by large clinical trials confirms our experience that the signals generated by data mining have high enough specificity to deserve further investigation. The application of these tools may ultimately improve usage recommendations.

After several drugs were removed from the market in recent years because of death due to ventricular tachycardia resulting from drug-induced QT prolongation (Khongphatthanayothin et al., 1998; Lasser et al., 2002; Pratt et al., 1994; Wysowski et al., 2001), the ICH Regulatory agencies requested all sponsors of new drugs to conduct a clinical study, named a Thorough QT/QTc (TQT) study, to assess any possible QT prolongation due to the study drug. The final version of the ICH E14 guidance (ICH, 2005) for "The Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Nonantiarrhythmic Drugs" was released in May 2005. The purpose of the ICH E14 guidance (ICH, 2005) is to provide recommendations to sponsors concerning the design, conduct, analysis, and interpretation of clinical studies to assess the potential of a drug to delay cardiac repolarization. The guideline, however, is not specific on several issues. In this paper, we try to address some statistical issues, including study design, primary statistical analysis, assay sensitivity analysis, and the calculation of the sample size for a TQT study.

Three breast cancer risk factors were evaluated in terms of their interactions with radiation dose in a case-control interview study of Japanese A-bomb survivors. Cases and controls were matched on age at the time of the bombings and radiation dose, and dose-related risk was estimated from cohort rather than case-control data. Each factor--age at first full-term pregnancy, number of deliveries, and cumulative lactation period summed over births--conformed reasonably well to a multiplicative interaction model with radiation dose (the additive interactive model, in which the absolute excess risk associated with a factor is assumed to be independent of radiation dose, was rejected). An important implication of the finding is that early age at first full-term pregnancy, multiple births, and lengthy cumulative lactation are all protective against radiation-related, as well as baseline, breast cancer. Analyses by age at exposure to radiation suggest that, among women exposed to radiation in childhood or adolescence, a first full-term pregnancy at an early age following exposure may be protective against radiation-related risk.

Pharmacologists and other biologists frequently use methods based on the interpretation of isobolograms to quantify the extent of synergy or antagonism between drugs used in combination in pre-clinical studies. Most methods have been unsatisfactory from a statistical viewpoint, many because they have relied solely on visual evaluation, others because the methods have not taken into account the variability of the measurements. We describe a direct approach for quantifying the joint potency of two drugs, a central feature being the use of simple isobole models that lead directly to response surface models for the expected experimental outcomes. The approach is general in the sense that one can use it for discrete or continuous responses, different underlying probability distributions, linear or non-linear dose-response functions of the drugs used singly, and a variety of experimental designs. Our approach extends the suggestions made by Hewlett for measuring the joint potency of drugs, and is similar in spirit to the approaches proposed by Greco et al. and Weinstein et al. We describe the analysis of data from an in vitro experiment conducted to evaluate the efficacy of the antiviral drugs AZT and ddI used in combination.