Philip Jarvis

The cynomolgus monkey as a model for developmental toxicity studies: variability of pregnancy losses, statistical power estimates, and group size considerations

BACKGROUND This work evaluates pregnancy and infant loss in 1,069 vehicle-treated cynomolgus monkeys from 78 embryo-fetal development (EFD) studies and 14 pre-postnatal development (PPND) studies accrued during 1981-2007. METHODS Losses were analysed by survival function and hazard ratio using logistic regression for influence of year, study type (e.g., dose duration), and test item route of administration (ig, im, iv, sc). RESULTS Neither study type nor route of dosing affected pregnancy outcome. Losses were higher pre-1990 (104 losses/347 pregnancies) compared to 1990 onwards (94 losses/722 pregnancies). Losses were greatest before gestation day 50 and at parturition. Using post-1989 data, Monte-Carlo simulations of pregnancy outcomes were created. The power associated with the comparison of vehicle survival curves and simulated adverse survival curves was examined. This showed that EFD studies with initial vehicle group sizes of 16 and 20 have an 80% probability of having 13 and 16 ongoing pregnancies at gestational day 100, respectively. For PPND studies with initial vehicle group sizes of 16, 20, or 28, there is an 80% likelihood of having 9, 11, or 16 infants at day 7 post-partum, respectively. A PPND study initiated with group size 20 could detect a threefold increase of test item-related pregnancy or infant loss. CONCLUSIONS For designing and managing primate developmental toxicity studies, this type of analysis provides an objective tool to facilitate decisions either by supplementing groups with additional pregnant animals or stopping a group because an adverse effect on offspring survival has already been adequately revealed.

Recommendations on the statistical analysis of the Comet assay

In 2010, the Statisticians in the Pharmaceutical Industry (PSI) Toxicology Special Interest Group met to discuss the design and analysis of the Comet assay. The Comet assay is one potential component of the package of safety studies required by regulatory bodies. As these studies usually involve a three-way nested experimental design and as the distribution of the measured response is usually either lognormal or lognormal plus a point mass at zero, the analysis is not straightforward. This has led to many different types of analysis being proposed in the literature, with several different methods applied within the pharmaceutical industry itself. This article summarises the PSI Toxicology Groups discussions and recommendations around these issues.

Translational pharmacokinetic-pharmacodynamic modeling of QTc effects in dog and human.

INTRODUCTION Preclinical assessment of the heart rate corrected QT interval (QTc) is an important component of the cardiovascular safety evaluation in drug discovery. Here we aimed to quantify the translational relationship between QTc prolongation and shortening in the conscious telemetered dog and humans by a retrospective pharmacokinetic-pharmacodynamic (PKPD) analysis. METHODS QTc effects of 2 proprietary compounds and 2 reference drugs (moxifloxacin and dofetilide) were quantified in conscious dogs and healthy volunteers via a linear and Emax pharmacokinetic-pharmacodynamic models. The translational relationship was quantified by correlating the QTc response from dog and human at matching free drug concentrations. RESULTS A consistent translational relationship was found at low delta-QTc intervals indicating that a QTc change of 2.5-8 ms in dog would correspond to a 10 ms change in human. DISCUSSION The translational relationship developed here can be used to predict the QTc liability in human using preclinical dog data. It could therefore help protect the health of human volunteers, for example by appropriate clinical study design and dose selection, as well as improve future decision-making and help reduce compound attrition due to changes in QT interval.

A statistician's perspective on biomarkers in drug development

Biomarkers play an increasingly important role in many aspects of pharmaceutical discovery and development, including personalized medicine and the assessment of safety data, with heavy reliance being placed on their delivery. Statisticians have a fundamental role to play in ensuring that biomarkers and the data they generate are used appropriately and to address relevant objectives such as the estimation of biological effects or the forecast of outcomes so that claims of predictivity or surrogacy are only made based upon sound scientific arguments. This includes ensuring that studies are designed to answer specific and pertinent questions, that the analyses performed account for all levels and sources of variability and that the conclusions drawn are robust in the presence of multiplicity and confounding factors, especially as many biomarkers are multidimensional or may be an indirect measure of the clinical outcome. In all of these areas, as in any area of drug development, statistical best practice incorporating both scientific rigor and a practical understanding of the situation should be followed. This article is intended as an introduction for statisticians embarking upon biomarker-based work and discusses these issues from a practising statisticians perspective with reference to examples.

Review of the statistical analysis of the dog telemetry study.

In 2008, the PSI Toxicology Special Interest Group met to discuss the design and analysis of dog telemetry studies. The dog telemetry study is one component of the integrated cardiovascular assessment required by regulatory bodies. Although there are guidelines for these studies, little is said about the statistical analysis. With parameters of interest measured continually over time, in studies typically involving four dogs, the analysis is not straightforward. This has led to many different types of analysis being proposed in the literature, with many different methods applied within the pharmaceutical industry itself. This paper summarizes the PSI Toxicology groups discussions and recommendations around these issues.

On the Relationship between Molecular Hit Rates in High-Throughput Screening and Molecular Descriptors

High-throughput screening (HTS) is widely used in the pharmaceutical industry to identify novel chemical starting points for drug discovery projects. The current study focuses on the relationship between molecular hit rate in recent in-house HTS and four common molecular descriptors: lipophilicity (ClogP), size (heavy atom count, HEV), fraction of sp3-hybridized carbons (Fsp3), and fraction of molecular framework (fMF). The molecular hit rate is defined as the fraction of times the molecule has been assigned as active in the HTS campaigns where it has been screened. Beta-binomial statistical models were built to model the molecular hit rate as a function of these descriptors. The advantage of the beta-binomial statistical models is that the correlation between the descriptors is taken into account. Higher degree polynomial terms of the descriptors were also added into the beta-binomial statistic model to improve the model quality. The relative influence of different molecular descriptors on molecular hit rate has been estimated, taking into account that the descriptors are correlated to each other through applying beta-binomial statistical modeling. The results show that ClogP has the largest influence on the molecular hit rate, followed by Fsp3 and HEV. fMF has only a minor influence besides its correlation with the other molecular descriptors.

Profiling biological samples using ultra performance liquid chromatography–inductively coupled plasma–mass spectrometry (UPLC-ICP-MS) for the determination of phosphorus and sulfur-containing metabolites

In this preliminary study UPLC-ICP-MS has been utilized to profile a range of different bio-fluids and tissue extracts for sulfur and phosphorus-containing metabolites. Particular attention has been given to the livers, plasma and urine from lean and obese Zucker rats, with a view to differentiating between them based solely on their respective sulfur or phosphorus profiles and/or their total sulfur and phosphorus content. In addition, bile and tumour extracts have been analysed to observe the nature of their profiles. To the best of our knowledge this is the first time ICP-MS has been used in a non-targeted metabonomic study. Results have shown lower limits of quantification for sulfur and phosphorus methods of 0.25 and 0.15 ng on column with CVs of 14.7% and 10.9% respectively. Total phosphorus analysis of the Zucker rat aqueous liver extracts, plasma and urine has shown the pattern of phosphorus concentrations to be statistically significantly different in the lean and obese Zucker rats. Chromatographic separation of the Zucker rat organic liver extracts and plasma allowed further differentiation between the lean and obese rats using their phosphorus profiles alone. In conclusion, this preliminary study has shown the potential of UPLC-ICP-MS to quantitatively discriminate between different species biofluids, fluids and tissues based solely on their phosphorus or sulfur concentrations and/or metabolomes.

Phosphorus and sulfur metabonomic profiling of tissue and plasma obtained from tumour‐bearing mice using ultra‐performance liquid chromatography/inductively coupled plasma mass spectrometry

RATIONALE Metabonomic studies use complex biological samples (blood plasma/serum, tissues, etc.) that when analysed with high-performance liquid chromatography/mass spectrometry (HPLC/MS) or nuclear magnetic resonance (NMR) generate profiles that may contain many thousands of features. These profiles can be difficult to interpret with the majority of the features contributing little to the study. As such there is an argument for the development of techniques that can simplify the problem by targeting particular classes of compounds. METHODS In this study ultra-performance liquid chromatography/inductively coupled plasma mass spectrometry (UPLC/ICP-MS) was used to profile tumour tissue and plasma samples for phosphorus- and sulfur-containing metabolites. These samples were xenograft tumours (derived from breast, lung and colon cell lines) and plasma obtained from nude mice. Plasma was also obtained from non-tumour-bearing mice as a control. Due to isobaric interferences this method took advantage of the dynamic reaction cell within the ICP-MS system to react the phosphorus and sulfur ions with oxygen. The PO+ and SO+ ions were then monitored free of interferences. The total phosphorus and sulfur within each sample was also recorded using flow injection ICP-MS. A robust quality control system based on pooled sample replicate analysis was used throughout the study. RESULTS Determination of the total phosphorus and sulfur content of each sample was sufficient in itself for statistical differentiation between the majority of the cell lines analysed. Subsequent reversed-phase chromatographic profiling of the organic tumour and plasma extracts revealed the presence of a number of well-retained phosphorus-containing compounds that showed tumour-specific profiles. Reversed-phase profiling was not suitable for the sulfur-containing compounds which eluted with the solvent front. CONCLUSIONS This study has shown the potential use of UPLC/ICP-MS to differentiate between tumour cell lines, using both plasma and tumour tissue samples, based solely on metabolites that contain phosphorus or sulfur. Whilst further work is required to identify these compounds this methodology shows the ability of the described methods to provide targets for future biomarker discovery studies. Copyright

Assessment of male rodent fertility in general toxicology 6-month studies.

An outcome and statistical review of male reproductive performance assessed by including a mating phase within 6-month general toxicity studies in the Han Wistar rat was undertaken. The basic study design was 16-20 animals per group dosed for approximately 9 weeks before pairing the male rats with undosed females. This design provides opportunity for remating and automatically includes general toxicity parameters. The dose levels used in the 1- and 6-month studies show that male reproduction was assessed at generally similar doses. The majority of males (compound-dosed and controls) mated within 7 days. All vehicle-dosed males mated and 98.5% of these females were pregnant. Modeling shows that a pregnancy rate of less than 14 out of 16 pregnant animals is very unlikely to occur due to biological variability. Power calculations based on vehicle control data show that group sizes of >10 males have a >80% power of detecting a decrease in median of three embryos per group compared with the control group. Even if the number of pregnancies decreased by a third, a group size of ≥12 would still detect a decrement in the median of three embryos with >80% power. Based on the statistical modeling and inherent strengths of the study design, this review indicates that decrements in male reproductive function can be successfully detected by incorporating a mating phase into a 6-month rat study and that a group size of 12-16 is generally adequate rather than the 16-20 group size indicated as a generic default within ICHS5(R2).

An assessment of the statistical methods used to analyse toxicology studies.

The Statisticians in the Pharmaceutical Industry Toxicology Special Interest Group has collated and compared statistical analysis methods for a number of toxicology study types including general toxicology, genetic toxicology, safety pharmacology and carcinogenicity. In this paper, we present the study design, experimental units and analysis methods.