Pretreatment prediction of response to ursodeoxycholic acid in primary biliary cholangitis: development and validation of the UDCA Response Score.

BACKGROUND: Treatment guidelines recommend a stepwise approach to primary biliary cholangitis: all patients begin treatment with ursodeoxycholic acid (UDCA) monotherapy and those with an inadequate biochemical response after 12 months are subsequently considered for second-line therapies. However, as a result, patients at the highest risk can wait the longest for effective treatment. We determined whether UDCA response can be accurately predicted using pretreatment clinical parameters. METHODS: We did logistic regression analysis of pretreatment variables in a discovery cohort of patients in the UK with primary biliary cholangitis to derive the best-fitting model of UDCA response, defined as alkaline phosphatase less than 1·67 times the upper limit of normal (ULN), measured after 12 months of treatment with UDCA. We validated the model in an external cohort of patients with primary biliary cholangitis and treated with UDCA in Italy. Additionally, we assessed correlations between model predictions and key histological features, such as biliary injury and fibrosis, on liver biopsy samples. FINDINGS: 2703 participants diagnosed with primary biliary cholangitis between Jan 1, 1998, and May 31, 2015, were included in the UK-PBC cohort for derivation of the model. The following pretreatment parameters were associated with lower probability of UDCA response: higher alkaline phosphatase concentration (p<0·0001), higher total bilirubin concentration (p=0·0003), lower aminotransferase concentration (p=0·0012), younger age (p<0·0001), longer interval from diagnosis to the start of UDCA treatment (treatment time lag, p<0·0001), and worsening of alkaline phosphatase concentration from diagnosis (p<0·0001). Based on these variables, we derived a predictive score of UDCA response. In the external validation cohort, 460 patients diagnosed with primary biliary cholangitis were treated with UDCA, with follow-up data until May 31, 2016. In this validation cohort, the area under the receiver operating characteristic curve for the score was 0·83 (95% CI 0·79-0·87). In 20 liver biopsy samples from patients with primary biliary cholangitis, the UDCA response score was associated with ductular reaction (r=-0·556, p=0·0130) and intermediate hepatocytes (probability of response was 0·90 if intermediate hepatocytes were absent vs 0·51 if present). INTERPRETATION: We have derived and externally validated a model based on pretreatment variables that accurately predicts UDCA response. Association with histological features provides face validity. This model provides a basis to explore alternative approaches to treatment stratification in patients with primary biliary cholangitis. FUNDING: UK Medical Research Council and University of Milan-Bicocca.

Surveillance versus ablation for incidentally diagnosed small renal tumours: the SURAB feasibility RCT.

BACKGROUND: There is uncertainty around the appropriate management of small renal tumours. Treatments include partial nephrectomy, ablation and active surveillance. OBJECTIVES: To explore the feasibility of a randomised trial of ablation versus active surveillance. DESIGN: Two-stage feasibility study: stage 1 - clinician survey and co-design work; and stage 2 - randomised feasibility study with qualitative and economic components. METHODS: Stage 1 - survey of radiologists and urologists, and development of patient information materials. Stage 2 - patients identified across eight UK centres with small renal tumours (< 4 cm) were randomised (1 : 1 ratio) to ablation or active surveillance in an unblinded manner. Randomisation was carried out by a central computer system. The primary objective was to determine willingness to participate and to randomise a target of 60 patients. The qualitative and economic data were collected separately. RESULTS: The trial was conducted across eight centres, with a site-specific period of recruitment ranging from 3 to 11 months. Of the 154 patients screened, 36 were eligible and were provided with study details. Seven agreed to be randomised and one patient was found ineligible following biopsy results. Six patients (17% of those eligible) were randomised: three patients received ablation and no serious adverse events were recorded. The 3- and 6-month data were collected for four (67%) and three (50%) out of the six patients, respectively. The qualitative substudy identified factors directly impacting on the recruitment of this trial. These included patient and clinician preferences, organisational factors (variation in clinical pathway) and standard treatment not included. The health economic questionnaire was designed and piloted; however, the sample size of recruited patients was insufficient to draw a conclusion on the feasibility of the health economics. CONCLUSIONS: The trial did not meet the criteria for progression and the recruitment rate was lower than hypothesised, demonstrating that a full trial is presently not possible. The qualitative study identified factors that led to variation in recruitment across the sites. Implementation of organisational and operational measures can increase recruitment in any future trial. There was insufficient information to conduct a full economic analysis. TRIAL REGISTRATION: Current Controlled Trials ISRCTN31161700. FUNDING: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 21, No. 81. See the NIHR Journals Library website for further project information.

A comparison of methods for inferring causal relationships between genotype and phenotype using additional biological measurements.

Genome wide association studies (GWAS) have been very successful over the last decade at identifying genetic variants associated with disease phenotypes. However, interpretation of the results obtained can be challenging. Incorporation of further relevant biological measurements (e.g. 'omics' data) measured in the same individuals for whom we have genotype and phenotype data may help us to learn more about the mechanism and pathways through which causal genetic variants affect disease. We review various methods for causal inference that can be used for assessing the relationships between genetic variables, other biological measures, and phenotypic outcome, and present a simulation study assessing the performance of the methods under different conditions. In general, the methods we considered did well at inferring the causal structure for data simulated under simple scenarios. However, the presence of an unknown and unmeasured common environmental effect could lead to spurious inferences, with the methods we considered displaying varying degrees of robustness to this confounder. The use of causal inference techniques to integrate omics and GWAS data has the potential to improve biological understanding of the pathways leading to disease. Our study demonstrates the suitability of various methods for performing causal inference under several biologically plausible scenarios.

Using gene expression data to identify causal pathways between genotype and phenotype in a complex disease: application to Genetic Analysis Workshop 19.

We explore causal relationships between genotype, gene expression and phenotype in the Genetic Analysis Workshop 19 data. We compare the use of structural equation modeling and a Bayesian unified framework approach to infer the most likely causal models that gave rise to the data. Testing an exhaustive set of causal relationships between each single-nucleotide polymorphism, gene expression probe, and phenotype would be computationally infeasible, thus a filtering step is required. In addition to filtering based on pairwise associations, we consider weighted gene correlation network analysis as a method of clustering genes with similar function into a small number of modules. These modules capture the key functional mechanisms of genes while greatly reducing the number of relationships to test for in causal modeling.

A Variant in the BACH2 Gene Is Associated With Susceptibility to Autoimmune Addison's Disease in Humans.

CONTEXT: Autoimmune Addison's disease (AAD) is a rare but highly heritable condition. The BACH2 protein plays a crucial role in T lymphocyte maturation, and allelic variation in its gene has been associated with a number of autoimmune conditions. OBJECTIVE: We aimed to determine whether alleles of the rs3757247 single nucleotide polymorphism (SNP) in the BACH2 gene are associated with AAD. DESIGN, SETTING, AND PATIENTS: This case-control association study was performed in two phases using Taqman chemistry. In the first phase, the rs3757247 SNP was genotyped in 358 UK AAD subjects and 166 local control subjects. Genotype data were also available from 5154 healthy UK controls from the Wellcome Trust (WTCCC2) for comparison. In the second phase, the SNP was genotyped in a validation cohort comprising 317 Norwegian AAD subjects and 365 controls. RESULTS: The frequency of the minor T allele was significantly higher in subjects with AAD from the United Kingdom compared to both the local and WTCCC2 control cohorts (58% vs 45 and 48%, respectively) (local controls, P = 1.1 × 10-4; odds ratio [OR], 1.68; 95% confidence interval [CI], 1.29-2.18; WTCCC2 controls, P = 1.4 × 10-6; OR, 1.44; 95% CI, 1.23-1.69). This finding was replicated in the Norwegian validation cohort (P = .0015; OR, 1.41; 95% CI, 1.14-1.75). Subgroup analysis showed that this association is present in subjects with both isolated AAD (OR, 1.53; 95% CI, 1.22-1.92) and autoimmune polyglandular syndrome type 2 (OR, 1.37; 95% CI, 1.12-1.69) in the UK cohort, and with autoimmune polyglandular syndrome type 2 in the Norwegian cohort (OR, 1.58; 95% CI, 1.22-2.06). CONCLUSION: We have demonstrated, for the first time, that allelic variability at the BACH2 locus is associated with susceptibility to AAD. Given its association with multiple autoimmune conditions, BACH2 can be considered a "universal" autoimmune susceptibility locus.

Investigation of maternal effects, maternal-fetal interactions and parent-of-origin effects (imprinting), using mothers and their offspring.

Many complex genetic effects, including epigenetic effects, may be expected to operate via mechanisms in the inter-uterine environment. A popular design for the investigation of such effects, including effects of parent-of-origin (imprinting), maternal genotype, and maternal-fetal genotype interactions, is to collect DNA from affected offspring and their mothers (case/mother duos) and to compare with an appropriate control sample. An alternative design uses data from cases and both parents (case/parent trios) but does not require controls. In this study, we describe a novel implementation of a multinomial modeling approach that allows the estimation of such genetic effects using either case/mother duos or case/parent trios. We investigate the performance of our approach using computer simulations and explore the sample sizes and data structures required to provide high power for detection of effects and accurate estimation of the relative risks conferred. Through the incorporation of additional assumptions (such as Hardy-Weinberg equilibrium, random mating and known allele frequencies) and/or the incorporation of additional types of control sample (such as unrelated controls, controls and their mothers, or both parents of controls), we show that the (relative risk) parameters of interest are identifiable and well estimated. Nevertheless, parameter interpretation can be complex, as we illustrate by demonstrating the mathematical equivalence between various different parameterizations. Our approach scales up easily to allow the analysis of large-scale genome-wide association data, provided both mothers and affected offspring have been genotyped at all variants of interest.