Atopic dermatitis (AD) has long been associated with Staphylococcus aureus skin colonization or infection and is typically managed with regimens that include antimicrobial therapies. However, the role of microbial communities in the pathogenesis of AD is incompletely characterized. To assess the relationship between skin microbiota and disease progression, 16S ribosomal RNA bacterial gene sequencing was performed on DNA obtained directly from serial skin sampling of children with AD. The composition of bacterial communities was analyzed during AD disease states to identify characteristics associated with AD flares and improvement post-treatment. We found that microbial community structures at sites of disease predilection were dramatically different in AD patients compared with controls. Microbial diversity during AD flares was dependent on the presence or absence of recent AD treatments, with even intermittent treatment linked to greater bacterial diversity than no recent treatment. Treatment-associated changes in skin bacterial diversity suggest that AD treatments diversify skin bacteria preceding improvements in disease activity. In AD, the proportion of Staphylococcus sequences, particularly S. aureus, was greater during disease flares than at baseline or post-treatment, and correlated with worsened disease severity. Representation of the skin commensal S. epidermidis also significantly increased during flares. Increases in Streptococcus, Propionibacterium, and Corynebacterium species were observed following therapy. These findings reveal linkages between microbial communities and inflammatory diseases such as AD, and demonstrate that as compared with culture-based studies, higher resolution examination of microbiota associated with human disease provides novel insights into global shifts of bacteria relevant to disease progression and treatment.

When trying to learn a model for the prediction of an outcome given a set of covariates, a statistician has many estimation procedures in their toolbox. A few examples of these candidate learners are: least squares, least angle regression, random forests, and spline regression. Previous articles (van der Laan and Dudoit (2003); van der Laan et al. (2006); Sinisi et al. (2007)) theoretically validated the use of cross validation to select an optimal learner among many candidate learners. Motivated by this use of cross validation, we propose a new prediction method for creating a weighted combination of many candidate learners to build the super learner. This article proposes a fast algorithm for constructing a super learner in prediction which uses V-fold cross-validation to select weights to combine an initial set of candidate learners. In addition, this paper contains a practical demonstration of the adaptivity of this so called super learner to various true data generating distributions. This approach for construction of a super learner generalizes to any parameter which can be defined as a minimizer of a loss function.

Stratification of women according to their risk of breast cancer based on polygenic risk scores (PRSs) could improve screening and prevention strategies. Our aim was to develop PRSs, optimized for prediction of estrogen receptor (ER)-specific disease, from the largest available genome-wide association dataset and to empirically validate the PRSs in prospective studies. The development dataset comprised 94,075 case subjects and 75,017 control subjects of European ancestry from 69 studies, divided into training and validation sets. Samples were genotyped using genome-wide arrays, and single-nucleotide polymorphisms (SNPs) were selected by stepwise regression or lasso penalized regression. The best performing PRSs were validated in an independent test set comprising 11,428 case subjects and 18,323 control subjects from 10 prospective studies and 190,040 women from UK Biobank (3,215 incident breast cancers). For the best PRSs (313 SNPs), the odds ratio for overall disease per 1 standard deviation in ten prospective studies was 1.61 (95%CI: 1.57–1.65) with area under receiver-operator curve (AUC) = 0.630 (95%CI: 0.628–0.651). The lifetime risk of overall breast cancer in the top centile of the PRSs was 32.6%. Compared with women in the middle quintile, those in the highest 1% of risk had 4.37- and 2.78-fold risks, and those in the lowest 1% of risk had 0.16- and 0.27-fold risks, of developing ER-positive and ER-negative disease, respectively. Goodness-of-fit tests indicated that this PRS was well calibrated and predicts disease risk accurately in the tails of the distribution. This PRS is a powerful and reliable predictor of breast cancer risk that may improve breast cancer prevention programs.

Importance Germline pathogenic variants in BRCA1 and BRCA2 predispose to an increased lifetime risk of breast cancer. However, the relevance of germline variants in other genes from multigene hereditary cancer testing panels is not well defined. Objective To determine the risks of breast cancer associated with germline variants in cancer predisposition genes. Design, Setting, and Participants A study population of 65 057 patients with breast cancer receiving germline genetic testing of cancer predisposition genes with hereditary cancer multigene panels. Associations between pathogenic variants in non-BRCA1 and non-BRCA2 predisposition genes and breast cancer risk were estimated in a case-control analysis of patients with breast cancer and Exome Aggregation Consortium reference controls. The women underwent testing between March 15, 2012, and June 30, 2016. Main Outcomes and Measures Breast cancer risk conferred by pathogenic variants in non-BRCA1 and non-BRCA2 predisposition genes. Results The mean (SD) age at diagnosis for the 65 057 women included in the analysis was 48.5 (11.1) years. The frequency of pathogenic variants in 21 panel genes identified in 41 611 consecutively tested white women with breast cancer was estimated at 10.2%. After exclusion of BRCA1, BRCA2, and syndromic breast cancer genes (CDH1, PTEN, and TP53), observed pathogenic variants in 5 of 16 genes were associated with high or moderately increased risks ofbreast cancer: ATM (OR, 2.78; 95% CI, 2.22-3.62), BARD1 (OR, 2.16; 95% CI, 1.31-3.63), CHEK2 (OR, 1.48; 95% CI, 1.31-1.67), PALB2 (OR, 7.46; 95% CI, 5.12-11.19), and RAD51D (OR, 3.07; 95% CI, 1.21-7.88). Conversely, variants in the BRIP1 and RAD51C ovarian cancer risk genes; the MREHA, RAD50, and NBN MRN complex genes; the MLH1 and PMS2 mismatch repair genes; and NF1 were not associated with increased risks of breast cancer. Conclusions and Relevance This study establishes several panel genes as high- and moderate-risk breast cancer genes and provides estimates of breast cancer risk associated with pathogenic variants in these genes among individuals qualifying for clinical genetic testing.

No abstract