Winham, Stacey J.

Abstract Background: Clinical use of breast cancer risk prediction requires simplified models. We evaluate a simplified version of the validated Rosner–Colditz model and add percent mammographic density (MD) and polygenic risk score (PRS), to assess performance from ages 45–74. We validate using the Mayo Mammography Health Study (MMHS). Methods: We derived the model in the Nurses' Health Study (NHS) based on: MD, 77 SNP PRS and a questionnaire score (QS; lifestyle and reproductive factors). A total of 2,799 invasive breast cancer cases were diagnosed from 1990–2000. MD (using Cumulus software) and PRS were assessed in a nested case–control study. We assess model performance using this case–control dataset and evaluate 10-year absolute breast cancer risk. The prospective MMHS validation dataset includes 21.8% of women age <50, and 434 incident cases identified over 10 years of follow-up. Results: In the NHS, MD has the highest odds ratio (OR) for 10-year risk prediction: ORper SD = 1.48 [95% confidence interval (CI): 1.31–1.68], followed by PRS, ORper SD = 1.37 (95% CI: 1.21–1.55) and QS, ORper SD = 1.25 (95% CI: 1.11–1.41). In MMHS, the AUC adjusted for age + MD + QS 0.650; for age + MD + QS + PRS 0.687, and the NRI was 6% in cases and 16% in controls. Conclusion: A simplified assessment of QS, MD, and PRS performs consistently to discriminate those at high 10-year breast cancer risk. Impact: This simplified model provides accurate estimation of 10-year risk of invasive breast cancer that can be used in a clinical setting to identify women who may benefit from chemopreventive intervention. See related commentary by Tehranifar et al., p. 587

Abstract Background: Accumulating evidence suggests a relationship between endometrial cancer and ovarian cancer. Independent genome-wide association studies (GWAS) for endometrial cancer and ovarian cancer have identified 16 and 27 risk regions, respectively, four of which overlap between the two cancers. We aimed to identify joint endometrial and ovarian cancer risk loci by performing a meta-analysis of GWAS summary statistics from these two cancers. Methods: Using LDScore regression, we explored the genetic correlation between endometrial cancer and ovarian cancer. To identify loci associated with the risk of both cancers, we implemented a pipeline of statistical genetic analyses (i.e., inverse-variance meta-analysis, colocalization, and M-values) and performed analyses stratified by subtype. Candidate target genes were then prioritized using functional genomic data. Results: Genetic correlation analysis revealed significant genetic correlation between the two cancers (rG = 0.43, P = 2.66 × 10−5). We found seven loci associated with risk for both cancers (PBonferroni < 2.4 × 10−9). In addition, four novel subgenome-wide regions at 7p22.2, 7q22.1, 9p12, and 11q13.3 were identified (P < 5 × 10−7). Promoter-associated HiChIP chromatin loops from immortalized endometrium and ovarian cell lines and expression quantitative trait loci data highlighted candidate target genes for further investigation. Conclusions: Using cross-cancer GWAS meta-analysis, we have identified several joint endometrial and ovarian cancer risk loci and candidate target genes for future functional analysis. Impact: Our research highlights the shared genetic relationship between endometrial cancer and ovarian cancer. Further studies in larger sample sets are required to confirm our findings.

Abstract Statistically significant findings are more likely to be published than non-significant or null findings, leaving scientists and healthcare personnel to make decisions based on distorted scientific evidence. Continuously expanding ´file drawers’ of unpublished data from well-designed experiments waste resources creates problems for researchers, the scientific community and the public. There is limited awareness of the negative impact that publication bias and selective reporting have on the scientific literature. Alternative publication formats have recently been introduced that make it easier to publish research that is difficult to publish in traditional peer reviewed journals. These include micropublications, data repositories, data journals, preprints, publishing platforms, and journals focusing on null or neutral results. While these alternative formats have the potential to reduce publication bias, many scientists are unaware that these formats exist and don’t know how to use them. Our open source file drawer data liberation effort (fiddle) tool (RRID:SCR_017327 available at: http://s-quest.bihealth.org/fiddle/) is a match-making Shiny app designed to help biomedical researchers to identify the most appropriate publication format for their data. Users can search for a publication format that meets their needs, compare and contrast different publication formats, and find links to publishing platforms. This tool will assist scientists in getting otherwise inaccessible, hidden data out of the file drawer into the scientific community and literature. We briefly highlight essential details that should be included to ensure reporting quality, which will allow others to use and benefit from research published in these new formats.

AbstractProblemPrevious studies identified circulating CD14+HLA‐DRlo/− monocytic cells as an immune suppressive subset in solid malignancies, such as prostate, renal cell carcinoma, and pancreatic cancer. Such monocytic cells have been implicated not only in tumour progression but also as a potential barrier for immunotherapy. This study examined the relationship between the frequency of circulating monocytic cells and epithelial ovarian cancer (EOC) progression pre‐ and post‐frontline chemotherapy, defined by disease stage, which is a leading prognostic factor for this malignancy.Method of studyIncident cases of 236 women with EOC were recruited and comprehensive flow cytometry was utilized to assess the frequency of peripheral blood CD33+CD11b+HLA‐DR−/lowCD14+CD15− monocytic cells, henceforth termed CD14+HLA‐DRlo/− monocytic cells, prior to and after completion of frontline chemotherapy. Multivariable odds ratios (OR) were used to estimate the association between CD14+HLA‐DRlo/− monocytic cell percentages and disease stage. Wilcoxon signed‐rank tests evaluated changes in these monocytic cell levels pre‐ and post‐chemotherapy in a patient subset (n = 70).ResultsPatients with elevated frequencies of circulating CD14+HLA‐DRlo/− monocytic cells at diagnosis were at 3.33‐fold greater odds of having advanced stage (III/IV) EOC (CI: 1.04‐10.64), with a significant trend in increasing CD14+HLA‐DRlo/−monocytic cell levels (P = .04). There was a 2.02% median decrease of these monocytic cells post‐chemotherapy among a subset of patients with advanced stage disease (P < .0001).ConclusionThese findings support the potential clinical relevance of CD14+HLA‐DRlo/−monocytic cells in EOC for prognosis and may indicate a non‐invasive biomarker to measure disease progression.

Abstract Copy number alterations and loss of heterozygosity (LOH) are common in ovarian tumors, including those on the X chromosome. In females, one of the two copies of the X chromosome is transcriptionally silenced during embryonic development in the epigenetic process of X chromosome inactivation. Specifically, the long non-coding RNA XIST is expressed from the inactive X (Xi) and triggers chromosome-wide DNA methylation. In breast and ovarian cancer cell line studies, Xi deletion with concomitant duplication of the active X (Xa) has been observed. We sought to examine which copy of the X chromosome, Xi or Xa, was gained or lost in a set of primary ovarian tumors, using levels of XIST expression as a marker of X chromosome inactivation. To determine which copy of the X chromosome harbors the alterations, we compared local copy number and LOH in Xq13.2 to the levels of XIST expression in a set of N=134 primary tumors from Mayo Clinic ovarian cancer patients. Associations with XIST expression were examined using ANOVA, with copy number of two and absence of LOH as the reference level. Associations between expression of XIST and expression of other genes were examined with a mixed effects model, to account for the repeated observations across patients, after adjusting for local copy number and LOH. Associations with XIST expression and overall survival were performed with Cox regression. Additional analyses were performed in a subset of 24 patient tumors with X chromosome-wide alterations to simplify interpretation, including a visual examination of chromosome-wide DNA methylation patterns. Analyses were replicated in N=106 ovarian tumors from TCGA with X chromosome-wide alterations. Across N=134 ovarian tumors from Mayo Clinic, abnormal copy number (not equal to two) was associated with reduced XIST expression (p=0.0016). Levels of XIST expression were similar for tumors with one copy and tumors with three or more copies. Similarly, levels of XIST expression were also lower for tumors with LOH (p=5.2E-5). Reduced XIST expression was also associated with poorer survival, after adjusting for age (p=0.02). Results were consistent in the subset of tumors with whole chromosome alterations (pCN=0.0013, pLOH=0.0019). For these tumors, when LOH was present, XIST expression was low and the DNA methylation distribution is bi-modal, suggesting that the inactive X was lost and the active X was gained. Furthermore, after adjusting for copy number and LOH, XIST expression was weakly positively correlated with the median expression of other genes (R2=0.04, p=0.026). In N=106 TCGA tumors with whole X chromosome alterations, XIST expression was also reduced in tumors with abnormal copy number (p=0.006) and with LOH (p=1.5E-10). Taken together, because XIST is expressed from Xi and not Xa, associations of reduced XIST expression with abnormal copy number and presence of LOH suggest that Xi is frequently lost and Xa is gained in ovarian tumors, consistent with cell line studies. These results should be further validated in larger sample sets and examined for association with clinical outcome. Citation Format: Yilin Song, Bryan M. McCauley, Melissa C. Larson, Sebastian M. Armasu, Kate Lawrenson, Ellen L. Goode, Nicholas B. Larson, Stacey J. Winham. Evidence of decreased X chromosome inactivation in primary ovarian tumors [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2452.

Abstract Background: Biopsy diagnosis of benign breast disease (BBD) based on the most severe lesion in a sample predicts future breast cancer risk and has implications for screening and management. Lobules are the functional unit of the breast and the structures from which BBD arises. We developed and preliminarily validated an automated computational pathology algorithm to discriminate normal from BBD breast lobules as a step toward automated comprehensive characterization of benign biopsies. Methods: 152 BBD biopsies (27 training, 125 validation) from the Mayo Clinic were examined using scanned digital images of H&E stained sections. For each image, a pathologist annotated up to 10 representative lobules as normal or BBD using standard pathology criteria. A deep learning algorithm to quantify lobular features was developed using 129 lobules from 27 subjects. Nine features were identified that discriminate normal vs BBD lobules, expressing lobule size, acini size and number, acinar lumen size, proportion of lobular stroma, and capillaries. Here, we validate their performance to discriminate normal vs BBD lobules in a set of 1250 lobules from 125 subjects using area under the ROC curve (AUC) analysis. Random forest analysis was used for multivariable modeling; model performance was assessed with a tenfold cross-validation approach. Results: Median subject age was 52 years. Among the 125 validation subjects, BBD findings were nonproliferative in 39%, proliferative in 45%, and atypical hyperplasia in 16%. Sections included 552 (44%) normal lobules and 698 BBD (56%) lobules, with representation of both lobule types on each section. In univariate analyses, four individual features showed good discrimination between normal vs. BBD lobules, yielding the following AUCs: lobule size= 0.74, mean acini size= 0.75, epithelial area= 0.75, and number of acini with large lumens= 0.76. Lobule size and epithelial area were highly correlated (Spearman rank correlation r = 0.95), but both of these features showed lower correlation with mean acini size and number of acini with large lumens (each r &lt; 0.5). With random forest modeling, number of acini with large lumens was the strongest discriminating feature, followed by mean acini size, lobule size, and epithelial area. With ten-fold cross validation of the multivariable random forest model, the overall AUC was 0.82 (95% CI: 0.79-0.85). Conclusion: Our validation showed that automated quantitative computational pathology assessment of breast lobules can discriminate normal versus BBD on a per lobule basis. This finding supports the feasibility of developing automated algorithms to classify every lesion in breast biopsies, expanding beyond visual assessment. Further studies using deep learning may reveal novel pathology features for classifying BBD biopsies with the potential to strengthen estimation of breast cancer risk. Citation Format: Amy C. Degnim, Thomas de Bel, Mark E. Sherman, Derek C. Radisky, Stacey J. Winham, Tanya L. Hoskin, Melody L. Stallings Mann, Marlene Frost, Robert A. Vierkant, Brendan T. Broderick, Ethan P. Heinzen, Rushin Brahmbhatt, Muhammad Arshad, Celine M. Vachon, Jodi M. Carter, Lori A. Denison, Daniel W. Visscher, Jeroen van der Laak. Discrimination of benign breast disease from normal lobules using an automated computational pathology algorithm [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2113.

Abstract Over one million women in the United States receive biopsy diagnoses of benign breast disease (BBD) each year, which confer a 1.5–4.0-fold increase in breast cancer risk. Studies in the general population suggest that nonsteroidal anti-inflammatory agents (NSAID) lower breast cancer risk; however, associations among women with BBD are unknown. We assessed whether NSAID use among women diagnosed with BBD is associated with lower breast cancer risk. Participants included 3,080 women (mean age = 50.3 ± 13.5 years) in the Mayo BBD surgical biopsy cohort diagnosed between January 1, 1992 and December 31, 2001 who completed breast cancer risk factor questionnaires that assessed NSAID use, and whose biopsies underwent detailed pathology review, masked to outcome. Women were followed from date of BBD biopsy to breast cancer diagnosis (main outcome) or censoring (death, prophylactic mastectomy, reduction mammoplasty, lobular carcinoma in situ or last contact). Median follow-up time was 16.4 ± 6.0 years. Incident breast cancer was diagnosed among 312 women over a median follow-up of 9.9 years. Regular non-aspirin NSAID use was associated with lower breast cancer risk [HR = 0.63; 95% confidence interval (CI) = 0.46–0.85; P = 0.002] with trends of lower risk (highest tertiles of use vs. nonuse) for greater number of years used [HR = 0.55; 95% CI = 0.31–0.97; Ptrend = 0.003), days used per month (HR = 0.51; 95% CI = 0.33–0.80; Ptrend = 0.001) and lifetime number of doses taken (HR = 0.53; 95% CI = 0.31–0.89; Ptrend = 0.003). We conclude that nonaspirin NSAID use is associated with statistically significant lower breast cancer risk after a BBD biopsy, including a dose–response effect, suggesting a potential role for NSAIDs in breast cancer prevention among patients with BBD.

Abstract Purpose: Gene expression–based molecular subtypes of high-grade serous tubo-ovarian cancer (HGSOC), demonstrated across multiple studies, may provide improved stratification for molecularly targeted trials. However, evaluation of clinical utility has been hindered by nonstandardized methods, which are not applicable in a clinical setting. We sought to generate a clinical grade minimal gene set assay for classification of individual tumor specimens into HGSOC subtypes and confirm previously reported subtype-associated features. Experimental Design: Adopting two independent approaches, we derived and internally validated algorithms for subtype prediction using published gene expression data from 1,650 tumors. We applied resulting models to NanoString data on 3,829 HGSOCs from the Ovarian Tumor Tissue Analysis consortium. We further developed, confirmed, and validated a reduced, minimal gene set predictor, with methods suitable for a single-patient setting. Results: Gene expression data were used to derive the predictor of high-grade serous ovarian carcinoma molecular subtype (PrOTYPE) assay. We established a de facto standard as a consensus of two parallel approaches. PrOTYPE subtypes are significantly associated with age, stage, residual disease, tumor-infiltrating lymphocytes, and outcome. The locked-down clinical grade PrOTYPE test includes a model with 55 genes that predicted gene expression subtype with &gt;95% accuracy that was maintained in all analytic and biological validations. Conclusions: We validated the PrOTYPE assay following the Institute of Medicine guidelines for the development of omics-based tests. This fully defined and locked-down clinical grade assay will enable trial design with molecular subtype stratification and allow for objective assessment of the predictive value of HGSOC molecular subtypes in precision medicine applications. See related commentary by McMullen et al., p. 5271

Purpose The known epithelial ovarian cancer (EOC) susceptibility genes account for less than 50% of the heritable risk of ovarian cancer suggesting that other susceptibility genes exist. The aim of this study was to evaluate the contribution to ovarian cancer susceptibility of rare deleterious germline variants in a set of candidate genes. Methods We sequenced the coding region of 54 candidate genes in 6385 invasive EOC cases and 6115 controls of broad European ancestry. Genes with an increased frequency of putative deleterious variants in cases versus controls were further examined in an independent set of 14 135 EOC cases and 28 655 controls from the Ovarian Cancer Association Consortium and the UK Biobank. For each gene, we estimated the EOC risks and evaluated associations between germline variant status and clinical characteristics. Results The ORs associated for high-grade serous ovarian cancer were 3.01 for PALB2 (95% CI 1.59 to 5.68; p=0.00068), 1.99 for POLK (95% CI 1.15 to 3.43; p=0.014) and 4.07 for SLX4 (95% CI 1.34 to 12.4; p=0.013). Deleterious mutations in FBXO10 were associated with a reduced risk of disease (OR 0.27, 95% CI 0.07 to 1.00, p=0.049). However, based on the Bayes false discovery probability, only the association for PALB2 in high-grade serous ovarian cancer is likely to represent a true positive. Conclusions We have found strong evidence that carriers of PALB2 deleterious mutations are at increased risk of high-grade serous ovarian cancer. Whether the magnitude of risk is sufficiently high to warrant the inclusion of PALB2 in cancer gene panels for ovarian cancer risk testing is unclear; much larger sample sizes will be needed to provide sufficiently precise estimates for clinical counselling.

Objective Our aim was to evaluate the relationship between evening chronotype, a proxy marker of circadian system dysfunction, and disordered eating behavior and poor dietary habits in individuals with bipolar disorder (BD). Methods In this cross‐sectional study, we evaluated 783 adults with BD. Chronotype was determined using item 5 from the reduced Morningness‐Eveningness Questionnaire. The Eating Disorder Diagnostic Scale (EDDS) and the Rapid Eating Assessment for Participants—Shortened Version (REAP‐S) were used to assess disordered eating behavior and dietary habits respectively. General linear models and logistic regression models were utilized to evaluate differences between chronotype groups. Results Two hundred and eight (27%) BD participants self‐identified as having evening chronotypes. Compared to non‐evening types, evening types were younger ( P < 0.01) and, after controlling for age, had higher mean EDDS composite z‐scores ( P < 0.01); higher rates of binge‐eating (BE) behavior ( P = 0.04), bulimia nervosa ( P < 0.01), and nocturnal eating binges ( P < 0.01); and a higher body mass index ( P = 0.04). Compared to non‐evening types, evening chronotypes had a lower REAP‐S overall score ( P < 0.01) and scored lower on the ‘healthy foods’ and ‘avoidance of unhealthy food’ factors. Evening types also skipped breakfast more often ( P < 0.01), ate less fruit ( P = 0.02) and vegetables ( P = 0.04), and consumed more fried foods ( P < 0.01), unhealthy snacks ( P = 0.02), and soft drinks ( P = 0.01). Conclusions Our findings suggest that the circadian system plays a role in the disordered eating and unhealthy dietary behaviors observed in BD patients. The circadian system may therefore represent a therapeutic target in BD‐associated morbidity that warrants further investigation.