Winham, Stacey J.

Abstract We evaluated the joint associations between a new 313-variant PRS (PRS313) and questionnaire-based breast cancer risk factors for women of European ancestry, using 72 284 cases and 80 354 controls from the Breast Cancer Association Consortium. Interactions were evaluated using standard logistic regression and a newly developed case-only method for breast cancer risk overall and by estrogen receptor status. After accounting for multiple testing, we did not find evidence that per-standard deviation PRS313 odds ratio differed across strata defined by individual risk factors. Goodness-of-fit tests did not reject the assumption of a multiplicative model between PRS313 and each risk factor. Variation in projected absolute lifetime risk of breast cancer associated with classical risk factors was greater for women with higher genetic risk (PRS313 and family history) and, on average, 17.5% higher in the highest vs lowest deciles of genetic risk. These findings have implications for risk prevention for women at increased risk of breast cancer.

Abstract Background Prior epidemiologic studies have shown an inverse relationship of aspirin and NSAID use with breast cancer (BC) risk in the general population. Women with benign breast disease (BBD) are at an increased BC risk compared with women in the general population, and we have shown that BBD biopsies contain increased numbers of immune cells compared with normal breast tissues donated for research in the Komen Tissue Bank. We hypothesized that NSAID use among women with BBD is associated with reduced risk of BC. Methods We evaluated BC risk in relation to aspirin and NSAID use in a single institution retrospective cohort of 4,498 women who received a biopsy diagnosis of BBD between 1/1/1992 and 12/31/2001. Information on usual medication use, including average number of days used per month (0, 1-7, 8-28 and >28), was collected via questionnaires mailed after identification of participants. NSAIDs were categorized into three classes: 1) Aspirin, 2) Ibuprofen (plus other prostaglandin synthesis inhibitors), and 3) COX inhibitors. BC events were captured via medical record, questionnaires, and Tumor Registry records. Medication use was evaluated for associations with clinical and pathology features. Hazard ratios (HR) and 95% confidence intervals (CI) examining associations between medication use and BC risk were estimated using Cox regression analyses, with adjustment for age and BMI at biopsy, years between biopsy and questionnaire, and pathologic features related to BC risk (severity of BBD and degree of involution in background lobules). Tests for trend examining frequency of use were calculated from ordered values (lowest to highest) in Cox models with a one degree-of-freedom linear term. We assessed BC associations with any type of NSAID use, as well as the three subclasses. Results Information on NSAID use was available for 3,089 (69%) of the participants in the cohort; these women had a median age of 49 and did not differ in BC risk from women who did not provide data on NSAID use (p=0.46). Of the 3,089 respondents, 313 (10.1%) subsequently developed BC with median follow-up of 17 years. Age was significantly associated with use of all types of NSAIDs, with increasing use in older women. BMI at the time of biopsy, severity of BBD findings, and involution in background lobules were also associated with NSAID use; thus these factors and age were adjusted for in risk association analyses. Ever users of NSAIDs (any type) had a significantly lower risk of BC than never users (HR 0.76, CI 0.59-0.98, p=0.03), and increasing frequency of NSAID use was significantly associated with lower BC risk (HRs 0.81, 0.79, and 0.70 for regular use 1-7, 8-28, and >28 days/month), trend p=0.02. Analyses of drug classes revealed that ever use of ibuprofen (HR 0.61, CI 0.42-0.89, p=0.01) and ever use of COX inhibitors (HR 0.59, CI 0.39-0.89, p=0.01) were associated with decreased BC risk. COX inhibitors in particular showed a strong dose response effect with increasing use (HRs 0.97, 0.52, and 0.47 for regular use 1-7, 8-28, and >28 days/month), trend p<0.01. Use of aspirin was not associated with BC risk (HR 0.88, CI 0.64-1.21, p=0.44). Conclusions In this exploratory analysis, we found use of ibuprofen and other NSAIDs to be associated with decreased BC risk among women who had undergone a BBD biopsy. These findings suggest that anti-inflammatory approaches may have value in chemoprevention among women with BBD. Citation Format: Amy Degnim, Robert Vierkant, Stacey Winham, Marlene Frost, Daniel Visscher, Jodi Carter, Suneetha Kaggal, Julie Cunningham, Teresa Allers, Tanya Hoskin, Ethan Heinzen, Celine Vachon, Keith Knutson, Keith Knutson, Derek Radisky, Mark Sherman. NSAID use and breast cancer risk in a retrospective cohort of women with benign breast disease [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P5-08-06.

Abstract Background: Alcohol intake and breast density are established risk factors for breast cancer (BC). A few studies suggest breast density is on the causal pathway between alcohol and BC but the literature is not consistent. We examine the interrelationship between alcohol, volumetric breast density and BC risk, specifically whether there is a stronger alcohol and BC association among women with dense breasts and/or whether breast density mediates the alcohol and BC association. Methods: We conducted a nested case-control study within the San Francisco Mammography Registry consisting of four breast screening centers. From 2006-2013, through linkages with the California cancer registry, we identified 2572 women diagnosed with BC who had screening mammograms performed at least six months prior to diagnosis. Controls (5119) were matched to cases on age, date of earliest mammogram, race/ethnicity, facility, and mammography machine. We obtained the raw format of digital mammograms on average 3.1 (standard deviation (SD) =1.7) years prior to diagnosis or corresponding date for controls. We ascertained usual daily alcohol intake and other risk factors from a clinical questionnaire at time of mammography. We obtained BI-RADS density from clinical records and used VolparaTM to assess volumetric percent density, dense volume and non-dense volume from the mammogram. We examined the associations of daily alcohol intake (none, one or less, two or more drinks per day) and volumetric density phenotypes (per 1 SD) with BC risk using logistic regression (odds ratios, OR; 95% confidence intervals, CI; and trend tests). We examined deviation from multiplicative interaction using chi-squared tests. We evaluated mediation of the alcohol and BC association by volumetric density measures using logistic regression to estimate the association between alcohol use and BC with and without adjustment for density measures. Percent mediation was estimated using the differences in the log OR estimates from the two models. All models were adjusted for age, 1/BMI and menopause and matching factors. Analyses were also stratified by menopausal status. Results: Alcohol intake was available on 2233 cases and 4562 controls, 88% of those eligible. BC cases and controls had similar age (57.2 years (SD=11.5) vs. 57.1 years (SD=11.5)), BMI (25.3 kg/m2 (SD=5.3) vs. 24.9 kg/ m2 (SD=5.1)) and race (69.0% vs. 68.2% Caucasian). Cases were more likely to drink alcohol daily than controls (52.1% vs. 49.0%), in particular two or more drinks per day (14.8% vs. 13.2%). Alcohol was associated with increased BC risk (OR=1.14, 95% CI:1.02-1.27, for one or less drinks per day and OR=1.22, 95% CI:1.05-1.42 for 2 or more drinks per day) compared to non-drinkers (p-trend=0.004). Percent volumetric density (OR=1.45 per SD, 95%CI: 1.36-1.56,) and dense volume (OR=1.30, 95% CI: 1.24-1.37) were also positively associated with BC risk; non-dense volume was inversely associated (OR=0.93, 95%CI: 0.86-1.01). Associations were similar by menopausal subgroup. There was no evidence for a differential association of alcohol and breast cancer risk by dense breasts assessed using any of the density phenotypes examined (all P’s>0.1). However, the association between alcohol and overall risk of BC was partially mediated by dense volume among all women (percent mediated=25%, P=0.01) and postmenopausal women (percent mediated=19%, P=0.03). Conclusions: The association of daily alcohol intake and breast cancer risk was similar among women with dense and non-dense breasts. However, dense volume partially mediated the association between alcohol and risk of breast cancer, particularly among postmenopausal women, suggesting that alcohol partially influences breast cancer risk through changes in breast tissue composition. Citation Format: Celine Marie Vachon, Christopher G. Scott, Stacey J Winham, John A Shepherd, Kathleen R Brandt, Matthew R Jensen, Carrie B Hruska, John J Heine, V. Shane Pankratz, Karla Kerlikowske. Association of daily alcohol intake, volumetric density and breast cancer risk [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P5-08-02.

Abstract Background: Polygenic risk scores (PRS) based on a large number of SNPs associated with breast cancer have recently been made available in combination with multigene panel testing for clinical management of breast cancer risk. Breast cancer risk assessment has historically been performed using various non-genetic risk models, which predict lifetime risk of breast cancer in unaffected women based on personal and family history information. While clinical use of PRS in combination with these models continues to increase, the predictive performance of a PRS combined with clinical information in real-world patient populations has not been thoroughly investigated. Methods: We assessed a 100-SNP breast cancer PRS in combination with the Tyrer-Cuzick model-estimated lifetime breast cancer risk in 5,374 Caucasian women from the case-control Mayo Clinic Breast Cancer Study (MCBCS), comprised of 3,373 breast cancer cases and 2,001 unaffected controls seen at Mayo Clinic from 2004 to 2016. Clinical, family history, and demographic information was collected on all MCBCS participants via questionnaires and medical record review. SNPs were selected for the PRS if they exceeded genome-wide significance in ≥2 large-scale breast cancer genome wide association studies (GWAS) in women of European descent. Selected SNPs were then used to construct a population standardized PRS, accounting for per-allele odds ratios reported in the largest GWAS to date and population-specific allele frequencies. PRS and Tyrer-Cuzick model associations with breast cancer and related clinical features were tested using multivariate logistic regression. Results: Among cases and controls, mean±SD age at diagnosis and study enrollment was 56.4±12.4 and 56.2±14.1, respectively. Approximately 36.9% of cases and 16.9% of controls had ≥1 first- or second-degree relative with breast or ovarian cancer, respectively. Among unaffected women, the median (IQR) estimated lifetime risk of breast cancer to age 85 based on Tyrer-Cuzick alone was 8.3% (6.9%), and 2.6% had a clinically actionable lifetime risk of ≥20%. The mean±SD sum of the risk alleles across 100 SNPs was significantly higher for cases vs. controls (94.8±6.3 vs. 92.7±6.2, p<0.001). Compared to women in the lowest quartile of PRS, those in the 2nd, 3rd and 4th quartiles were 1.47 (95% CI: 1.26-1.72), 1.87 (95% CI: 1.60-2.19), and 2.22 (95% CI: 1.89-2.61) times more likely to have breast cancer after adjustment for age and family history. The AUC was 0.59 (95% CI: 0.58-0.61) for PRS and 0.60 (95% CI: 0.58-0.61) for Tyrer-Cuzick model estimated risk alone; but when combined, the model achieved an AUC of 0.63 (95% CI: 0.62-0.65). Among unaffected women, 6.7% had ≥20% lifetime risk of breast cancer based on the combined PRS-clinical model. Conclusions: These findings suggest that the 100-SNP PRS significantly improves estimation of breast cancer risk based on non-genetic models, and can be used to further identify women at increased lifetime risk who would otherwise not be identified by clinical assessment alone. Citation Format: Mary Helen Black, Shuwei Li, Holly LaDuca, Sebastian Armasu, Eric C. Polley, Patrick Fitz-Gibbon, Chris G. Scott, Stacey J. Winham, Janet E. Olson, Kathryn J. Ruddy, Celine M. Vachon, Fergus J. Couch. Performance of a polygenic risk score combined with clinical assessment for breast cancer risk [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P5-07-06.

Reports highlighting the problems with the standard practice of using bar graphs to show continuous data have prompted many journals to adopt new visualization policies. These policies encourage authors to avoid bar graphs and use graphics that show the data distribution; however, they provide little guidance on how to effectively display data. We conducted a systematic review of studies published in top peripheral vascular disease journals to determine what types of figures are used, and to assess the prevalence of suboptimal data visualization practices. Among papers with data figures, 47.7% of papers used bar graphs to present continuous data. This primer provides a detailed overview of strategies for addressing this issue by (1) outlining strategies for selecting the correct type of figure depending on the study design, sample size, and the type of variable; (2) examining techniques for making effective dot plots, box plots, and violin plots; and (3) illustrating how to avoid sending mixed messages by aligning the figure structure with the study design and statistical analysis. We also present solutions to other common problems identified in the systematic review. Resources include a list of free tools and templates that authors can use to create more informative figures and an online simulator that illustrates why summary statistics are meaningful only when there are enough data to summarize. Last, we consider steps that investigators can take to improve figures in the scientific literature.

Abstract Background: Archived formalin fixed paraffin embedded (FFPE) samples are valuable clinical resources to examine clinically relevant morphology features and also to study genetic changes. However, DNA quality and quantity of FFPE samples are sub-optimal, and resulting NGS-based genetics variant detections are prone to false positives. Evaluations of wet-lab and bioinformatics approaches are needed to optimize variant detection from FFPE samples. Results: As a pilot study, we designed within-subject triplicate samples of DNA derived from paired FFPE and fresh frozen breast tissues to highlight FFPE-specific artifacts. For FFPE samples, we tested two FFPE DNA extraction methods to determine impact of wet-lab procedures on variant calling: QIAGEN QIAamp DNA Mini Kit ("QA"), and QIAGEN GeneRead DNA FFPE Kit ("QGR"). We also used negative-control (NA12891) and positive control samples (Horizon Discovery Reference Standard FFPE). All DNA sample libraries were prepared for NGS according to the QIAseq Human Breast Cancer Targeted DNA Panel protocol and sequenced on the HiSeq 4000. Variant calling and filtering were performed using QIAGEN Gene Globe Data Portal. Detailed variant concordance comparisons and mutational signature analysis were performed to investigate effects of FFPE samples compared to paired fresh frozen samples, along with different DNA extraction methods. In this study, we found that five times or more variants were called with FFPE samples, compared to their paired fresh-frozen tissue samples even after applying molecular barcoding error-correction and default bioinformatics filtering recommended by the vendor. We also found that QGR as an optimized FFPE-DNA extraction approach leads to much fewer discordant variants between paired fresh frozen and FFPE samples. Approximately 92 % of the uniquely called FFPE variants were of low allelic frequency range (<5%), and collectively shared a “C>T|G>A” mutational signature known to be representative of FFPE artifacts resulting from cytosine deamination. Based on control samples and FFPE-frozen replicates, we derived an effective filtering strategy with associated empirical false-discovery estimates. Conclusions: Through this study, we demonstrated feasibility of calling and filtering genetic variants from FFPE tissue samples using a combined strategy with molecular barcodes, optimized DNA extraction, and bioinformatics methods incorporating genomics context such as mutational signature and variant allelic frequency.

Abstract Background Archived formalin fixed paraffin embedded (FFPE) samples are valuable clinical resources to examine clinically relevant morphology features and also to study genetic changes. However, DNA quality and quantity of FFPE samples are often sub-optimal, and resulting NGS-based genetics variant detections are prone to false positives. Evaluations of wet-lab and bioinformatics approaches are needed to optimize variant detection from FFPE samples. Results As a pilot study, we designed within-subject triplicate samples of DNA derived from paired FFPE and fresh frozen breast tissues to highlight FFPE-specific artifacts. For FFPE samples, we tested two FFPE DNA extraction methods to determine impact of wet-lab procedures on variant calling: QIAGEN QIAamp DNA Mini Kit ("QA"), and QIAGEN GeneRead DNA FFPE Kit ("QGR"). We also used negative-control (NA12891) and positive control samples (Horizon Discovery Reference Standard FFPE). All DNA sample libraries were prepared for NGS according to the QIAseq Human Breast Cancer Targeted DNA Panel protocol and sequenced on the HiSeq 4000. Variant calling and filtering were performed using QIAGEN Gene Globe Data Portal. Detailed variant concordance comparisons and mutational signature analysis were performed to investigate effects of FFPE samples compared to paired fresh frozen samples, along with different library preparations. In this study, we found that five times or more variants were called with FFPE samples, compared to their paired fresh-frozen tissue samples even after applying molecular barcoding error-correction and default bioinformatics filtering recommended by the vendor. We also found that QGR as an optimized FFPE-DNA extraction approach leads to much fewer discordant variants between paired fresh frozen and FFPE samples. Approximately 92% of the uniquely called FFPE variants were of low allelic frequency range (<5%), and collectively shared a “C>T|G>A” mutational signature known to be representative of FFPE artifacts resulting fromcytosine deamination. Based on control samples and FFPE-frozen replicates, we derived an effective filtering strategy with associated empirical false-discovery estimates. Conclusions Through this study, we demonstrated feasibility of calling and filtering genetic variants from FFPE tissue samples using a combined strategy with molecular barcodes, optimized DNA extraction, and bioinformatics methods incorporating genomics context such as mutational signature and variant allelic frequency.

Abstract Existing risk models underperform among women who have undergone a benign breast disease (BBD) biopsy (&gt; one million performed annually in the US) and with respect to estimating the risk of aggressive BCs. We've shown that molecular pathologic analysis of BBD biopsies can improve individual risk prediction, compared with standard risk models relying on self-reported factors, and provide insights into mechanisms mediating BC risk. BC risk factors such as obesity and ethanol use are proposed to increase production of cytokines and chemokines, resulting in chronic inflammation, and leading to production of DNA damaging free radicals and growth factors (e.g., VEGF, IGFs) that activate pro-carcinogenic pathways (e.g. NF-KB and JAK/STAT). The effects of most BC risk factors on immunity are poorly understood. Further, whereas experimental models implicate immunity throughout carcinogenesis, knowledge of immune markers and mechanisms related to BC development among women is limited. Therefore, we aim to define tissue immune cell content throughout BC development, and particularly at the earliest stages, to improve risk assessment and discover immune-based prevention strategies. Our study combines novel resources and technologies to define the immune landscape in: 1) normal breast tissues in relation to BC risk factors, 2) BBD with respect to BC risk, and 3) invasive BC in relation to neoantigen expression and molecular subtype. We will perform NanoString GeoMxTM Digital Spatial Profiling (DSP) to quantitatively map expression of key immune proteins in healthy breast tissues donated to the Komen Tissue Bank (KTB) in relation to BC risk factors and in BBD biopsies from two cohorts to predict BC risk. We will re-analyze top prognostic markers in BBD biopsies using Vectra multiplex IF staining with machine learning algorithms to refine how immune cell content and organization impacts prognosis, and assess critical immune pathways related to BBD progression with the NanoString PanCancer IO 360 panel, which provides RNA expression of 770 immune genes (13 signatures). Finally, we will evaluate immune responses in BCs categorized for predicted neoantigen load and underlying mutation type with our novel bioinformatics pipeline that enables accurate prediction of MHC class I and class II missense, fusion and frameshift mutation neoantigens generated through faulty DNA repair, aberrant DNA and RNA splicing, insertions, and deletions. Our project demonstrates a novel approach to defining the immune processes in early stages of BC development which may improve BC risk prediction and prevention. We present a plan to comprehensively analyze immunity throughout BC development and define changes that accompany transitions from normal to BBD and BBD to BC. Citation Format: Doug Hinerfeld, Keith Knutson, Derek Radisky, E. Aubrey Thompson, Yan Asmann, Kim McCoy, Amy Degnim, Jodi Carter, Stacey Winham, Michele Cote, Jeroen van der Laak, Mark Sherman. Analysis of the immune microenvironment to advance breast cancer risk prediction and prevention [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 651.

Abstract Background: Macrophages are well-established components of the inflammatory repertoire in breast cancer (BC) progression. The role of macrophages in early breast carcinogenesis is less understood. Macrophages adopt microenvironment-driven pro- or anti-inflammatory phenotypes. We evaluated macrophagic subsets in benign breast stroma to evaluate their association with future or concurrent BC. Methods: With institutional approval, FFPE tissues were identified from age-matched subjects, [with index benign breast biopsies and (a subset) interval-matched second benign tissues taken at the time of one of three events: ER+ BC, ER- BC or no cancer event (Controls) (index N=32, 26 and 30 respectively)]. Tissues were immunostained (CD68 or CD163) and stromal CD68+ and CD163+ macrophagic cell densities (MCD) (cells/mm2 stromal fibroadipose area) were digitally quantified. Statistical analysis was performed using Wilcoxon signed-rank tests for univariate analysis and conditional logistic regression models to adjust for BMI. Results: In the index age-matched biopsies (median age 54y), within subjects in all groups, the median CD163+ MCD (54.5; range: 13.9-371.3) was significantly higher than CD68+ MCD (19.6, range: 3.5-78.5) but the two were strongly correlated (r = 0.76); neither correlated with BMI (r&lt; 0.1 each). Across the groups, index benign biopsies of ER- BC had significantly lower CD68+ MCD and CD163+ MCD compared to ER+ BC (CD163: median 47.2 vs. 57.8, p=0.004; and median CD68+ MCD: 15.8 vs. 29.7, p=0.03); and ER- BC had significantly lower CD163+ MCD density than Controls (p=0.02), (all by univariate analysis and after adjusting for BMI). Comparing the index biopsy and second biopsy/excision within subjects, all showed significant increases in CD68+ MCD (median increase 27.9 for ER- BC, 16.1 for ER+ BC, and 17.8 for controls, p&lt;0.01), and only ER-BC showed an increase in CD163+ MCD (median increase 26.6, p=0.01). Thus, in the second biopsies/excisions, the CD163+ MCD and CD68+ MCD were similar across groups (median CD163+ MCD: 69.3, 75.4, and 59.3 and median CD68+ MCD: 47.4, 45.2, and 36.0 for ER+ BC, ER- BC, and controls); however, all groups had decreased CD163+:CD68+ MCD ratios compared to the index biopsies. Conclusion: Benign breast tissues have higher stromal CD163+ MCD than CD68+ MCD irrespective of future BC events and independent of BMI. However, between index and second biopsies/excisions, the stromal CD68+ MCD differentially increased in all groups, lowering CD163+:CD68+ MCD ratios, and likely shifting pro- and anti-inflammatory stromal macrophage functionality. Across groups, ER- BC index biopsies had significantly lower stromal MCD of both types compared to ER+ BC, and were the only group to increase stromal CD163+ MCD between index biopsy and cancer event. These data suggest that ER+ and ER- BC may have different pre-carcinogenic stromal inflammatory microenvironments. Citation Format: Jodi M. Carter, Ethan P. Heinzen, Tanya L. Hoskin, Stacey J. Winham, Derek C. Radisky, Daniel W. Visscher, Amy C. Degnim. Differential stromal macrophagic profiles precede ER+ and ER- breast carcinogenesis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2705.