Citations

Abstract Abstract 3881 Background: Chronic lymphocytic leukemia (CLL) is largely incurable neoplasm of the blood and has one of the highest familial risks of disease among cancers. Monoclonal B-cell lymphocytosis (MBL) with a CLL immunophenotypic profile (i.e., CLL-like MBL) is a precursor state to CLL. In the general adult population, the prevalence rate of MBL is 5–9% and increases with age. This is in contrast to the 13–18% rate of MBL in first-degree relatives of CLL patients from high-risk families (defined here as families with at least two confirmed living cases of CLL). Few longitudinal studies of MBL incidence exists. One study evaluated MBL status over time in a cohort of individuals with MBL from a rural valley in Northern Italy. Our work herein will be the first to evaluate the status of MBL over time in relatives of CLL patients from high-risk families. Methods: Using a cohort of 41 high-risk CLL families, we evaluated the MBL status and clonal blood B-cell counts in 94 relatives of CLL patients at two time points at least 1 year a part. High sensitivity, multi-color flow cytometry was performed on fresh or previously frozen blood samples in three laboratories: Duke University, the US Food and Drug Administration, and Mayo Clinic. The number of cell events collected for flow cytometry evaluation varied from 1 to 5 x105 cells. To assure concordance across flow cytometry laboratories, 15 samples were evaluated at multiple labs. The concordance of MBL calling was a 100%. We defined three monotypic B-cell subsets as CLL-like MBL (CD5+ CD20dim), atypical MBL (CD5+ CD20+), and non-CLL like MBL (CD5−CD20+). Results: Of the 94 relatives, the median age at consent was 57 years (range 42–93) and 39% were male. At the initial MBL screening, 70 (74%) relatives had normal immunophenotype (NIPT) and 24 (26%) had MBL (1 atypical MBL, 3 non-CLL like MBL, and 20 CLL-like MBL). Seven of the 24 individuals with MBL had monoclonal B-cell counts available at baseline screening; median = 147.4 cells/uL (range 22.7–1663). The 94 relatives underwent repeat evaluation with a median follow up time of 1.8 years (range 1–6). At the second evaluation, we found an increase in the prevalence of MBL, with 11 of the 70 (12%) initially NIPT relatives developing a new MBL clone (10 CLL-like, 1 non-CLL like). These 11 incident MBL cases were from 9 families with 2 families each having 2 relatives changing MBL status. The median number of monoclonal B-cells in these 11 incident MBL cases was 0.58 cells/uL (range 0.38–60.3). The 23 out of the 24 relatives with baseline MBL (20 CLL-like MBL and 3 non-CLL like MBL) had a persistent MBL clone. Only 6 of these 23 individuals had monoclonal B-cell counts available at both time points, and among these 6, the clonal B-cell counts decreased at follow-up screening; median difference −64.1 cells/uL (range −1204.3– 10.5). The second evaluation of the initial atypical MBL relative had too few clonal cells for flow assessment. Conclusion: We found that our CLL-like MBL clones persisted over the period of observation in relatives from high-risk CLL families; this finding is similar to that previously reported using a population-based study conducted in Northern Italy. Notably, 12% of relatives from high-risk CLL families developed incident MBL during just 2 years of follow up. Although these MBL clones are all low count (< 1500 cells/uL), these data provide further evidence that MBL is a marker of inherited predisposition to CLL. It will be important to see how our rate of change of MBL status compares to that of the general population as we continue to follow our cohort of CLL families. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 2443 Background: There is strong and consistent evidence that a genetic component contributes to the etiology of chronic lymphocytic leukemia (CLL). A recent genome-wide association (GWA) study of CLL identified genetic variants located on chromosomes 2q13, 2q37.1, 6p25, 11q24, 15q23, and 19q13 that increased the risk of CLL within a European population. We replicated 5 of these 6 loci in an independent sample of CLL cases and controls from the United States. We now investigate whether these loci also influences MBL, a reported precursor condition of CLL. In addition, a follow-up analysis of the initial GWA study identified four more CLL-susceptibility loci on 2q37.3, 8q24.21, 15q21.3, and 16q24.1. Herein, we also evaluate the association of these four loci with risk of CLL. Methods: Peripheral blood samples were obtained from three ongoing studies: the Genetic Epidemiology CLL (GEC) Consortium, the Mayo Clinic non-Hodgkin lymphoma (NHL)/ CLL study, and the Mayo Clinic Biobank. We implemented rigorous genotyping quality-control measures, and successfully genotyped a total of 407 CLL patients, 965 controls, and 60 MBLs from these studies. Within each locus, the previously reported single nucleotide variants (SNPs) or variants in high linkage disequilibrium (LD) with the previously reported SNPs were evaluated with risk of MBL or CLL. Tests for association was done using the Cochran-Armitage trend test, and unconditional logistic regression was used to estimate odds ratios (ORs) and corresponding 95% confidence intervals (CIs) for CLL or MBL risk Results: In our evaluation of the six initially reported CLL-susceptibility loci (2q13, 2q37.1, 6p25, 11q24, 15q23, and 19q13) with MBL risk, we found three of the six had suggestive associations (p-value < 0.20) with ORs comparable to and in the same direction as those observed from CLL risk. Our strongest finding was with rs13397985 at locus 2q37.1 (OR= 1.56; 95% CI: 1.05, 2.31; p-trend = 0.041), followed by rs17483466 at locus 2q13 (OR= 1.49; 95% CI: 0.99, 2.24; p-trend = 0.074). As expected given our previously reported findings with CLL risk, the association between rs11083846 on chromosome 19q13 and MBL risk was not significant (p-trend = 0.70). Of the four recently reported CLL-susceptibility loci SNPs located on 2q37.3, 8q24.21, 15q21.3, and 16q24.1, we found all to be associated with CLL risk but one. Specifically, the strongest association was seen for locus 8q24.21 (best tagged SNP rs1021955; OR = 1.37; 95% CI: 1.10, 1.70; p-trend = 0.005), followed by locus 16q24.1 (best tagged SNP rs305065; OR= 0.77; 95% CI: 0.61, 0.97; p-trend = 0.024). However, we found no associations for locus 15q21.3 for the previously reported SNP nor for any SNPs in LD with the previously reported SNP. Conclusions: Our MBL results provide additional robust genetic evidence that MBL is a precursor to CLL and that it shares similar underlying genetic predisposition. Also our results confirm three of the four recently reported CLL-susceptibility loci and further support the role of a genetic basis in the etiology of CLL. More research is needed to elucidate the potential manner in which these genetic loci function in CLL or MBL. Disclosures: No relevant conflicts of interest to declare.

Abstract Background: It is well established that the incidence of skin cancers in patients with CLL are significantly elevated compared to age- and sex- matched controls. However, little is known about the characteristics of CLL patients who develop skin cancer. Herein, we evaluate the associations of CLL clinical and prognostic characteristics, along with UV radiation exposure, with risk of first skin cancer following CLL diagnosis. Methods: Newly diagnosed CLL patients from Minnesota, Iowa, and Wisconsin were enrolled in the Mayo Clinic case-control study from 2002-2015 and systematically followed in the Iowa/Mayo Lymphoma SPORE. Clinical and prognostic CLL data were obtained from the Mayo Clinic CLL database, and skin cancer clinical data were abstracted from medical records using a standard protocol. The CLL international prognostic index (CLL-IPI) was computed using a weighted average of five independent CLL prognostic factors (IGHV mutational status, serum b2-microglobulin, Rai stage, age, and FISH 17p deletion/TP53 status). Self-reported history of midday sun exposure at various ages (birth to age 12; 13 to 21 years; 22 to 40 years, and 41+ years) was obtained from a risk factor questionnaire. For each age, we asked the extent of mid-day sun as: practically no exposure (under 3 hours per week), little exposure (4-7 hours per week), moderate exposure (8 to 14 hours per week) and extensive exposure (15+ hours per week). Midday sun exposure was modeled as an ordinal covariate. To evaluate associations with risk of skin cancer following CLL diagnosis, we calculated time from date of CLL diagnosis to date of first skin cancer, death, or last known follow-up. We used Cox regression analysis to estimate hazard ratios (HRs) and 95% CIs. CLL treatment was considered a time-dependent covariate. Results: Among 846 CLL patients enrolled, the median age at diagnosis was 63 years (range 28-91), 68% were male, 7% had Rai stage III-IV at diagnosis. Based on the CLL-IPI, 42% were categorized as low risk, 33% as intermediate risk, and 25% as high or very-high risk. 109 CLL cases (13%) had one or more reported skin cancers at or prior to CLL diagnosis. Melanoma was observed in 19 (2%) cases and non-melanoma was in 90 (11%) cases. At a median follow-up of 7 years from CLL diagnosis, 165 patients (20%) had one or more skin cancers after CLL diagnosis. Among these patients, 49 had skin cancer before CLL diagnosis. The most frequent skin cancer was squamous cell carcinoma (59%), followed by basal cell (31%), melanoma (5%), and Merkel cell (1%). 552 (65%) of the 846 patients returned a questionnaire. Significant associations of clinical and prognostic characteristics with risk of first skin cancer were observed for age (HR=1.35 per 10 year increase, 95% CI=1.17-1.56, P<0.001), male sex (HR=1.38, 95% CI 0.98-1.96, p=0.07), prior history of skin cancer (HR=4.19, 95% CI=2.98-5.88, P<0.001), and CLL-IPI (HR=1.26, 95% CI= 1.03-1.54, P=0.026, after adjusting for age, sex, and prior skin cancer). Of note, the risk of first skin cancer in those CLL patients categorized as very high via CLL-IPI had 2.28 fold risk (95% CI 1.02-5.11). Midday sun exposures for each of the ages considered showed no evidence of association with risk of first skin cancer (all P>0.05). 50 CLL patients were treated prior to first skin cancer following CLL diagnosis; we observed no evidence of association between treatment and risk of first skin cancer (HR=1.44, 95% CI= 0.93-1.92, P=0.12). Conclusion: CLL patients who are at an increased risk of skin cancer following CLL diagnosis are those who either have had a prior history of skin cancer or a more aggressive CLL disease at diagnosis, according to CLL IPI. Routine skin cancer screening is currently recommended for CLL patients. Our data suggest that more frequent screening would be particularly important among patients with aggressive CLL and who have a prior history of skin cancer. Unexpectedly, we found no evidence of association of skin cancer risk with UV radiation following CLL diagnosis or with CLL treatment. Further investigation is needed to determine whether other factors increase the risk of skin cancer following CLL diagnosis. Disclosures Shanafelt: Genentech: Research Funding; GlaxoSmithkKine: Research Funding; Janssen: Research Funding; Celgene: Research Funding; Cephalon: Research Funding; Pharmacyclics: Research Funding; Hospira: Research Funding.

Abstract We and others previously demonstrated an increased risk of monoclonal gammopathy of undetermined significance (MGUS) in first-degree relatives of multiple myeloma (MM) and MGUS patients, underscoring a familial component to this disease. However, MM is known to be a collection of cytogenetically distinct diseases, with 40-50% of MM cases having a reciprocal translocation involving the IgH locus on chromosome 14q32 and in many of the remaining patients, trisomies. In a small proportion of cases, both IgH abnormalities and trisomies are found in the same clone. These MM cytogenetic subtypes are associated with differential prognosis. Whether there are differences in familial risk by cytogenetic subtypes is not known. Patients with MM were recruited from the Mayo Clinic Hematology practice in Rochester, Minnesota (MN) between 2005 and 2015. Serum samples were collected from first-degree relatives and subjected to electrophoresis and immunofixation for MGUS determination. Age- and sex-specific rates of MGUS from the population-based Olmsted County, MN, prevalence study were used as expected rates for the calculation of risk ratios (RRs) for comparisons of first-degree relatives. Cytogenetic subtypes were classified by FISH testing performed in the clinic and categorized into groups defined by IgH translocation , trisomy, intermediate/ high-risk disease (t(4;14), t(14;16), t(14;20), and del17p). Further, the distribution of cytogenetic subtypes was compared between MM cases with and without evidence of MGUS in first degree relatives using Chi square tests. Analyses were also subset to MM probands diagnosed in MN. There were a total of 442 MM probands, with 1235 first-degree relatives recruited and eligible for the study. MGUS was detected in 78 relatives, for an age- and sex-adjusted prevalence of 6.1% (4.7%-7.5%). Analyses comparing prevalence of MGUS in first degree relatives to population-based rates from Olmsted County showed a 2.5-fold (95% CI=1.9-3.0) increased prevalence. When performing analyses among MN probands only (RR=2.9; 95% CI=2.1-3.8), results were similar. Of the 442 probands, 310 had FISH performed on plasma cells (identified by cIG staining; cIg-FISH) at diagnosis. Of these cases, 49% had a trisomy, 23% had an IgH translocation, and 9% had both. Although the proportion of trisomies and IgH translocations were 4-6% higher among MM probands with a relative with MGUS compared to those without, these differences were not statistically significant (Table 1). MM probands with a first degree relative with MGUS also had a similar proportion of cytogenetic subtypes that are considered of intermediate or high risk compared to those MM without a family history (26.9% vs 25.2%) (p=0.79) (Table 1). When restricted to MM probands from MN, results remained non-statistically significant (Table 1). We showed an increased risk of MGUS in first-degree relatives of MM cases, implying shared environment and/or genetics underlying this malignancy. However, we did not find an association between family history of MGUS in MM probands and the cytogenetic subtypes examined. Larger cohorts are needed to examine if a relationship exists between cytogenetic subtypes of MM and familial risk. Disclosures Kumar: Noxxon Pharma: Consultancy, Research Funding; Array BioPharma: Consultancy, Research Funding; AbbVie: Research Funding; Onyx: Consultancy, Research Funding; Millennium: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; Glycomimetics: Consultancy; Kesios: Consultancy; Janssen: Consultancy, Research Funding; Skyline: Honoraria, Membership on an entity's Board of Directors or advisory committees; Sanofi: Consultancy, Research Funding; BMS: Consultancy.

Abstract Multiple myeloma (MM) is a malignancy of plasma B-cells that is invariably preceded by monoclonal gammopathy of undetermined significance (MGUS), a pre-neoplastic plasma cell proliferative disorder. C57Bl/KaLwRij (KaLwRij) is a spontaneous inbred mouse strain predisposed to benign idiopathic paraproteinemia (BIP), a plasma cell dyscrasia that parallels many features of human pathology including progression to myeloma. In both humans and mice, the genetic basis for MGUS/MM risk is unknown. We sought to identify the genetic basis for MGUS/MM risk in KaLwRij mice to shed light on genes or pathways that mediate disease risk in humans. Using an integrative approach, we combined KaLwRij x B6 haplotype analysis (418 genes) and a MM patient association study (180 genes) to identify five candidate genes likely to contribute to both murine BIP-susceptibility and human MM risk. Surprisingly, we found KaLwRij mice had homozygous germline deletion of one of the five candidate genes, Samsn1, a putative negative regulator of B-cell activation. Consistent with the reported inhibitory function of SAMSN1 in B-cells, KaLwRij mice had enhanced naïve B-cell response in vitro following stimulation with IL4 and LPS and in vivo following immunization. We found that re-expression of Samsn1 in KaLwRij-derived myeloma cell line 5TGM1 decreased MM tumor growth in vitro, suggesting that loss of Samsn1 contributes to tumor growth in part through a cell intrinsic mechanism. Importantly, germline risk alleles may also influence cell types beyond the pre-malignant B-cells. It is established that the host environment also plays a critical role in supporting myeloma growth in KaLwRij mice. We found that Samsn1 is highly expressed in macrophages, but not in other MM-supportive cell types including osteoclasts and bone marrow stromal cells. Myeloma-associated macrophages play critical roles in supporting tumor growth and are broadly classified as anti-tumor M1 macrophages and tumor-promoting M2 macrophages. Microarray analysis of macrophages from B6 and KaLwRij mice prior to onset of BIP revealed three differentially expressed genes, but only one, Samsn1, is conserved in the human genome. We found that both unpolararized primary peritoneal and MCSF cultured bone marrow macrophages from healthy KaLwRij mice had markedly elevated expression of pro-tumorigenic M2 macrophage markers FIZZ1 and YM1 compared to B6 controls, despite the absence of activating cytokines. To confirm that loss of Samsn1 influenced macrophage activation and M2 polarization, we obtained bone marrow macrophages from mice with a targeted deletion of Samsn1 and found that unpolarized Samsn1-/- cells had similarly increased pro-tumor M2 macrophage marker expression. To evaluate the biologic significance of Samsn1 loss in macrophages on MM tumor growth, we established 5TGM1 MM tumors and injected M2 polarized macrophages from B6, KaLwRij, or Samsn1-/- mice. We found that MM tumor burden was markedly enhanced by the addition of a Samsn1-/- or KaLwRij macrophages compared to wild type macrophages. Thus, we found that loss of Samsn1 in macrophages enhances M2 polarization, prior to MGUS/MM development and can promote established myeloma tumor growth. Together, this data suggests that Samsn1 is a gene contributing to BIP-susceptibility in KaLwRij mice through mutually exclusive functions in multiple cell types including B-cells and macrophages. It is likely that SAMSN1 and its pathway members contribute to human MM progression, and may represent a point for preventative therapeutic intervention. Disclosures No relevant conflicts of interest to declare.

Abstract Introduction: Several B-cell malignancies, including chronic lymphocytic leukemia (CLL) and multiple myeloma (MM), are known to have a familial component to disease risk. Although several common inherited genetic variants have been found to be associated with risk of these malignancies, a large proportion of the heritability remains unexplained. Here we set out to identify rare high-penetrance susceptibility variants for CLL and MM using the established Mayo Clinic family study of B-cell malignancies. Methods: We performed exome sequencing on germ line DNA of 106 families with either two or more members with CLL (n=93 families) or two or more members with MM or monoclonal gammopathy of undetermined significance (MGUS) (n=13 families) using Aglient capture kits. All samples were sequenced on an Illumina HiSeq2000. Bioinformatics analyses leveraged the following software packages: Novoalign, Picard, The Genome Analysis Toolkit (GATK), and the Biological Reference Repository (bioR). Quality control filters were implemented; subjects with mis-specified relationships were removed as were variants with <95% call rate or <20x coverage, and those variants identified as sequencing artifacts. To identify potentially functional single nucleotide variants, the following was required a) uncommon in public databases, b) cosegregating variants in multiple families, c) variants to be highly conserved and in coding regions, and d) functional prediction status of deleterious (SIFT score), damaging (PolyPhen score), and a moderate, or high variant impact (SNPEffect). The association of these variants with risk was then evaluated in large samples of cases and controls in genome-wide association studies (GWAS). Results: In the CLL familial cohort, there were 443 individuals (160 with CLL, 73 with monoclonal B cell lymphocytosis (MBL), and 210 relatives without CLL or MBL) sequenced; and in the MM cohort, there were 39 individuals (13 with MM, 13 with MGUS and 13 without MM or MGUS) sequenced. A total of 61,992 single nucleotide variants passed quality control filters in the CLL cohort and 39,456 single nucleotide variants in the MM cohort. From these, we identified 32 variants in the CLL cohort and 39 variants in the MM cohort that were cosegregating in multiple CLL or MM families, respectively, and were predicted to affect protein structure or function. Of these, one CLL variant was within 1 Mb of a known GWAS-discovered locus. Moreover, there were seven CLL and four MM variants that were within 1 Mb of single nucleotide polymorphisms (SNPs) that had statistical significance between p<10-4 and p<10-6 in GWAS. Finally, there were 3 variants found to be common between CLL and MM families (see Table). Conclusions: Through whole exome sequencing, we identified a number of rare, highly penetrant predisposition variants for CLL and MM. Several of these variants were common to both B-cell malignancies providing evidence of shared genetic components; one variant may be a functional driver of the GWAS-discovered loci. These identified variants provide a list for future validation studies. Table: Exome variants common in both Chronic Lymphocytic Leukemia and Multiple Myeloma SNP ID rs41314099 rs61754114 rs143025033 Chromosome 17 6 7 Gene WSCD1 BACH2 TNS3 Penetrance (CLL) 0.70 0.75 0.80 Penetrance (MM) 1.00 0.80 0.75 Allele Frequency (CLL) 0.015 0.025 0.017 Allele Frequency (MM) 0.077 0.077 0.058 1000 Genome Frequency (European) 0.010 0.020 0.010 SIFT Term Deleterious Deleterious Deleterious PolyPhen Term Probably Damaging Probably Damaging Probably Damaging SNP: Single Nucleotide Polymorphism; CLL: Chronic Lymphocytic Leukemia; MM: Multiple Myeloma; Penetrance is defined as proportion of individuals carrying the alternate allele that are also affected; SIFT: Sorting Intolerant From Tolerant, predicts whether an amino acid substitution affects protein function; PolyPhen: Polymorphism Phenotyping, predicts possible impact of an amino acid substitution on the structure of a human protein Disclosures No relevant conflicts of interest to declare.

Abstract Abstract 4000 Background: We have previously reported that there is an increased risk of monoclonal gammopathy of undetermined significance (MGUS) in first-degree relatives of patients with multiple myeloma (Vachon CM. Blood 2009 114: 785–790). There are several cytogenetic subtypes of myeloma, and there are no data on whether certain cytogenetic subtypes of myeloma are more frequently associated with familial MGUS. Methods: We studied patients with myeloma who participated in the familial MGUS study in whom presence or absence of MGUS in the first-degree relatives had been ascertained (Vachon CM. Blood 2009 114: 785–790). Probands were seen at the Mayo Clinic Hematology/Oncology practice (Rochester, MN, USA) between February 2006 and September 2007. Cytogenetic data was acquired via electronic medical record abstraction of fluorescence in-situ hybridization (FISH) lab reports, and used to categorize patients into one of nine cytogenetic subgroups: trisomy(ies), t(6;14), t(11;14), t(4;14), t(14;16), t(14;20), Mixed (those with trisomy(ies) and an IgH translocation), other cytogenetic abnormalities (in the absence of trisomy(ies) or IgH translocation), and normal (Kumar S. Blood 2012;119:2100–2105). We examined whether a difference in the distribution of the six primary cytogenetic categories of myeloma existed between probands with a family history of MGUS and/or myeloma and those without. Results and Conclusions: Of the 248 patients invited to participate, FISH data (with sufficient plasma cells) was available on 119 participants to establish the primary molecular cytogenetic classification of myeloma. All had available information regarding family history of MGUS and multiple myeloma. 27 had an affected first-degree relative with MGUS, and 92 did not. Distributions of cytogenetic subtypes in the two groups are shown in Table 1. IgH translocated MM was more common in myeloma patients who lacked an affected first-degree relative compared to those with familial MGUS (19% vs 30%, P=0.32) Interestingly, the t(11;14) subtype was more common in myeloma patients without familial MGUS compared to those with an first-degree relative with MGUS (19.6% vs 7.4%, P=0.24). The differences in Table 1 did not reach statistical significance, possibly due to the small numbers of individuals with a family history in this sample. However, the distribution (Table 1) suggests that the distribution of cytogenetic subtypes may be different in myeloma that is not associated with familial MGUS compared with myeloma in which a familial tendency is detected. IgH translocated MM appears to have a lower risk of familial tendency. Further investigation is needed to estimate the risk of familial MGUS within each cytogenetic subtype. Disclosures: Kumar: Merck: Consultancy, Honoraria; Millennium: Research Funding; Celgene: Consultancy, Research Funding; Novartis: Research Funding; Genzyme: Research Funding.

Abstract Background: Prospective validation of breast cancer risk models integrating classical risk factors and genetic variants is required for risk-stratified prevention and screening strategies. The objective of this study was to validate a breast cancer risk model integrating classical risk factors and a 313-variant polygenic risk score (PRS) in multiple prospective cohort studies, and to project five-year risk of breast cancer in six different countries. Methods: The study population included 7,529 cases and 230,103 controls from 14 prospective cohort studies in Australia, Germany, the Netherlands, Sweden, UK, and USA. We used the Individualized Coherent Absolute Risk Estimator (iCARE) tool for risk model building, validation, and risk projection. Expected five-year risk of invasive or in situ breast cancer was compared to observed risk, overall and within deciles of expected risk using goodness of fit statistics. We evaluated calibration of the relative risk through meta-analysis across cohorts, and of the absolute risk within each cohort. Model discrimination was evaluated using the area under the curve (AUC), and percentages of women crossing risk thresholds. Projections of five-year risk distributions were estimated for women of European ancestry aged 50-70 years in the general populations of these six countries. Results: Analysis showed overall good calibration of the integrated iCARE-based model relative risk for both women younger than 50 years (χ2=14.9, P=0.09) and aged 50 years or older (χ2=14.6, P=0.10), with a small overestimation of risk for women in the highest decile of expected risk (RR = 3.5 expected vs 2.3 (95% CI 1.6 to 3.2) observed for women &lt;50 years; and 2.8 expected vs 2.3 (95% CI 2.0 to 2.7) observed for women 50+ years). The age-adjusted AUCs for the integrated model were 63.1 (95% CI 60.9 to 65.3) and 62.9 (95% CI 61.8 to 64.0), for the two age groups respectively. The calibration of absolute risk showed substantial variation across cohorts, particularly for the older group, but had no systematic bias. Model based projections in the general populations showed that compared to the population average, women in the 1st and 99th percentiles of the integrated risk score had relative risks 0.19 and 3.56 respectively. The proportion of women of European ancestry aged 50-70 years with a five-year risk greater than 3% (threshold for consideration of risk-lowering drugs by U.S. Preventive Services Task Force) ranged from 7.1% in Germany to 18.2% in the US, which corresponds to ~5.5 million women in the US. Conclusions: Five-year risk predictions from a model with classical risk factors and PRS are well calibrated and provide substantial risk stratification across multiple cohorts in six different countries. Further studies are needed to evaluate the clinical utility of the validated model for risk stratified screening and prevention of breast cancer. Citation Format: Parichoy Pal Choudhury, Amber Wilcox, Chi Gao, Brian Carter, Anika Husing, Mark Brook, Mikael Eriksson, Kara Martin, Chris Scott, Min Shi, Thomas Ahearn, Michael Jones, Nick Orr, Minouk Schoemaker, Kamila Czene, Jenny Chang-Claude, Jacques Simard, Doug Easton, Marjanka K. Schmidt, Dale Sandler, Clarice R. Weinberg, Celine Vachon, Roger Milne, Per Hall, Anthony Swerdlow, Rudolph Kaaks, Myrto Barrdahl, Mia Gaudet, Antonis Antoniou, Peter Kraft, Montserrat Garcia-Closas, Nilanjan Chatterjee. Validation of breast cancer risk model incorporating classical risk factors and polygenic risk scores in 14 prospective cohort studies in 6 countries [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 962.

Abstract Background: High mammographic breast density is a strong, well-established breast cancer risk factor. A recent study found increased expression of stem cell markers in the areas of high breast density, but, whether stem cells may explain high breast cancer risk due to high breast density is unknown. We investigated the association between breast density and the risk of breast cancer by the status of stem cell markers CD44, CD24, and ALDH1A1 in the tumor. Methods: We included 223 postmenopausal women with primary invasive or in-situ breast cancer and 399 age-matched controls from Mayo Clinic Mammography Study. Percent breast density, absolute dense area and non-dense area were assessed using computer-assisted thresholding technique. Immunohistochemical analysis of the markers was performed on tumor tissue microarrays according to a standard protocol with commercial antibodies. Cases with positive staining (≥1% positively stained cells) on one or more cores were considered marker-positive. We used polychotomous logistic regression to quantify the associations of breast density measures with breast cancer risk by the status of each of the markers. Heterogeneity tests were conducted to compare the differences in associations by marker-defined tumor subtypes. Results: Of the 223 cancers in the study, 182 (82%) were positive for CD44, 83 (37%) for CD24 and 52 (23%) for ALDH1A1. Associations of percent density were not significantly different across any of the marker-defined subtypes (density 51%+ vs. 11-25%: OR=2.83, 95% CI 1.49-5.37 for CD44+ vs. OR=1.87, 95% CI 0.47-7.51 for CD44-, p-heterogeneity=0.66; OR=2.80, 95% CI 1.27-6.18 for CD24+ vs. OR=2.44, 95% CI 1.14-5.22 for CD24-, p-heterogeneity=0.61; OR=3.04, 95% CI 1.14-8.10 for ALDH1A1+ vs. OR=2.57. 95% CI 1.30-5.08 for ALDH1A1-, p-heterogeneity=0.94). Positive associations of absolute dense area and inverse associations of non-dense area with breast cancer risk were similar across marker-defined subtypes. Similar associations of breast density measures with breast cancer were seen when cancer subtypes were defined using combined marker expression status (none, one positive marker, or two and more positive markers). Conclusions: We found no evidence of differential associations of breast density measures with breast cancer risk by the status of stem cell markers. However, our power was limited and further studies in larger study populations are warranted to confirm these associations in a larger sample. Citation Format: Lusine Yaghjyan, Ashwini Esnakula, Christopher Scott, Akemi Wijayabahu, Matthew Jensen, Celine Vachon. Associations of mammographic breast density with breast stem cell marker-defined breast cancer subtypes [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 592.

Abstract Genome-wide association studies (GWAS) conducted among populations of European ancestry (EA) have identified 23 common single nucleotide polymorphisms (SNPs) associated with multiple myeloma (MM) risk. We hypothesize that the combination of these SNPs in a polygenic risk score (PRS) is likely to be a strong risk factor for MM. However, it is unclear whether the genetic variation associated with MM susceptibility also predisposes to monoclonal gammopathy of undetermined significance (MGUS). Thus, we calculated a PRS and evaluated the association with risk of MM and its precursor, MGUS. We pooled genotype data for 2434 MM and 3446 controls from ten MM GWAS of individuals of EA within the Interlymph Consortium, for 23 MM risk SNPs identified by prior GWAS. An additional 754 MGUS cases were ascertained from Mayo Clinic and MD Anderson clinical practices. To calculate the PRS, we used the risk estimates corresponding to the 23 SNP associations from the largest published MM GWAS. The log of the odds ratio (OR) for each SNP was multiplied by the respective number of risk alleles and summed to generate a PRS for each individual. The PRS was examined continuously, per one standard deviation (SD), and as quintiles, based on the PRS distribution in the controls. Associations of PRS with MM and MGUS risk were examined separately, using multivariable logistic regression assuming an additive model to assess ORs and 95% confidence intervals adjusted for age, sex, and site. We also evaluated age and sex stratified models. The distribution of sex within MM cases, MGUS cases and controls were each ~60% male and ~40% female. The median age was 61, 66, and 66 years for MM cases, MGUS cases and controls, respectively. PRS ranged from 1.52-4.91, with a median PRS of 3.21 for MM cases, 3.19 for MGUS cases, and 3.05 for controls. PRS was significantly associated with MM risk when assessed continuously (OR=1.19 per SD, p=2.2x10-16) and categorically; compared with the middle quintile (Q3), individuals in the highest quintile (Q5) had a 66% increased MM risk (OR=1.66, p=2.3x10-9) and those in the lowest quintile (Q1) had a 38% decreased MM risk (OR=0.62, p=1.3x10-6). PRS was also significantly associated with MGUS risk (OR=1.19 per SD, p=1.7x10-11); individuals with the highest PRS (Q5) had a 77% increased risk (OR=1.77, p=4.0x10-4) and those with lowest PRS (Q1) had 30% decreased risk (OR=0.70, p=0.04), compared with Q3. When stratified by age and sex, similar associations and trends were found. Using an independent sample of MM / MGUS cases and controls, we showed that a PRS constructed from 23 common genetic variants for MM risk is associated with risk of both MM and MGUS, regardless of age or sex. A future direction of this work is testing associations with PRS and clinical characteristics of the MM cases, as well as differences between MGUS cases that progress and those that do not. Our results suggest that common genetic variation may predispose to MGUS as the precursor to MM. Citation Format: Alyssa I. Clay-Gilmour, Michelle A. Hildebrandt, Nicola J. Camp, Elad Ziv, Elizabeth E. Brown, Jonathan N. Hofmann, John J. Spinelli, Graham G. Giles, Parveen Bhatti, Wendy Cozen, Xifeng Wu, Dennis P. Robinson, Aaron D. Norman, Jason P. Sinnwell, Shaji K. Kumar, S Vincent Rajkumar, Susan L. Slager, Celine M. Vachon. Associations between a polygenic risk score and risk of multiple myeloma and its precursor [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 2686.