Mean Number Of Variants Research Articles

Utilizing Cell-Free Urinary and Plasma Tumor DNA to Predict Pathologic Stage at Radical Cystectomy

ObjectiveTo assess the ability of cell-free urinary and plasma tumor DNA (cfDNA) to predict pathologic stage at radical cystectomy for patients with clinical muscle-invasive bladder cancer. MethodsA total of 25 patients with clinical muscle-invasive bladder cancer were enrolled before undergoing radical cystectomy. Blood and urine were collected before surgery. The 600-gene PredicineATLAS panel was used to sequence blood buffy-coat germline DNA, plasma cfDNA, and urine cfDNA samples. Low-pass whole genome sequencing was performed on plasma- and urine-derived cfDNA. CfDNA tumor fraction (TF), genome-wide copy number burden (CNB), and estimated tumor mutational burden (TMB) were measured in both plasma and urine samples and their correlation with pathologic T-stage was examined. ResultsThree of 25 plasma samples had insufficient cfDNA. In 22 of 22 plasma samples and 24 of 25 urine samples, at least one nonsynonymous somatic variant was detected. Across the cohort, 44% of plasma variants were concordant with paired urine variants. The mean number of variants did not differ between noninvasive (< pT1/pN0) and invasive disease (≥ pT1 or N+) for both plasma (8 vs. 9.5 variants; P = 0.85) and urine (33.7 vs. 30 variants; P = 0.45). A strong correlation was observed between urine TF and urine CNB score within patients (rv = 0.92). Plasma TF (r = 0.38), urine TF (r = 0.21), and urine CNB score (r = 0.16) exhibited positive correlations with pT stage. Patients with carcinoma in situ (CIS) had higher mean urine TF and CNB scores ( P = 0.07 and P = 0.05, respectively). Plasma TF and CNB score did not correlate with the presence of CIS.ConclusionsCombining plasma- and urine-based cfDNA analysis may help identify patients with residual disease at radical, although we were unable to predict pathologic T-stage based on these metrics.The presence of CIS may contribute to greater urinary CNB and TF levels. Considering CIS in the analysis may improve the ability to correlate tumor metrics with pathologic stage. Low-pass whole genome sequencing–derived urinary CNB correlates strongly with urinary TF and may provide a less resource-intensive method for future longitudinal disease monitoring.

Mutational landscape and clinical characterization of over 17,000 patients with myeloid malignancies using real-world data.

7022 Background: Myeloid neoplasms represent a broad spectrum of hematological disorders for which somatic mutation status in key driver genes is important for diagnosis, prognosis and treatment. Here we summarize the findings from 17,181 clinical samples analyzed by a targeted next generation sequencing (NGS) laboratory developed test. Samples were analyzed comprehensively and as part of individual cohorts specific to acute myeloid leukemia (AML), myelodysplastic syndrome (MDS), and myeloproliferative neoplasms (MPN). Methods: Whole blood or bone marrow samples from patients with cause-for-testing for hematological symptoms were submitted for analysis by a referring clinician. DNA was extracted and assayed by a targeted, NGS panel to detect and report single nucleotide variants and small indels within 50 genes associated with myeloid malignancies and sequenced on Illumina DNA sequencers (Illumina, San Diego, CA). Results were reviewed, orthogonally confirmed unless previously validated, and reported by clinical laboratory directors. Disease status or symptoms were taken from test requisitions for each patient. Results: Overall, 34,581 Tier I, II, or III somatic variants were detected. The mean number of variants reported per patient was 1.98. This number increased with age (p < 0.001) and was highest in patients with AML. Mutations in genes associated with FDA-approved therapies ( FLT3, IDH1, IDH2, KIT, and JAK2) were observed in 22.3% (3836) patients while 28.0% of samples had mutations considered diagnostically relevant by the WHO. Clinical trials are in progress for therapies targeting TP53, NPM1, and CEBPA mutations and were observed in 12.5% (2144) of samples. FLT3, NPM1 and RUNX1 mutation which are diagnostic for AML were found in 29.0% (646 of 2225) of AML samples. Pair-wise mutation analyses found 21 mutually exclusive pairs including between genes associated with RNA splicing ( SF3B1, SRSF2 and U2AF1; OR < 0.29, p < 0.001) suggesting possible candidates for targeted therapy. The clinically favorable co-mutation of NPM1 with FLT3 internal tandem duplicate was significantly enriched (OR = 7.4, p < 0.001) in the AML population. Co-occurrence of ASXL1 with RUNX1 mutations which independently confer adverse risk was also enriched in AML (OR = 3.9, p < 0.001). Some patients (1021; 6.3%) were tested at multiple time-points with 48.8% showing loss or gain of a mutation between sample dates, potentially the result of tumor evolution and/or therapeutic intervention. Conclusions: This study shows that parallel testing of multiple genes in addition to the canonical driver mutations encompasses the mutations contributing to the etiology of these diseases. Despite the breadth of different mutations observed for myeloid neoplasms, consistent patterns are routinely observed that can help the clinician tailor the treatment and chart the progression of these diseases for each patient.

Nationwide Laboratory Network for AML Cross-Validated NGS Studies: Results from a Real-Life Cohort of the Pethema Group

Nationwide Laboratory Network for AML Cross-Validated NGS Studies: Results from a Real-Life Cohort of the Pethema Group

Next generation sequencing in therapy-related myeloid neoplasms compared to de novo myeloid neoplasms

ABSTRACT Introduction Therapy-related myeloid neoplasms (t-MN) are frequently categorized according to previous therapy or pattern of cytogenetic abnormalities. Our objective was to evaluate and compare the mutational profile of de novo and t-MN by next generation sequencing. Methods Sixty-four samples from patients with t-MN, previously treated for a solid tumor (mainly breast), or de novo AML, MDS, MDS/MPN were selected for our study. The library was prepared using diagnostic samples and the TruSight Myeloid sequencing panel targeting 54 genes. Samples were sequenced on a MiSeq. The classification system of the Belgian ComPerMed Expert Panel was used for the biological variant classification. Results Taking only pathogenic, probably pathogenic variants and variants of unknown significance into account 141 variants in 33 genes were found in 52 of 64 samples (81%; mean number of variants per patient = 2; range = [1–11]; 67 variants in 25 genes in t-MN and 74 variants in 25 genes in de novo MN). Overall, the most frequently detected variants included TET2 (n = 22), TP53 (n = 12), DNMT3A (n = 10) and FLT3, NPM1, RUNX1 (n = 8 each). Conclusion Our study revealed a high variety of variants both in t-MN and de novo MN patients. There was a higher incidence of FLT3 and TP53 variants in t-MN compared to de novo MN.

Abstract P5-03-04: Co-segregation of rare possibly-damaging variants in cancer-related genes correlates with phenotypic homogeneity in familial breast cancer

Abstract BACKGROUND Breast cancer (BC) is a complex disease. While Mendelian mutations in high-penetrance genes predispose to some familial forms, others are due to multiple common low-penetrance polymorphisms. However, it is unclear if the combination of some rare coding variants has an effect. As some Mendelian forms demonstrate genotype-phenotype correlations, we hypothesized co-segregation of rare variants in cancer-related genes would be more frequent in high-risk families with a uniform BC phenotype among relatives. METHODS Whole-exome sequencing was performed on germline DNA from unrelated BC patients referred for genetic testing but without a causative mutation and for which DNA was available from at least one second relative with BC. We retained rare (< 1.5% in ExaC) in silico possibly damaging variants (splice alteration, nonsense or frameshift mutation, or missense variant considered deleterious by at least 2 different algorithms) in a list of 735 cancer or DNA repair related genes. We assessed correlation between segregation of these variants in affected relatives (studied by Sanger sequencing) and tumor phenotypes. A familial uniform BC phenotype was denoted as sharing the same BC histology and the same clinical molecular classification, based on the expression of the hormone receptors and HER2. RESULTS Of the 54 families, 23 (43%) presented a uniform phenotype across the affected relatives. We identified 251 variants. The mean number of variants tested did not differ between families with a similar BC phenotype and families with different BC phenotypes (4.3 vs 4.9 respectively, two-sided T-test p=0.19). The proportion of disease co-segregating variants inside each family was however significantly higher in families with a uniform BC phenotype than in families with a heterogeneous BC phenotype (57.8% vs 30.2%, p<0.01 adjusted for the number of relatives and meioses within each family). Sequencing of matched tumor DNA, obtained for 44 probands, revealed co-segregating variants were more frequently enriched by a somatic event (18.6 vs 4.5%, Mann-Whitney test p=0.042 and 8.3 vs 1.1%, p=0.031 for the families with a uniform or a heterogeneous breast cancer phenotype, respectively). CONCLUSION In this cohort of families at high risk of BC, the proportion of co-segregating variants was significantly higher in the families in which the affected relatives shared a uniform BC phenotype. Investigation of oligogenic models could be more pertinent in this subgroup of families. The theory of bi-allelic gene inactivation required to drive cancer, shown for several Mendelian genes, could also be pertinent in the oligogenic model of inheritance. This study should be replicated in an independent cohort for confirmation. Citation Format: Cedric Van Marcke, Raphaël Helaers, Céline A Schoonjans, Jérôme Ambroise, Martine Berlière, Jean-Luc Canon, Peter Vuylsteke, Jean-Pascal Machiels, Miikka Vikkula, Francois P Duhoux. Co-segregation of rare possibly-damaging variants in cancer-related genes correlates with phenotypic homogeneity in familial breast cancer [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 P5-03-04.

Abstract 2513: Mutational and transcriptional landscape of nodal marginal zone lymphoma

Abstract Nodal marginal zone lymphoma (NMZL) is a rare B-cell neoplasm lacking disease-defining markers. In this study, we aimed to determine the mutational profile of Asian NMZLs and report novel genetic features of this disease. We performed WES on FFPE samples of eight Korean NMZLs and two matched non-neoplastic tissue samples. For unmatched cases, additional variant filtering was performed to minimize germline contamination. RNAseq was performed on six NMZL FFPE samples with acceptable samples quality. A total of 394 non-synonymous variants in 370 genes were found with the mean number of variants per case of 49.3 (range 14 – 98). We selected 93 candidate NMZL genes (CNGs) based on the following criteria: (i) genes with mutations in ≥ 2 patients; (ii) genes with mutations in either of two cases with matched control; (iii) genes with mutations in at least one case and reported to have biological roles in B-cell lymphomas. Mutations in genes involved in chromatin remodeling including KMT2D (25%), HIST1H1E (25%), and HIST1H1B (13%) were observed, which was consistent with previous studies. Genes encoding key components of B-cell receptor (BCR) signaling and related pathways were also affected, including CD79B (25%), CARD11 (13%) and BCL6 (13%). Of note, NFKBIE (c.759_762delTTAC) frameshift mutation was found in one patient; this mutation was proven to be loss-of-function in nature, resulting in deregulation of NF-κB signaling. Identical mutation was reported in Hodgkin lymphoma and primary mediastinal B-cell lymphoma, however, this is the first report of this mutation in NMZL. Genes involved in immune response including MYD88 (13%), PIM1 (13%), CD70 (13%) and TNFRSF14 (13%) were also mutated, suggesting the linkage between immune dysregulation and NMZL pathogenesis. At the transcriptome level, three samples were consistently clustered together. This group of patients was characterized by higher Ki-67 proliferation index compared to the others, and was associated with older age, higher serum LDH level and higher IPI score. Gene set analysis (GSA) based on lymphoma-specific gene sets from SignatureDB (https://lymphochip.nih.gov/signaturedb/) showed significant association between XBP1 target genes and these high Ki-67 group (P = 0.024). In summary, this study showed high prevalence of mutations in chromatin remodeling and BCR signaling pathway, readdressing the importance of these pathways in NMZL. Several novel genetic alterations in NMZL were found including NFKBIE c.759_762delTTAC, warranting further validation to clarify the significance of these alterations. We also found that NMZLs with high Ki-67 index were associated with specific gene expression profile, providing clue to the explanation for the heterogeneity of NMZLs. Citation Format: Jiwon Koh, Seongmin Choi, Insoon Jang, Sehui Kim, Cheol Lee, Jin Ho Paik, Kwangsoo Kim, Yoon Kyung Jeon. Mutational and transcriptional landscape of nodal marginal zone lymphoma [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 2513.

Deep sequencing shows that oocytes are not prone to accumulate mtDNA heteroplasmic mutations during ovarian ageing.

Does ovarian ageing increase the number of heteroplasmic mitochondrial DNA (mtDNA) point mutations in oocytes? Our results suggest that oocytes are not subject to the accumulation of mtDNA point mutations during ovarian ageing. Ageing is associated with the alteration of mtDNA integrity in various tissues. Primary oocytes, present in the ovary since embryonic life, may accumulate mtDNA mutations during the process of ovarian ageing. This was an observational study of 53 immature oocyte-cumulus complexes retrieved from 35 women undergoing IVF at the University Hospital of Angers, France, from March 2013 to March 2014. The women were classified in two groups, one including 19 women showing signs of ovarian ageing objectified by a diminished ovarian reserve (DOR), and the other, including 16 women with a normal ovarian reserve (NOR), which served as a control group. mtDNA was extracted from isolated oocytes, and from their corresponding cumulus cells (CCs) considered as a somatic cell compartment. The average mtDNA content of each sample was assessed by using a quantitative real-time PCR technique. Deep sequencing was performed using the Ion Torrent Proton for Next-Generation Sequencing. Signal processing and base calling were done by the embedded pre-processing pipeline and the variants were analyzed using an in-house workflow. The distribution of the different variants between DOR and NOR patients, on one hand, and oocyte and CCs, on the other, was analyzed with the generalized mixed linear model to take into account the cluster of cells belonging to a given mother. There were no significant differences between the numbers of mtDNA variants between the DOR and the NOR patients, either in the oocytes (P = 0.867) or in the surrounding CCs (P = 0.154). There were also no differences in terms of variants with potential functional consequences. De-novo mtDNA variants were found in 28% of the oocytes and in 66% of the CCs with the mean number of variants being significantly different (respectively 0.321, SD = 0.547 and 1.075, SD = 1.158) (P < 0.0001). Variants with a potential functional consequence were also overrepresented in CCs compared with oocytes (P = 0.0019). N/A. Limitations may be due to the use of immature oocytes discarded during the assisted reproductive technology procedure, the small size of the sample, and the high-throughput sequencing technology that might not have detected heteroplasmy levels lower than 2%. The alteration of mtDNA integrity in oocytes during ovarian ageing is a recurring question to which our pilot study suggests a reassuring answer. This work was supported by the University Hospital of Angers, the University of Angers, France, and the French national research centers, INSERM and the CNRS. There are nocompeting interests.

Abstract 4298: Defects in DNA repair genes and sensitivity to cisplatin based neoadjuvant chemotherapy (NAC) for bladder cancer

Abstract Background: Cisplatin based NAC prior to cystectomy is standard of care for MIBC, with 40-50% expected to respond with ≤pT1N0M0. Biomarkers predictive of response are lacking. Methods: MIBC pts who received 3 cycles of cisplatin based NAC on 1 of 2 prospective multicenter clinical trials were included. Pts treated with accelerated methotrexate, vinblastine, doxorubicin + cisplatin (AMVAC) provided the discovery set [n = 34, 15/34 (44%) ≤pT1N0M0]. Pts treated with dose dense gemcitabine + cisplatin (DDGC) provided the validation set [n = 24, 11/24 (46%) ≤pT1N0M0]. DNA from pre-treatment tumor tissue underwent sequencing for all coding exons of 287 cancer related genes and was analyzed for presence of base substitutions, indels, copy number alterations, and selected re-arrangements. The mean number of variants and variant status for each gene were correlated with response using two-sample t-test and Fisher's exact tests. Variant data were used to create a classification tree to discriminate responders vs. non-responders in the AMVAC discovery cohort. The resulting decision rule was then tested in the independent DDGC validation set. Overall survival analysis was performed using Kaplan-Meier. Results: Pts with pT0 had significantly more alterations than those with residual tumor in both the AMVAC discovery (p = .024) and DDGC validation (p = 0.018) set. In the AMVAC discovery set, alteration in ≥1 of the three DNA repair genes ATM, RB1 or FANCC predicted for ≤pT1N0M0 (p&amp;lt;0.001, 87% sensitivity, 100% specificity) and improved overall survival (OS) (p = 0.007). This test remained predictive for ≤pT1N0M0 in the DDGC validation set (p = 0.033), with a trend towards improved OS (p = 0.07) at short median follow up of 14.3 mo. Conclusions: Alterations in ≥1 of ATM, RB1 and FANCC predict response to cisplatin based chemotherapy defined as ≤pT1N0M0 in both our AMVAC discovery and DDGC validation sets. We hypothesize that defects in these genes, which are important for maintenance of chromatin structure and DNA repair, confer sensitivity to DNA damaging chemotherapy and explain the accumulation of alterations seen among pts with pT0. External validation in collaboration with the cooperative groups is planned. Citation Format: Elizabeth R. Plimack, Roland L. Dunbrack, Timothy A. Brennan, Mark D. Andrake, Yan Zhou, Ilya Serebriiskii, Essel Dulaimi Al-Saleem, Jean Hoffman-Censits, Marijo Bilusic, Yu-Ning Wong, Alexander Kutikov, Rosalia Viterbo, Richard Greenberg, David Chen, Costas D. Lallas, Edouard J. Trabulsi, Roman Yelensky, Vincent A. Miller, Erica Golemis, Eric Ross. Defects in DNA repair genes and sensitivity to cisplatin based neoadjuvant chemotherapy (NAC) for bladder cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4298. doi:10.1158/1538-7445.AM2015-4298

Abstract 56: Fine-Mapping of Metabochip Lipid Regions in Global Populations Identifies Signals Unique to Hispanic Descent Populations and Refines Previously Identified Lipid Loci

INTRODUCTION: Genome wide association studies (GWAS) have identified over 150 loci associated with lipids traits. The majority of these GWAS were performed in European Americans (EA); no large-scale studies exist for Hispanic descent populations. Additionally, in many cases, the genetic architecture of these trait-influencing loci remains largely unknown. To address these gaps in knowledge, we performed one of the most ethnically diverse fine-mapping genetic studies on HDL-C, LDL-C, and triglycerides (TG) to-date. HYPOTHESIS: Here we aimed to identify variants with the strongest association at each locus, detect population-specific signals, and refine previously identified EA GWAS loci. METHODS: We used Metabochip data from African American (AA, ~21,000), Hispanic American (HA, ~20,000), Asian (AS, ~2,000), and Native American (NA, ~550) participants from the Population Architecture using Genomics and Epidemiology (PAGE) Study. We applied multiple linear regression models and assumed an additive mode of inheritance to test for association between genotypes and HDL-C, LDL-C, or log-transformed TG levels; lipid levels were corrected for lipid-lowering medication use. Model covariates included age, sex, and principal components of ancestry. We first conducted a meta-analysis within each ethnic group separately and then performed a combined trans-ethnic fixed effects meta-analysis. Significance was defined as p &lt; 1 x 10 -6 ; equivalent to 0.05/ the mean number of variants at each Metabochip lipid locus. RESULTS: For HDL-C, 19 loci significantly associated in the trans-ethnic meta-analysis; the top signals at 5 of these loci, APOB, LIPC, STARD3, LIPG, and APOC1, have not been reported in EA. We identified a signal unique to HA at APOA5. In addition, we refined the set of candidate functional variants at PPP1R3B, LPL, and PLTP. For LDL-C, 16 loci significantly associated in the trans-ethnic meta-analysis; the top signals at 5 of these loci, PCSK9, APOB, APOA5, CLIP2, and APOC1, have not been reported in EA. We identified a signal unique to HA at SLC22A1. In addition, we refined the set of candidate functional variants at TIMD4 and LDLR. For TG, 15 loci significantly associated in the trans-ethnic meta-analysis; the top signals at 3 of these loci, APOB, APOA5, and LIPC, have not been reported in EA. In addition, we refined the set of candidate functional variants at ANGPTL3, MLXIPL, PPP1R3B, and LPL. CONCLUSIONS: By taking advantage of the genetic architecture of ethnically diverse populations, we identified novel lipid-influencing variants in HA and refined the set of candidate functional variants at GWAS lipid loci. Anticipated conditional analyses will provide further insight into secondary and ethnic-specific signals. Our results can guide the creation of more informed risk models, which can then be used for targeted prevention efforts, especially for underrepresented populations.

Defects in DNA repair genes and sensitivity to cisplatin based neoadjuvant chemotherapy (NAC) for bladder cancer.

320 Background: Cisplatin based NAC prior to cystectomy is standard of care for MIBC, with 40-50% expected to respond with ≤pT1N0M0. Biomarkers predictive of response are lacking. Methods: MIBC pts who received 3 cycles of cisplatin based NAC on 1 of 2 prospective multicenter clinical trials were included. Pts treated with accelerated methotrexate, vinblastine, doxorubicin + cisplatin (AMVAC) provided the discovery set [n=34, 15/34 (44%) ≤pT1N0M0]. Pts treated with dose dense gemcitabine + cisplatin (DDGC) provided the validation set [n=24, 11/24 (46%) ≤pT1N0M0]. DNA from pre-treatment tumor tissue underwent sequencing for all coding exons of 287 cancer related genes and was analyzed for presence of base substitutions, indels, copy number alterations, and selected re-arrangements. The mean number of variants and variant status for each gene were correlated with response using two-sample t-test and Fisher’s exact tests. Variant data were used to create a classification tree to discriminate responders vs. non-responders in the AMVAC discovery cohort. The resulting decision rule was then tested in the independent DDGC validation set. Overall survival analysis was performed using Kaplan-Meier. Results: Pts with pT0 had significantly more alterations than those with residual tumor in both the AMVAC discovery (p=.024) and DDGC validation (p=0.018) set. In the AMVAC discovery set, alteration in ≥1 of the three DNA repair genes ATM, RB1 or FANCC predicted for ≤pT1N0M0 (p&lt;0.001, 87% sensitivity, 100% specificity) and improved overall survival (OS) (p=0.007). This test remained predictive for ≤pT1N0M0 in the DDGC validation set (p=0.033), with a trend towards improved OS (p=0.07) at short median follow up of 14.3 mo. Conclusions: Alterations in ≥1 of ATM, RB1 and FANCC predict response to cisplatin based chemotherapy defined as ≤pT1N0M0 in both our AMVAC discovery and DDGC validation sets. We hypothesize that defects in these genes, which are important for maintenance of chromatin structure and DNA repair, confer sensitivity to DNA damaging chemotherapy and explain the accumulation of alterations seen among pts with pT0. External validation in collaboration with the cooperative groups is planned.

Comparison of anatomical abnormalities in patients with limited and diffuse chronic rhinosinusitis

The role of anatomical abnormalities in the pathogenesis of chronic rhinosinusitis (CRS) remains unclear. We hypothesized that anatomical abnormalities causing obstruction of mucus drainage would be more prevalent in patients with limited CRS than in patients with pansinusitis, who are more likely to have a generalized mucosal abnormality. The computed tomography (CT) scans of patients with limited sinusitis (maxillary sinus and/or the ostiomeatal unit), patients with diffuse bilateral disease, and asymptomatic controls were studied. Lund-Mackay scores were calculated and the frequency of anatomical abnormalities that could affect mucus drainage of the ostiomeatal region was determined. The limited disease group comprised 22 patients with 96 total anatomical variants (4.4 ± 2.0; mean number of variants per patient ± standard deviation [SD]). The diffuse disease group comprised 28 patients with 70 anatomical variants (2.6 ± 2.0). The control group had 27 patients with 68 variants (2.5 ± 1.4). The frequency of total anatomical variants in the limited group was significantly higher than in the pansinusitis and control groups (p < 0.003). There was no significant difference in the total number of anatomical variations between the diffuse disease and control groups. Concha bullosae and accessory ostia were more prevalent in the local disease group than the diffuse disease group. (p < 0.01). These results suggest that patients with limited CRS may be predisposed by anatomical variants impairing drainage of the ostiomeatal region, and pansinusitis (likely resulting from generalized mucosal abnormality) is associated with a lower frequency of anatomical variants similar to the general population.

Epistatic Effects of Potassium Channel Variation on Cardiac Repolarization and Atrial Fibrillation Risk

The aim of this study was to evaluate the role of cardiac K(+) channel gene variants in families with atrial fibrillation (AF). The K(+) channels play a major role in atrial repolarization but single mutations in cardiac K(+) channel genes are infrequently present in AF families. The collective effect of background K(+) channel variants of varying prevalence and effect size on the atrial substrate for AF is largely unexplored. Genes encoding the major cardiac K(+) channels were resequenced in 80 AF probands. Nonsynonymous coding sequence variants identified in AF probands were evaluated in 240 control subjects. Novel variants were characterized using patch-clamp techniques and in silico modeling was performed using the Courtemanche atrial cell model. Nineteen nonsynonymous variants in 9 genes were found, including 11 rare variants. Rare variants were more frequent in AF probands (18.8% vs. 4.2%, p < 0.001), and the mean number of variants was greater (0.21 vs. 0.04, p < 0.001). The majority of K(+) channel variants individually had modest functional effects. Modeling simulations to evaluate combinations of K(+) channel variants of varying population frequency indicated that simultaneous small perturbations of multiple current densities had nonlinear interactions and could result in substantial (>30 ms) shortening or lengthening of action potential duration as well as increased dispersion of repolarization. Families with AF show an excess of rare functional K(+) channel gene variants of varying phenotypic effect size that may contribute to an atrial arrhythmogenic substrate. Atrial cell modeling is a useful tool to assess epistatic interactions between multiple variants.