Germline Whole Exome Sequencing Research Articles

Potential Role for the RASD1 Glucocorticoid-Responsive Gene in Corticotroph Tumorigenesis

Introduction: Originally identified due to its dexamethasone inducibility in mouse corticotropinoma AtT20 cells, RASD1 is a receptor-independent activator of G-proteins, via guanine nucleotide exchange factor (GEF) activity. It remains unclear, however, whether, and if so, how RASD1 mediates the effects of glucocorticoids on corticotroph cells. We identified a rare germline RASD1 variant and investigated its functional effects in vitro. Methods: We screened 209 CD patients (94.3% pediatric) studied at the at the National Institutes of Health Clinical Research Center between 1997 and 2018 by germline whole-exome sequencing (WES) only (n=157), germline and tumor WES (n=27), and/or RASD1 droplet digital PCR germline copy number variant (CNV) analysis (n=201). Corticotropinoma DNA was available in 72 patients to screen for USP8 hotspot variants by Sanger sequencing. A RASD1 variant was identified and functionally characterized. Results: We studied 119 female (56.9%) and 90 (43.1%) male CD cases, including 197 pediatric (≤18 years at disease onset) and 12 adult patients. USP8 defects were present in 19.4% (14/72) of cases. No RASD1 CNVs were found. A rare (with a minor allele frequency of 0.0022% in gnomAD v3) heterozygous germline missense RASD1 variant, c.580A>C, p.M194L was detected in one male sporadic case. Neither USP8 variants nor loss of heterozygosity at the RASD1 variant position were observed in the patient’s microadenoma. The wild type and p.M194L RASD1 transiently overexpressed proteins displayed similar short half-lives (<1 h) by cycloheximide chase in HEK293 cells, as well as cytoplasmic localization by immunocytofluorescence in AtT20 cells. A CRISPR/Cas9 Rasd1 knockout AtT20 cell line displayed reduced Pomc expression compared with the parental cell line at the mRNA level (Actb-normalized absolute quantification 5.80±0.92 vs 9.62±0.7, P=0.005). Viability of the cell lines did not differ significantly by MTT assay. Overexpression of p.M194L resulted in increased accumulation of phospho-CREB S133 (1.83±0.8 vs 1±0.2 in empty vector control, P=0.0390) as well as a non-significant increase in Pomc expression in wild type, but not in Rasd1 knockout AtT20 cells by immunoblot band densitometry. Conclusions: We found an infrequent RASD1 variant in one CD patient. Rasd1 seems to have a role within the intracellular signaling pathways controlling Pomc expression. Overexpression of the p.M194L variant caused phospho-CREB S133 activation, suggesting increased GEF activity for this variant. Interestingly, another variant at the same position, p.M194I, was found in the COSMIC database (COSS2121715) as a somatic change in cutaneous malignant melanoma. Further studies are required to better define the role of RASD1 in corticotroph physiology and its possible involvement in tumorigenesis.

Circulating Tumor Cell Genomic Evolution and Hormone Therapy Outcomes in Men with Metastatic Castration-Resistant Prostate Cancer.

Men with circulating tumor cell (CTC) AR-V7-positive metastatic castration-resistant prostate cancer (mCRPC) have worse outcomes when treated with enzalutamide/abiraterone. However, most men lack CTC AR-V7 detection, and additional predictive biomarkers are needed. We conducted a retrospective secondary analysis of the prospective PROPHECY trial (NCT02269982) of men with mCRPC undergoing treatment with enzalutamide/abiraterone, analyzing pooled CTC and germline DNA for whole-genome copy-number alterations (CNA) in 73 samples from 48 men over time along with pooled CTC and germline whole-exome sequencing on 22 paired samples before and following progression on androgen receptor (AR) inhibitor therapy to identify somatic genomic alterations associated with acquired resistance. We observed broad interpatient and longitudinal CTC genomic heterogeneity from AR-V7-negative men with mCRPC, including common gains of KDM6A, MYCN, and AR, and loss of ZFHX3, BRCA1, and PTEN. Men who had progression-free survival of ≤3 months despite enzalutamide/abiraterone treatment were more likely to have baseline CTC genomic loss of CHD1, PTEN, PHLPP1, and ZFHX3 and gains of BRCA2, KDM5D, MYCN, and SPARC. After progression on abiraterone/enzalutamide, we observed clonal evolution of CTCs harboring TP53 mutations and gain of ATM, KDM6A, and MYC, and loss of NCOR1, PTEN, RB1, and RUNX2. CTC genomic findings were independently confirmed in a separate cohort of mCRPC men who progressed despite prior treatment with abiraterone/enzalutamide (NCT02204943). IMPLICATIONS: We identified common and reproducible genomic alterations in CTCs from AR-V7-negative mCRPC men associated with poor outcomes during enzalutamide/abiraterone treatment, including CNAs in genes linked to lineage plasticity and epigenetic signaling, DNA repair, AR, TP53/RB1, PTEN, and WNT pathways.

MBRS-73. AN EXPLORATORY ANALYSIS LOOKING AT THE ASSOCIATION OF GERMLINE CODING MUTATIONS WITH IMPAIRED DEVELOPMENT AND ADAPTIVE BEHAVIOR FUNCTION IN PEDIATRIC MEDULLOBLASTOMA PATIENTS TREATED ON HEAD START 4

Children with brain tumors often carry germline mutations known to contribute to tumorigenesis and treatment response; however, little is known about how these mutations impact developmental and behavioral outcomes. As the molecular mechanisms governing cancerous and normal tissues expand, we hypothesize that specific germline variants may impact baseline neurocognitive function and/or treatment-induced toxicities. In this pilot study, ten children on the Head Start 4 (HS4) clinical trial diagnosed with medulloblastoma were assessed for baseline adaptive functioning using the Adaptive Behavior Assessment System Third Edition (ABAS-III) and germline whole-exome sequencing was performed. After filtering for high impact variants, Welch’s T-tests were used to identify mutations associated with lower ABAS-III General Adaptive Composite (GAC) scores, reflecting developmental and adaptive behavior delays compared with peers their age. We found twenty genes with alterations associated with lower scores with p-values less than 0.05. Genes found to be significant included LAMC1 (p=0.04) and KRTAP1-1 (p=0.045), which encode members of the laminin and keratin family respectively and are involved in extracellular matrix adhesion. Mutations in PITX1, a known suppressor of RAS, were also associated with lower ABAS-III GAC scores (p=0.007). We hypothesize that additional analyses of HS4 patients will reveal alterations in cell-to-cell communication and signal transduction pathways, common molecular perturbations in tumors that would likely impact central nervous system function. Validation studies are essential to improve our understanding of the functional impact of germline variants on both tumor and regular tissue biology, allowing for novel strategies to circumvent these delays.

Common deleterious germline variants shape the urothelial cancer genome

The prevalence and biological consequences of deleterious germline variants (DGVs) in urothelial cancer (UC) are unknown. We performed whole-exome sequencing (WES) of 158 tumors and corresponding germline DNA from 80 UC patients at Weill-Cornell Medicine (WCM). We developed a novel computational framework (DGVar) to detect DGVs from germline WES data and predict their biological functions. We used strict criteria to identify truncating variants in 1604 tumor suppressor genes (TSGs) from germline WES data. We assessed germline-somatic interactions over the lifetime of each tumor. We confirmed our findings by applying DGVar to germline WES from 398 patients from the Cancer Genome Atlas (TCGA). We performed extensive validation of our results in other UC cohorts and whole-exome sequencing data from more than 13,000 non-cancer subjects. DGVs were identified in 45/80 patients (56%) of the WCM UC cohort and 315 DGVs in 48% (190/398) of patients in the TCGA UC cohort. DGVs were significantly more common in UC patients of WCM and TCGA compared to 2,504 subjects from the 1000 genome project (1KGP) population (p<0.0001). Similarly, our DGVs were more common in WCM and TCGA UC compared to 11,209 adult parents of European ancestry from the SPARK study of autism. KLK6, POLQ, and ITGA7 variants were the most common DGVs in WCM (15%) and TCGA UC (5%) cohorts. DGVs in the DNA repair pathway genes, including BRCA2, FANCA, POLQ, and XPA were identified in 7.5% of patients. Select DGVs were validated by Sanger sequencing of germline DNA including a novel XPA Leu200* variant, which eliminated the DNA-binding domain. DGVar identified 60 DGVs which are not detectable using currently used germline clinical targeted sequencing panels. Functional modeling showed that 96% of DGVs clustered with somatic variants within 15 Angstroms in genes with known crystal structures. Consistent with the pathogenic role of identified DGVs, we discovered frequent bi-allelic inactivation events arising from somatic losses of wild-type alleles. Furthermore, we observed deepening loss of heterozygosity of DGV genes in serial tumor samples as UC progressed from primary to metastatic sites. Finally, we identified extensive intra-patient heterogeneity in germline-somatic variant interactions and characterized their biological consequences. We show that close half of UC patients harbor DGVs, which potentially play a critical role in UC initiation and progression. Our results redefine the landscape of germline variants in urothelial cancer and provide a pivotal new understanding of their unique role in the biology of the disease.

NCOG-46. A PILOT STUDY OF GERMLINE CODING MUTATIONS ASSOCIATED WITH IMPAIRED DEVELOPMENT AND ADAPTIVE BEHAVIOR IN PEDIATRIC MEDULLOBLASTOMA PATIENTS TREATED ON HEAD START 4

Abstract Children with brain tumors often carry germline mutations that have long been known to contribute to tumorigenesis and treatment response. However, less is known about how germline mutations may impact developmental and behavioral outcomes for children with tumors of the central nervous system (CNS). As the molecular mechanisms governing cancerous and normal tissue expand, we hypothesize that particular germline variants may impact baseline neurocognitive function and/or treatment-induced toxicities. In this preliminary study, 10 children on the Head Start 4 trial diagnosed with medulloblastoma were assessed for baseline adaptive functioning using the Adaptive Behavior Assessment System Third Edition (ABAS-III) and germline whole-exome sequencing was performed. After filtering for high impact variants, Welch’s T-tests were used to identify mutations associated with lower ABAS-III General Adaptive Composite (GAC) scores, reflecting developmental and adaptive behavior delays compared with peers their age. We found 20 genes with alterations associated with lower scores with P-values less than 0.05, although none met FDR significance cutoffs. Genes found to be significant included LAMC1 (P=0.04) and KRTAP1-1 (P=0.045), which encode members of the laminin and keratin family and are involved in extracellular matrix adhesion. Mutations in PITX1, a known suppressor of RAS, were also associated with lower ABAS-III GAC scores (P=0.007). We suspect that follow-up studies with more patients will reveal alterations in cell-to-cell communication and signal transduction pathways, as these are common molecular perturbations in tumors that would likely impact regular CNS function. Future research with larger cohorts is essential to validate these findings and to improve our understanding of the functional impact of germline variants on both tumor and regular tissue biology, allowing for a more comprehensive view of the spectrum of neurodevelopmental impairment and novel strategies to circumvent these impairments. Funding: Thrasher Research Fund (JF, AC). R01CA108633, RC2CA148190, U10CA180850(NCI), and OSUCCC (AC).

Inherited Rare, Deleterious Variants in ATM Increase Lung Adenocarcinoma Risk.

Lung cancer is the leading cause of cancer deaths in the world, and lung adenocarcinoma (LUAD) is its most prevalent subtype. Symptoms are often found in advanced disease in which treatment options are limited. Identifying genetic risk factors will enable better identification of high-risk individuals. To identify LUAD risk genes, we performed a case-control association study for gene-level burden of rare, deleterious variants (RDVs) in germline whole-exome sequencing data of 1083 patients with LUAD and 7650 controls, split into discovery and validation cohorts. Of these, we performed whole-exome sequencing on 97 patients and acquired the rest from multiple public databases. We annotated all rare variants for pathogenicity conservatively, using the guidelines of the American College of Medical Genetics and Genomics and ClinVar curation, and investigated gene-level RDV burden using penalized logistic regression. All statistical tests were two-sided. We discovered and replicated the finding that the burden of germline ATM RDVs was significantly higher in patients with LUAD versus controls (combined cohort OR= 4.6; p= 1.7e-04; 95% confidence interval= 2.2-9.5; 1.21% of cases; 0.24% of controls). Germline ATM RDVs were also enriched in an independent clinical cohort of 1594 patients from the MSK-IMPACT study (0.63%). In addition, we observed that an Ashkenazi Jewish (AJ) founder ATM variant, rs56009889, was statistically significantly more frequent in AJ cases versus AJ controls in our cohort (combined AJ cohort OR= 2.7, p= 6.9e-03, 95% confidence interval= 1.3-5.3). Our results indicate that ATM is a moderate-penetrance LUAD risk gene and that LUAD may be a part of the ATM-related cancer syndrome spectrum. Individuals with ATM RDVs are at an elevated LUAD risk and can benefit from increased surveillance (particularly computed tomography scanning), early detection, and chemoprevention programs, improving prognosis.

Expanding the diagnostic yield of germline genetic testing in cancer patients using deep learning.

1518 Background: Germline genetic analysis is an essential tool for implementing precision cancer prevention and treatment. However, only a small fraction of cancer patients, even those with features suggestive of a cancer-predisposition syndrome, have detectable pathogenic germline events, which may in part reflect incomplete pathogenic variant detection by current gold-standard methods. Here, we leveraged deep learning approaches to expand the diagnostic utility of genetic analysis in cancer patients. Methods: Systematic analysis of the detection rate of pathogenic cancer-predisposition variants using the standard clinical variant detection method and a deep learning approach in germline whole-exome sequencing data of 2367 cancer patients (n = 1072 prostate cancer, 1295 melanoma). Results: Of 1072 prostate cancer patients, deep learning variant detection identified 16 additional prostate cancer patients with clinically actionable pathogenic cancer-predisposition variants that went undetected by the gold-standard method (198 vs. 182), yielding higher sensitivity (94.7% vs. 87.1%), specificity (64.0% vs. 36.0%), positive predictive value (95.7% vs. 91.9%), and negative predictive value (59.3% vs. 25.0%). Similarly, germline genetic analysis of 1295 melanoma patients showed that, compared with the standard method, deep learning detected 19 additional patients with validated pathogenic variants (93 vs. 74) with fewer false-positive calls (78 vs. 135) leading to a higher diagnostic yield. Collectively, deep learning identified one additional patient with a pathogenic cancer-risk variant, that went undetected by the standard method, for every 52 to 67 cancer patients undergoing germline analysis. Superior performance of deep learning, for detecting putative loss-of-function variants, was also seen across 5197 clinically relevant Mendelian genes in these cohorts. Conclusions: The gold-standard germline variant detection method, universally used in clinical and research settings, has significant limitations for identifying clinically relevant pathogenic disease-causing variants. We determined that deep learning approaches have a clinically significant increase in the diagnostic yield across commonly examined Mendelian gene sets.

Germline CDKN1B Loss-of-Function Variants Cause Pediatric Cushing's Disease With or Without an MEN4 Phenotype.

Germline loss-of-function CDKN1B gene variants cause the autosomal dominant syndrome of multiple endocrine neoplasia type 4 (MEN4). Even though pituitary neuroendocrine tumors are a well-known component of the syndrome, only 2 cases of Cushing's disease (CD) have so far been described in this setting. To screen a large cohort of CD patients for CDKN1B gene defects and to determine their functional effects. We screened 211 CD patients (94.3% pediatric) by germline whole-exome sequencing (WES) only (n = 157), germline and tumor WES (n = 27), Sanger sequencing (n = 6), and/or germline copy number variant (CNV) analysis (n = 194). Sixty cases were previously unpublished. Variant segregation was investigated in the patients' families, and putative pathogenic variants were functionally characterized. Five variants of interest were found in 1 patient each: 1 truncating (p.Q107Rfs*12) and 4 nontruncating variants, including 3 missense changes affecting the CDKN1B protein scatter domain (p.I119T, p.E126Q, and p.D136G) and one 5' untranslated region (UTR) deletion (c.-29_-26delAGAG). No CNVs were found. All cases presented early (10.5 ± 1.3 years) and apparently sporadically. Aside from colon adenocarcinoma in 1 carrier, no additional neoplasms were detected in the probands or their families. In vitro assays demonstrated protein instability and disruption of the scatter domain of CDKN1B for all variants tested. Five patients with CD and germline CDKN1B variants of uncertain significance (n = 2) or pathogenic/likely pathogenic (n = 3) were identified, accounting for 2.6% of the patients screened. Our finding that germline CDKN1B loss-of-function may present as apparently sporadic, isolated pediatric CD has important implications for clinical screening and genetic counselling.

Abstract PD8-01: The metastatic breast cancer project: Generating the clinical and genomic landscape of metastatic breast cancer through patient-partnered research

Abstract Background: The Metastatic Breast Cancer Project (MBCproject) is a research study that directly engages patients (pts) through social media and advocacy groups, and empowers them to share their samples, clinical information, and experiences. The goal is to create a publicly available dataset of linked genomic, clinical, and pt-reported data to enable research. Methods: In collaboration with pts, advocates, and advocacy groups, a website (MBCproject.org) was developed that allows pts with metastatic breast cancer (MBC) anywhere in the U.S. or Canada to register. Registered pts are sent an online consent form that asks for permission to obtain and analyze their medical records and samples. Once enrolled, pts are sent a saliva kit and a blood kit and asked to mail back a saliva sample, which is used to extract germline DNA, and/or a blood sample, which is used to extract germline DNA and cell free DNA (cfDNA). We contact participants’ medical providers and obtain medical records and a portion of their stored tumor biopsies. Whole exome sequencing (WES) is performed on tumor DNA, germline DNA, and cfDNA; transcriptome sequencing (RNA-seq) is performed on tumor RNA. Medical records and pt-reported data are abstracted to create a detailed clinical record for each pt. All de-identified data are shared regularly via public databases (cbioportal.org, mbcproject.org, dbGaP, NCI Genomic Data Commons) without restrictions. Study updates are shared with participants regularly. Results: From 10/20/15-7/8/19, 5357 women and men with MBC registered. 3290 pts receiving care at over 1700 different institutions consented to share medical records and tumor/saliva/blood samples, and to have genomic analysis performed. Details of clinical data collection, biospecimen acquisition, and genomic data generation to date are outlined in the Table. WES from 463 tumors obtained from 326 pts have been generated (with matched germline WES), including 61 pts with 2 timepoints, 19 pts w 3 timepoints, and 11 pts w 4+ timepoints. 278 tumor exomes were from the breast/regional lymph nodes, 63 from distant metastatic sites and 122 from cfDNA. 110 tumor exomes were from samples obtained before the diagnosis of MBC, 258 from after the diagnosis of MBC, and 95 to be determined (TBD). 161 tumor exomes were obtained prior to any therapy, 204 following some therapy, and 98 TBD. Clinically annotated genomic data are used to study specific pt cohorts (including rare subsets and outliers) and to identify mechanisms of response and resistance to therapies. Examples of the clinical and genomic analyses that will be presented include: - Pts diagnosed &amp;lt;40 yrs of age (1108 pts enrolled; 120 with tumor WES) - de novo MBC (1122 pts enrolled; 121 with tumor WES) - Late recurrence, &amp;gt;5 yrs after diagnosis (830 pts enrolled; 77 with tumor WES) - Long-term survivors, &amp;gt;10 yrs with MBC (159 pts enrolled; 11 with tumor WES) - Resistance to CDK4/6 inhibitors (709 pts enrolled; 148 with tumor WES) Conclusions: Partnering directly with pts enables rapid identification of thousands of pts willing to share tumors, blood, saliva, and medical records to accelerate research. This approach allows for identification of patients with specific phenotypes, who have been challenging to identify with traditional approaches. Remote acquisition of medical records and saliva/blood/tumor samples is feasible. This clinically annotated dataset is a shared resource for the research community. Table 1Clinical data collection, biospecimen acquisition, and genomic data generation:NumberConsent signed3290 ptsSurvey #1 submitted3290 pts(demographics, diagnosis details, receptor status, clinical experiences)Survey #2 submitted1435 pts(pathology details, sites of metastasis, treatments with start and stop dates)Medical record received1307 ptsSaliva sample received1976 ptsBlood sample received1121 ptsTumor samples received482 tumor samples from 346 ptsDigital image of tumor slide H&amp;E generated482 tumor samplesWES from germline complete310 germline samplesWES from tumor sample complete341 tumor samplesRNA-seq from tumor sample complete229 tumor samplesULP-WGS from cfDNA complete947 blood samplesWES from circulating tumor DNA complete122 blood samples Citation Format: Nikhil Wagle, Corrie Painter, Elana Anastasio, Michael Dunphy, Mary McGillicuddy, Esha Jain, Tania G Hernandez, Sara Balch, Beena Thomas, Dewey Kim, Alyssa L. Damon, Shahrayz Shah, Brett N. Tomson, Rachel Stoddard, Colleen Nguyen, Jorge Buendia-Buendia, Ofir Cohen, Jorge Gomez Tejeda Zanudo, Netsanet Tsegai, Lauren Sterlin, Ulcha Fergie Ulysse, Kathryn Sine, Oyin Alao, Jacqueline Lucia, Eric S. Lander, Todd R. Golub. The metastatic breast cancer project: Generating the clinical and genomic landscape of metastatic breast cancer through patient-partnered research [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 PD8-01.

Unconventional Diagnosis Based on Somatic Findings through Germ Line Whole-Exome Sequencing.

Unconventional Diagnosis Based on Somatic Findings through Germ Line Whole-Exome Sequencing.

Integrated Somatic and Germline Whole-Exome Sequencing Analysis in Women with Lung Cancer after a Previous Breast Cancer.

Women treated for breast cancer (BC) are at risk of developing secondary tumors, such as lung cancer (LC). Since rare germline variants have been linked to multiple cancer development, we hypothesized that BC survivors might be prone to develop LC as a result of harboring rare variants. Sixty patients with LC with previous BC (the study population; SP) and 53 women with either BC or LC and no secondary cancer (control population; CP) were enrolled. Whole exome sequencing was performed in both tumors and unaffected tissues from 28/60 SP patients, and in germline DNA from 32/53 CP. Candidate genes were validated in the remaining individuals from both populations. We found two main mutational signature profiles: S1 (C>T) in all BCs and 16/28 LCs, and S2 (C>A) which is strongly associated with smoking, in 12/28 LCs. The burden test over rare germline variants in S1-LC vs CP identified 248 genes. Validation confirmed GSN as significantly associated with LC in never-smokers. In conclusion, our data suggest two signatures involved in LC onset in women with previous BC. One of these signatures is linked to smoking. Conversely, regardless of smoking habit, in a subgroup of BC survivors genetic susceptibility may contribute to LC risk.

3323 Juvenile Polyposis Syndrome Patients Without a Mutation in SMAD4 or BMPR1A: Clinical Presentation and Novel Drivers of Disease

OBJECTIVES/SPECIFIC AIMS: Juvenile Polyposis Syndrome (JPS) is an inherited cancer predisposition syndrome sometimes attributed to a germline mutation in SMAD4 or BMPR1A. However, many patients meet clinical criteria for JPS without having a pathogenic alteration in either gene. Herein, we perform a cross-sectional analysis of JPS patients at a pediatric and adult tertiary referral center to understand potential differences in the clinical presentation and outcomes of patients with or without a known causative gene mutation. Additionally, we conduct whole exome sequencing (WES) on a subset of the pediatric patients to evaluate for novel genomic drivers of disease. METHODS/STUDY POPULATION: Data were abstracted from medical charts using IRB-approved protocols at the Children’s Hospital of Philadelphia (CHOP) and the University of Pennsylvania (Penn). Records were reviewed for patients with a clinical diagnosis of JPS and genetic testing result and seen at either institution in the last 10 years (2008-2018). Patients recruited for sequencing were consented for blood draw through the CHOP IRB protocol, and had whole exome sequencing completed at 70X depth, with data analyzed through institutional pipeline. RESULTS/ANTICIPATED RESULTS: Records were reviewed for 41 patients at CHOP and 19 patients at Penn, for a total of 60 JPS patients. Mean age of CHOP cohort was 11 years: 58.5% male, mean length of follow up 3.9 years. Mean age Penn cohort was 33 years: 47.4% male, mean length of follow up 9.3 years. In the pediatric cohort, 7 patients (17%) had a mutation in BMPR1A (n=6) or SMAD4 (n=1); in the adult cohort, 15 patients (79%) had a mutation in BMPR1A (n=3) or SMAD4 (n=12). The average number of polyps in the pediatric cohort was not significantly higher in patients with a SMAD4 or BMPR1A mutation (9.3 polyps/year of surveillance with a SMAD4 or BMPR1A mutation, vs 5.7 polyps/year; p=0.19). In combined cohort review, all individuals that required gastrectomy and/or colectomy (n=8) as well as all those who developed gastrointestinal cancer (n=3) had a mutation in SMAD4 or BMPR1A. Of the patients who underwent whole exome sequencing (n=13), potential causative germline mutations were identified in four patients (30.8%); all potential drivers identified were within the TNF/BMP pathway. DISCUSSION/SIGNIFICANCE OF IMPACT: This data from a dual-institution review demonstrates that the rate of SMAD4/BMPR1A mutation in JPS is lower in a pediatric cohort compared to an adult cohort. Furthermore, although individuals with JPS may have similar clinical presentations in childhood regardless of whether or not a causative mutation is present, the presence of a mutation in SMAD4 or BMPR1A is associated with a more severe course of disease in adulthood. Further study and a larger cohort will be required to fully validate these findings. Approximately 30% of patients who underwent germline WES had a potential novel driver identified, with further validation underway.

Abstract 5378: Whole exome and RNA sequencing identify novel somatic mutations in gangliogliomas

Abstract Gangliogliomas (GG) are rare, well differentiated, slow growing neuroepithelial tumors, composed of neoplastic mature ganglion cells and neoplastic glial cells. GG comprises 0.4% of all CNS primary tumors, and 70% of GG cases are diagnosed in patients with long-term temporal lobe epilepsy. The molecular pathogenesis of GG is poorly understood. Few groups have described a high frequency of a hotspot BRAF p.V600E mutation. More recently, a co-occurrence of histone H3 p.K27M and BRAF p.V600E mutations was reported in a pediatric midline grade I GG. However, the rarity of these tumors poses challenges to uncovering a more comprehensive landscape of somatic alterations. In this study, we describe the somatic mutational profiles of 12 GG from the University of Sao Paulo Medical School in one of the largest GG case series reported to date. We performed paired germline and somatic whole exome sequencing (WES) and RNA-Seq in all cases: 9 males and 3 females with ages ranging from 1 to 49 years (mean 17 ± 13 years). These tumors were located in the temporal lobe (8), frontal lobe (1), occipital lobe (1), intraventricular region (1), and intramedullary region (1). We used the Nextera Exome Enrichment Kit (Illumina) to perform WES, and the TruSeq Stranded Total RNA (Illumina) to perform RNA-Seq. Sequencing was performed in the Illumina HiSeq 2000. We aligned reads from WES and RNA-Seq libraries to the reference genome (hg19) using BWA and STAR, respectively. We used freebayes to perform variant call from germline WES paired with the respective tumor library (WES or RNA-Seq). We used the vcfcompare tool from vcflib to filter out germline variants and annovar to annotate the final list of variants. We identified mutations in cancer-related genes in 6/12 GGs: hotspot BRAF p.V600E in 4/6 cases, biallelic NF1 mutations (p.Y333X/p.R2258X) co-occurring with hotspot H3F3A (p.K27M) and FGFR1 (p.N546K) mutations in a 13-yo female with intraventricular GG, and, finally, biallelic hotspot TP53 mutations (p.V173M, p.P172T) combined with a hotspot IDH1 (p.R132H) mutation in a 26-yo male with frontal GG. Interestingly, the same combination of mutations observed in one of our patients (FGFR1 p.N546K/H3F3A p.K27M) was recently described in a diffuse leptomeningeal tumor with glial and neuronal markers. In summary, we identified novel mutations in cancer-related genes in 6 GG (50% of our cohort), including cases with mutations in more than one known cancer driver gene. In one of the largest GG case series reported to date, this study expands the spectrum of somatic alterations in GGs to include mutations in genes recurrently altered in higher-grade gliomas, such as IDH1, NF1, and TP53. Citation Format: Suely K. Marie, Sueli M. Oba-Shinjo, Antonio M. Lerario. Whole exome and RNA sequencing identify novel somatic mutations in gangliogliomas [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5378.

Abstract 1243: Elucidating the molecular pathogenesis of familial glioma

Abstract INTRODUCTION: In recent years, the molecular characterization of sporadically arising diffuse gliomas has identified recurrent driving alterations and delineated molecularly and clinically distinct subclasses of disease. However, less is known about the molecular nature of gliomas that are familial in origin. To address this question, we integrated germline and somatic genomic data to characterize the molecular pathogenesis of 20 tumors arising in unrelated individuals with a family history of glioma collected through the Gliogene International Consortium. METHODS: FFPE tumor specimens were sectioned and reviewed to localize neoplastic tissue for DNA extraction. Library preparation, exome plus targeted capture, and paired-end sequencing on the Illumina HiSeq 2000 platform was performed at the Baylor College of Medicine Human Genome Sequencing Center. Single-nucleotide variants and indels were called with respect to germline DNA sequencing data for each case using MuTect2. Copy number profiling was performed on the Illumina HumanOmniExpress BeadChip and analyzed using GenomeStudio v2.0. Genotypes at known glioma risk polymorphisms were determined from germline DNA profiled on the Illumina Infinium OncoArray and rare, predicted deleterious germline mutations were identified from germline whole-exome sequencing data. RESULTS: Tumor exome sequencing was completed at an average read depth of 116X and we detected a median of 54 non-silent somatic mutations per tumor across the 20 tumors profiled. All three molecular subtypes of sporadic glioma were represented, including IDH-mutant, 1p/19q codeleted (n=3), IDH-mutant, 1p/19q intact (n=7), and IDH-wildtype tumors (n=10). Characteristic subtype-specific mutations and copy number alterations (e.g., TP53 and ATRX mutations among IDH-mutant, 1p/19q intact tumors) were observed, and the frequencies of recurrent alterations were comparable to sporadic glioma cases analyzed by The Cancer Genome Atlas. Notably, all 20 cases had alterations in genes regulating telomere length; 17 had acquired mutations in ATRX or the TERT promoter as typically seen in sporadic glioma, while three instead had germline mutations in telomere shelterin complex genes POT1 or TERF2. Frequencies of known common glioma risk alleles were similar to those among sporadic cases and correlations between risk alleles and specific somatic mutations were not observed. CONCLUSIONS: This study illustrates: 1) the complementarity of inherited and acquired alterations in driving gliomagenesis in some individuals with a familial predisposition to the disease; and 2) that the molecular characteristics of familial tumors profiled largely recapitulate what is known about sporadic glioma. In the majority of cases, the source of germline genetic susceptibility is not known but does not appear to be conferred by common risk polymorphisms. Citation Format: Daniel I. Jacobs, Kazutaka Fukumura, Matthew N. Bainbridge, Georgina N. Armstrong, Donna M. Muzny, Beatrice Melin, Jason T. Huse, Melissa L. Bondy. Elucidating the molecular pathogenesis of familial glioma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1243.

A pilot study of exome sequencing in a diverse New Zealand cohort with undiagnosed disorders and cancer

ABSTRACTWe report the results of a pilot project for clinical DNA sequencing in New Zealand. This project aimed to estimate the diagnostic yield of next generation sequencing in the New Zealand clinical environment. Trio whole exome sequencing (WES) was performed on germline DNA of 40 individuals from 12 families with presumptive Mendelian disorders. In addition, both WES and deep targeted sequencing (DTS) was performed on tumours, metastases and corresponding normal blood leukocytes from two cancer patients. For the rare Mendelian disorder cohort, the diagnostic yield was 6/12, including previously recognised pathogenic mutations and novel mutations. In tumour sequence analysis, WES identified somatic single nucleotide mutations and copy number aberrations in both cancer patients; however, DTS was required to obtain clinically informative information. This study showed that diagnostic germline and tumour WES and DTS could be easily undertaken in New Zealand, and identified specific infrastructural challenges that must be solved to facilitate its clinical use.

Unraveling genetic predisposition to familial or early onset gastric cancer using germline whole-exome sequencing

Recognition of individuals with a genetic predisposition to gastric cancer (GC) enables preventive measures. However, the underlying cause of genetic susceptibility to gastric cancer remains largely unexplained. We performed germline whole-exome sequencing on leukocyte DNA of 54 patients from 53 families with genetically unexplained diffuse-type and intestinal-type GC to identify novel GC-predisposing candidate genes. As young age at diagnosis and familial clustering are hallmarks of genetic tumor susceptibility, we selected patients that were diagnosed below the age of 35, patients from families with two cases of GC at or below age 60 and patients from families with three GC cases at or below age 70. All included individuals were tested negative for germline CDH1 mutations before or during the study. Variants that were possibly deleterious according to in silico predictions were filtered using several independent approaches that were based on gene function and gene mutation burden in controls. Despite a rigorous search, no obvious candidate GC predisposition genes were identified. This negative result stresses the importance of future research studies in large, homogeneous cohorts.

Abstract 1115: Germline single nucleotide polymorphisms in DNA repair genes in urothelial cancer patients

Abstract Introduction: Germline single nucleotide polymorphisms (SNPs) have been investigated in several cancers. Repair of DNA damage is a key cellular process involved in the development of chemotherapy-resistance. The role of germline SNPs in DNA damage repair genes (DRGs) in determining resistance to DNA-damaging agents in urothelial carcinoma (UC) patients is not completely understood. Methods: We examined a cohort of 53 UC patients (median age 67, 42 males) enrolled in our IRB-approved Precision Medicine program. Patients had histologically confirmed UC (43 bladder, 10 upper tract UC) and received treatment with platinum-based chemotherapy. We isolated germline DNA from peripheral blood lymphocytes or buccal swabs, and used whole exome sequencing (WES) to examine germline SNPs. As a reference for SNP frequencies we used the Exome Aggregation Consortium (ExAC) database, which represents the largest, randomly selected, germline WES database in the general population, including 60,706 individuals, among which 7,601 patients with multiple cancer types from the Tumor Cancer Genome Atlas (TCGA) cohort. Results: Twelve different DRG SNPs were identified in germline DNA samples from 53 patients, affecting genes involved in non-homologous end-joining (RECQL4, n = 18, 54.50%; POLQ, n = 2, 6%), nucleotide excision repair (ERCC6, n = 2, 6%; XPA, n = 1, 3%; CCNH, n = 1, 3%; POLK, n = 1, 3%), homologous recombination (RNF168, n = 1, 3%; RAD17, n = 1, 3%; POLE, n = 1, 3%), Fanconi anemia pathway (POLN, n = 1, 3%), mismatch repair (EXO1, n = 1, 3%) and mitochondrial DNA repair (POLG, n = 2, 6%). The frequency of rs11342077 of the DNA helicase RECQL4 was significantly higher in our cohort (18/53, 34%) compared to its frequency in the ExAC database (p&amp;lt;0.01%). There was no significant difference in overall survival (OS) between patients with and without DRG SNPs (log-rank p=0.46). There was no significant association between the most commonly identified SNP, rs11342077, and overall survival among UC patients with DRG SNPs (log-rank p=0.39). Overall, the presence of DRG SNPs in our cohort occurred at a significantly higher frequency (33/53, 62.3%) compared to the ExAC database (0.56%, chi-square p&amp;lt;0.01). Conclusions: Germline SNPs in DNA repair pathway genes are common in UC. Additional study of the role of these SNPs as potential biomarkers of response DNA damaging chemotherapeutic agents, including platinum-based chemotherapy or/and PARP inhibitors is needed in a larger cohort. Citation Format: Bishoy M. Faltas, Panagiotis J. Vlachostergios, Linda Lam, Tuo Zhang, Olivier Elemento, Mark A. Rubin. Germline single nucleotide polymorphisms in DNA repair genes in urothelial cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1115. doi:10.1158/1538-7445.AM2017-1115

Parental Perspectives on Whole Exome Sequencing in Pediatric Cancer: A Typology of Perceived Utility.

To explore how parents of pediatric cancer patients perceived the utility of clinical tumor and germline whole-exome sequencing (WES) results. We conducted longitudinal interviews with parents of a diverse pediatric cancer population before disclosure of WES results (n=64), then one to eight months (n=33) after disclosure. Interview transcripts were analyzed using a thematic qualitative approach. Parents identified a broad range of types of utility for their child's WES results. Even when results did not affect their child's current treatment, they expressed optimism about future clinical utility for their child, themselves, and other family members. Parents also reported experiencing psychological utility including peace of mind, relief of guilt, and satisfaction of curiosity. Pragmatic utility, such as the ability to plan for the future and make better reproductive decisions, was also described. Parents of pediatric cancer patients perceive WES to have broad utility, including psychological and pragmatic utility, even if there is no direct impact on clinical care. Further work will need to consider how the value of genomic information should be characterized, how risks and benefits should be described, and how these results should inform recommendations and decisions about using WES.

Defining Clonal Plasticity Using Whole Exome Sequencing in Single Cells in an Index Case of amp1q21 Multiple Myeloma

Introduction: Genomic-wide analysis of multiple myeloma (MM) reveals subclonal evolution as a major mechanism in disease evolution and progression, and interrogation of the entire functional genome at the single cell (SC) level is essential to fully map intra- clonal variation (ICV) and competition as the driver mechanism. To progress this, we have sought to establish whole exome sequencing (WES) in SCs in MM, initially in an index case of amp1q21 MM. Methods: CD138+ tumour cells and CD3+ T-cells were isolated from a presentation case of amp1q21 MM. Single MM and T cells were isolated for single cell WES. Whole genome amplification was performed using Qiagen REPLI-g Mini kit, and exome capture was performed using Agilent SureSelect. Libraries were 90 bp PE sequenced on an Illumina HiSeq2000 (BGI, China). Data was produced for bulk (1000 cells) MM and T cells, 20 MM SCs and 5 T cell SCs. Fastq were aligned to hg19 ref sequence using NovoAlignMPI (v3.02.03). Variant calling was performed using VarScan (v2.3.6) and variants were annotated using ANNOVAR. High confidence variants were called in the bulk tumour WES by pairwise comparison with bulk germline WES and by annotation against variant databases to exclude germline variants. Multiple quality control measures were employed to minimise false positives. Results and Discussion: SC WES generated raw data that were similar to bulk WES, with comparable mapping to target (69-76% SC vs. 70% bulk) and mean fold coverage (56.8-59.1x vs. 59.7x bulk). On average, 82% of the exome was covered sufficiently for somatic variant (SV) calling (! 5x), matching seminal published SC WES studies (70-80%) (Hou et al., Cell, 2012; Xu et al., Cell, 2012). We called 33 high confidence potentially deleterious SVs in the bulk tumour, 21 of which were also called in ! 1 SC exomes. SVs include mutations in genes involved in plasma cell differentiation, the MAPK pathway and known pathways in origins of B cellmalignancy. Random SVs were validated by Sanger sequencing with 100% concordance in the bulk (15/15) and SCs (55/55). ICV was apparent from the SC exome variant data. Total variant counts varied considerably among SCs and most variant positions had at least several cells where no evidence of the variant existed. Bulk WES lacks crucial information: We called 23 variants in ! 2 SCs that were absent in the bulk. Of these, 7/7 amplifiable variants were re-sequenced to obtain 100% concordance. These variants are of high interest as they reveal a marked occurrence of subclonal mu- tations in the MM tumour that are not identified by bulk WES. They indicate that the mutational status of the MM genome may be substantially underestimated by bulk WES. Conclusion: We establish the feasibility of SC WES as a method for defining true intraclonal genetic variation as a driver of cancer plasticity in MM.

Assigning clinical meaning to somatic and germ-line whole-exome sequencing data in a prospective cancer precision medicine study.

Implementing cancer precision medicine in the clinic requires assessing the therapeutic relevance of genomic alterations. A main challenge is the systematic interpretation of whole-exome sequencing (WES) data for clinical care. One hundred sixty-five adults with metastatic colorectal and lung adenocarcinomas were prospectively enrolled in the CanSeq study. WES was performed on DNA extracted from formalin-fixed paraffin-embedded tumor biopsy samples and matched blood samples. Somatic and germ-line alterations were ranked according to therapeutic or clinical relevance. Results were interpreted using an integrated somatic and germ-line framework and returned in accordance with patient preferences. At the time of this analysis, WES had been performed and results returned to the clinical team for 165 participants. Of 768 curated somatic alterations, only 31% were associated with clinical evidence and 69% with preclinical or inferential evidence. Of 806 curated germ-line variants, 5% were clinically relevant and 56% were classified as variants of unknown significance. The variant review and decision-making processes were effective when the process was changed from that of a Molecular Tumor Board to a protocol-based approach. The development of novel interpretive and decision-support tools that draw from scientific and clinical evidence will be crucial for the success of cancer precision medicine in WES studies.Genet Med advance online publication 26 January 2017.