Tumor-normal Pairs Research Articles

Development and extensive sequencing of a broadly-consented Genome in a Bottle matched tumor-normal pair.

The Genome in a Bottle Consortium (GIAB), hosted by the National Institute of Standards and Technology (NIST), is developing new matched tumor-normal samples, the first to be explicitly consented for public dissemination of genomic data and cell lines. Here, we describe a comprehensive genomic dataset from the first individual, HG008, including DNA from an adherent, epithelial-like pancreatic ductal adenocarcinoma (PDAC) tumor cell line and matched normal cells from duodenal and pancreatic tissues. Data for the tumor-normal matched samples comes from thirteen distinct state-of-the-art whole genome measurement technologies, including high depth short and long-read bulk whole genome sequencing (WGS), single cell WGS, and Hi-C, and karyotyping. These data will be used by the GIAB Consortium to develop matched tumor-normal benchmarks for somatic variant detection. We expect these data to facilitate innovation for whole genome measurement technologies, de novo assembly of tumor and normal genomes, and bioinformatic tools to identify small and structural somatic mutations. This first-of-its-kind broadly consented open-access resource will facilitate further understanding of sequencing methods used for cancer biology.

Comprehensive Analysis of Microsatellite Instability in Canine Cancers: Implications for Comparative Oncology and Personalized Veterinary Medicine.

Microsatellite instability (MSI) is a crucial feature in cancer biology, yet its prevalence and significance in canine cancers remain largely unexplored. This study conducted a comprehensive analysis of MSI across 10 distinct canine cancer histotypes using whole-exome sequencing data from 692 tumor-normal sample pairs. MSI was detected in 64% of tumors, with prevalence varying significantly among cancer types. B-cell lymphomas exhibited the highest MSI burden, contrasting with human studies. A novel "MSI-burden" score was developed, correlating significantly with tumor mutational burden. MSI-high (MSI-H) tumors showed elevated somatic mutation counts compared to MSI-low and microsatellite stable tumors. The study identified 3632 recurrent MSI-affected genomic regions across cancer types. Notably, seven of the ten cancer types exhibited MSI-H tumors, with prevalence ranging from 1.5% in melanomas to 37% in B-cell lymphomas. These findings highlight the potential importance of MSI in canine cancer biology and suggest opportunities for targeted therapies, particularly immunotherapies. The high prevalence of MSI in canine cancers, especially in B-cell lymphomas, warrants further investigation into its mechanistic role and potential as a biomarker for prognosis and treatment response.

DeepSomatic: Accurate somatic small variant discovery for multiple sequencing technologies.

Somatic variant detection is an integral part of cancer genomics analysis. While most methods have focused on short-read sequencing, long-read technologies now offer potential advantages in terms of repeat mapping and variant phasing. We present DeepSomatic, a deep learning method for detecting somatic SNVs and insertions and deletions (indels) from both short-read and long-read data, with modes for whole-genome and exome sequencing, and able to run on tumor-normal, tumor-only, and with FFPE-prepared samples. To help address the dearth of publicly available training and benchmarking data for somatic variant detection, we generated and make openly available a dataset of five matched tumor-normal cell line pairs sequenced with Illumina, PacBio HiFi, and Oxford Nanopore Technologies, along with benchmark variant sets. Across samples and technologies (short-read and long-read), DeepSomatic consistently outperforms existing callers, particularly for indels.

Comprehensive genomic profiling of infiltrative follicular variant of papillary thyroid carcinoma.

Infiltrative follicular variant of papillary thyroid carcinoma (IFVPTC) exhibits nuclear characteristics typical of papillary thyroid carcinoma (PTC) but demonstrates a follicular growth pattern. The diagnosis of IFVPTC presenting with atypical nuclear features of PTC poses challenges for both preoperative cytopathology and postoperative histopathology. In such cases, molecular markers are needed to serve as diagnostic aids. Given the limited knowledge of IFVPTC's genomic features, this study aimed to characterize its genetic alterations and identify clinically relevant molecular markers. Whole-exome sequencing of 50 IFVPTC tumor-normal pairs identified single-nucleotide variants, somatic copy number alterations (sCNAs), and subclonal architecture. Key mutations were verified via polymerase chain reaction and Sanger sequencing, whereas valuable biomarkers were validated via immunohistochemistry (IHC). This study found that endogenous processes rather than exogenous mutagens dominated the shaping of the genome of IFVPTC during tumorigenesis. BRAF V600E was the only common trunk mutation and significantly mutated gene in IFVPTC. Subcloning analysis found that most IFVPTC samples harbored two or more coexisting clones. sCNA analysis revealed that human leukocyte antigen C (HLA-C) and HLA-A were significantly amplified. Subsequent IHC investigations indicated that HLA-C shows promise in averting the misclassification of challenging-to-interpret IFVPTC and invasive encapsulated follicular variant of PTC (I-EFVPTC) as noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP). Although there were several similarities between classic PTC and IFVPTC, they differed significantly in their sCNA patterns. This study provides valuable insights into IFVPTC's genetic alterations and highlights the potential of HLA-C IHC to distinguish challenging-to-interpret IFVPTC and I-EFVPTC from NIFTP, which will enhance the understanding of its molecular features for improved diagnosis and management.

Abstract 3551: Continued analysis of extensive data towards Genome in a Bottle benchmarks for a new tumor normal pair

Abstract Here, we detail in-progress genome-scale measurements from a variety of technologies of the first tumor normal benchmark from the Genome in a Bottle (GIAB) consortium. We created the first broadly-consented tumor cell line from a pancreatic ductal adenocarcinoma with matched normal pancreatic and duodenal tissue. Data is being collected from a large homogeneous batch of the tumor cell line and paired normal tissues. As we receive data we make it publicly available on the NCBI hosted GIAB FTP site, https://ftp.ncbi.nlm.nih.gov/ReferenceSamples/giab/data_somatic/. When complete, this dataset will include WGS measurements from Illumina, ONT, Hi-C, Bionano, single cell WGS, PacBio HiFi and Onso, and Element. Analysis and development of a variant calling benchmark using this data is ongoing in collaboration with an open working group of the GIAB consortium. Initial Hi-C and optical mapping data from the tumor cell line indicates substantial aneuploidy from translocations that cause large deletions. We found roughly 17 large inversions and translocations and 16 chromosomes with extensive loss of heterozygosity due to missing >30% of one copy, in addition to a few smaller duplications. Low coverage single cell sequencing that provides ploidy estimates across chromosomes showed that most of the large deletions appear in all or nearly all cells with some variation in a few cells. These observations are consistent with bulk WGS analyses from two batches of cells. Additionally, many of the observed large deletions correspond to deletions seen in the population of TCGA chromosomal Loss of Heterogeneity samples. Examining somatic SNVs in non-repetitive regions, we find that close to 60% occur in almost all cells in diploid regions, 30% occur in almost all cells in haploid regions, and 5% in only some cells in diploid and haploid regions, respectively. We take these results to indicate the cell line is relatively homogeneous and stable with most CNVs being deletions. As such, we plan to explore using long reads, ultralong reads, and Hi-C data to generate a near-complete genome assembly of the dominant tumor clone as well as a complete diploid assembly of the normal. With this personalized genome assembly, we will explore aligning tumor and normal reads to each haplotype of the normal to characterize somatic variants, including variants in minor clones. Building on the methods GIAB and T2T have developed to polish diploid assemblies of GIAB’s normal genomes and mosaic variant characterization, we will develop benchmarks for somatic variants against the normal assembly as well as standard references. Citation Format: Justin Wagner, Jennifer McDaniel, Gail L. Rosen, Nathan D. Olson, Vaidehi Patel, Chunlin Xiao, Andrew Liss, Justin Zook. Continued analysis of extensive data towards Genome in a Bottle benchmarks for a new tumor normal pair [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3551.

Abstract 4926: Advancements in somatic variant calling from UG100 whole genome and whole exome sequencing data

Abstract Somatic variant calling involves the identification of genomic alterations that occur in somatic cells, requiring deep coverage to enable high sensitivity for low-frequency variants. Characterizing somatic variants across the entire genome therefore benefits from novel cost-efficient sequencing platforms, such as UG100. Here, we present optimization of variant calling tools for short and structural variants on WGS and WES data from UG100. For calling short variants, we optimized DeepVariant (DV) for somatic calling using data from matched tumor-normal sample pairs, improving both variant calling accuracy and pipeline running time (up to 10-fold). We defined the task of somatic variant calling as deciding if the pileup image containing reads from the tumor and normal samples represents a true somatic variant (vs a germline variant or artifact). The challenge of robust variant calling using deep learning models is exacerbated in somatic calling, where sequencing depth and coverage variability are typically high. Our optimized DV overcomes these challenges by several data sampling strategies. First, allele-frequency preserving down-sampling reduces randomness of read sub-sampling in high coverage regions. Second, alternative allele prioritization samples alt-allele supporting reads first allowing to call variants at very high coverage loci without sacrificing sensitivity and computational efficiency. Finally, a Panel-of-Normals based on targeted WES data provides an additional improvement of precision for this assay type. We used these strategies to train two models, one for tumor characterization using WGS (T/N coverage: 40x-150x/40x-100x), and one for deep WES (T/N coverage: >500x/>120x). We called variants on simulated tumors using the WGS model. For VAF>10% the model showed SNV recall >98% and indel recall >95% with false-positive rate of 0.2/Mb. For VAF range of 5-10%, indel recall was 67% and SNV recall was 86%. To demonstrate the utility of our somatic variant calling, we applied the models to call somatic variants from well characterized cancer cell lines: COLO829, HCC1395 and HCC1143. Results showed F1>90% for variants with VAF>10%. The WES model was used to reliably call variants at VAF>5% on simulated tumors with average SNV recall of 99% with precision >99% and indel recall >86% with precision >94%. To analyze structural and copy-number variations, we optimized the assembly engine of GRIDSS to enable fast calling of structural variations and demonstrate that Control-FREEC can be used to call copy number variants. SV calling on COLO829/COLO829BL achieved sensitivity >95%. In conclusion, our research highlights the utility of UG100 within the field of oncology, demonstrating its capacity for comprehensive and precise somatic variant detection, both on WGS and WES data. Citation Format: Doron Shem-Tov, Maya Levy, Gil Hornung, Ilya Soifer, Hila Benjamin, Ariel Jaimovich, Adam Blattler, William Brandler, Robert Sugar, Isaac Kinde, Omer Barad, Doron Lipson. Advancements in somatic variant calling from UG100 whole genome and whole exome sequencing data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4926.

Abstract 2946: Characterizing the genomic landscapes of breast and lung tumors using cost-effective whole genome sequencing

Abstract Whole Genome Sequencing (WGS) has emerged as a pivotal tool for unraveling the intricate genomic alterations underlying cancer, enabling the detection of a broad spectrum of somatic changes in cancer genomes, including single nucleotide variants, insertions, deletions, copy number variations, and structural rearrangements. Characterizing somatic variants across the entire genome significantly benefits from novel cost-efficient sequencing platforms, such as UG100. In the current study, we aim to demonstrate the utility of UG100 WGS in characterizing the landscape of somatic events in breast and lung cancer. Over 35 tumor-normal cancer sample pairs were profiled using UG100 WGS (mean tumor coverage: >100x; normal: >50x). Variant calling was done using a somatic deep-variant algorithm, optimized to UG100 data. Results were compared to Illumina sequencing and high concordance was observed, with over 97% of SNVs and over 93% of indels detected by Illumina identified also by UG100. UG100 sequencing was deeper and allowed for identification of a larger number of genomic variations, including known driver events in cancer-related genes. The somatic mutational signatures by UG100 were concordant with the Illumina platform, with a cosine similarity higher than 99%. Copy number variation and structural variation analysis was performed on all sample pairs, revealing a wide range of genomic aberrations. The integration of genetic sequencing into routine clinical oncology practices has proven instrumental in identifying actionable mutations, predicting treatment responses, and monitoring minimal residual disease. Moving from targeted sequencing to WGS provides a more comprehensive and unbiased assessment of the entire genome, uncovering rare or unexpected mutations that might be missed by targeted approaches. WGS allows for the simultaneous evaluation of both coding and non-coding regions of the genome, providing a deeper understanding of the regulatory elements influencing gene expression. This broader scope enhances our understanding of cancer biology and contributes to a more accurate assessment of the tumor's functional landscape. Citation Format: Hila Benjamin, Ilya Soifer, Doron Shem-Tov, Maya Levy, Islam Oguz Tuncay, Baek-Lok Oh, Jong-Yeon Shin, Young Seok Ju, Sangmoon Lee, Omer Barad, Doron Lipson. Characterizing the genomic landscapes of breast and lung tumors using cost-effective whole genome sequencing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2946.

Abstract 6218: Haplotype-resolved analysis of cancer genomes and epigenomes using Oxford Nanopore sequencing

Abstract Cancer is a complex and dynamic disease driven by somatic genomic and epigenomic alterations that accumulate over time. These changes give rise to heterogeneous collections of cells or clones, each with distinct (epi)genomic profiles within a single tumour. Accurate characterisation of these changes is crucial for understanding the mechanisms driving the disease, identifying potential therapeutic targets, and personalising treatment strategies. Due to the technological constraints of short-read and array-based approaches, cancer research has historically had a strong focus on detecting small genomic changes like SNPs and small indels (SNVs) as well as the broad characterisation of large-scale copy number changes (CNVs), mostly ignoring other important variant classes and epigenetic modifications. Here we demonstrate how Oxford Nanopore native long-read sequencing enables the direct detection of not only SNVs, but also break point-resolved simple and complex structural variants (SVs and CNVs), haplotype phasing of all variant types, and identification of DNA modifications like 5-methylcytosine (5mC) and 5-hydroxy-methylcytosine (5hmC) from a single tumour-normal dataset. We illustrate how our end-to-end somatic workflow streamlines analysis for the different variant classes and benchmark somatic SNV and SV calling performance using well characterised cancer cell lines and in-silico benchmarking datasets at different sequencing depths. Finally, we use real-world tumour-normal pairs to showcase a comprehensive sample to answer tumour-normal analysis using Nanopore sequencing. This includes the characterisation of complex patterns of somatic SVs in the different cancer samples, the identification of unique 5mC methylation patterns, exploring characteristic differences in 5hmC levels between tumour and normal samples as well as showing how Nanopore long reads enable haplotype and clone specific CNV calling as well as phasing of genetic and epigenetic variation. Citation Format: Philipp Rescheneder, Phill James, Sean McKenzie, Andrea Talenti, Sergey Aganezov, Dan Turner, Sissel Juul. Haplotype-resolved analysis of cancer genomes and epigenomes using Oxford Nanopore sequencing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6218.

Unraveling the role of the mitochondrial one-carbon pathway in undifferentiated thyroid cancer by multi-omics analyses

The role of the serine/glycine metabolic pathway (SGP) has recently been demonstrated in tumors; however, the pathological relevance of the SGP in thyroid cancer remains unexplored. Here, we perform metabolomic profiling of 17 tumor-normal pairs; bulk transcriptomics of 263 normal thyroid, 348 papillary, and 21 undifferentiated thyroid cancer samples; and single-cell transcriptomes from 15 cases, showing the impact of mitochondrial one-carbon metabolism in thyroid tumors. High expression of serine hydroxymethyltransferase-2 (SHMT2) and methylenetetrahydrofolate dehydrogenase 2 (MTHFD2) is associated with low thyroid differentiation scores and poor clinical features. A subpopulation of tumor cells with high mitochondrial one-carbon pathway activity is observed in the single-cell dataset. SHMT2 inhibition significantly compromises mitochondrial respiration and decreases cell proliferation and tumor size in vitro and in vivo. Collectively, our results highlight the importance of the mitochondrial one-carbon pathway in undifferentiated thyroid cancer and suggest that SHMT2 is a potent therapeutic target.

Genomic profiling of bladder cancer in waterpipe and cigarette smokers: Implications for carcinogenicity and genetic landscape.

657 Background: While cigarette smoking is a well-established risk factor for bladder cancer (BC), limited research has explored the contribution of waterpipe smoking (WPS) to BC. WPS is gaining popularity among Middle Eastern youth, often perceived as a safer alternative to cigarettes. Alarmingly, Lebanese males had an age-standardized incidence rate of 54.0 per 100,000 in 2019, the highest in the Arab World and five times the global average, prompting a critical investigation into potential risk factors. Mounting evidence suggests that WPS exposes users to toxicants and carcinogens that can induce chromosomal damage contributing to cancer. This study aims to assess the genomic and mutational profiles of BC in waterpipe smokers compared to cigarette smokers. Methods: This cross-sectional study enrolled adults with urothelial carcinoma (UC) who exclusively smoked waterpipe at the American University of Beirut Medical Center (AUBMC) between April 2017 and September 2019. Next-generation sequencing by MSK-IMPACT characterized mutational profiles, tumor mutation burden, and microsatellite instability (MSI) status in tumor-normal pairs. For comparison, a cohort from Memorial Sloan Kettering Cancer Center (MSKCC), comprising previously profiled BC tumors from cigarette smokers, was used. Results: A total of 14 exclusively WPS-BC patients were enrolled, with 12 males (85.7%) and a median age at diagnosis of 62.9 years (range: 30.7 to 78.9). Among these patients, 9 (64.3%) had non-muscle invasive BC, with 8 (57.2%) classified as high-grade UC. MSK-IMPACT sequencing successfully analyzed 13 WPS samples, revealing genomic patterns similar to the cigarette smoking cohort. Common genetic alterations in the WPS cohort included mutations in TERT promoter (62%), TP53 and FGFR3 (46% each), and PIK3CA (38%). Mutations in chromatin modifiers were less prevalent, with the highest frequencies observed in KDM6A (31%) and KMT2C (15%). The tumor mutation burden was higher in the WPS group (22 per MB) compared to the cigarette smoking cohort (9 per MB), with no definite evidence of MSI. PD-L1 expression was generally low, with only 2 samples exhibiting combined positivity scores (CPS) of 15 and 10, primarily associated with immune cells PD-L1 expression. Only one tumor demonstrated 3% PD-L1 expression on tumor cells, while 7 tumors displayed no expression on either tumor or immune cells (CPS of 0). Conclusions: This study underscores the genomic similarities between BC in waterpipe and cigarette smokers. The preliminary findings, limited by sample size, highlight the need for further investigations to elucidate the potential role of WPS in BC development. Future efforts will focus on cohort expansion to validate initial observations, raising awareness of health risks associated with WPS, and emphasizing the importance of continued research to inform preventive strategies.

Whole-Exome Sequencing Reveals the Genomic Profile and IL6ST Variants as a Prognostic Biomarker of Paraneoplastic Pemphigus–Associated Unicentric Castleman Disease

Unicentric Castleman disease (UCD) is a rare lymphoproliferative disorder. Paraneoplastic pemphigus (PNP) is a major complication associated with poor UCD prognosis. However, the genomic profiles and prognostic biomarkers of PNP-associated UCD remain unclear. Here, we performed whole-exome sequencing analysis for 28 matched tumor-normal pairs, and nine tumor-only samples to define the genomic landscape of Chinese patients with PNP-associated UCD. An integrative analysis was performed to identify somatic mutations, the mutational signatures, and key pathways in tumors. Besides, we analyzed the relationship among mutated genes, clinical characteristics, and prognosis. Sixty-one somatic mutant genes were identified in more than one patient with PNP-associated UCD. Specifically, IL6ST and PDGFRB were the most frequently mutated genes (32%), followed by DPP6 (18%) and MUC4 (18%). Signaling molecules and interactions, cellular processes and signal transduction pathways were enriched. Furthermore, we found that poor overall survival was related to IL6ST mutations (p=0.02). Finally, we classified PNP-associated UCD into four genomic subgroups: IL6ST, PDGFRB, IL6ST-PDGFRB, and an unknown subgroup. In summary, we defined the molecular profile of PNP-associated UCD and identified a potential molecular biomarker for predicting prognosis, which may provide therapeutic targets for treating this severe disorder.

UV-radiation and MC1R germline mutations are risk factors for the development of conventional and spitzoid melanomas in children and adolescents

Genomic characterisation has led to an improved understanding of adult melanoma. However, the aetiology of melanoma in children is still unclear and identifying the correct diagnosis and therapeutic strategies remains challenging. Exome sequencing of matched tumour-normal pairs from 26 paediatric patients was performed to study the mutational spectrum of melanomas. The cohort was grouped into different categories: spitzoid melanoma (SM), conventional melanoma (CM), and other melanomas (OT). In all patients with CM (n=10) germline variants associated with melanoma were found in low to moderate melanoma risk genes: in 8 patients MC1R variants, in 2 patients variants in MITF, PTEN and BRCA2. Somatic BRAF mutations were detected in 60% of CMs, homozygous deletions of CDKN2A in 20%, TERTp mutations in 30%. In the SM group (n=12), 5 patients carried at least one MC1R variant; somatic BRAF mutations were detected in 8.3%, fusions in 25% of the cases. No SM showed a homozygous CDKN2A deletion nor a TERTp mutation. In 81.8% of the CM/SM cases the UV damage signatures SBS7 and/or DBS1 were detected. The patient with melanoma arising in giant congenital nevus (CNM) demonstrated the characteristic NRAS Q61K mutation. UV-radiation and MC1R germline variants are risk factors in the development of conventional and spitzoid paediatric melanomas. Paediatric CMs share genomic similarities with adult CMs while the SMs differ genetically from the CM group. Consistent genetic characterization of all paediatric melanomas will potentially lead to better subtype differentiation, treatment, and prevention in the future. Found in Acknowledgement.

Profiling endogenous, environmental, and infectious disease mutational signatures in blastic plasmacytoid dendritic cell neoplasms.

Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare hematopoietic disease derived from plasmacytoid dendritic lineage cells. The disease typically shows skin as well as frequent bone marrow and peripheral blood involvement. However, the pathogenesis of this disease is still not well understood. While somatic point mutations and genetic rearrangements have been described in BPDCN, the types and origins of these mutations and relationships to other cancer types is not well understood. To probe the origins of BPDCN, we analyzed the exome sequence data of 9 tumor-normal pair cases of BPDCN. We utilized SignatureAnalyzer, SigProfiler and a custom microbial analysis pipeline to understand the relevance of endogenous and environmental mutagenic processes. Our results identified a significant tobacco exposure and aging genetic signature as well as signatures related to nucleotide excision repair deficiency, ultra violet (UV) exposure, and endogenous deamination in BPDCN. We also assessed the samples for microbial infectious disease organisms but did not find a link to a microbial etiology. The identification of a tobacco exposure and aging genetic signature in patients with BPDCN suggests that environmental and endogenous genetic changes may be central to the oncogenesis of BPDCN.

Genomics of ERBB2-Positive Breast Cancer in Young Women Before and After Exposure to Chemotherapy Plus Trastuzumab.

Erb-B2 receptor tyrosine kinase 2 (ERBB2)-positive breast cancer (BC) is particularly common in young women. Genomic features of ERBB2-positive tumors before and after chemotherapy and trastuzumab (chemo + H) have not been described in young women and are important for guiding study of therapeutic resistance in this population. From a large prospective cohort of women age 40 years or younger with BC, we identified patients with ERBB2-positive BC and tumor tissue available before and after chemo + H. Whole-exome sequencing (WES) was performed on each tumor and on germline DNA from blood. Tumor-normal pairs were analyzed for mutations and copy number (CN) changes. Twenty-two women had successful WES on samples from at least one time point; 12 of these had paired sequencing results from before and after chemo + H and 10 had successful sequencing from either time point. TP53 was the only significantly recurrently mutated gene in both pre- and post-treatment samples. MYC gene amplification was observed in four post-treatment tumors. Seven of 12 patients with paired samples showed acquired and/or clonally enriched alterations in cancer-related genes. One patient had an increased clonality putative activating mutation in ERBB2. Another patient acquired a clonal hotspot mutation in TP53. Other genomic changes acquired in post-treatment specimens included alterations in NOTCH2, STIL, PIK3CA, and GATA3. There was no significant change in median ERBB2 CN (20.3 v 22.6; Wilcoxon P = .79) between paired samples. ERBB2-positive BCs in young women displayed substantial genomic evolution after treatment with chemo + H. Approximately half of patients with paired samples demonstrated acquired and/or clonally enriched genomic changes in cancer genes. ERBB2 CN changes were uncommon. We identified several genes warranting exploration as potential mechanisms of resistance to therapy in this population.

Association of KMT2C loss-of-function mutations in circulating tumor DNA and prolonged response to the combination of PARPi and PI3Ki.

e13003 Background: The emergence of somatic KMT2C mutations has been associated with resistance to various treatments, including chemotherapy and endocrine therapy in patients with estrogen receptor-positive breast cancers. The impact of these mutations in patients treated with poly-ADP ribose polymerase inhibitors (PARPi) and phosphatidylinositol-3-kinase inhibitors (PI3Ki) is unknown. Methods: Circulating tumor DNA (ctDNA) was obtained from plasma and peripheral blood at different time points from 38 patients prior to treatment with the combination of the PARPi olaparib with the alpha-specific PI3Ki alpelisib (NCT01623349). CtDNA was also collected immediately before the first cycle and at the time of disease progression. CtDNA underwent deep targeted sequencing (tumor-normal pairs, 397 genes (2Mb) at 25,000x somatic and 10,000x germline) at the Broad Institute of MIT and Harvard. Variants that were exonic, non-synonymous, and predicted to lead to loss-of-function were considered mutations. Cohorts were stratified according to the presence or absence of the genomic alteration, and progression-free survival was compared with the Kaplan-Meier method and log-rank test. A Cox regression was performed using age, tumor type, germline BRCA1/2 mutation status, performance status, and number of previous lines of treatment as covariates. Results: The KMT2C gene was one of the 3 genes most frequently mutated in our cohort, affecting 5 patients (15%) after the exclusion of those with variants with non-deleterious alterations. The other 2 commonly mutated genes were TP53 (76%) and PIK3CA (15%). In the comparison of patients based on these most common mutations, we identified that patients with somatic KMT2C mutations (n=5) have longer mPFS than those that were wild type (n=28): mPFS 386 vs. 151 days, log-rank p = 0.025. Cox proportional hazards model accounting for age, tumor type, germline BRCA1/2 mutation status, performance status, and number of previous lines revealed a hazard ratio of 0.16; p = 0.007. All somatic KMT2C mutations were proximal to the SET domain critical for methyltransferase activity. Upon progression, the KMT2C-mutated clones were suppressed to below the detection level of mutation calling algorithms. Analysis of the raw files revealed that the clones were suppressed but not eliminated. Conclusions: While the emergence of KMT2C has been reported to be associated with cancer progression and resistance to different therapies, patients with deleterious KMT2C mutations in the ctDNA may benefit from therapy with PARPi and PI3Ki.

Engagement of patient participation in genomics research by the Osteosarcoma and Leiomyosarcoma Count Me In Projects of the Cancer Moonshot funded PE-CGS Network.

TPS11588 Background: Osteosarcoma (OS) and Leiomyosarcoma (LMS) are sarcomas with complex genomes for which there has been limited progress in identifying new treatments and improving outcomes. Slow progress in OS and LMS is partially due to insufficient characterization of the genomic landscape. Generating large genomic datasets in OS and LMS is challenging because of the rarity of these sarcomas and recruitment barriers such as care fragmentation between institutions and specialties. The OS and LMS Project research studies aim to: 1) establish a network of engaged pediatric and adult participants with OS and LMS who will co-create a shared database of clinical, genomic, molecular, and patient reported data to enable research; 2) define the clinicogenomic landscape of OS and LMS; and 3) optimize the approach to direct patient engagement in cancer research. Methods: Count Me In, a research initiative with prior success in angiosarcoma, working with patients and advocates created websites (OSProject.org and LMSProject.org) where patients register and consent to participation. Projects were launched in September, 2022. After 5 months of accrual, 306 patients age 6-79 from 149 Institutions have consented. Blood and saliva are collected from consented participants, tumor samples are obtained from pathology departments and medical records are requested from treating hospitals. WES and WGS of tumor and normal, and RNASeq of tumor is performed. ctDNA is obtained and sequenced. Results are shared with patient, advocacy, physician and research communities in several ways. Individual participants receive a shared learning report describing the somatic variants identified in their tumor from paired tumor-normal WES and are offered genetic counseling and clinical germline testing. Registered participants receive updates via email and Project websites. There are regular pre-publication data releases to the genomic data commons and to cBioPortal. A physician engagement committee meets regularly to discuss clinical insights and conundrums from shared learning reports and germline testing. Patient accrual over the next 3 years is anticipated to result in sequencing of 750 tumor-normal pairs and 500 ctDNA samples.

Whole-exome and targeted gene sequencing of large-cell lung carcinoma reveals recurrent mutations in the PI3K pathway.

Large cell lung carcinoma (LCLC) is an exceptionally aggressive disease with a poor prognosis. At present, little is known about the molecular pathology of LCLC. Ultra-deep sequencing of cancer-related genes and exome sequencing were used to detect the LCLC mutational in 118 tumor-normal pairs. The cell function test was employed to confirm the potential carcinogenic mutation of PI3K pathway. The mutation pattern is determined by the predominance of A > C mutations. Genes with a significant non-silent mutation frequency (FDR) < 0.05) include TP53 (47.5%), EGFR (13.6%) and PTEN (12.1%). Moreover, PI3K signaling (including EGFR, FGRG4, ITGA1, ITGA5, and ITGA2B) is the most mutated pathway, influencing 61.9% (73/118) of the LCLC samples. The cell function test confirmed that the potential carcinogenic mutation of PI3K pathway had a more malignant cell function phenotype. Multivariate analysis further revealed that patients with the PI3K signaling pathway mutations have a poor prognosis (P = 0.007). These results initially identified frequent mutation of PI3K signaling pathways in LCLC and indicate potential targets for the treatment of this fatal type of LCLC.

Expression and prognostic value of ING3 in advanced laryngeal squamous cell carcinoma

Objectives: Laryngeal squamous cell carcinomas (LSCC) is one of the most common aggressive neoplasms of the head and neck region. There is a significant need for identification of successful and accurate prognostic markers to better estimate the clinical outcomes for LSCC patients. In this study, we aimed at analyzing the differential expressions of inhibitor growth (ING) family members and to evaluate the prognostic values of deregulated ING genes in LSCC. Methods: We investigated the relative expressions of ING genes in laryngeal tumor-normal tissue pairs in the mRNA level using quantitative real-time polymerase chain reaction and relative expression of ING3 in the protein level using Western Blot analysis. Results: The rate of genetic alterations of ING3 was relatively higher in head and neck cancer including LSCC. ING3 expression was significantly upregulated in LSCC tissue samples in both mRNA and protein level. Higher expression of ING3 was also correlated with poor disease-free survival of patients with head and neck cancer. Conclusions: Our findings assigned an oncogenic feature for ING3 in laryngeal cancer with a significant upregulation detected in advanced cases and suggested a vital prognostic potential for ING3.

Abstract 5356: Collaborating to ensure data-driven drug discovery on the Cancer Genomics Cloud: Realizing the possibilities for MoDaC and ATOM

Abstract Introduction. The NCI-funded Cancer Genomics Cloud (CGC) and the NCI Predicative Oncology Model and Data Clearinghouse (MoDaC) advance NCI computing infrastructure and tools that aim to reduce the burden of cancer on patients. The CGC provides a collaborative cloud base computation infrastructure that collocates computation, bioinformatics workflows, and 3+ PB data to researchers. MoDaC provides a publicly available resource generated from the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C) Program and the Accelerating Therapeutics for Opportunities in Medicine (ATOM Consortium). The NCI-sponsored MoDaC program aims to add machine learning toolsets to identifying novel treatments for Cancer Patients. We present our progress collaborating with MoDaC to make machine learning models available on the Cancer Genomics Cloud. Methods. Our Bioinformatics teams migrated MoDaC tools on the CGC, defined standards for releasing models, and collected recommendations for decreasing the time required to make MoDaC tools available on the Cancer Genomics Cloud. We mirrored the ATOM Modeling Pipeline (AMPL), a drug discovery platform, on the CGC by making the GIT repository cloud accessible through Jupyter Notebook access, supporting interactive analysis. We translated AMPL and JDACS4C models into CWL. Converting these ML models into CWL supports reproducible execution, scalable deployment, and computational portability. Results. The AMPL drug discovery platform on the CGC includes data ingestion &amp; curation, featurization, model training &amp; tuning, prediction generation, visualization &amp; analysis functionality as Jupyter notebooks and CWL workflows. The release consists of chemo-informatics tools for integrating cancer treatment features in deep-learning graph models. JDACS4C ML Models ported to the CGC include classifiers (tumor and normal-tumor pairs), autoencoders (Gene Expression), drug response predictors (single and combination), and Multitask Convolutional Neural Networks (extract information from cancer pathology reports). Conclusion. We optimized the MoDaC Drug Discovery and Machine Learning tools into cloud-native resources on the CGC, supporting interactive and GUI-driven analysis. The release supports technical and newer users to machine learning, allowing access to a broader user base than those who traditionally have access to ML toolsets. These MoDaC toolsets will support pre-clinical study evaluation, treatment identification, and experimental design. Moreover, existing MoDaC-AMPL tutorials on the CGC support distributed ML-Drug Discovery training. Lastly, collaborating with MoDaC teams identified standardization approaches that can reduce the time and effort to make these tools widely available across the NIH-NCI computational infrastructure. Citation Format: Soner Koc, Vojislav Varjacic, Miona Rankovic, Marijeta Slavkovic-Ilic, Aleksandar Danicic, Sean Black, Naomi Ohashi, Titli Sarkar, Zelia Worman, Jack DiGiovanna, Brandi Davis-Dusenbery, Dennis A. Dean. Collaborating to ensure data-driven drug discovery on the Cancer Genomics Cloud: Realizing the possibilities for MoDaC and ATOM. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5356.

Abstract 1513: A multi-sample study reveals the evolution and heterogeneity in high-grade serous ovarian cancer

Abstract Ovarian cancer is a heterogeneous disease and high-grade serous carcinoma (HGSC) accounts for 70% of them. The pathogenesis of HGSC had remained obscure until the identification of its tubal origin; however, the carcinogenesis process remains largely underexplored. Meanwhile, targeted therapy using poly(ADP-ribose) polymerase (PARP) inhibitors have revolutionized the treatment of HGSCs and emphasizes the importance of patient stratification. Nonetheless, current clinical trials use different assays and a consensus approach for patient stratification is still lacking. In this study, whole genome sequencing technique was used to profile tumor-normal sample pairs, including some related tumors collected from different anatomical sites. This yields a multi-sample cohort comprising 55 pre-treatment samples from 33 HGSC patients and is suitable for addressing questions about molecular stratification and tumor pathogenesis. We first recapitulated two reported HGSC subgroups (H-HRD and H-FBI) in our cohort. Comprehensive analyses were conducted on genomic scars ranging from mutations, indels, structural variants and copy numbers. Our data suggests that the genomic subgroups are characterized by different extent and onset timing of homologous recombination repair defect (HRD) as well as CCNE1 activation. Of note, H-HRD group showed higher levels of copy number-based HRD score, which often serves as a surrogate biomarker for PARP inhibitor treatment response. On the other hand, tumor evolutionary trajectory was reconstructed based on in silico methods capable of inferring relative timing of genetic alterations. This highlights an early bifurcation of carcinogenesis paths in this HGSC dichotomy, despite a common scenario of a very early TP53 mutation, an often early whole genome duplication and a chromosomal instability phenotype seen eventually. Overall, these findings corroborate the concept of tumor-intrinsic genomic phenotypes by providing mechanistic underpinnings from the aspect of tumor evolution. This provides the rationale for studying HGSC disease biology in different contexts and formulating questions about subtype-specific pathogenesis and vulnerabilities. Translationally, it also holds the promise for better identifying the patient subset that might benefit from PARP inhibitor treatment. Citation Format: Siao-Han Wong, Kai Doberstein, Frederik Marmé, Benedikt Brors. A multi-sample study reveals the evolution and heterogeneity in high-grade serous ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1513.