Search

Genomic profiling of glioblastoma uncovers fundamental differences in the driver events of malignant transformation and progression, yet the nuances driving the particular growth rate of glioblastoma in individual patients remain poorly defined. We exploited the availability of fourteen histocompatibility-matched mouse hosts from the collaborative cross (CC) [PMID: 15514660] as hosts for a non-germline, engineered model of murine glioblastoma [PMCID: PMC4210043] to interrogate cell types, functional states and proximity in situ in tumors with varying growth rates. Specifically, Kaplan-Meier survival data was used to determine the rates of Nf1:TP53 null tumor growth in different cc host groups. Spatial transcriptomics and single nucleus RNA seq (sn-seq) analyses were conducted in tumor-bearing brain tissue from C57Bl/6 (syngeneic) and 2 CC histocompatible allogeneic mouse strains with differential survival outcomes. Spatial analysis was conducted using a disease-focused multiplexed mRNA probe set assembled from annotations of GBM/glioma specific genes reported in the literature. Cell segmentation of the data was performed to investigate the spatial pattern of single-cell level expression. The spatial transcriptomic dataset was also analyzed with a segmentation-free pipeline. In addition, to detect cell clusters, the probability of genes detected in close proximity (genes within a cell 5px radius [0.5 um]) was determined using Pdgfra as the tumor marker (anchor). We identified differential levels of spatially correlated expression of cell clusters expressing hypoxia markers such as HIF-1a and PDK1 between hosts with fast and slow growing tumors. Tumors from the 3 hosts were also processed for sn-seq. Integrated analysis of cell clustering was used to determine differences in clusters between the fast growing and slow growing groups. Enrichment of expression of Top2a, a gene associated with proliferation and poor prognosis in glioma, was detected in a unique cluster of cells in the slow-growing tumor when compared to the fast-growing group. This suggests that the host microenvironment may play a role in influencing the decelerated growth of this tumor. These results support the utility of spatial transcriptomics and sn-seq in understanding how differences in tumor:host cell interactions contribute to tumor growth. Future opportunities enabled by histocompatible allogeneic hosts for GEM and nGEM glioma models include unraveling host effects on drug sensitivity, tumor heterogeneity, molecular plasticity, angiogenesis, invasion, and lymphoid and myeloid microenvironmental dependencies, each with profound opportunities for clinical translation. Citation Format: Angela S. Baker. Orthotopic nGEM glioma growth rates in allogenic hosts associate with hypoxia and proximity to host cells [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 10.

Purpose: Prostate cancer is the most common cancer in males, but also exhibits a ∼1.5-2-fold higher incidence in African American men compared with whites. Epidemiologic evidence supports a large heritable contribution to prostate cancer, with over 100 susceptibility loci identified to date that can explain ∼33 % of the familial risk. A portion of the undefined risk may be due to rare susceptibility variants. The African American Hereditary Prostate Cancer (AAHPC) Study, established in 1997, enrolled 77 AA families from seven clinical sites across the United States (Ann Epidemiol, 2000). The aim of this study is to identify rare, predictive, deleterious variants through exome sequencing of 99 cases from 26 families selected from the AAHPC families (2 to 3 affected men sequenced per family) and three female 1000 Genome controls. Methods: To explore the contribution of rare variation in coding regions of 38 known cancer-causing genes to prostate cancer risk (PCa), we sequenced the exomes of 99 AAHPC cases at a mean coverage of 30x. Post-variant calling quality control (QC) was implemented using Golden Helix SVS 8 software with filters set for removal of variants with Read Depth >10, Quality Score >10, and Quality Score: Read Depth Ratio > 0.5. Mendelian inconsistency was checked using PLINK. Prioritization of all candidate genes/variants were evaluated using online databases including 1000 Genomes and ANNOVAR for non-reference allele frequency and predictions of functional impact. Conclusions. Through exome sequencing of 99 AAHPC cases and 3 female 1000 Genome controls, we identified 37 non-synonymous single nucleotide variants that are considered damaging by at least one predictive scoring tool in our 38 candidate genes. However, most of these variants were common and thus unlikely to be causal. We observed three rare variants that were considered damaging by three predictive scoring tools in two known PCa genes, PCNT and ZFHX3. These 3 variants were each observed in a single different family (in 1 of 3 affected individuals). Interesting rare candidate variants (MAF ≤ 0.01) rs190517526 & rs114692729 were found in known cancer susceptibility loci (CD82 and EZH2). The CD82 variant was observed in all 3 sequenced affecteds in one family and the EZH2 variant was observed in 2 of 3 sequenced affecteds in a different family. These predicted damaging variants in these four genes represent potentially novel causal candidates for some of the 26 AAHPC families sequenced here. Future work will carefully analyze the remainder of the genome in the other 21 sequenced families including planned sequencing of additional family members. Citation Format: Deyana D. Lewis, Shukmei Wong, Angela S. Baker, Isaac Powell, John D. Carpten, Joan E. Bailey-Wilson, Cheryl Cropp. Rare candidate variants shared among affected family members in the African American Hereditary Prostate Cancer Study families [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 4240.

Lung cancer remains the leading cause of cancer deaths in the US and throughout the world. Never-smokers with lung cancer constitute an understudied subset of these patients, though recent estimates show that ∼10% of men and women with lung cancer in the US are never-smokers. Independently, lung cancer in never-smokers ranks among the ten most common causes of cancer mortality, and thus the causes, driver pathways, and potential therapeutics must be investigated for this clinically relevant subpopulation. We hypothesize that novel mutations and pathways identified by whole genome sequencing (WGS) and whole transcriptome sequencing (WTS) drive tumorigenesis in adenocarcinomas of never-smoker patients, and represent potential therapeutic targets. We have completed WGS and WTS on two lung adenocarcinomas from female, never-smokers, one early-stage and one advanced-stage (metastatic), and one female smoker patient with early-stage lung adenocarcinoma. Approximately 100 short nucleotide variants (SNV) causing non-synonymous DNA sequence changes were discerned from the early- and advanced-stage adenocarcinomas from never-smokers. Of interest, these never-smoker patients lacked alterations in common genes associated with lung cancer such as EGFR, KRAS, and EML/ALK translocations, making them ideal cases for the discovery of new mutations that may drive lung adenocarcinomas in never-smokers. In comparison, the adenocarcinoma from a smoker patient contained 78 SNVs, including a well-characterized KRAS mutation. Mutations in MAGEC1, a tumor marker in melanoma and multiple myeloma, were observed in common between the early-stage, never-smoker tumor and the smoker tumor. The mutations observed in the never-smoker cases included genes suspected to play a role in lung carcinogenesis. The early-stage, never-smoker patient contained a mutation in PIK3C3 and DOCK10, known to play a role in cancer cell migration, and CSNK1E, a casein kinase involved Wnt signaling and in beta catenin-induced cancer cell proliferation. The late-stage never-smoker patient demonstrated a p53 mutation, a mutation in the tumor suppressor LATS2, and a mutation in the DNA damage checkpoint gene, ATM. The mutations discovered in known tumor suppressor genes tended to be in the late-stage, never-smoke patient. Ingenuity Pathway Analysis implicated G-protein coupled receptor signaling for the early-stage never-smoker, Hif1α signaling for the smoker, and Sonic Hedgehog Signaling for the late-stage, never-smoker in both WGS and WTS. Currently, we are validating the identified mutations discovered by WGS using Sanger sequencing and surveying their frequency in ∼30 and ∼60 additional cases of lung adenocarcinoma from never-smokers and clinically matched smokers, respectively. Validation of these mutations and pathways will lead to a better understanding of tumorigenesis and tumor progression in never-smokers and identify nodes for therapeutic vulnerability.