Embedded Files
1. Cancer genomics and tumor evolution
1. Cancer genomics and tumor evolution
Next-generation sequencing (NGS) has become widely accessible in research and clinical laboratories. By using whole-genome, whole-exome and transcriptomes sequencing, we can detect various types of variants such as single-nucleotide variants (SNVs), small insertions/deletions (Indels), as well as large variants including structural variants (SVs) and copy-number variants (CNVs). Integrative analysis of all types of variants can reveal frequently mutated genes and pathways in specific cancer types, as well as potential association with clinical phenotypes and treatment outcomes. These findings provide new insights into the mechanisms of various types of cancers, and highlight potential biomarkers for therapeutic development.
With many tools available, accuracy is the key to high-quality data analysis. Due to various types of potential issues in the process of sampling/sequencing/mapping, it is not uncommon to see both false positive and false negative generated in automatic predictions. After fixing these errors, the mutational profile may look dramatically different (right).
Selected papers:
a) Wei L*, Wang J*, Lampert E, Schlanger S, DePriest AD, Hu Q, Gomez EC, Murakami M, Glenn ST, Conroy J, Morrison C, Azabdaftari G, Mohler JL, Liu S, Heemers HV. Intratumoral and Intertumoral Genomic Heterogeneity of Multifocal Localized Prostate Cancer Impacts Molecular Classifications and Genomic Prognosticators. European Urology. 2017 Feb;71(2):183-192. (Editorial Highlights)
b) Wei L*, Liu S*, Conroy J*, Wang J*, Papanicolau-Sengos A*, Glenn ST, Murakami M, Liu L, Hu Q, Conroy J, Miles KM, Nowak DE, Liu B, Qin M, Bshara W, Omilian AR, Head K, Bianchi M, Burgher B, Darlak C, Kane J, Merzianu M, Cheney R, Fabiano A, Salerno K, Talati C, Khushalani NI, Trump DL, Johnson CS, Morrison CD. Whole-genome sequencing of a malignant granular cell tumor with metabolic response to pazopanib. Cold Spring Harb Mol Case Stud. 2015 Oct;1(1):a000380.
c) Holmfeldt L*, Wei L*, Diaz-Flores E, Walsh M, Zhang J, Ding L, Payne-Turner D, Churchman M, Andersson A, Chen SC, McCastlain K, Becksfort J, Ma J, Wu G, Patel SN, Heatley SL, Phillips LA, Song G, Easton J, Parker M, Chen X, Rusch M, Boggs K, Vadodaria B, Hedlund E, Drenberg C, Baker S, Pei D, Cheng C, Huether R, Lu C, Fulton RS, Fulton LL, Tabib Y, Dooling DJ, Ochoa K, Minden M, Lewis ID, To LB, Marlton P, Roberts AW, Raca G, Stock W, Neale G, Drexler HG, Dickins RA, Ellison DW, Shurtleff SA, Pui CH, Ribeiro RC, Devidas M, Carroll AJ, Heerema NA, Wood B, Borowitz MJ, Gastier-Foster JM, Raimondi SC, Mardis ER, Wilson RK, Downing JR, Hunger SP, Loh ML, Mullighan CG. The genomic landscape of hypodiploid acute lymphoblastic leukemia. Nature Genetics. 2013 Mar;45(3):242-52. (Cover Story)
2. Ultra-sensitive detection of mutations in tumor and normal tissues
2. Ultra-sensitive detection of mutations in tumor and normal tissues
We have conducted innovative studies aiming to expand our detection limits to very-low-frequency (<5%) mutations, which are important for cancer early detection and residual disease monitoring. The technical challenge comes from distinguishing real somatic mutations from background noise caused by sequencing, mapping, or library preparation errors. We were able to overcome this challenge by using bioinformatics algorithms with in-silico error suppression. The accuracy of our detection method was validated using spike-in experiments (right, from Wei L, et. al. Journal of Urology. 2019 Jun;201(6):1105-1114.)
Selected papers:
a) Wei L, Fitzgerald M, Graham J, Hutson N, Zhang C, Huang Z, Hu Q, Zhan F, Xie J, Zhang J, Liu S, Remenyik E, Gellen E, Colegio OR Christensen S, Lin H, Bax M, Xu J, Huss WJ, Foster BA, Paragh G. Ultra-deep sequencing differentiates patterns of skin clonogenic mutations associated with sun-exposure status. Version: 1. BioRxiv [Preprint]. 2020 January 11. Available from: https://www.biorxiv.org/content/10.1101/2020.01.10.902098v1. DOI: https://doi.org/10.1101/2020.01.10.902098.
Develop an objective assessment of skin cancer risk. Most cancers are caused by a lifetime accumulation of somatic mutations. Despite the important roles they play in tumor initiation and progression, mutations in physiologically normal tissues have been under-studied due to their low abundance and random patterns. We sequenced a large number (n = 450) of normal skin samples to identify early clonal mutations (CMs). The skin samples were collected in an individual-matched manner containing equal numbers of sun-exposed and non-sun-exposed samples from every donor, which allowed us to exclude mutations caused by other factors such as aging. This allowed us to then precisely identify the mutation patterns introduced by UV light. The results not only confirmed the previous findings of widespread mutations in aged skin, as well as elevated mutational burden after UV-exposure, but also identified several associated novel mutational patterns. For instance, we found UV-induced mutations were enriched in specific genomic regions, including several mutational hotspots where mutations are highly recurrent only in UV-exposed skin samples. These findings pave the road for future development of a quantitative measurement of UV-induced DNA damage, which allows objective assessment of skin cancer risk as well as early intervention such as field treatment
Collaboration with Dr. Gyorgy Paragh (Dermatology), Drs. Wendy Huss and Barbara Foster (Pharmacology and Therapeutics and Urologic Oncology), Sean Christensen (Dermatology, Yale)
More mutations, higher VAFs, in sun-exposed (SE) than non-sun-exposed (NE) normal skin.
Identification of "UV sensitive" genomic regions.
b) Wei L*, Hussein AA*, Ma Y, Azabdaftari G, Ahmed Y, Wong LP, Hu Q, Luo W, Cranwell VN, Bunch BL, Kozlowski JD, Singh PK, Glenn ST, Smith G, Johnson CS, Liu S, Guru KA. Accurate quantification of residual cancer cells in pelvic washing reveals association with cancer recurrence following robot-assisted radical cystectomy. Journal of Urology. 2019 Jun;201(6):1105-1114. doi: 10.1097/JU.0000000000000142. (Editorial Highlights & Cover story)
Collaboration with Dr. Khurshid Guru, Urology
Prediction of bladder cancer recurrence by sequencing surgery washings. Bladder cancer is a common cancer type with high recurrent rate after surgery, while the mechanism of recurrence remains unclear. One major hypothesis is that the tumor spillage during surgery may result in residual tumor cells (RTCs), which may be responsible for tumor recurrence. However, RTCs could not be detected in previous tests using conventional mRNA markers or cytology. In collaboration with the surgical urology team, we designed a comprehensive study that collected pelvic washings at multiple time points during the surgery, followed by ultra-deep sequencing to measure RTCs in each washing. The results were remarkable: all washings before the surgery were negative for RTCs; however, most patients’ post-surgery washings tested positive, confirming the presence of RTCs after surgery. Importantly, we found a significant association between the levels of RTCs and the aggressiveness of the tumor’s histology, suggesting the identified RTCs were unlikely to be caused by random tumor spillage during surgery, but were probably related to preexisting infiltrating microscopic tumor clones. In the current data, we found that the levels of RTCs were more significantly associated with cancer recurrence than conventional markers, such as tumor grade or surgical margin, which may open a new area of developing novel, highly sensitive biomarkers to predict recurrence
High levels of residual tumor cells (pRCCF) are associated with aggressive histology and cancer recurrence.
Pre-existing, microscopic tumor clones potentially contribute to bladder cancer recurrence after surgery.
We made the cover!
We made the cover!
c) Wei L*, §, Papanicolau-Sengos A*, Liu S*, Wang J*, Conroy JM*, Glenn ST, Brese E, Hu Q, Miles KM, Burgher B, Qin M, Head K, Omilian AR, Bshara W, Krolewski J, Trump DL, Johnson CS, Morrison CD§. Pitfalls of improperly procured adjacent non-neoplastic tissue for somatic mutation analysis using next-generation sequencing. BMC Med Genomics. 2016 Oct 19;9(1):64.
Collaboration with Dr. Carl Morrison, Pathology
Identify the source of tumor contamination in adjacent normal tissue. The normal tissue adjacent to tissue is often used as the normal DNA source in somatic mutation detection for filtering out germline polymorphisms. One problem is these normal tissues frequently contain low-level tumor cells, which would lead to false-negative calls as somatic mutations can be misclassified as germline SNVs when also present in normal tissue. By using ultra-deep sequencing, we confirmed the presence of low-level tumor contamination in normal tissue, and found the observed contamination was related to inadvertent handling during the surgical pathology gross assessment and tissue procurement process. This finding highlights the acquisition of high-quality normal tissue as one important but often overlooked area in somatic mutation detection (right).
Inadvertent handling during sample grossing and tissue procurement results in tumor contamination in the adjacent normal tissue.
Ongoing collaboration
Ongoing collaboration
With Drs. Philip McCarthy, Jens Hillengass, Swapna Thota to characterize Clonal hematopoiesis of indeterminate potential (CHIP)identify biomarkers for secondary leukemia and other blood disorders. Stay tuned for updates as the project progresses.
3. Identify tumor-specific neoantigens for immunoprofiling
3. Identify tumor-specific neoantigens for immunoprofiling
We designed the Christmas Light Plot (CLP) for easy visualization of identified neoantigens (below). The CLP incorporates pre-defined criteria for neoantigen prioritization, including MHC binding affinities, expression level, HLA class types, and the mutant clonal status. X-axis - Variant allele fraction (VAF) in WES, which can be used to infer clonal status; Y-axis - the predicted binding affinity of the mutant peptide. Each dot represents a neoantigen with the following characteristics displayed: size – the gene expression level by RNASeq; shape – HLA binding classes (I, II, or both); vertical bar –the difference between mutant and wildtype binding affinities; color –stratified based on the mutant versus wildtype binding and mutant expression level. Gene symbols are displayed for neoantigens selected for screening.
Collaboration with Dr. Kunle Odunsi, Gynecologic Oncology.
Selected papers:
a) Liu S*, Matsuzaki J*, Wei L*, Tsuji T*, Battaglia S*, Hu Q, Cortes E, Wong L, Yan L, Long M, Miliotto A, Bateman NW, Lele SB, Chodon T, Koya RC, Yao S, Zhu Q, Conrads TP, Wang J, Maxwell GL, Lugade AA, Odunsi K. Efficient identification of neoantigen-specific T-cell responses in advanced human ovarian cancer. J Immunother Cancer. 2019 Jun 20;7(1):156. doi: 10.1186/s40425-019-0629-6.
b) Yang Y, Jain RK, Glenn ST, Xu B, Singh PK, Wei L, Hu Q, Long M, Hutson N, Wang J, Battaglia S, George S. Complete response to anti-PD-L1 antibody in a metastatic bladder cancer associated with novel MSH4 mutation and microsatellite instability. J Immunother Cancer. 2020 Mar;8(1). pii: e000128. doi: 10.1136/jitc-2019-000128.
4. Characterize the mutational landscapes in mouse models
4. Characterize the mutational landscapes in mouse models
Cancer mouse models, include genetically engineered mouse (GEM) models and xenografts (PDX) are popular systems for studying human cancers in research and clinical laboratories. One essential question for using any model is to determine what degree the model faithfully represents the human counterparts, especially on the mutational level. Analyses of mouse somatic mutations pose a unique challenge due to the diverse genetic background in GEM and mouse stromal contamination in PDX models. We overcame these challenges using specifically designed computational approaches. The identified mutations in mice re-capitulate human cancer hotspot mutations. An example is shown below. From: Dang J*, Wei L*,et. al. Blood. 2015 Jun 4;125(23):3609-17.
Selected papers:
Selected papers:
a) Dang J*, Wei L*, de Ridder J, Su X, Rust AG, Roberts KG, Payne-Turner D, Cheng J, Ma J, Qu C, Wu G, Song G, Huether RG, Schulman B, Janke L, Zhang J, Downing JR, van der Weyden L, Adams DJ, Mullighan CG. Pax5 is a tumor suppressor in mouse mutagenesis models of acute lymphoblastic leukemia. Blood. 2015 Jun 4;125(23):3609-17. (GEM)
b) Wei L, Murphy, BL, Wu G, Parker M, Easton J, Gilbertson RJ, Zhang J, Roussel MF. Exome sequencing analysis of murine medulloblastoma models identifies Wdr11 as a potential tumor suppressor in Group 3 tumors. Oncotarget. 2017 Jul 27;8(39):64685-64697 (GEM)
c) Wei L, Chintala S, Ciamporcero E, Ramakrishnan S, Elbanna M, Wang J, Hu Q, Glenn ST, Murakami M, Liu L, Cortes Gomez E, Sun Y, Conroy J, Miles KM, Malathi K, Ramaiah S, Anbarasu A, Woloszynska-Read A, Johnson CS, Conroy J, Liu S, Morrison CD, Pili R. Genomic profiling is predictive of response to cisplatin treatment but not to PI3K inhibition in bladder cancer patient-derived xenografts. Oncotarget. 2016 Nov 22;7(47):76374-76389. (PDX)
Page updated
Google Sites
Report abuse