Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Multi-omic profiling of E-cadherin deficient breast cancers and their in-vitro models towards enabling precision medicine

Shah, Osama Shiraz (2023) Multi-omic profiling of E-cadherin deficient breast cancers and their in-vitro models towards enabling precision medicine. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

[img] PDF
Restricted to University of Pittsburgh users only until 8 December 2025.

Download (9MB) | Request a Copy

Abstract

Breast cancer (BC) is the most common cancer in women. Most cases (~75%) are histologically classified as "no special type" (NST). Invasive lobular carcinoma (ILC), the most common histological “special type”, accounts for ~10-15% of all BC diagnoses, is characterized by E-cadherin loss/deficiency and unique traits including lower grade, hormone receptor positivity, larger diffuse tumors, and specific metastatic patterns. Given these unique characteristics, ILC is recognized as a distinct disease necessitating tailored and precision medicine treatments.
Besides ILC, numerous understudied histological “special types” exist, such as mixed invasive ductal-lobular breast carcinoma (mDLC). mDLC represents ~3-6% of all BC diagnoses, characterized by mixture of E-cadherin proficient ductal and E-cadherin deficient lobular tumor regions. Despite its increasing incidence, mDLC remains understudied leading to diagnostic and prognostic uncertainty and making its clinical management challenging. We addressed this by curating a well-annotated mDLC cohort, demonstrating its histomorphic and E-cadherin heterogeneity, and proposed three distinct subclasses. Utilizing spatially-resolved and single-cell techniques, we dissected the molecular complexity of mDLC and uncovered significant biological differences between its ductal and lobular tumor regions. All lobular tumor regions showed genetic or epigenetic inactivation of the CDH1 gene. Moreover, ductal and lobular regions showed distinct clinically actionable alterations, oncogenic signatures and intrinsic molecular subtype heterogeneity, which could indicate differences in prognosis.
On the other hand, the translational research focused on ILC is currently impeded by lack of well-characterized in vitro models. To address this, we generated the ILC Cell Line Encyclopedia (ICLE), providing a comprehensive multi-omic characterization of ILC and ILC-like cell lines. ILC/ILC-like cell lines recapitulated pathognomonic E-cadherin deficiency and retained key ILC molecular alterations seen in ILC patient tumors. Whole-genome analysis uncovered novel structural rearrangements including deletions in CDH1 gene, which encodes E-cadherin, functional gene fusions and breast cancer specific patterns of chromothripsis. Integrative analysis of DNAm and RNA datasets revealed epigenetic activation of TFAP2B – an emerging biomarker of lobular disease. Analysis of publicly available RNAi screen datasets revealed numerous putative druggable vulnerabilities in cytoskeletal components, focal adhesion and PI3K/AKT pathway. Finally, we established a scheme based on molecular similarities to ILC patient tumors, enabling rational selection of ILC/ILC-like cell lines. In summary, we generated ICLE – the first ILC cell line encyclopedia, highlighting key molecular features, cataloging druggable targets and presenting a scheme for rational-model-selection.
In conclusion, we have curated an extensively annotated mDLC cohort, revealing three unique subclasses. Moreover, we conducted the first spatially-resolved multi-omic profiling of mDLC revealing its molecular heterogeneity with implications for disease prognosis. Furthermore, we utilized multi-omic characterization to generate ICLE – a critical resource for ILC disease model selection and precision drug discovery. Both efforts emphasize the importance of multi-omic profiling of E-cadherin deficient (ILC) and heterogeneous (mDLC) breast cancers and pave the way for precision medicine research, enabling us to design more effective treatments for these unique subtypes of breast cancer.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Shah, Osama Shirazosamashirazshah@gmail.comoss100000-0003-4756-4535
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairKostka, Denniskostka@pitt.edukostka0000-0002-1460-5487
Committee MemberSoong, Rindarsoong@pitt.edursoong0000-0002-1565-992X
Committee MemberLucas, Peterpcl8@pitt.edupcl80000-0003-4880-7172
Committee MemberMann, Mellissamannmr@mwri.magee.edumannmr0000-0002-8075-3824
Thesis AdvisorLee, Adrianavl10@pitt.eduavl100000-0001-9917-514X
Thesis AdvisorOesterreich, Steffisto16@pitt.edusto160000-0002-2537-6923
Date: 8 December 2023
Date Type: Publication
Defense Date: 3 August 2023
Approval Date: 8 December 2023
Submission Date: 11 August 2023
Access Restriction: 2 year -- Restrict access to University of Pittsburgh for a period of 2 years.
Number of Pages: 175
Institution: University of Pittsburgh
Schools and Programs: School of Medicine > Integrative Systems Biology
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: Breast Cancer Multiomics Genomics Epigenomics Transcriptomics scRNAseq Spatial Profiling Cancer cell line characterization Invasive lobular carcinoma Mixed invasive ductal-lobular carcinoma
Date Deposited: 08 Dec 2023 16:18
Last Modified: 08 Dec 2023 16:18
URI: http://d-scholarship.pitt.edu/id/eprint/45305

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item