Date of Award

Summer 8-14-2025

Embargo Period

8-14-2025

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Medical Science)

Department

Pharmacology

College

College of Graduate Studies

First Advisor

Peggi Angel

Abstract

Hepatocellular carcinoma (HCC) remains a major global health challenge, with rising incidence, high mortality, and a five-year survival rate below 20%. Approximately 90% of HCC cases arise in cirrhosis, a stromal disease characterized by excessive deposition and remodeling of collagenous extracellular matrix (ECM). As HCC progresses, alterations in ECM composition and spatial distribution influence disease progression, emphasizing the potential of the ECM as a biomarker for HCC prognosis and treatment response. However, ECM proteomic analysis is challenging due to complex post-translational modifications, crosslinking, and insolubility. This dissertation molecularly and spatially characterizes the stromal landscape in and around HCC tumors in the context of patient outcome and improves current proteomic techniques for ECM biomarker discovery. The central hypothesis is that the composition, post translational modifications, and spatial distribution of ECM can predict clinical HCC outcomes and that ECM fragments released into the circulation can serve as noninvasive biomarkers for patient stratification.

Aim 1 uses mass spectrometry imaging (MSI) to identify ECM signatures in formalin-fixed paraffin-embedded (FFPE) resected HCC tumors that differ by patient outcome. Stromal infiltration and proline-hydroxylated fibrillar collagen peptides differentiated between tumor subtypes (S1: poor prognosis, S3: better prognosis) with high sensitivity and specificity (AUROC > 0.8). Machine learning algorithms were able to identify multiple peptide signatures (n = 4-6) that could distinguish classes (AUROC > 0.97). Aim 2 developed collagenase-based serum proteomics to identify circulating ECM markers for patients with liver cirrhosis and HCC. This approach enriched proteomic readouts for ECM proteins, curating a panel of ECM proteins unique to patients with liver cirrhosis and HCC. Aim 3 optimized collagenase-based proteomics workflows to enhance peptide identification for mass spectrometry imaging. Collagenase preferentially cleaved collagen at the N-terminus of glycine resides in G-P-X-G-P-X repeats, enriched for ECM proteins in FFPE tissues and generated more singly-charged peptides. Expansion of TIMS windows to include these singly-charged peptides improved MALDI-MSI peptide identifications by 30%. The combined optimization of sample preparation and mass spectrometry increased MALDI-MSI peptide identifications by 7-fold.

Overall, this research advances ECM-based biomarker discovery and noninvasive screening for HCC, with implications for personalized patient management and improved clinical outcomes.

Rights

Copyright is held by the author. All rights reserved.

Download

Share

COinS