Peritoneal mesothelioma is a rare and aggressive malignancy that arises from the mesothelial lining of the abdominal cavity. Mesothelioma in situ, a recently defined pre-invasive state, offers a critical window into the earliest molecular events of malignant transformation. However, the scarcity of cases and the limited availability of high-quality tissue, particularly from early-stage disease, has hindered comprehensive molecular characterization. This thesis investigates the progression from in situ to invasive peritoneal mesothelioma using an integrated transcriptomic and proteomic approach. RNA sequencing was performed on 9 samples from 7 unique cases: 3 in situ and 6 invasive tumors. Matched mass spectrometry-based proteomics was carried out on 9 FFPE samples; 2 in situ, 3 invasive, and 4 non-tumor peritoneal controls, including two normal cases not profiled by RNA-seq. Despite technical challenges posed by formalin-fixed tissues and low RNA yields from thin peritoneal samples, robust datasets were generated. Differential expression and pathway analyses revealed consistent enrichment of genes involved in RNA processing (HNRNPA2B1, HNRNPK), translation (RPL21, EIF3 subunits), and cell cycle regulation (CNOT2, NRF1) in invasive tumors. A marked enrichment of cell adhesion–related pathways was also observed, pointing to a functional shift facilitating tumor dissemination. Strikingly, COL5A2, a collagen gene associated with extracellular matrix remodeling and invasive behavior, was significantly upregulated in invasive tumors and was validated across transcriptomic and proteomic datasets. This highlights the central role of matrix reorganization in mesothelioma progression. Validation against external cohorts using TCGA mesothelioma samples (n = 87) and normal omental tissue from GTEx (n = 541) confirmed the reproducibility of key transcriptional signatures, including genes involved in splicing, adhesion, and ribosome biogenesis. These data support the hypothesis that mesothelioma in situ represents an early molecular stage preceding full malignancy. Together, these findings provide new insight into the molecular programs driving peritoneal mesothelioma progression and identify potential biomarkers and therapeutic targets. This work also underscores the feasibility, and limitations, of applying multi-omics to rare tumor samples, and emphasizes the importance of studying early disease states in cancer research.
Sana Shahidi (Thu,) studied this question.