Comprehensive Molecular Profiling of Cancer Progression and Rare Cancers

Abstract

Comprehensive molecular profiling of the genomic, transcriptomic, and epigenetic landscape of cancers is rapidly evolving due to increased accessibility to next-generation sequencing (NGS) technology and inter-institutional collaborations such as The Cancer Genome Atlas (TCGA). Currently, precision medicine approaches in oncology are being guided by NGS due to the clinical implications of genomic data on directing patient care. Here, we use targeted NGS on routine formalin-fixed, paraffin-embedded (FFPE) tissue to conduct comprehensive molecular profiling to address translational research opportunities. Targeted NGS of FFPE material provides greater access to patient samples, the ability to select for a specific tissue region to maximize tumor content in cases of small lesions, and a more cost-effective method focused on cancer-related genes. By using this approach, we were able to answer a wide-range of questions regarding tumor progression and the genomic landscape of rare cancers. Since the molecular events driving low-grade endometrioid endometrial carcinoma (LGEC) and squamous cell carcinoma (SCC) development are incompletely understood—like in many cancers—we assessed tumor progression in these two cancers by profiling a series of presumed precursor and invasive lesions. Multi-region profiling of LGEC populations using a highly scalable approach demonstrated clinically-relevant multiclonality and intratumoral heterogeneity. Additional DNA profiling of two models of invasive SCCs and their precursors suggested that the presumed genomic complexity primarily found in invasive disease was also found at the precursor stage. Although we used transcriptomic data to compare precursors and invasive disease in LGECs and SCCs to expand our study, our conclusions were limited due to the need for a greater sample size and additional functional studies. Importantly, however, we demonstrated that our methodology is broadly scalable to enable high-throughput genomic and transcriptomic characterization of precursor and invasive cancer populations from FFPE specimens. Additionally, we characterized three types of rare cancers where driving alterations that could have diagnostic, prognostic, and therapeutic implications were still largely unknown. First, we focused on a rare class of soft tissue sarcomas defined by a gene fusion between CIC and DUX4, which resemble Ewing sarcomas at the histological level. Like Ewing sarcomas, we identified limited somatic driver mutations. However, we identified recurrent known chromosome 8 gain, a deleterious mutation in ARID1A (chr 1p36), and novel copy-number alterations (CNAs) including chromosome 1p loss, which is also the locus of ARID1A. Therefore, we identified genomic aberrations that can be used to refine the classification of CIC-DUX4 sarcomas. We also focused on Merkel cell carcinoma (MCC), an aggressive cutaneous neuroendocrine carcinoma, where the development of an additional cutaneous MCC tumor can clinically be interpreted as a second primary MCC tumor or a cutaneous metastasis. Due to the important treatment and prognostic implications of this distinction, we assessed clonality. We identified cases with tumors that were non-clonal (second primary) and clonal (metastasis) and observed that tumors from the same patient arose via the same mechanism (via virus or UV-damage). Lastly, olfactory neuroblastomas (ONBs), also known as esthesioneuroblastomas, are aggressive round-cell tumors. Despite their aggressive course, molecular studies of ONBs have been limited, and targeted therapies are lacking. Here, we identified recurrent potential targetable FGFR3 alterations associated with overexpression that may represent a novel therapeutic target in ONBs. Through these projects involving the molecular characterization of cancer progression and rare cancers, we identified important basic and clinical insights with the potential to improve patient care.