Molecular subtyping and vulnerability exploitation in Merkel cell carcinoma

Abstract

Merkel cell carcinoma (MCC) is an aggressive neuroendocrine (NE) carcinoma of the skin with limited therapeutic options. Currently, MCC is divided into two subtypes based upon the presence or absence of clonally integrated Merkel cell polyomavirus (MCPyV) within the host-cell genome. These subtypes are genetically and etiologically distinct: virus-negative MCC (MCCN) are heavily mutated including specific inactivation of the RB transcriptional corepressor 1 (RB1) and tumor protein p53 (TP53) tumor suppressors while virus-positive MCC (MCCP) contain few mutations but express viral oncoproteins that functionally inhibit RB1 and p53. Despite these independent etiologies, MCCP and MCCN are considered clinically indistinguishable due to overlapping morphological and immunohistological characteristics paired with similar prognoses and responses to therapy. As such, increased molecular characterization of MCC could lead to the identification of therapeutically-relevant subtypes which may be exploitable for more targeted and efficacious treatment.

In Chapter 2, we identify inosine monophosphate dehydrogenase (IMPDH) proteins as critical and druggable targets in TP53-wildtype MCC. We found that inhibition of IMPDH led to the rapid depletion of DNA, but not RNA, synthesis in addition to the accumulation of replication stress (RS) and p53-dependent apoptosis. Building on these results, we combined IMPDH and ataxia telangiectasia and Rad3-related (ATR) inhibitors to uncouple RS from the ATR-dependent replication checkpoint and induce genome-wide double-stranded DNA (dsDNA) breaks and cellular death independent of TP53-status. In this context, we identified TP53-dependent and -independent therapeutic vulnerabilities exploitable in MCC with currently available therapeutics.

In Chapter 3, we used bulk and single-cell sequencing modalities to characterize the transcriptional landscape of patient-derived tumor biopsies. In these samples, we identified heterogeneity in NE gene expression that was negatively correlated with Hippo pathway transcription factors Yes1-associated transcriptional regulator 1 (YAP1) and WW-domain containing transcription regulator 1 (WWTR1). We observed this reciprocal expression at the protein and RNA levels, in patient-derived tissues, patient-derived and established cell lines, in bulk samples as well as in single-cells. This led us to categorize MCC into either NE-hi or NE-lo subtypes based on the expression of NE genes or YAP1 and WWTR1, respectively. Mechanistically, we found that induction of YAP1 and WWTR1 in NE-hi MCCP cells suppressed expression of MCPyV oncoproteins to reinstate RB1-dependent cell-cycle arrest. As a result, we identified additional NE-based subtypes of MCC which could be exploited with current NE- or non-NE-targeted therapeutics in addition to uncovering a novel YAP1- and WWTR1-dependent MCPyV oncoprotein restriction mechanism.

Taken together, this work has increased the repertoire of molecular subtypes in MCC to include TP53- and NE-status in addition to identifying and validating clinically targetable subtype-specific vulnerabilities. These findings support the growing push towards personalized medicine in MCC and suggest the importance of accounting for TP53- and NE-status throughout clinical course. Unexpectedly, this work has also unlocked a potential model for the cell-of-origin for MCCP oncogenesis and MCPyV infection, a better understanding of which could lead to the development of clinically-relevant disease models.