Uveal melanoma (UM) is a highly aggressive and frequently fatal cancer of the eye. While only 2-4% of
patients will have detectable metastases at diagnosis, up to 50% of patients later develop metastatic disease,
overwhelmingly localized to the liver. Gene expression profiling separates UM to into two sub-groups, low risk
Class 1 UM and high risk Class 2 UM. Almost all UMs harbor an initiating mutation in the Gαq signaling
pathway, most commonly in GNAQ. Class 2 UMs harbor “progression mutations” in one of several genes that
drive tumor progression, the most common being BAP1. Previous studies from our group identified biallelic
inactivation of BAP1 as the strongest predictor of liver metastasis in UM. However, the cellular and molecular
mechanisms underlying liver metastasis in UM remain undetermined. This gap in our knowledge limits our
ability to develop effective adjuvant treatment for high risk patients and systemic treatments for disseminated
disease.
Gαq signaling activates the MAPK pathway in UM cells, and our preliminary single cell data show that Gαq-
mutant UMs are enriched for FOS/JUN transcriptional states. Loss of BAP1 function was associated with
escape from senescence, and a de-differentiated phenotype in UM cells that allowed their interaction with
hepatic stellate cells, leading to progression of liver metastases and drug resistance. The dedifferentiated state
of UM cells was specifically dependent on HDAC1, with HDAC inhibition restoring the differentiated state of
UM cells, and increasing the efficacy of MEK inhibition in UM liver metastasis models. The overarching goal of
this grant is to define the mechanism by which mutant Gαq and BAP1 loss co-operate to drive UM liver
metastasis development. In Aim 1 we will define how BAP1 loss alters the signaling, transcriptional and
epigenetic state of Gαq-mutant UM cells to allow senescence to be overcome. In Aim 2, we will address how
the cellular state of Gαq-mutant/BAP1-loss UM activates hepatic stellate cells to generate a pro-survival niche
in the liver through increased angiogenesis and upregulation of MAPK signaling. In Aim 3, we will explore the
HDAC1 dependency of the Gαq-BAP1 loss UM cell state and will validate hits from recent CRISPR screens to
identify novel therapeutic approaches to treat UM liver metastases, that we will validate in animal models. At
completion of this study, we expect to have defined new approaches to prevent and treat UM liver metastases,
that we will explore in clinical trials.