Salivary gland cancer (SGC) is a rare disease making up fewer than five percent of head and neck tumors. SGCs are further divided into multiple histological subtypes, of which mucoepidermoid carcinoma (MEC) and adenoid cystic carcinoma (ACC) make up the majority of cases, and have a combined five year survival rate of >75%. However, more than 20% of patients that are disease free after five years have a late recurrence, most commonly as distant metastases. This suggests the persistence of tumor cell population following treatment that is long-lived, resistant to therapy, can self-renew, and recapitulate the heterogeneity of the original tumor upon relapse or metastasis. These characteristics are commonly used as criteria to define cancer stem cell (CSC) populations in solid tumors. We have engrafted patient-derived xenografts of MEC, ACC and acinic carcinoma cases and conducted preliminary CSC implantation studies for tumor cells stained with Aldefluor (ALDH) and CD44. We find that as few as 100 ALDH+CD44high cells initiate tumors in immunocompromised mice, but these studies need to be expanded to multiple cases for each histotype. Confirming previous reports, we have demonstrated that the stemness factor SOX2 is associated with PI3K/mTOR activity in SGC tumors. However, the pathways and factors that maintain the CSC phenotype and promote tumorigenesis have yet to be investigated directly. We hypothesize that ALDH+CD44high defines the CSC population across SGC histotypes and that PI3K/mTOR signaling and stem cell factors SOX2 and OCT4 cooperate to maintain stemness and promote tumor repopulation. We will first define the SGC CSCs by conducting limiting dilution implantation and sphere assays using 3-4 PDX cases for each major histotype. Second, we will simultaneously characterize the gene expression (RNA-seq) and translation (Ribo-seq) profile of CSCs and non-CSCs sorted directly from PDX tumors. Third, we will mechanistically determine how PI3K/mTOR signaling and SOX2/OCT4 maintain the CSC population through genetic manipulation of ex vivo sphere cultures of tumor-derived CSCs. Characterization of SGC CSCs will ultimately identify targetable signaling pathways that can be inhibited therapeutically to block recurrence.
The immediate goal of this proposal is to define the salivary gland cancer stem cell (CSC) population across disease histotypes using patient-derived xenograft models. The second objective is to conduct gene expression and translation analysis to define enriched signaling profiles specific to CSCs. The overall objective of this proposal is to mechanistically identify the factors that maintain CSCs and promote tumor initiation and repopulation.