PROJECT SUMMARY/ABSTRACT
PD1 checkpoint blockade with nivolumab or pembrolizumab has given new hope for poor prognosis
recurrent/metastatic head and neck squamous cell carcinoma (HNSCC) patients, but response rates of only
15-20% highlight the need for continued investigation. In this application, we have assembled a
multidisciplinary team to address tumor cell intrinsic and adaptive resistance programs that are a critical barrier
to immunotherapy. Together, this targeted and discovery based approach coupled with human functional
immunophenotyping aims to generate new translational approaches to improving response rates for HNSCC
patients. The foundation of this proposal is based on supporting data showing robust preclinical modeling
platforms that span the spectrum of clinical responses to anti-PD1 therapy including response, intrinsic and
adaptive resistance. Using high dimensional time of flight mass cytometry (CyTOF) and an isogenic model of
response and resistance, we identified distinct tumor microenvironmental remodeling with anti-PD1 and anti-
CTLA4 therapies. We also completed a genome scale CRISPR screen for T cell recognition in a PD1 resistant
HNSCC model and identified 355 candidate genes that may regulate this interaction. Candidates included
ones that may be relevant in all cancers as well as ones specific to HNSCCs. A major category to emerge from
this and other screens represents targets in the chromatin accessibility pathway. We validated a top screen
candidate that modulates chromatin accessibility, the histone methyltransferase and transcriptional regulator
Ezh2, and identified reduced anti-PD-1 resistance in combination with Ezh2 targeting. These data illustrate the
validity of our approach as well as potential for clinical translation, as agents targeting Ezh2 are currently in
clinical development. In a separate study of immunotherapy treated HNSCC patients, we identified mutations
in SMARCA4, a catalytic subunit key for the PBAF nucleosome remodeling complex, in immunotherapy
responsive HNSCC patients, highlighting chromatin remodelers as modulators of immunotherapy response in
patients and the potential for improved clinical management by targeting these activities in conjunction with
immunotherapy. In the proposed work, we will complete analysis of Ezh2 targeting with further validation of
candidate immunotherapy resistance modulators. Coupled with these preclinical studies we will complete
functional immunophenotyping of human HNSCC patients using CyTOF and a novel patient derived
organotypic platform. In parallel to these preclinical approaches, we have a strong clinical trial portfolio
including a novel neoadjuvant pembrolizumab clinical trial that has shown surprising pathologic response in a
subset of patients and highlights this approach as an ideal setting in which to address resistance mechanisms
in previously untreated patients. We will interrogate genomic data from our neoadjuvant trial to assess
resistance mechanisms defined in preclinical models. In conclusion, this integrated approach will define
mechanistic and therapeutic approaches to address the intrinsic immunotherapy resistance in HNSCC.
