PROJECT ABSTRACT/SUMMARY
Substantial pre-clinical evidence has demonstrated stress-induced tumor progression in animal models of head
and neck cancer (HNC). Clinically, patients with locoregionally advanced Human Papillomavirus (HPV)-negative
HNC are exposed to a broad range of chronic stressors. The link between cancer progression and chronic stress
may be one contributing explanation for the persistently poor oncologic outcomes of the HPV-negative HNC
population. To evaluate mechanism and develop therapeutics, animal models of chronic stress are needed.
However, existing models use either immunodeficient mice or syngeneic murine tumors, neither of which
recapitulate the human tumor microenvironment (TME). To overcome these limitations, we developed a novel
mouse model of chronic stress engineered to have a human immune system and carry human HNC xenografts.
In preliminary studies using this model, chronic stress led to increased tumor growth and metastases associated
with higher levels of circulating stromal-derived factor 1 (SDF-1) and immune alterations in the TME.
Taken together, we hypothesize that chronic stress leads to SDF-1/CXCR4 pathway upregulation which induces
an immunotolerant TME and increased tumor growth and metastasis in HNC. To test this hypothesis and
determine clinical relevance, we will investigate the mechanisms underlying stress-induced tumor growth in both
cell line-derived xenograft (CDX) and patient-derived xenograft (PDX) humanized models of HNC. We will further
evaluate clinical relevance through a prospective observational study in patients undergoing HNC surgery.
In Aim 1, humanized CDX mice will be randomized to chronic social isolation stress or control conditions. Three
arms will be designed including (1) animals treated with a CXCR4 antagonist, (2) HNC xenografts with a CXCR4
deletion, and (3) a control arm with wild-type HNC xenografts. Tumor growth and lung metastases will be
measured. Intra-tumoral immunophenotyping and RNA sequencing will be performed. In Aim 2, we will recruit
patients with locoregionally advanced HPV-negative HNC planned for surgical resection, a population with a
poor prognosis and significant baseline exposure to chronic stressors. Patients will undergo plasma cytokine
assessment and tumor resection specimens will undergo immunophenotyping and RNA sequencing. In a subset
of these patients, autologous humanized patient-derived xenograft (PDX) models will be generated, in which the
human immune system and tumor are derived from the same patient. This is uniquely feasible for HNC surgery
with fibula reconstruction, as a segment of fibula bone is routinely discarded from which hematopoietic stem cells
can be harvested for autologous humanization. These autologous PDX models will be randomized to chronic
stress and control conditions and treated with a CXCR4 antagonist as compared with control injection.
This study will be the first to investigate the mechanism of stress-induced HNC growth in the context of the SDF-
1/CXCR4 axis and the human TME. Results from the humanized pre-clinical models and parallel prospective
clinical analyses will lay the foundation for development of precision therapeutics and rational clinical trial design.