Detecting PIK3CA gene amplification using fluorescent in situ hybridization (FISH)
LZAP as a Tumor Suppressor
Head and Neck Squamous Cell Carcinoma (HNSCC) is the sixth most common cancer worldwide. Exploring molecular contributors to HNSCC, we identified LZAP as a tumor suppressor with activity inhibit NF-KB by dephosphorylation of RelA and activate p53. Mechanisms through which LZAP performs these activities are not well understood, but our studies suggest that LZAP inhibits tumor formation by inactivation of NF-kB while simultaneously activating p53. Loss of LZAP protects wild-type p53 cells but sensitizes p53 mutant or null cells to radiation or chemotherapy. Thus, temporary inhibition of LZAP activity may be beneficial for avoiding toxic side effects of anticancer therapies by sensitizing tumors while protecting normal surrounding tissues.
Crystal violet staining for invasive cells after PPM1A depletion
NF-kB transcriptional activity is regulated by phosphorylation of RelA and while kinases responsible for phosphorylation and activation are described much less is known about dephosphorylation and inhibition of RelA. We identified that a phosphatase, PPM1A has tumor suppressor-like activity at least partially due to dephosphorylation of RelA. We are now exploring how LZAP regulates PPM1A and its family members and restoration of PPM1A activity as a means to inactivated RelA and prevent metastases.
In addition to exploring LZAP activities using standard molecular techniques, we created a mouse model of LZAP loss. Because homozygous loss of LZAP results in embryonically lethality, spontaneous tumorigenesis was evaluated and found to be increased in heterozygous (LZAP+/-) mice. We are now exploring carcinogen- and oncogene-induced murine models. These studies are designed to determine the relative contribution of LZAP regulation of p53 and/or NF-kB to LZAP tumor suppressor activities
Salivary Gland Cancers
Salivary gland cancers are the least studied subtype of head and neck cancers that are in urgent need of targeted therapies. Among salivary cancers, adenoid cystic carcinoma is the most notorious due to its propensity for neuroinvasion, recurrence, and metastases to lung and skull. We identify novel biological markers and clinically important genes/pathways using expression profiling and new generation sequencing. In parallel, we develop and profiled xenograft models of salivary cancers for pre-clinical studies. Once clinically important genes and signaling pathways are determined, we use our mouse models to explore drugs or siRNAs that block these pathways. Using this approach, we identified in salivary adenoid cystic carcinoma a distorted neural stem cell gene signature.
Two most interesting genes are NTRK3, which encodes a neurotrophin receptor TrkC, and SOX10, a transcriptional factor that regulates differentiation and migration of neural stem cells. Blocking TrkC activity by clinically approved small molecule inhibitors (e.g. AZD7451) produces anti-tumorigenic effects in vitro and in vivo. As SOX10 has been recently linked with melanoma progression, molecular insights into its activation in salivary adenoid cystic may provide new therapeutic approaches.
Activation of neurotropic signaling in salivary gland cancers provides novel therapeutic targets. Overexpression of TrkC and Bcl2 as well as activation of Erk 1/2 (A), and production of the TrkC ligand NT-3 (B) in adenoid cystic carcinoma (ACC) and normal salivary gland (SAL).
HNSCC Therapeutic Targets
Patients with HPV-associated head and neck tumors respond better to radiation- and chemo-therapy and have a survival advantage compared to patients with HPV-negative cancer. We focused our research on finding molecular pathways responsible for different sensitivity of HPV-positive and negative head and neck tumors to the treatment, and found that HPV-positive head and neck cancer cells have elevated levels of DNA damage coupled with defects in DNA repair. Currently, we are investigating a role of HPV proteins in DNA repair and exploiting DNA repair defects for rationally designed therapy.
Treatment of HNSCC has not benefitted as greatly as other tumor types from molecular tumor characterization. Obstacles that have impeded therapeutic advances for HNSCC include:
- poor understanding of molecular defects
- absence of meaningful molecular subtypes (excluding HPV status)
- inability to identify tumors that will respond to new therapies
- paucity of molecularly characterized cell lines
- paucity of robust pre-clinical models
- inadequate support from funding agencies or industry
- poor understanding of the mechanisms through which HPV drives a subset of HNSCC.
HNSCC that is not associated with HPV (HPV-negative = HPV(-)) have been particularly hard to target. In addition, therapy for HNSCC has stagnated with no major advances since acceptance of concurrent chemoradiation, and this advance that improved organ preservation was not associated with improved patient survival. We are pursuing targets in HNSCC that are activated by mutation or amplification such as PIK3CA, FGFR1, and HRAS and are partnering with David Stern’s lab to discover potential combination therapies that will be effective. Once identified, we will test these therapies in patient-derived xenografts with the ultimate goal of testing them in clinical trials.
Modeling of HNSCC and Salivary Cancers
We have developed a patient-derived xenograft (PDX) model of short-term culture followed by implantation into immune deficient mice. Our model is unique because it uses an implanted scaffolding to encourage tumor growth. Using this model, we have successfully created xenografts from over one hundred patients. The goal is to expand these tumors so that virtual clinical trials can be performed using clones of a single tumor to determine the best treatment option.