RNA from Brush Cytology to Identify Aggressive Squamous Cell Carcinoma

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Squamous cell carcinoma (SCC), a cancer which afflicts sites including the gastrointestinal and airway mucosa, is by far the predominant form of oral cancer. Despite current advances in treatment, survival from oral SCC (OSCC) remains poor, chiefly due to recurrence of the primary tumor within 3-5 years after diagnosis. Advanced stage tumors, or those of aggressive nature, recur much earlier. Various clinical and pathological parameters of the tumor, such as invasive growth, and state of differentiation have been tested for their ability to predict tumor aggressiveness. Lymph node metastasis to multiple nodes, or spread outside the node capsule, are key indicators of aggressive tumors at high risk for recurrence but detection of these factors is not always accurate or an easy task. Recent work has highlighted the usage of RNA from surgically obtained primary tumor tissue to allow global gene expression analysis that identifies gene expression patterns associated with lymph node metastasis - thus identifying high risk tumors. We would like to extend this approach to take advantage of cells that can be obtained noninvasively with a cytology brush without the use a of scalpel biopsy. We previously published this method and have identified 2 specific markers for OSCC, in hamsters and humans, beta 2 microglobulin (B2M) and cytochrome p450 1B1 (CYP1B1). We have gone on to identify a pilot gene expression classifier for OSCC using RNA from brush cytology with an accuracy of 91%. While examination of cell morphology from brush cytology cells has already proven helpful in tumor detection, RNA analysis of these cells has the potential to contribute to true diagnosis and prognosis. Our long term goal is to develop a platform that will allow the clinician to develop a prognosis for malignant lesions in a noninvasiv manner. In the time frame of a two year study, we propose to develop a testable oral cytology based gene expression classifier that will differentiate tumors with multiple lymph node metastasis, extracapsular spread, perineural invasion and/or poor differentiation- four clinical/pathological indicators of aggressive tumor growth that are seldom found in less aggressive tumors that are less prone to spread. We will use the same patient gene expression data to create a pilot gene expression based classifier that directly predicts tumors that recur and cause mortality the first year after treatment. These gene expression based classifiers will ultimately aid in treatment decisions. 1 PUBLIC HEALTH RELEVANCE: These results will deliver a testable noninvasive method for differentiation of the most aggressive oral cancers from those that are less prone to metastasize. While many clinical and histological parameters have been used to predict tumor recurrence it is difficult to standardize them for widespread usage. The most accepted marker, high level lymph node metastasis, requires neck surgery for greatest accuracy, a procedure with high long-term morbidity. There is a need for a noninvasive method to determine what kind of treatment is best for an individual oral tumor as early as possible. We will begin to test if RNA from tumor cells collected with a brush will be able to yield a gene expression signature that can be used for tumor classification, prognosis and treatment decisions.