Elaboration of multiparametric marker integration platforms for relapse and therapy response prediction in oral cancer

Elaboration of multiparametric marker integration platforms for relapse and therapy response prediction in oral cancer

Oral Squamous Cell Carcinoma (OSCC), the 7th tumor by diffusion worldwide, is of considerable public health significance and is associated with substantial mortality and morbidity. Despite that numerous prognostic/predictive markers have been proposed for the clinical monitoring of OSCC, based upon molecular comparisons between healthy and neoplastic tissues, up to now, none of these have clinical practice to predict therapeutic response. Moreover, they fail to define the molecular differences between discrete classes of OSCC patients with different degrees of aggressiveness. Therefore, this line of research aims at unveiling more accurate and stringent molecular markers to further classify patients and their disease using biological specimens, robust experimental methods and statistical analysis to support clinical decision making.

To pursue this task we make use of a prospective case study of OSCC in which a selected panel of markers is examined at the transcript and protein level. Furthermore, approaches of global gene profiling are exploited to disclose gene signatures for distinct OSCC patients categories. Both the evaluation of discrete panels of functional markers and the global gene screenings are conducted on well-defined groups of patients: young patients developing OSCC in the absence of overt risk factors; patients with early and advanced stage tumors which, unexpectedly, show reversed clinical evolution.

Because a late diagnosis and the aggressive nature of the disease cause tumor relapse in 25%-50% of cases and because currently methods to predict disease reoccurrence, such as staging and grading (TNM system) do not show sufficient accuracy in discriminating relapsing versus non-relapsing patients, a second goal of the study is to elaborated an integrated platform for prospective prognostic clusterization of OSCC patients that comprehensively takes into account clinical parameters, lifestyle risk factors, radiological tumor portraits, histo-pathological traits and gene expression patterns of their primary tumor lesions. The study is performed on OSCC patients surgically treated and followed up for at least 18 months. Clinical and imaging data are collected at the time of diagnosis and during follow-up; smoke, alcohol consumption and other risk factors are noted. Surgical specimens of primary lesions are collected and processed for histological diagnoses and classified according to the TNM staging system. Moreover, fresh tumor tissue samples are immediately frozen and used to extract DNA and RNA for molecular studies. Genome-wide expression profile is obtained by DNA microarray and the most representative genes are validated by qPCR, followed by immunohistochemistry were this is applicable. Through bioinformatic algorithms and dedicated software tools, gene expression profiles from relapsing and non-relapsing patients are compared and integrated with clinical, diagnostic imaging and histo-pathological data to extract a multi-parametric platform for relapse prediction.

We expect to identify novel remarkable clinical and biological traits categorizing and stratifying high and low risk OSCC patients with a superior accuracy of any currently reported system and, thereby, to open for the implementation of an individualized and more tailored management of this disease.