Abstract
BACKGROUND
Conventional prognostic factors in squamous cell carcinomas are tumor stage, tumor size, evidence of lymph node metastases, extracapsular spread of lymph node metastases, and Broder's grading. Unfortunately these parameters are only of limited value in predicting the biological behavior and ultimately the prognosis of a particular tumor. The present study was conducted for determining objective prognostic factors based on tumor biologic examinations in patients with squamous cell carcinomas of the oral cavity. These parameters were compared to the conventional prognostic factors.
PATIENTS AND METHODS
Operative specimens of fourty-two patients who underwent surgery for a squamous cell carcinoma of the oral cavity with clinical TNM stage T1N0M0 were examined. All tumors were radically excised with histologic clear margins, which were confirmed by serial sections. The examinations included morphologic multifactorial tumor front grading, quantitative DNA analysis, and immunohistochemical assessment of proliferation markers (i.e. proliferating cell nuclear antigen [PCNA] and MIB1) and of oncogene products (i.e. p53; nm23). Prognostic significance of particular parameters was evaluated by univariate and multivariate Cox regression models.
RESULTS
In clinical follow-up of 70 months on average, 6 patients developed local recurrences and 5 patients lymph node metastases. Three patients died of disease. Tumors which recurred had increased values for 2 c Deviation Index, 5 c Exceeding Rate along with high tumor front grading scores and proliferation scores. Using multivariate Cox regression analysis, parameters which were highly significant for prognosis were 5 c Exceeding Rate, tumor front grading score and PCNA score. None of the clinical parameters achieved statistical significance in the multivariate model. Tumors which recurred had also increased expression rates for p53 and nm23. Nevertheless this was statistically not significant.
CONCLUSIONS
Tumor biologic examinations provide important informations about the clinical aggressiveness and ultimately about the prognosis of a particular tumor. Tumors with aggressive behavior can already be identified during initial diagnosis, which has consequences for the therapeutic management of the patients.
Collapse