ABSTRACT
Background: Histologic grading has been used as a prognostic factor and for clinical behavior evaluation of oral squamous cell carcinoma for the past several decades. At the same time, the prognostic value of different grading classifications remains controversial. A major problem for most histopathological grading systems is the intraobserver and interobserver disagreement. Aims and Objectives: To validate the prognostic efficiency of the histologic assessment of the primary tumor in predicting cervical metastasis, to identify those histologic features in the tumor most closely associated with cervical metastasis and to evaluate the reliability of the multifactorial grading (MFG) system by measuring intraobserver and interobserver agreement using kappa statistic. Materials and Methods: A set of 60 cases were chosen at random out of the 292 squamous cell carcinoma cases reported in our institution from 2007 to 2010. All cases were graded according to: Modified Broders' descriptive system and Anneroth et al., MFG system. Two Pathologists independently graded the tumor forefront blinded to the node metastasis. Results: The MFG showed a significant relation between the degree of histologic malignancy and presence of metastasis in the nodes. Among the components of grading, a significant difference was observed in the nuclear polymorphism and the pattern of invasion between the metastatic and non-metastatic patients. Intraobserver and interobserver agreement was acceptable and satisfactory. Conclusions: Moderate to good agreement between observers greatly increases the validity of the MFG system. The multifactorial malignancy grading could serve as a predictor for metastasis in the cervical lymph nodes.