Stage grouping of oropharyngeal cancer: evaluation of three systems by means of survival analysis

J Craniofac Surg. 1999 Jan;10(1):73-8. doi: 10.1097/00001665-199901000-00014.

Abstract

Stage grouping is a method of summarizing multiple categories generated from the tumor, node, metastasis (TNM) classification system. Recently three different systems, the T and N integer score (TANIS), Hart, and 1997 Union Internationale Contre le Cancer systems, have been proposed. To verify their correlation to survival, a series of 64 patients affected by primary squamous cell carcinoma of the oropharynx was considered in this retrospective study. The data set was classified according to Union Internationale Contre le Cancer T stage and then grouped as recommended by the three systems. Results showed a crude survival rate of 28.13%. Univariate analysis by means of the logrank test yielded significant P values for the T and N integer score and Hart systems (0.0452 and 0.0179, respectively) and a borderline P value (0.0728) for the stratification based on the 1997 Union Internationale Contre le Cancer system. Multivariate analysis (Cox regression adjusted for age and sex) showed a significant correlation between the three staging systems and the mortality rate. Odds ratios were 1.36 (95% confidence interval, 1.12-1.66), 1.58 (95% confidence interval, 1.18-2.12), and 1.63 (95% confidence interval, 1.08-2.48) for the T and N integer score, Hart, and 1997 Union Internationale Contre le Cancer systems, respectively. The T and N integer score system showed the best statistical correlation, but a conclusive result could not be achieved because of the low number of patients in this study.

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Carcinoma, Squamous Cell / mortality
  • Carcinoma, Squamous Cell / pathology*
  • Confidence Intervals
  • Evaluation Studies as Topic
  • Female
  • Humans
  • Male
  • Middle Aged
  • Neoplasm Staging / methods*
  • Odds Ratio
  • Oropharyngeal Neoplasms / mortality
  • Oropharyngeal Neoplasms / pathology*
  • Prognosis
  • Proportional Hazards Models
  • Retrospective Studies
  • Survival Analysis