Background and objective: The Chinese 2008 staging system for nasopharyngeal carcinoma (NPC) was a common set of recommendations by initial revision of the previous'92 staging system. This study was to verify the Chinese 2008 staging system and to provide evidence for its further revision.
Methods: Between January 2003 and December 2004, 924 consecutive patients with newly diagnosed, nondisseminated biopsy-proven NPC, presented at the Department of Radiation Oncology, Cancer Center, Sun Yat-sen University, were entered into our study. All patients received magnetic resonance imaging examination of the neck and nasopharynx before treatment. According to the principle of the staging system, the indices of hazard consistency, hazard discrimination, prognostic value, and distribution were used to evaluate the Chinese 2008 staging system.
Results: According to the Chinese 2008 staging system, the distribution of stage group for the whole series was 4.9% for the stage I, 22.6% stage II, 38.0% stage III, and 34.5% stage IVA. As for hazard discrimination, the 4-year local relapse-free survival (LRFS) rates for T1-T4 patients were 95.4%, 93.7%, 90.5% and 79.1%, respectively. Although the separation of LRFS for all subclassifications of T-stage showed reasonable, the differences among the subgroups T1, T2, and T3 were still lack of significant statistically. The 4-year distant metastasis failure-free survival (DMFS) rates for N0-N3 patients were 89.4%, 84.3%, 73.6, and 59.2%, respectively. The 4-year overall survival (OS) rates for stage I-IV patients were 96.7%, 94.1%, 82.6% and 67.1%, respectively. As for hazard consistency, the distinctions between T3 with skull base involvement and T3 with medial pterygoid muscle in the hazard ratios of local failure (1.628 vs. 3.905) and disease failure (1.630 vs. 3.288) were large. Multivariate analyses showed that cervical nodal level, extranodal neoplastic spread, and laterality were independent prognostic factors for OS and DMFS, but maximal axial diameter was not.
Conclusion: Using the Chinese 2008 staging system for NPC produces an acceptable distribution of patient numbers. Stage group, T classification, and N classification represent independent prognosis factors for major endpoints. However, our study reveals several deficiencies in the current system yet.