Stage specific survival is altered by the changes in the 2002 AJCC staging for melanoma

Abstract

7510 Purpose: To examine how stage specific survival is altered in the 2002 American Joint Committee on Cancer (AJCC) melanoma staging system compared to the 1997 AJCC staging system. Patient and Methods: 5,847 consecutive patients presented with melanoma to Memorial Sloan Kettering Cancer Center (MSKCC) from 1996 to 2004 and were prospectively entered into a database. These patients were staged under both the 1997 and 2002 AJCC staging criteria. Overall survival estimates were determined using the Kaplan-Meier method. Results: 1,035 patients were shifted to a lower stage in the 2002 staging system while only 15 patients were upstaged. The number of stage I patients increased greatly under the 2002 system (n=2,166) as compared to the 1997 system (n=1,463) although stage I survival probability estimates were virtually the same with both classification systems. As a result of the changes in 2002, the stage II survival estimates decreased considerably with a 15% lower 5 year overall survival (79% (1997) versus 64% (2002)). With the initiation of subgroups in 2002, it became apparent that stage III patients were very heterogeneous in terms of their survival probabilities (5 year overall survival ranged from 70% in IIIA to 24% in IIIC). Conclusions: Stage specific survival was substantially altered by the changes made in the 2002 AJCC staging for melanoma, particularly for stage II. It is very important to be aware of this effect so that accurate conclusions can be drawn with regard to treatment efficacy or comparisons across different studies. In addition, stage subgroups were added in 2002 that demonstrated the diversity of risk within stage group and did not improve prediction ability. This highlights the importance of developing individualized risk prediction models. No significant financial relationships to disclose.