1793 - External validation of the 8th Edition Melanoma Staging System of the American Joint Committee on Cancer (AJCC) using the Surveillance, Epidemiology and End Results (SEER) Program

Background

Changes to the AJCC melanoma staging system eighth edition (AJCC-8) should be independently validated to assess its prognostic performance compared with the seventh edition (AJCC-7) in accurately staging melanoma.

Methods

We used the SEER-18 registry from 2010 to 2015 to extract patient-, tumour-related and survival data. Kaplan-Meier analysis was used for overall survival (OS) and cancer-specific survival (CSS) for AJCC-7 and AJCC-8. Cumulative hazard functions were computed using Nelson-Aalen function.

Results

Of 126,408 individuals, 59,989 (47%) and 60,411 (48%) had available data for pathological and clinical stage OS analysis, respectively. The 12-month OS for AJCC-7 among pathologically staged patients were: stage IA 99%, stage IB 99%, stage IIA 96%, stage IIB 94%, stage IIC 87%, stage IIIA 98%, stage IIIB 94%, stage IIIC 82% and stage IV 41%. The 12-month OS for AJCC-8 patients was similar to AJCC-7 but was 88% for stage IIIC and 65% for stage IIID. The 12-month risk of dying for pathological stage IIIC was 13% compared to 41% for pathological stage IIID (p < 0.001). Stage IV individuals with an elevated LDH had worse OS and CSS at all other measured time-points up to 60 months compared to those with a normal LDH. Stage IV individuals with brain metastases (M1d) also had worse OS and CSS at all other measured time-points up to 60 months compared to other stage IV subgroups.

Conclusions

The discriminatory ability of the AJCC-7 and AJCC-8 melanoma staging system appear comparable. Changes in AJCC-8 identifies individuals with a poorer prognostic within new subgroups. These include the subgroups of stage IIID and M1d within stage IV individuals, and the addition of elevated LDH as a prognostic marker in stage IV disease. However, advanced T stage, node-negative tumours experienced worse survival compared with earlier T stage, node-positive tumours requiring further research to evaluate the underlying biology of these tumour subgroups.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.