Send to

Choose Destination
J Am Coll Surg. 2012 Apr;214(4):608-17; discussion 617-9. doi: 10.1016/j.jamcollsurg.2011.12.027. Epub 2012 Feb 17.

A novel and accurate computer model of melanoma prognosis for patients staged by sentinel lymph node biopsy: comparison with the American Joint Committee on Cancer model.

Author information

Department of Surgery, University of Louisville, Louisville, KY 40202, USA.



We found that a computer model developed by the American Joint Committee on Cancer (AJCC) melanoma staging committee had limitations for predicting prognosis of patients staged by sentinel lymph node (SLN) biopsy. We sought to develop a model that more accurately predicts prognosis in this population.


Using a data set obtained from a prospective multi-institutional study of 2,507 patients with clinically node-negative melanomas ≥1.0 mm Breslow thickness, we developed a prognostic model using a Cox regression formula incorporating a number of significant clinicopathologic factors. The AJCC model and our model were used to predict 5-year survival from this test data set. The concordance correlation coefficient (CCC) was determined and chi-square tests were performed. Our new prognostic model was validated using an independent data set of 1,001 patients.


Using the test data set, the CCC for the AJCC model was 0.875; chi-square tests demonstrated statistically significant differences between observed and predicted survivals for numerous clinicopathologic factors. The CCC for our model was 0.976 and none of the chi-square tests was statistically significant. Our model performed similarly well in SLN-negative patients (CCC 0.929) and SLN-positive patients (CCC 0.889). The AJCC model performed well in SLN-negative patients (CCC 0.854), but not in SLN-positive patients (CCC 0.626). Using the validation data set, similar findings were obtained.


Our prognostic model provides superior survival estimates compared with the AJCC model for patients undergoing SLN biopsy. This online tool is available at, and will provide important information that can be used to guide adjuvant therapy decisions and stratification in clinical trials.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center