TY - JOUR
T1 - Analysis and comparison of the 7th Edition American Joint Committee on Cancer (AJCC) nodal staging system for metastatic cutaneous squamous cell carcinoma of the head and neck
AU - Clark, Jonathan R.
AU - Rumcheva, Pavlina
AU - Veness, Michael J.
PY - 2012/12
Y1 - 2012/12
N2 - Background. The American Joint Committee on Cancer (AJCC) substantially changed the staging of cutaneous squamous cell carcinoma (cSCC) in the 7th edition of its staging manual. We aim to compare the 7th edition AJCC staging of nodal metastases from cSCC with the N1S3 staging system. Methods. Analysis of 603 patients from two prospective cancer center databases was performed. Multivariable analysis was performed using a Cox proportional hazards competing risk model adjusting for the effect of immuno-suppression, treating institution, adjuvant radiotherapy, nodal margins, and extracapsular spread. Criteria used for comparing staging systems were distribution of patients, stratification of patients according to risk of death from cSCC, and model performance. Results. The N1S3 staging system functioned well in terms of distribution and stratification of patients. The distribution of patients within the AJCC staging system was problematic with three groups (N2a, N2c, and N3) containing less than 10 % of patients without any prognostic relevance. Stratification of patients within the AJCC staging system was poor in terms of monotonicity (N2c) and distinctiveness (N2a). The performance of the AJCC and N1S3 staging systems was similar despite the AJCC staging being more complex. Conclusions. The N1S3 staging system for cSCC is preferred on the grounds of better distribution, stratification, and parsimony.
AB - Background. The American Joint Committee on Cancer (AJCC) substantially changed the staging of cutaneous squamous cell carcinoma (cSCC) in the 7th edition of its staging manual. We aim to compare the 7th edition AJCC staging of nodal metastases from cSCC with the N1S3 staging system. Methods. Analysis of 603 patients from two prospective cancer center databases was performed. Multivariable analysis was performed using a Cox proportional hazards competing risk model adjusting for the effect of immuno-suppression, treating institution, adjuvant radiotherapy, nodal margins, and extracapsular spread. Criteria used for comparing staging systems were distribution of patients, stratification of patients according to risk of death from cSCC, and model performance. Results. The N1S3 staging system functioned well in terms of distribution and stratification of patients. The distribution of patients within the AJCC staging system was problematic with three groups (N2a, N2c, and N3) containing less than 10 % of patients without any prognostic relevance. Stratification of patients within the AJCC staging system was poor in terms of monotonicity (N2c) and distinctiveness (N2a). The performance of the AJCC and N1S3 staging systems was similar despite the AJCC staging being more complex. Conclusions. The N1S3 staging system for cSCC is preferred on the grounds of better distribution, stratification, and parsimony.
UR - http://www.scopus.com/inward/record.url?scp=84876495970&partnerID=8YFLogxK
U2 - 10.1245/s10434-012-2504-2
DO - 10.1245/s10434-012-2504-2
M3 - Article
C2 - 22805870
AN - SCOPUS:84876495970
VL - 19
SP - 4252
EP - 4258
JO - Annals of Surgical Oncology
JF - Annals of Surgical Oncology
SN - 1068-9265
IS - 13
ER -