Clinical and prognostic factors for melanoma of the skin using SEER registries: collaborative stage data collection system, version 1 and version 2.

BACKGROUND: The objectives of this article are to assess the completeness of the data collected on site-specific factors (SSFs) as a part of Collaborative Stage (CS) version 2 and the impact of the transition from the American Joint Committee on Cancer's (AJCC) 6th to 7th edition guidelines on stage distribution. METHODS: Incidence data for melanomas of the skin from 18 Surveillance, Epidemiology, and End Results (SEER) registries (SEER-18) were analyzed. Percentages of unknown cases for 7 SSFs were examined, along with staging trends from 2004 to 2010 and differences in AJCC 6th and 7th edition stage distributions for 2010 cases. RESULTS: Fewer than 10% of cases were coded as unknown for SSFs 1 (measured thickness), 2 (ulceration), and 3 (lymph node metastasis). For the remaining SSFs, 36-81% of cases were coded as unknown. Stage distributions were relatively consistent across time and between the AJCC 6th and 7th editions, with the exception of stage IA and stage INOS (not otherwise specified), for which a shift in cases was observed between the AJCC 6th and 7th edition guidelines fOR 2010 cases. CONCLUSIONS: A shift of cases out of stage IA and into stage INOS was observed between the AJCC 6th and 7th edition guidelines for 2010 cases. This was attributed to the high number of cases coded as unknown for SSF7 (primary tumor mitotic count/rate). The percentage of cases coded as unknown varied by SSF. Data completeness presents an issue for SSFs introduced in CS version 2.

Incidence of cancers of the oral cavity and pharynx among American Indians and Alaska Natives, 1999-2004.

BACKGROUND: Previous studies identified disparities in incidence rates of cancers of the oral cavity and pharynx between American Indians/Alaska Natives (AI/AN) and non-Hispanic whites (NHW) and differences between various AI/AN populations. Reporting among AI/AN has been hampered by: 1) heterogeneity among various anatomic sites of oral cavity and pharyngeal cancers obscuring unique patterns of individual anatomic sites; 2) race misclassification and underreporting of AI/AN; and 3) sparseness of data needed to identify regional variations. METHODS: To improve race classification of AI/AN, data from US central cancer registries were linked with Indian Health Service (IHS) records. AI/AN incidence data from 1999 to 2004 were stratified by sex, age, stage at diagnosis, and anatomic subsite for 6 IHS geographic regions and compared with NHW populations. RESULTS: For all oral cavity and pharynx cancers combined, among residents of Contract Health Service Delivery Area counties, AI/AN overall had significantly lower incidence rates than NHW (8.5 vs 11.0). However, AI/AN rates were significantly higher in the Northern Plains (13.9 vs 10.5) and Alaska (16.3 vs 10.6), significantly lower in the Pacific Coast (7.7 vs 11.6) and Southwest (3.3 vs 10.4), and similar in the Southern Plains (11.4). Overall AI/AN males had higher incidence rates than AI/AN women. Nasopharyngeal cancer was more frequent (1.1AI/AN vs 0.4 NHW), and tongue cancer less frequent (1.6 AI/AN vs 2.9 NHW) in AI/AN than NHW populations; however, rates varied by region. Stage distribution was modestly less favorable for AI/AN compared with NHW populations. CONCLUSIONS: Variation by region, anatomic site, and sex indicates a need for research into etiologic factors and attention to regional risk factor profiles when planning cancer control programs.