Abstract

Abstract A82 Changing trends over recent decades in squamous cell carcinomas (SCCs) compared to adenocarcinomas (ACs) at various organs/sites have been reported in the U.S. and many other countries. The phenomenon has no clear explanation. This dynamic, however, is important for developing future cancer prevention strategies. Our aim is to study aspects important for general cancer prevention strategies: incidence trends of SCCs and ACs (for each cancer histotype, regardless of tumor site), and parameters characterizing differences in the mechanisms of ACs compared to SCCs. Data and methods: The most prevalent ACs and SCCs for 12 cancer sites were selected, and their time trends for age-adjusted incidence rates analyzed using SEER data for a 31-year period (1973-2003). Their age-specific incidence rates were analyzed using biologically motivated models of carcinogenesis, such as the Armitage-Doll model, two-stage clonal expansion model, and models with hidden frailty. In total, each of 186 sex-, race-, time period-, and histology-specific age patterns of incidence rates were fit to 10 models. The model providing the best fit for most age-patterns (i.e., frailty model with Weibull baseline and gamma-distributed frailty) was generalized to include parameters characterizing carcinogenesis (e.g., latency, number of stages of carcinogenesis, and others). Results: The incidence of all SCCs analyzed (lung, esophageal and cervical) decreased. In contrast, the incidence of seven out of twelve ACs (lung, esophageal, hepatic, renal, breast, prostate, cervical) increased (e.g., ACs of lung and esophagus are now predominant over SCCs - incidence shifted about 15 years ago). Only two out of twelve ACs (uterine and gastric) decreased. SCCs and ACs with corresponding time trends for cancer histotypes where checked for histotype-related similarity in underlying carcinogenesis mechanisms - expected differences were detected. The number of carcinogenesis stages did not differ for most ACs, regardless of cancer site, such as lung, stomach, esophagus, colon, rectum, pancreas, kidney, corpus uteri, and breast duct carcinomas. SCCs of lung, esophagus, and cervix uteri did not differ from each other by number of carcinogenesis stages but differed substantially from ACs (SCCs had more stages than ACs). Conclusions: Over the past decades, SCCs incidence continued to decline, while incidence of ACs, and their proportion compared to SCCs, increased. That may reflect changes associated with histotype-specific risk factors (e.g., smoking and HPV - for SCCs) occurring over the past several decades, including the effects of preventive strategies (e.g., anti-smoking strategies, screening effectiveness to detect precursor lesions). SCCs and ACs may differ not only by their time trends and associated risk factors, but by mechanisms these factors influencing tumorigenesis. Cancer risk factors have been clearly established for SCCs, but remain unclear for most ACs (e.g., obesity, diet, sex hormones, physical activity), making it more difficult to develop preventive strategies for ACs. Analyses of SCCs and ACs not only per site/organ, but also by histotype in general from various perspectives could be important to understand mechanisms of tumor development and thus planning of future preventive measures (e.g., risk factor intervention, screening, chemoprevention). Citation Information: Cancer Prev Res 2008;1(7 Suppl):A82.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call