Trends in lung adenocarcinoma incidence and survival

Abstract

Dear Sir, A rise in lung cancer adenocarcinoma has been observed in several European countries and in the US [1] and it has been attributed to the shift to filter cigarette smoking [2]. Also a decrease in the adenocarcinoma survival rates recently observed in the Netherlands has been attributed to the increased consumption of low-tar filter cigarettes and to different inhalation behavior [3]. Such a decrease was not evidenced in the US where survival rates for lung adenocarcinoma were stable from 1973 to 1997 (seerstat@ims.nci.nih.gov). In the provinces of Florence and Prato, central Italy (approximately 1 180 000 inhabitants at the 1991 census) the Tuscany Cancer Registry (RTT) is active since 1984 [4]. During 1985–1997, 10 400 lung cancer cases (8572 males and 1828 females) were diagnosed in the RTT area, 2662 squamous cell carcinomas (25.6%), 1748 (16.8%) adenocarcinoma, 882 (8.5%) small cell carcinomas, 176 (1.7%) large cell cancers, 37 other and 4895 (47.1%) without histology. In central Italy, we evidenced from 1985 to 1997 an increasing incidence trend for lung adenocarcinoma. The expected annual percent change (EAPC) computed on standardized rates (European population) in a loglinear model with the SEER*Stat software (seerstat@ims.nci.nih.gov) showed a statistically significant yearly increase not only in males +3.0% [95% confidence intervals (CI) +0.5/+5.5] but also in females +5.4 (95% CI +0.6/+10.5). Meanwhile, there was a decreasing incidence trend for squamous cell carcinomas, significant among males (−3.8, 95% CI − 5.6/−2.0), and quite stable trends for the other types. In Italy, the proportion of lung cancer cases without histology verification is rather high and the values for the RTT are on the Italian average [5]. The trend for cases without verification was quite stable (EAPC=− 1.2%). The increase in the incidence of adenocarcinoma can not be attributed in the RTT area to bronchiolaolveolar carcinoma (only 43 cases diagnosed in the period). Each patient was actively followed-up at 31/12/1998 (about 0.5% lost to follow-up) and relative survival rates were computed. The survival rates showed an increasing trend from cases diagnosed in 1985–1990 and those diagnosed in 1991–1997. The increasing survival tendency was evident at 1-, (48 vs. 51%), 3(23 vs. 26%) and 5-years (17 vs. 20%) after diagnosis; in males (at 5-years 17 vs. 20%) and in females (15 vs. 21%), and for subjects younger (21 vs. 22%) and older than 60 years (14 vs. 19%). According to the cancer extension at diagnosis, the increase in the incidence was due to a significant increase for both localized and distant cancers, and to a not significant increase for regionally diffused and unspecified cases. For all the classes of extension there was a not statistically significant increasing survival trend from 1985–1990 to 1991–1997. In conclusion, in central Italy there was a statistically significant increase from mid-1980s to late-1990s in the incidence rates of lung adenocarcinoma. The survival increasing tendency, although not statistically significant, seemed intriguing being evident in both genders, among young and old patients and in different groups of disease extension. Presumably, survival has been positively affected by an anticipation of diagnosis and an improvement in the quality of diagnoses. From 1985–1987 to 1995–1997 the proportion of localized cases increased from 11 to 16% and that for metastatic cases from 11 to 19%. This may probably be the effect of a stage shift according to a more accurate definition of cancer extension at diagnosis; in fact, regional cases decreased from 47 to 34%. Cases without diffusion definition were stable (32%). Therefore, survival trends for lung adenocarcinoma evidenced in central Italy * Corresponding author. Tel.: +39-055-566-2693; fax: +39-055679-954. E-mail address: e.crocetti@cspo.it (E. Crocetti).