Colorectal cancer screening in Ferrara: Results in early (2005-2009) and late (2009-2013) screening periods and surgical treatment of the screenees versus pre-screening (2003-2005) unselected population.

Background and aims: The Region of Emilia-Romagna (Italy), and particularly Ferrara, has the top national incidence and mortality rates for colorectal cancer (CRC). In March 2005, a population-based CRC screening campaign was started in Emilia-Romagna. The aim of this study was two-fold: (1) to analyse the results of the screening program in Ferrara in both early (March 2005 - March 2009) and late (April 2009 - March 2013) screening periods and (2) to compare the cohort of screenes undergoing CRC surgical treatment in the early and late screening periods with an unselected pre-screening (April 2003 - March 2005) control population who underwent CRC surgical resection at the same institution. Patients and methods: All residents in Ferrara aged 50-69 years took part of the screening program receiving an invitation letter for immunological Faecal Occult Blood Test (i-FOBT). If i-FOBT positive, the screenees were offered a colonoscopy, while if i-FOBT negative were invited by letter to repeat the test after two years. Adenomas were classified in: (a) low-risk adenomas (diameter <10 mm, villous component <25%, and low-grade dysplasia) and (b) high-risk adenomas (diameter 10 mm, at least 25% of villous component or high-grade dysplasia). If colonoscopy was negative, the screenees were invited to repeat endoscopy after 5 years, while if adenomatous polyps were found endoscopic follow-up was started accordingly. In case of CRC diagnosis, necessary surgery was performed within 30 days. Demographic, surgical, and pathology data were reviewed for: (a) screenees undergoing surgical treatment for CRC in early and late screening periods and (b) all patients submitted to elective surgical resection for CRC at the S. Anna University Hospital in Ferrara during the pre-screening period April 2003 - March 2005. Statistical analysis. Data are presented as median (interquartile range 25-75) for normal distribution. Kruskal-Wallis, chi-square, and log rank tests were used to compare non-parametric data, proportions, and hospital length of stay (LOS), respectively. Results: From March 2005 to April 2009 (early screening period), 76,767 people in Ferrara were invited for i-FOBT. Of these, 36,930 (48%) adhered to the screening program and executed the test; 2,004 (5%) were i-FOBT positive, and 1,688 (84%) subsequently agreed to undergo colonoscopy. Endoscopy revealed 853 tumours (adenoma or cancer) confirmed on hystopathologic analysis. In the late screening period (April 2009 - March 2013), 86,763 subjects were invited for i-FOBT and 20,384 (46%) adhered; 1,827 (5%) were positive and 1,506 (82%) agreed to undergo colonoscopy. The endoscopy detected 731 histology proven adenomas or cancers. Data in early and late screening and pre-screening periods were as follows: Variables Pre-screening period (2003-2005) N=275 Early screening period (2005-2009) N=230 Late screening period (2009-2013) N=110 p Age (years) 72.3 (63.5-78.8) 64.0 (58.3-67.8) 64.1 (57.3-68.0) <0.001 Gender (Men:Women) 145 (53%):130 (47%) 149 (65%):81 (35%) 58 (53%):52 (47%) 0.014 Tumor stage 0 I IIa IIb IIIa IIIb IIIc IV 22 (8%) 31 (11%) 99 (36%) 18 (7%) 4 (1%) 56 (20%) 28 (10%) 17 (6%) 54 (24%) 48 (21%) 49 (21%) 2 (1%) 7 (3%) 37 (16%) 30 (13%) 3 (1%) 24 (22%) 38 (34%) 17 (15%) 3 (3%) 0 21 (19%) 6 (6%) 1 (1%) <0.001 . Tumor Site Right colon Transverse colon Left colon-rectum 103 (37%) 13 (5%) 159 (58%) 74 (32%) 7 (3%) 149 (65%) 35 (32%) 5 (4%) 70 (64%) 0.505 Hospital LOS (days) 12 (9-17) 10 (9-13) 9 (7-11) <0.001 Conclusions: The adherence to i-FOBT remained low throughout the screening periods (48% early period vs. 46% late period); adequate campaigns of sensitization to CRC screening are therefore needed. The highest adherence rate in Italy was observed for colonscopy adherence among i-FOBT positive screeenes (≥82%). The screening program did not detect an increased proportion of proximal colorectal cancers, in contrast to other Authors. As expected, decreased patients’ age as well as earlier stage CRC diagnosis was achieved with the screening program. Reduction in postoperative hospital LOS was likely due to younger patients’ age and increased adoption of minimally invasive techniques.