Immunochemical faecal occult blood test: number of samples and positivity cutoff. What is the best strategy for colorectal cancer screening?
We present the main results of the third survey of the Italian screening programmes for colorectal cancer carried out by the ONS (Osservatorio Nazionale Screening, National Centre for Screening Monitoring) on behalf of the Ministry of Health. During 2006, many new programmes were activated and by the end of the year, 44% of Italians aged 50-69 years were residing in areas covered by organised screening programmes (theoretical extension). Five regions had their entire population covered. In the South of Italy and Islands only one new programme was activated in 2006, with a 10% theoretical extension. The majority of programmes employ the faecal occult blood test (FOBT), while some have adopted the flexible sigmoidoscopy (PS) once in a lifetime, or a combination of both. Overall, about 2,107,000 subjects were invited to undergo FOBT, 70.3% of those to be invited within the year. The adjusted attendance rate was 44.6% and approximately 907,000 subjects were screened. Major differences in the attendance rate were observed among regions, with 10% of programmes reporting values lower than 26%. Positivity rate of FOBT programmes was 5.3% at first and 3.9% at repeat screening: it increased with age and was higher among males than females in all age groups. The average attendance rate to total colonoscopy (TC) was 81% (10th-9Oth percentiles: 64%, 93%). Completion rate of TC was 89%, with higher rates among males. At first screening, the detection rate (DR) per 1,000 screened subjects was 3.1 and 14.6 for invasive cancer and advanced adenomas (AA--adenomas with a diameter > or =1 cm, with villous/tubulo-villous type or with high-grade dysplasia), respectively; the corresponding figures at repeat screening were 1.3 per thousand for cancer and 7.7 per thousand for AA. The DR of cancer and adenomas increased with age and was higher among males; 25% of screen-detected cancers were in TNM stage III+. The positive predictive value (PPV) was 6.8% for cancer and 32.1% for AA at first screening, and 4.0% for cancer and 23.4% for AA at repeat screening. Given the high PPVof a positive FOBT, obtaining a high attendance at TC is therefore crucial. Seven programmes employed PS as the screening test: 51.4% of the target population (about 50,000 subjects) were invited and 7,589 subjects were screened, with an attendance rate of 29.2%. Overall, 88% of FS were classified as complete. TC referral rates ranged between 6.1 and 17.8%, due to different referral criteria. Among subjects referred to colonoscopy, the prevalence of proximal AA and cancer ranged from 6 to 18.2%. The overall DR (subjects with at least one advanced lesion) ranged from 4.9 to 7.5%. In conclusion, during 2006, organised colorectal cancer screening programmes in Italy grew considerably, covering almost half of the eligible population at a national level. Many programmes were activated in the second part of the year, thus their results should be evaluated with caution. However, reported experiences showed good results in terms of attendance and DR, although some critical aspects need to be carefully addressed when planning and implementing screening activity.