Comparing attendance and detection rate of different screening strategies for colorectal cancer
aCPO Piemonte, Torino, Firenze, bSurgery Unit II, Istituto Europeo di Oncologia, Milano, Firenze, cEpidemiology Unit. ASL "Citta' di Milano", Milano, Firenze, dGastroenterology Unit, CSPO, Firenze, eGastroenterology Unit, Maria Vittoria Hospital, Torino, Rimini, fGastroenterology Unit, Infermi Hospital, AUSL Rimini, Rimini, gGastroenterology Unit, Infermi Hospital, ASL 12, Biella, hGastroenterology Unit, Mauriziano "Umberto I"Hospital, Torino, Firenze, iGastroenterology Unit, S Giovanni Bosco Hospital, Torino, Firenze, jGastroenterology Unit, S Giovanni Battista-Molinette Hospital, Torino, Firenze, kGastroenterology Unit, S Giovanni Antica Sede, Hospital, Torino, Firenze, lCitopathology Unit, CSPO, Firenze, Italy
Aims: To estimate the degree of attendance and the detection rate (DR) achievable through different strategies of colorectal cancer (CRC) screening. Methods: Population based multicentre randomised trial comparing four screening protocols: 1) Biennial fecal occult blood test (FOBT) 2) "Once-only" sigmoidoscopy (FS) 3) Patient's choice between the previous modalities 4) FS followed by biennial FOBT for screenees with negative FS. With different sampling strategies a sample of men and women, aged 55 to 64, was drawn from the National Health Service lists in the study areas. General Practitioners (GPs) were asked to exclude their patients with a history of CRC, adenomas, inflammatory bowel disease, recent (ï£2 years) colorectal endoscopy or FOBT, or two cases of CRC among first degree relatives. Eligible subjects were randomised, within GP, and were mailed a personal invitation letter, signed by the GP. A reminder was mailed to non-responders. Subjects allocated to FS, who did not respond to the first invitation and reminder, were mailed two additional invitations (at 12 and 24 months). Immunological FOBT was adopted. The aim of FS was to advance the scope beyond the sigmoid-descending colon junction. Screenees with "high risk" distal polyps (villous component, severe dysplasia, CRC, size ï³ 10 mm, or > 2 adenomas), and those with positive FOBT, were referred for colonoscopy (TC). Results: Of 28,319 people included in the sample, 1,637 were excluded by the GP, 26,682 were randomised and invited for screening. After excluding subjects who could not be traced (456 undelivered letters; range: 1.3%- 2.7% across intervention groups) attendance rate was: 28.7% (2338/8138) for group 1, 26.9% (980/3641) for group 2, 24.1% (965/3583) for group 3, and 27.7% (3004/10864) for group 4. Mailing of FOBT kit was associated with a non significant increase in the uptake rate. Among subjects who were offered the option to choose between FS and FOBT, 445 of the attenders (46.1%) underwent FS and 520 (53.9%) underwent FOBT. Positivity rate of FOBT was 4.3% (122/2858) and the positive predictive value for advanced neoplasia (villous component, severe dysplasia, CRC, size ï³ 10 mm) was 44.9%: 11 patients were detected with CRC (3.8 per 1000) and 35 with advanced adenomas (1.2%). Among 4429 patients undergoing FS, 4371 were actually examined, while 58 with inadequate bowel preparation refused to repeat the test. Referral rate to TC was 8.1% (355/4371). CRC was detected in 19 patients (4.3 per 1000), while 216 (5.0%) harboured an advanced adenoma. Conclusions: Compliance was higher (RR:1.05; 95%CI 1.0-1.1; p=0.44) among people allocated to FOBT, as compared to FS arms. However, the 4.4% relative increase in compliance was associated with a 12% relative reduction in the DR of CRC; also DR for advanced adenomas was four times higher for FS, compared to FOBT. The evaluation of the impact of these strategies should take into account the cost and the number of tests required to achieve the same yield of advanced neoplasia.
Paper presented at the International Symposium on Predictive Oncology and Intervention Strategies; Nice, France; February 7 - 10, 2004; in oral session 992 (Screening & detection - Part I).