Although several modalities are being used in colorectal cancer (CRC) screening, only fecal occult blood testing (FOBT) and flexible sigmoidoscopy (FS) have been subjected to randomized trials and long-term follow-up. Both have been found to reduce CRC mortality compared to no screening (18% and 28%, respectively), but a direct comparison between the two has never been done with end-point CRC incidence and mortality. Also, it is not clear which modality is the most cost-effective in any given population. In Norway, a country with high colorectal cancer incidence and little prior screening against the disease, the government decided to start a comparative effectiveness research pilot in 2012 with these two screening modalities.

The pilot aims to randomize the entire population aged 50-74 in a defined geographical area in South-East Norway (approximately 140 000 individuals) to one of two screening modalities. Half will be randomized to biennial screening with immunochemical FOBT (OC-Sensor Diana, Eiken Ltd), and half to once only screening with FS. The enrollment will take four years and the pilot includes a number of sub-studies to determine how the screening is perceived in the target population. We report the results from the enrollment in the main trial after the first 18 months.

A total of 35746 women and men have been invited so far, 21495 to iFOBT and 14251 to FS. Participation rates have been 49% in the FS arm, and 57% in the iFOBT arm, with slightly higher rates among women than men. Positive FS was defined as advanced neoplasia or three or more adenomas. A total of 10.6% of the FS patients have been referred to colonoscopy. The cut-off value for positive iFOBT was set to 75 ug/L, and 780 patients (6.4%) have so far tested positive and referred to colonoscopy.

Forty-three cases of CRC have been detected in the FS group so far, at a rate of 6/1000 examined, somewhat higher in men (7.3/1000) than in women (5/1000). Thirty-nine cancers have been found in the FOBT group after first screening round (3.1/1000 examined), with 4.6/1000 in men and 1.9/1000 in women. Overall the adenoma detection rate at sigmoidoscopy is 14%. A total of 881 high-risk adenomas (adenomas >10 mm, or with high-grade dysplasia or villous features) have been detected. Adenoma detection rate at colonoscopy in the referred population is 60.5%. Complications at sigmoidoscopy have so far been 14/1000 exams, with the vast majority being vasovagal. At colonoscopy the complication rate is currently 5.8/1000, the main ones being reported are bleeding or vasovagal reactions.

Conclusion: participation rates in both arms are slightly below the expected 50% for FS and 60% for FOBT. Cancer rates among those screened are higher in the once-only FS than after first round of biennial iFOBT screening, but rates are within the expected range.

Citation Format: Giske Ursin, Anita Jørgensen, Ole Petter Børmer, Thomas de Lange, Per K. Sandvei, Christian Thorjussen, Geir Hoff. Colorectal cancer screening pilot: Comparative effectiveness research using two screening modalities. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3242. doi:10.1158/1538-7445.AM2014-3242