Preventing colorectal cancer or early diagnosis: Which is best? A re-analysis of the U.S. preventative services task force evidence report

Andrew W Swartz, Jan M Eberth, Scott M Strayer

Abstract

Flexible sigmoidoscopy (FS) is the only cancer screening test to lower the risk of death compared to usual care in randomized controlled trials (RCTs). We hypothesize that this unique death reduction is more attributable to prevention of colorectal cancer (CRC) than to early diagnosis. The systematic review of the 2016 US Preventive Services Task Force Evidence Report for CRC Screening was used for selection of RCT studies. A random-effects meta-analysis of five FS trials (N = 458,002) and four fecal occult blood test (FOBT) trials (N = 328,767) was performed using intention-to-screen outcomes for death, CRC incidence, and death attributed to CRC; correlation and linear regression analyses explored the relationships between these outcomes. At 10.5-11.9 years of follow-up FS reduces death (relative risk [RR], 0.975; 95% CI, 0.958-0.992 and reduces CRC incidence (RR, 0.79; 95% CI, 0.74-0.84). Within the FS trials death reduction shows a strong linear correlation with CRC incidence reduction (r, 0.95; 95% CI 0.42-0.99). At 15.6-30.0 years of follow-up FOBT does not reduce death (RR, 1.001; 95% CI, 0.992-1.010) or CRC incidence (RR, 0.96; 95% CI, 0.89-1.02) but does reduce deaths attributed to CRC (RR, 0.84; 95% CI, 0.78-0.91). Clinical trials of screening FS display a dose-response relationship between the magnitude of CRC prevention and the magnitude of death reduction. Prevention of CRC appears to be the major (or sole) mechanism of action for death reduction by FS in clinical trials. Conversely, early diagnosis of CRC does not appear to reduce death.