Survival of patients with symptom- and screening-detected colorectal cancer

// Hermann Brenner 1, 2, 3 , Lina Jansen 1 , Alexis Ulrich 4 , Jenny Chang-Claude 5 , Michael Hoffmeister 1 1 Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany 2 Division of Preventive Oncology, German Cancer Research Center (DKFZ), Heidelberg, Germany 3 German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ), Heidelberg, Germany 4 Department of General, Visceral and Transplantation Surgery, University of Heidelberg, Heidelberg, Germany 5 Unit of Genetic Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany Correspondence to: Hermann Brenner, e-mail h.brenner@dkfz.de Keywords: colonoscopy, colorectal cancer, fecal occult blood test, screening, survival Received: March 07, 2016     Accepted: April 26, 2016     Published: May 17, 2016 ABSTRACT Background: An increasing proportion of colorectal cancer (CRC) patients are diagnosed by screening rather than symptoms. Aims: We aimed to assess and compare prognosis of patients with screen-detected CRC and symptom-detected CRC. Methods: Overall and CRC specific mortality over a median follow-up of 4.8 years was assessed according to mode of diagnosis (symptoms, screening colonoscopy, fecal occult blood test FOBT, other) in a multi-center cohort of 2,450 CRC patients aged 50-79 years recruited in Germany in 2003-2010. Results: 68%, 11% and 10% were detected by symptoms, screening colonoscopy and FOBT, respectively. The screen-detected cancers had a more favorable stage distribution than the symptom-detected cancers (68% versus 50% in stage I or II). Age- and sex adjusted hazard ratios (HRs) of total mortality with 95% confidence intervals (95% CIs) compared to symptom-detected cancers were 0.35 (0.24-0.50) and 0.36 (0.25-0.53) for screening colonoscopy and FOBT detected CRCs, respectively. HRs were only slightly attenuated and remained highly significant after adjustment for stage and multiple other covariates (0.50 (0.34-0.73) and 0.54 (0.37-0.80), respectively). Even stronger associations were seen for CRC specific mortality. Patients with screen-detected stage III CRC had as good CRC specific survival as patients with symptom-detected stage I or II CRC. Conclusions: Patients with screen-detected CRC have a very good prognosis far beyond the level explained by their more favorable stage distribution. Mode of detection is an important, easy-to-obtain proxy indicator for favorable diagnosis beyond earlier stage at diagnosis and as such may be useful for risk stratification in treatment decisions.