• Public Health Sciences Division, Fred Hutchinson Cancer Research Center, Seattle, WA, USA.
• Discipline of Genetics, Faculty of Medicine, Memorial University, St. John's, NL, Canada.
• Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, The University of Melbourne, Melbourne, VIC, Australia.
• Department of Population Science, American Cancer Society, Atlanta, GA, USA.
• Lunenfeld Tanenbaum Research Institute, Mount Sinai Hospital, University of Toronto, Toronto, ON, Canada.
• Nutrition and Metabolism Section, International Agency for Research On Cancer, World Health Organization, Lyon, France.
• Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany.
• Program in Molecular Pathological Epidemiology, Department of Pathology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.
• Department of Internal Medicine, University of Utah, Salt Lake City, Utah, USA.
• Department of Radiation Sciences, Oncology Unit, Umeå University, Umeå, Sweden.

Identification of new genetic markers may improve the prediction of colorectal cancer prognosis. Our objective was to examine genome-wide associations of germline genetic variants with disease-specific survival in an analysis of 16,964 cases of colorectal cancer. We analyzed genotype and colorectal cancer-specific survival data from a consortium of 15 studies. Approximately 7.5 million SNPs were examined under the log-additive model using Cox proportional hazards models, adjusting for clinical factors and principal components. Additionally, we ran secondary analyses stratifying by tumor site and disease stage. We used a genome-wide p-value threshold of 5 × 10^-8 to assess statistical significance. No variants were statistically significantly associated with disease-specific survival in the full case analysis or in the stage-stratified analyses. Three SNPs were statistically significantly associated with disease-specific survival for cases with tumors located in the distal colon (rs698022, HR = 1.48, CI 1.30-1.69, p = 8.47 × 10^-9) and the proximal colon (rs189655236, HR = 2.14, 95% CI 1.65-2.77, p = 9.19 × 10^-9 and rs144717887, HR = 2.01, 95% CI 1.57-2.58, p = 3.14 × 10^-8), whereas no associations were detected for rectal tumors. Findings from this large genome-wide association study highlight the potential for anatomical-site-stratified genome-wide studies to identify germline genetic risk variants associated with colorectal cancer-specific survival. Larger sample sizes and further replication efforts are needed to more fully interpret these findings.