Impact of genetic testing on reducing colorectal cancer. D. W. Neklason1, 2, 7, H. A. Hanson2,3,7, C. Schaefer2,7, G. Mineau2, 6, 7, M. F. Leppert5, 7, R. W. Burt2,1,7, K. R. Smith2,4,7 1) Internal Medicine, University of Utah, Salt Lake City, UT., USA; 2) Huntsman Cancer Institute at University of Utah, Salt Lake City, UT, USA; 3) Family and Preventive Medicine, University of Utah, Salt Lake City, UT, USA; 4) Family and Consumer Studies, University of Utah, Salt Lake City, UT, USA; 5) Human Genetics, University of Utah, Salt Lake City, UT, USA; 6) Oncological Sciences, University of Utah, Salt Lake City, UT, USA; 7) Population Sciences, Huntsman Cancer Institute, Salt Lake City, UT, USA.

   In an attempt to measure the impact of genetic testing in disease prevention, we had a unique opportunity to follow a large family with >7000 individuals over 30 years. The family has an attenuated form of familial adenomatous polyposis leading to a 69% lifetime risk of colorectal cancer (CRC) due to mutation in the APC gene. Utah statewide cancer registry is linked to genealogies through the Utah Population Database which allowed measurement of hazard ratios for CRC over time. Subjects were enrolled in research starting in 1980s, after discovery of the involved gene in 1990s, and again in 2000s. The specific APC mutation was tested in 1005 participants (186 positive; 818 negative) and 669 went on to have clinical confirmation with genetic counseling around 1993 (144 positive; 525 negative). Two branches of the family, B and E had the mutation, whereas there was no evidence of the mutation in over 500 individuals tested from branch D. Analysis was restricted to individuals residing in Utah after 1963 (accurate cancer records) and excluded those < 30 years, deceased prior to age 30 or absence of a birth date. The final sample included 2,550 individuals (branch D=640, B=250, E=1660). Although not everyone in the family chose to participate in research, we assume that genetic counseling, education and communication within the family had potential to modify behaviors within each branch. We hypothesize that the risk will decrease over time with interventions. CRC risk was evaluated by family branch using Cox proportional hazard models. We show that when controlling for sex and birth year, individuals in branches B and E combined have a 5 fold increase in the risk of CRC (p=0.02) over branch D. We then evaluated separate hazards for 1963 to 1982 (pre-hereditary knowledge), 1983 to 1993 (pre-genetic diagnosis), and >1993 (post-genetic diagnosis) in branches B/E versus branch D. This study design is controlled with the comparison branch D undergoing the same intervention, but not harboring the mutation. The resulting hazard ratios decrease over these time windows from 4.41 to 2.39 to 2.23 with a p-value of 0.001. In conclusion, we are able to demonstrate the impact of the presence of a positive genetic test combined with education and counseling to reduce cancer risk in a family setting. Although the risk of CRC is dramatically reduced, it is not reduced to the level of the branch D, suggesting that there is opportunity for additional interventions.

You may contact the first author (during and after the meeting) at