No Evidence Seen of 'Genetic Anticipation' In Familial Colon Cancer

October 1, 1996

BUFFALO, NY--New analysis of familial colorectal cancer data suggests that the disease is not associated with genetic anticipation--the earlier onset of disease in successive generations--said Gloria M. Petersen, PhD, at the Eighth Annual Meeting of the ICG-HNPCC (International Collaborative Group-Hereditary Nonpolyposis Colorectal Cancer).

BUFFALO, NY--New analysis of familial colorectal cancer data suggeststhat the disease is not associated with genetic anticipation--theearlier onset of disease in successive generations--said GloriaM. Petersen, PhD, at the Eighth Annual Meeting of the ICG-HNPCC(International Collaborative Group-Hereditary Nonpolyposis ColorectalCancer).

"The study shows that a cohort effect is responsible forthe apparent observation of genetic anticipation in this disease,"she said, "and that when the data are corrected for thisbias, there is no statistical trend toward earlier age of onsetin succeeding generations."

Dr. Petersen, associate professor of epidemiology, and her colleaguesat the Johns Hopkins School of Hygiene and Public Health analyzeddata from family history questionnaires from the Johns HopkinsHereditary Colorectal Cancer Registry. Of 588 kindreds (1,293patients), 400 parent/offspring pairs in 260 pedigrees had completeinformation regarding age of diagnosis.

The 400 pairs were divided by their birth year into three cohorts:persons born before 1921, between 1921 and 1930, and after 1930.The analysis showed that the offspring and parents in the firstage group were diagnosed at ages 64 and 65, respectively; in thesecond age group, at ages 57 and 66; and in the third age group,at ages 44 and 61.

"The expected cohort effect was observed among the offspringin the latter two groups. However, in the cohort of subjects bornbefore 1921, a group that had the same window of risk, we observedno statistical difference between parents and offspring,"Dr. Petersen said. Further analysis of the cohort born before1921, by pairwise comparison and life table analysis, again showedno difference in age at diagnosis.

Subgroup Analysis

The researchers also analyzed two subgroups--52 parent/offspringpairs from 29 families who met the ICG criteria for hereditarynonpolyposis colorectal cancer (HNPCC) and 14 parent/offspringpairs from seven families with known germline mutations of DNAmismatch repair genes.

Neither group showed a statistical difference in mean age at diagnosisbetween generations: For the HNPCC families, parents and offspringwere ages 58 and 60, respectively, at diagnosis, and for the familieswith repair gene mutations, disease diagnosis was at ages 45 and44, respectively.

The study also found no apparent genetic imprinting associatedwith familial colorectal cancer, Dr. Petersen said. Genetic imprintingis a phenomenon in which a specific gene differs, ie, may be moreactive or less active, depending on whether it is inherited fromthe mother or the father.

"There is no observed difference in the age of diagnosisin this disease when examined by the gender of the parent or offspring,"Dr. Petersen said.