Environment and genetics in the etiology of gastrointestinal tract cancers

All biological diseases have a pathophysiological basis, but many are sporadic, i.e., patients lack close relatives with the same disease. Familial aggregation of a disease may indicate sharing of environmental risk factors between the family members or true heritable causation (or interaction of the two). However, for most common diseases the extent of heritable causation is not well known, excluding the rare Mendelian syndromes (which are 1% of all cancers [1]). Familial aggregation may account for a small proportion of the disease (some 10% of cancers [1]), which may be due to the lack of heritable effects or to their low penetrance. The “traditional” tool for assessing heritability of a disease is the twin study. Twinning is however rare and the statistical power is low. For example, only breast, prostate and colorectal cancers were found to have significant heritable components in a study covering Swedish, Danish and Finish twin registers extending to very old birth cohorts of the 1800s [2]. Using the same methodology on familial cancers from the Swedish Family-Cancer Database, even relatively rare cancers were found to have a heritable component, because of the superior sample size and statistical power [3]. The definition of the heritable proportion of a disease and the identification of the affected individuals is an important task for genetic epidemiology, because the search for heritable genes will only be successful for heritable diseases [4]. Gene-environment and gene-gene interactions and disease heterogeneity, i.e., many causative genes, are further complications, which may be overcome by a rigorous definition of study populations and disease phenotypes, some tasks for genetic epidemiology.

The Swedish Family-Cancer Database as a global resource

Family history is critically dependent on its accuracy. A recommendation for mutation testing for a disease gene, such as BRCA1/2 or mismatch repair gene, is usually based on the family history, and the result may turn out to be negative simply, if the family history has been wrong. Unfortunately patients tend often to be unaware of the medical histories of their relatives. Recently, data on the accuracy of family histories have been accumulated by comparing reported histories to those with medical documentation. Murff et al. collected literature on the accuracy of family history data for cancer [5]. The positive predictive value (positives of true positives) of the reported family history compared to medical diagnosis in a first-degree relative was 93% for breast, 85% for prostate, 81% for colon, 69% for ovarian and 37% for endometrial cancers when reported by cancer patients. When the family histories were reported by healthy controls, the positive predictive values were about 20% units lower; the worst were ovarian (25%) and endometrial cancers (17%). The values were lower for the second-degree relatives than for the first-year relatives. Many of the findings were similar in a Swedish study, which suggested that reporting is worse for rare cancers; somewhat surprisingly, the accuracy of reporting cancer in siblings was worse than that in parents [6].

Here we use the Family-Cancer Database to compare familial risks in gastrointestinal cancers in offspring who have either an affected parent. Additionally, we assess the level of environmental effect by comparing cancer risks in spouses.

Results and discussion

Familial risks are shown in Table I for offspring cancer when a parent was affected by a concordant cancer; only cancer sites affecting both sexes were included if at least five familial cases were recorded. SIRs for familial risk ranged from 3.95 for esophageal cancer to 1.95 for colorectal cancer. Cancer risks among spouses showed a modest variation around SIR 1.00. The only borderline significant risk was stomach cancer in husband 1.16 whose wives were diagnosed with this cancer.

Table I.  SIR for concordant cancer in offspring and spouses.

	Cancer in offspring by parental cancer				Cancer in husband by wife's cancer				Cancer in wife by husband's cancer
Cancer site	O	SIR	95%CI		O	SIR	95%CI		O	SIR	95%CI
Esophagus	7	3.95	1.57	8.18	3	0.83	0.16	2.45	3	0.75	0.14	2.22
Stomach	59	2.05	1.56	2.65	187	1.16	1.00	1.34	187	1.11	0.95	1.28
Colorectum	524	1.95	1.79	2.13	1003	1.02	0.96	1.09	1003	1.03	0.97	1.09

The results suggest that familial clustering of cancer at gastrointestinal sites is heritable, in line with previous analysis [7], [8].

The Family-Cancer Database was created by linking registers maintained at Statistics Sweden and the Swedish Cancer Registry.