Population-based family studies : genetic contribution to cancer development and survival?

Abstract

Cancer affects essentially everyone, directly or indirectly. The aim of this thesis was to study the genetic and environmental factors in cancer development and survival. Our studies were based on a record linkage between several Swedish population-based registries, principally the Multi-Generation Register, which records familial relationships, the Swedish Cancer Registry, and the Cause of Death Registry. In summary, the Swedish Family-Cancer database comprised over 11 million individuals organized into around three million families, including more than one million cancers.

In study I, we developed a generalized linear mixed model, enabling analyses of genetic and environmental effects in two- and three-generational families, considering all relationships in a family at once avoiding pairwise analyses of dependent family members. The two- and three-generational family design resulted in similar genetic and environmental estimates. In the two-generational families, no noteworthy differences were observed correcting for the unequal follow-up time in parents and children. Further, in our second study, the genetic contribution to melanoma was estimated at 18% (95% Confidence Interval [CI]=13% to 22%) in the analysis of all body sites. Contrasting the family-shared environment in sun-covered and sun-exposed body sites, the contribution was higher in covered sites, possibly conveying the benefit from cautious sunbathing on sensitive skin. The estimated childhood-shared environment for both melanoma and squamous cell carcinoma of the skin (SCC) elucidated the impact of sun habits and the avoidance of risk inflicted sunburns during infancy and youth. Moreover, in SCC, the familial shared environment at 18% (95% CI=16% to 19%) is important in defining the susceptibility to the disease. Genetic variability in individuals enhancing sensitivity to accumulated sun exposure will most probably also be involved in the aetiology. Finally, we propose that genetic factors are vital in the common liability to both melanoma and SCC. We estimated that 47% (95% CI=43% to 51%) of the susceptibility was estimated to be attributed to genetic factors.

It has been established that genetic variability influences the susceptibility to cancer; still little is known about the inheritance of cancer survival. In study III and IV, we present the first population-based comprehensive analyses of cancer survival concordance among family members. In study III, we noted a significantly increased risk of poor survival in children with poor parental survival compared to the risk in children with good parental survival in colorectal cancer (Hazard Ratio [HR]=1.44, 95% CI=1.01 to 2.01), lung cancer (HR=1.39, 95% CI=1.00 to 1.94), breast cancer (HR=1.75, 95% CI=1.13 to 2.71), ovarian cancer (HR=2.23, 95% CI=0.78 to 6.34) and prostate cancer (HR=2.07, 95% CI=1.13 to 3.79). All hazard ratio estimates, except for ovarian cancer, were statistically significant with trends of increasing risk of death among offspring by degree of survival outcome among parents. In study IV, lung cancer survival in children was associated with the lung cancer survival in their parents with a decreased hazard ratio for death in children with good parental survival (Hazard Ratio [HR]=0.71, 95% CI=0.51 to 0.99), compared to those with poor parental survival. We also found a strong protective effect (HR=0.14, 95% CI=0.030 to 0.65) for siblings, while no effect was seen on spouse survival. Genetic background of an individual may be more important than lifestyle factors such as smoking in lung cancer survival. The very strong protective effect in siblings compared to parent-child pairs further suggests a possible recessive pathway of inheritance. In light of study III and IV, we propose that genetic background is of importance in foreseeing an individual s cancer specific survival.

In conclusion, genetic factors are vital in the familial aggregation of melanoma in addition to the co-aggregation of melanoma and SCC. The ability to fight cancer disease and survive may also be inherited. In the future, I envision that population-based studies will help in identification of genetic variation influencing both the liability to cancer disease development and subsequent survival.