Genetic Susceptibility and Survival: Application to Breast Cancer

Abstract

Inherited mutations of the BRCA1 and BRCA2 genes are known to confer an elevated risk of both breast and ovarian cancers. The effect of carrying such a mutation on survival after developing breast or ovarian cancer is less well understood. We investigate the relationship between BRCA1 and BRCA2 carrier status and survival after breast cancer. We obtained data from the Cancer and Steroid Hormone Study, a large population-based study including more than 4,000 breast cancer cases. Patient data include extensive information about breast and ovarian cancer in relatives. We obtained follow-up information about patients via record linkage with the Surveillance, Epidemiology, and End Results registry, with maximum follow-up of 15 years. In the absence of genetic testing for each individual, presence or absence of mutation at a breast cancer susceptibility gene is captured by a pair of binary latent variables whose marginal probability depends on the patient's family history of breast and ovarian cancer. We estimate the effect of genotype on survival using a Cox proportional hazards model, treating genetic status as a latent variable and controlling for stage at diagnosis, histology, whether radiation treatment was administered, the individual's smoking history, body mass, race, and age at diagnosis. Inference is accomplished using a Markov chain Monte Carlo algorithm to draw a sample from the posterior distribution of model parameters accounting for sampling error, uncertainty in the genotype of study participants, and uncertainty in estimates of genetic parameters. An analysis that does not discriminate between BRCA1 and BRCA2 estimates the genetic effect on survival after breast cancer for women carrying a mutation at either site. We find evidence that survival of nonirradiated mutation carriers is better than that of noncarriers; we estimate the probability of improved survival to be .990. As a byproduct of this analysis, we estimate that a study based on 901 women with known mutation status would yield estimates of genetic effect on survival with comparable uncertainty to that obtained in the combined-gene analysis. An analysis assessing the separate effects of the two genes indicates that carriers of either mutation may have an improved prognosis. We estimate the probability of improved survival among nonirradiated BRCA1 carriers to be .844, and that among nonirradiated BRCA2 carriers to be .924. However, substantial uncertainty remains about the magnitude of the BRCA2 effect.

Key Words:

• Breast cancer susceptibility genes
• Latent variables
• Markov chain Monte Carlo
• Survival analysis.