The relationship between the efficacy of treatment for breast cancer and age at diagnosis of the disease has long been and still remains controversial. To assess the effect of age and other individual characteristics on survival of women diagnosed with breast cancer we propose to attack the problem with an extended arsenal of modern statistical methods of survival analysis. In doing so, we will use data on patients with breast cancer identified through the SEER Program. The foremost consideration in this population-based study will be estimation of the hazard function and cure rates for different categories of female patients with special emphasis on variations across age strata. The prognostic significance of age at detection of breast cancer will be studied using regression counterparts of nonparametric smoothing techniques and parametric procedures designed for estimation of the hazard function. These methods offer a way to interpret the effect of age and other covariates in terms of their predominant influence either on the probability of tumor cure (surviving fraction) or on the timing of failure occurrence. Such an interpretation provides a deeper insight into factors determining clinical outcome of currently practiced treatment methods.
| Hanin, L G; Yakovlev, A Y (2004) Multivariate distributions of clinical covariates at the time of cancer detection. Stat Methods Med Res 13:457-89 |
| Tsodikov, A (2002) Semi-parametric models of long- and short-term survival: an application to the analysis of breast cancer survival in Utah by age and stage. Stat Med 21:895-920 |
| Zaider, M; Zelefsky, M J; Hanin, L G et al. (2001) A survival model for fractionated radiotherapy with an application to prostate cancer. Phys Med Biol 46:2745-58 |
| Hanin, L G; Zaider, M; Yakovlev, A Y (2001) Distribution of the number of clonogens surviving fractionated radiotherapy: a long-standing problem revisited. Int J Radiat Biol 77:205-13 |
