Flexible Regression Models for Carcinogenesis Studies

Dynamic or flexible regression models are used more and more often in carcinogenesis studies to relate lifetime distribution to the time-depending explanatory variables. In addition to the classical regression models such as Cox model, AFT model, Linear transformation model, Frailty model, etc. In the paper exposed the so-called flexible regression models which well adapted to study the cross-effects of survival functions, which are sometimes observed in clinical trials. A classical example is the well-known data concerning effects of chemotherapy (CH) and chemotherapy plus radiotherapy (CH+R) on the survival times of gastric cancers patients, see Stablein and Koutrouvelis (1985), Kleinbaum (1996), Klei and Moeschberger (1997), Wu, Hsieh and Chen (2002), Bagdonavicius, Hafdi and Nikulin (2004), etc. In this paper we give examples to illustrate possible applications of the Hsieh model (2001) and the SCE model, proposed by Bagdonavicius and Nikulin (2005), adapted to treat survival data with one crossing point. We do the comparison of both models.