[生物书合集].Advances.in.Clinical.TBiostatisti

[生物书合集].Advances.in.Clinical.TBiostatisti

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 22 Babb and Rogatko As defined by Eichhorn and Zacks (1973), a dose sequence {xj}n=1 j is Bayesian feasible of level 1 À a if Fj (xj+1) V a, bj = 1, . . . , n À 1, where Fj the marginal posterior CDF of the MTD given Dj as defined in Eq. (12). Correspondingly, the design of a phase I clinical trial is said to be Bayesian feasible (of level 1 À a) if the posterior probability that each patient receives an overdose is no greater than the feasibility bound a. Zacks et al., (1998) showed that the dose sequence specified by Eq. (11) is consistent (i.e., under suitable conditions, the dose sequence converges in probability to the MTD) and is optimal among Bayesian feasible designs in the sense that it minimizes mG mV ðg À xk ÞIðÀl;xk Þ ðgÞC k ðg; NÞ dN dg, the expected amount by which any given patient is underdosed. Consequently, the method defined by equation (11) is referred to as the optimal Bayesian feasible design. Figure 3 Dose levels for patients 2–5 of the 5-FU trial conditional on the treatment-attributable toxicities observed. Copyright n 2004 by Marcel Dekker, Inc. All Rights Reserved. ...
View Full Document

This note was uploaded on 06/13/2011 for the course PHYSICS 101 taught by Professor Shu during the Spring '11 term at FIU.

Ask a homework question - tutors are online