according to traditional, non-Bayesian methods (e.g., the up-and-downschemes described in Storer, 1989), an empiric, data-based definition ismost often employed. Thus, the MTD is frequently taken to be the high-est dose utilized in the trial such that the percentage of patientsmanifesting DLT is equal to a specified level such as 33%. For example,patients are often treated in cohorts, usually consisting of three patients,with all patients in a cohort receiving the same dose. The dose is changedbetween successive cohorts according to a predetermined schedule typi-cally based on a so-called modified Fibonacci sequence (Von Hoff et al.,1984). The trial is terminated the first time at least some number ofpatients (generally 2 out of 6) treated at the same dose exhibit DLT. Thisdose level constitutes the MTD. The dose level recommended for phase IIevaluation of eﬃcacy is then taken to be either the MTD or one dose levelbelow the MTD (Kramar et al., 1999). Although this serves as an
This is the end of the preview.
access the rest of the document.