Devita, Hellman, and Rosenberg's Cancer

LWBK1006-16 LWW-Govindan-Review December 12, 2011 18:55

Chapter 16•Design and Analysis of Clinical Trials 191

ANSWERS

Answer 16.1. The answer is A.

Sample size may be altered during the course of a trial of any phase on

the basis of outcomes observed during the trial.

Answer 16.2. The answer is D.

Statistical power is not the focus of Phase I trial design, but the issues

(e.g., patient safety) are major, and long-term consequences (e.g., loss of

patient benefit and waste of resources) are severe if resources are incor-

rectly directed toward ineffective treatments or away from effective treat-

ments. Therefore, careful statistical design is needed if sound and sufficient

information is to be drawn from small and often heterogeneous patient

samples.

Answer 16.3. The answer is D.

The conventional 3+3 Phase I design is totally algorithm based, and

the dose recommended for Phase II trials is chosen among the fixed and

prespecified levels. The recommended dose, commonly referred to as the

maximum tolerated dose, is usually taken as the highest level for which

the dose-limiting toxicity rate is less than 33%.

Answer 16.4. The answer is C.

Traditional Phase I study design has several inherent limitations. Many

patients may receive subtherapeutic doses and it does not allow for dose

escalation in the individual patient. Furthermore, it may require longer

periods of time to complete than some of the newer trial designs and may

not be effective in determining cumulative toxicity.

Answer 16.5. The answer is A.

Phase II trials may be single-arm trials testing a single treatment, so no

direct comparison of treatments may be involved. The ideal sample size

depends on many factors, including the nature and variability of the pri-

mary outcome, the expected effect size, and the procedures to be used to

test hypotheses. An appropriate primary outcome is determined by the

nature of the treatment being tested and the goals of the study.

Answer 16.6. The answer is C.

Phase II trials are traditionally restricted to a particular tumor type and

are designed to establish the experimental drug’s activity in a broad pop-

ulation of unselected patients. The two phase study design is useful in

eliminating ineffective drugs at an early stage. With the entry of molec-

ularly targeted drugs, it is important to identify and develop assays for

predictive biomarkers. Furthermore, Phase II trials in which patient selec-

tion is driven by biomarker testing will have to be designed to identify

the target patient population.