Approaches to phase 1 clinical trial design focused on safety, efficiency, and selected patient populations: a report from the clinical trial design task force of the national cancer institute investigational drug steering committee

Abstract

The goals and objectives of phase 1 clinical trials are changing to include further evaluation of endpoints such as molecular targeted effects, in addition to dose-toxicity profile of the investigational agent. Because of these changes in focus, the National Cancer Institute and Investigational Drug Steering Committee's Task Force on Clinical Trial Design met to evaluate the most efficient ways to design and implement early clinical trials with novel therapeutics. Clinical approaches discussed included the conventional 3 + 3 cohort expansion phase 1 design, multi-institutional phase 1 studies, accelerated titration designs, continual reassessment methods, the study of specific target patient populations, and phase 0 studies. Each of these approaches uniquely contributes to some aspect of the phase 1 study, with all focused on dose and schedule determination, patient safety, and limited patient exposure to ineffective doses of investigational agent. The benefit of labor-intensive generation of preliminary biomarker evidence of target inhibition, as well as the value of molecular profiling of the study population, is considered. New drug development is expensive and the failure rate remains high. By identifying patient populations expected to respond to the study agent and tailoring the treatment with a novel drug, investigators will be one step closer to personalizing cancer treatment. The "fail early and fast" approach is acceptable if the appropriate patient population is evaluated in the phase 1 trial. The approaches outlined in this overview address the merits, advantages, disadvantages, and obstacles encountered during first in human studies.

Figure 1

Standard phase 1 cohort expansion design used to determine dose based on toxicity rate.

Figure 2

Theoretical dose-toxicity curves for continuous reassessment method with one patient per cohort. Solid line shows the prior dose-toxicity curve from which the dose for the first patient is selected. With a desired DLT rate of 0.25, the dose level for the first patient is level 4. The dashed line shows the estimated dose-toxicity curve after observing the first patient if the first patient experienced a DLT. If the first patient experienced a DLT at level 4, then patient 2 would receive dose level 3. The dotted line shows the estimated dose-toxicity curve if the first patient did not experience at DLT at level 4; patient 2 would receive dose level 5.

Figure 3

Objectives of phase 1 clinical trials. Conventional objectives of phase 1 trials are listed in the left column and more controversial objectives of phase 1 trials are listed in the right column.

Figure 4

(Top panel) Definition of PD “response” for an individual patient. Multipliers of the baseline SD are derived from asymptotic normal distribution theory. Significance levels are 1-sided. (Bottom panel) Definition of a promising observed response rate for a dose level with a 2-stage design. The target (true) PD response rate, across patients, is 60%. Power and false positive rate are derived from the binomial distribution. Adapted with permission from Murgo, et al (41)