Phase I cancer trials methodology

The main objective of phase 1 cancer trials is to determine precisely the recommended dose of an anticancer agent as a single agent or in a context of combinations of anticancer agents (including cytotoxic agents, immunotherapy, radiotherapy...), that is administered for the first time in man, to further proceed clinical development with phase 11 and 111 trials. The recommended dose must have the greatest efficiency with acceptable toxicity. For the anticancer agents, the ratio risk/benefit is high, since toxicities associated with many cancer therapeutic agents are substantial and because the efficacy is often limited. Thus, phase I cancer trials present unique challenges in comparison to other therapeutic areas. Indeed, it is essential to minimize the numbers of patients treated at sub efficient dose levels, and in the same time not to expose the patients to unacceptable toxicity. Historically, the first method that has been used is the Fibonacci escalation. The major problems raised with this method have been the lenghts of the trials and the risk to treat substantial numbers of patients at nontherapeutix doses. Thus, novel methods have been then developed modifying the numbers of patients included at each dose level and the rapidity of dose escalation. These methods include pharmacologically guided dose escalation, escalation with overdose control and the continual reassessment method which are both statistically based dose escalation methods, and the accelerated titration designs. Concerning the targeted anticancer therapies, the therapeutic effect on the target, due to their higher specificity, can be obtained using doses that have few toxicity. Using the toxicity to determine the recommended dose for phase II trials, as it is the case for classical anticancer agents, doer not seem to be sufficient. Alternatives to determine the optimal biological dose include measurement of target inhibition, pharmacokinetic analysis and functional imaging.