The challenge in developing Phase II/III disease-directed studies of FTIs arises from the difficulty in selection ofappropriate endpoints fortheir evaluation. Traditionally, the goal of Phase II trials have been the determination of efficacy, on the basis of objective tumor responses as described using standard criteria (e.g., World Health Organization), although it is clear that response rates are at best a surrogate for efficacy. The screening of compounds in Phase II studies using objective tumor response has evolved from the subsequent demonstration of efficacy as measured by prolonged survival in pivotal Phase III trials (65). Although FTIs have induced regression of tumors in some animal models, the principal therapeutic effect clinically may be tumor growth inhibition or "cytostasis," rather than an appreciable cytoreductive response, which would be anticipated for a conventional chemotherapeutic agent. Therefore, a developmental plan that provides for a clinical trial situation with adequate sensitivity to detect and measure tumor growth inhibition will be essential for disease-directed evaluations. Although experimental evidence exists indicating that FTIs may inhibit the growth of tumors with and without ras mutations, Phase III and earlier, exploratory (Phase II) evaluations may have the greatest likelihood of detecting meaningful clinical activity if the studies are performed in tumor types that are highly likely to have ras mutations. Following rigorous "proof of principle" trials, the scope of disease-directed evaluations can be broadened, and patient eligibility requirements can be less restrictive. The principal endpoints of such pivotal trials ought to be median survival, percentage of patients alive at relevant intervals, time to progression, clinical benefit (e.g., performance status, weight loss, pain control), and improvement in quality of life.
Practically, however, some type of "lead" or indication that the FTIs possess relevant clinical activity, with the ability to modify the natural history ofdisease progression will ultimately need to be observed before resource-intensive Phase III studies are commenced. One such way ofobtaining a "lead" prior to launching large randomized studies is to measure the relative time to tumor progression of patients receiving single-agent treatment with the FTI (period B) against that produced by treatment with a relevant standard therapy or supportive care, measured just before administration ofthe experimental agent (period A). On the basis of experience with agents that were later shown to have relevant clinical activity in randomized trials (i.e., 30% of patients had a longer time to progression on the new agent in earlier, single-arm studies than on the agent that they received prior to receiving the new agent), a 30% prolongation in the time to progression may be a reasonable threshold to use before proceeding to Phase III studies. Alternatively, "exploratory" single-arm or randomized Phase II studies that are designed with sufficient power to detect and quantify the relevant indices of tumor growth inhibition may provide meaningful leads about activity before randomized evaluations (66). For example, in advanced pancreatic cancer, the percentage of patients surviving for at least 1 yr in exploratory nonrandomized studies may be considered a reasonable endpoint to use in gauging whether or not to proceed with randomized Phase III evaluations (66, 67). Considering the results of Phase II and III studies of gemcitabine in patients with advanced pancreatic cancer, an FTI demonstrating a 1-yr survival rate with a lower limit of a 95% confidence interval of at least 20% might be viewed encouragingly as a candidate for Phase III development. Similarly, the proportion of patients having progressive disease as their best response appears to relate inversely to the ultimate utility of any particular agent in a specific clinical setting, and a maximum acceptable threshold ofpatients with progressive disease as their best response may be used to predict the potential usefulness ofthe agent. A retrospective analysis ofThe National Cancer Institute ofCanada, Clinical Trials Group Phase II studies of new agents indicates that the rates of disease progression of agents felt to be most promising in breast, lung carcinomas, and gliomas appear to be less than 20,30, and 40%, respectively (65). The use ofsuch thresholds, once validated, may be useful in screening FTIs before undertaking large randomized Phase III trials. Other surrogate endpoints which may be considered for efficacy in Phase II trials include assessment oftarget inhibition, the use of PET scanning to evaluate changes in tumor metabolism and, following changes in tumor markers. Although all of these potential end points remain intriguing possibilities for future Phase II trials, no proposed alternative endpoints have in fact been validated, and thus the challenge will be to integrate them successfully as we search for new paradigms for evaluating novel cytostatic agents (65).
Was this article helpful?