Purpose: To safely assess new drugs, cancer patients in initial cohorts of phase I oncology studies receive low drug doses. Doses are successively increased until the maximum tolerated dose (MTD) is determined. Because traditional chemotherapy is often more effective near the MTD, ethical concerns have been raised about administration of low drug doses to phase I patients. However, a substantial portion of oncology trials now investigate targeted agents, which may have different dose-response relationships than cytotoxic chemotherapies.
Experimental design: Twenty-four consecutive trials treating 683 patients between October 1, 2004, and June 30, 2008, at MD Anderson Cancer Center were analyzed. Patients were assigned to a low-dose (<or=25% MTD), medium-dose (25-75% MTD), or high-dose (>or=75% MTD) group, and groups were compared for response rate, time-to-treatment failure, progression-free survival, overall survival, and toxicity. To remove negatively biasing data from the high-dose group, in a second analysis, patients treated above the MTD were excluded (high-dose group, 75-100% MTD). Of the 683 patients, 97.7% received targeted agents.
Results: Even when excluding patients above the MTD, there was an early trend favoring the low- versus high-dose group in time-to-treatment failure, with 32.9% versus 25.2% of patients on therapy at 3 months (P = 0.08). In addition, the low-dose group fared at least as well as the other groups in all other outcomes, including response rate, progression-free survival, overall survival, and toxicity.
Conclusions: These data may help alleviate concerns that patients who receive low drug doses on contemporary phase I oncology trials fare worse and suggest targeted agents may have different dose-response relationships than cytotoxic chemotherapies.