Phase I trials are the building blocks of cancer therapeutics. Designed to determine the maximum tolerated dose (MTD) of a novel compound or regimen, it is through such trials that later investigations are built. Implied in this design is that the highest dose has the best chance of being the most effective dose. The inclusion criteria of these trials are generally broad, allowing for multiple tumor types and allowing for multiple prior therapies. There is no preconceived assumption as to what the “right” dose of a novel therapy is. As we move into the era of molecularly targeted treatments, we are faced with a new dilemma in the design of phase I trials. How do you incorporate these novel therapies in combination with standard chemotherapy? Does the concept of an MTD still apply, or should our goal be redefined?
In this issue, Pauer et al. present their phase I trial of the histone deacetylase inhibitor CI-994 in combination with carboplatin and paclitaxel (C + T). The purpose of the trial is to establish this regimen for eventual phase III consideration in non-small cell lung cancer. As such, the target doses of carboplatin (AUC = 6) and paclitaxel (225 mg/m2) were established at the onset.
Originally intended to be delivered as 14-day oral regimen every 21 days, CI-994 proved too toxic when given with C + T at the “standard” doses. Ultimately, the authors opted instead to reduce the duration of CI-994 to 7 days to deliver paclitaxel at 225 mg/m2, the pre-defined standard.
An issue that arises is the generalizability of the study. Certainly, Pauer and colleagues give us a starting point on which to proceed to subsequent clinical trial design using this combination. However, carboplatin and paclitaxel are a well-used regimen, and it is known that some variability in dosing exists as regards to primary tumor: 225 mg/m2 of paclitaxel may be standard treatment for lung cancer but would be considered too high in treating ovarian cancer.
This study illustrates the dilemma investigators face as we approach the evaluation of novel therapeutics. How important is the standard dosing of chemotherapy in combinaiton trials with a new targeted treatment? I would argue that trials that combine standard and novel therapies break free from the “established standard regimens.” Instead, phase I investigations incorporating a novel therapy should be based on the pharmacokinetics of the investigational compound, using scientific rationale to determine how best to incorporate a standard regimen. It may turn out that a lower dose of systemic chemotherapy is necessary when given with a targeted treatment. After all, if our “standardly dosed” treatments were so good, what need would there be for further refinement?