Cytochrome P450 enzymes and tumor therapy.

To the Editors: McFadyen et al. (1) have contributed a review on the aspects of cytochrome P450–related tumor therapy that is worth reading. It clearly shows the potential of such an approach for cancer therapy. Based on my own experience, I would kindly like to point out a few additional issues that may add to the excellent review. Ifosfamide and cyclophosphamide both use some more cytochrome P450 subenzymes (e.g., CYP2B1 and CYP3A4; refs. 2, 3). The problem for using both substances in solid tumors, namely in the elderly, arises from severe side effects that prevent their further clinical use despite promising response rates (e.g., in pancreatic cancer; refs. 4, 5). One approach therefore was to use the significant antitumoral potential but to spare the patient from the side effects. This can be achieved by establishing a second site of enzyme conversion at or in the tumor by either transfecting tumor cells in an experimental setting with a converting enzyme (6) or bringing cells transfected with one of the metabolizing enzymes close to the tumor (7). We have chosen the latter approach, demonstrating complete responses in the animal model and stable disease as well as some partial responses in a clinical phase I study in advanced pancreatic cancer (8). There were no side effects from the chemotherapy in these patients. As a strategy to ‘‘maximize efficacy and minimize toxicity,’’ as claimed by McFadyen et al., we have microencapsulated the genetically engineered cells to create ‘‘magic bullets’’ (9). This concept, although working, can be improved in several ways. One of them is to add the cytochrome reductase to the system as described by Jounaidi and Waxman (10). Another one would be the addition of less toxic antitumoral drugs metabolized by the P450 system such as AQ4N (11). This kind of ‘‘gene therapy’’ (gene-directed enzyme prodrug therapy) may not be very sophisticated but has the charm of simplicity and has shown proof of concept.