Reply

In Response: In comparing trough levels of fenretinide between the three clinical leukoplakia trials cited in the letter of Formelli et al. in this issue of the journal (1), one must carefully take into account the timing of blood sample collections. In our previous trial of low-dose fenretinide (200 mg once a day for 3 months, with a 3-day drug holiday at the end of each month to prevent ocular toxicity), trough levels were measured in samples obtained approximately 24 hours after the last dose of fenretinide at the end of 3 months of treatment (2). The Chiesa et al. trial of low-dose fenretinide (100 mg twice a day for 12 months, with a 3-day drug holiday at the end of each month) measured trough levels in samples obtained hours after the last dose of drug at the end of month 4 or 12 (1, 3). In the current trial of high-dose fenretinide [900 mg/m² twice daily for 7 days followed by a 14-day drug holiday (a 3-week cycle) for 12 weeks], trough levels were measured in samples obtained at the end of week 12, i.e., 14 days after the final drug dose of the study (4). Trough levels of fenretinide in the trials of Lippman et al., Chiesa et al., and William et al. were 0.23, 0.50, and 0.12 μmol/L, respectively, and the relatively low trough levels in the last trial are attributable to the 14-day interval between the last fenretinide dose and blood collection. Formelli et al. are correct, however, in stating that serum levels of fenretinide were lower at the end of each 21-day cycle of the intermittent high-dose schedule than they were on a daily basis during more-continuous administration of low-dose fenretinide (200 mg/day or 100 mg/b.i.d.). Nonetheless, the rationale for delivering high-dose fenretinide was to achieve peak levels above the threshold necessary for retinoid receptor–independent apoptosis. Indeed, the high-dose schedule we utilized had been previously shown to produce peak serum fenretinide levels of 3 to 5 μmol/L (5). Formelli and colleagues' following point is well-taken: given the results of the clinical trial (which closed early per a planned early stopping rule) and the preclinical data presented by us (4), exposure to intermittent (high dose) fenretinide that washes out during a prolonged drug holiday might be inferior to exposure to the more-continuous (low dose) fenretinide regimen tested in the trials of Chiesa et al. (3) and Lippman et al. (2). Whereas we observed that the in vitro data showed greater activity of lower-concentration fenretinide in premalignant than malignant cells and of higher-concentration fenretinide in malignant than premalignant cells (4), Formelli et al. make the interesting temporal observation that continuous (3 days) exposure to fenretinide was more effective compared with 1-day exposure (1). Both observations are consistent with our negative high-dose clinical finding, although the design of the 72-hour in vitro study does not allow a true comparison of intermittent versus continuous drug exposure.