Purpose: To evaluate study feasibility, toxicity, drug concentrations, and activity of escalating doses of the synthetic retinoid fenretinide [N-(4-hydroxyphenyl)retinamide (4-HPR)] in ovarian cancer by measuring serum CA125 and cytomorphometric biomarkers in cancer cells collected from ascitic fluid before and after treatment. Methods: Twenty-two naive patients with ascitic ovarian cancer were treated with escalating doses of 4-HPR at 0, 400, 600, and 800 mg/d for 1 to 4 weeks before surgery. Changes in the proportion of proliferating cells expressed by Ki67 and computer-assisted cytomorphometric variables (nuclear area, DNA index, and chromatin texture) were determined in ascitic cells. Drug levels were measured by high-performance liquid chromatography. Results: Doses up to 800 mg/d were well tolerated, and no adverse reactions occurred. There was no effect of 4-HPR on changes in serum CA125, Ki67 expression, which were assessed in 75% of subjects, and cytomorphometric variables, which were assessed in 80% of subjects. Plasma retinol levels were significantly lower in affected women than healthy donors. 4-HPR plasma concentrations increased slightly with increasing doses and attained a 1.4 Mmol/L concentration with 800 mg/d. Drug levels in malignant ascitic cells and tumor tissue were higher than in plasma but were 50 and 5 times lower, respectively, than in carcinoma cells treated in vitro with 1 Mmol/L 4-HPR. Conclusions: Cell biomarkers can be measured in ascitic cells to assess drug activity. Under our experimental conditions, 4-HPR did not show activity in advanced ovarian cancer cells. However, clinical evidence supports further investigation of fenretinide for ovarian cancer prevention. (Cancer Epidemiol Biomarkers Prev 2006;15(10):1914 -9)