This study was designed to investigate the efficacy of photodynamic therapy (PDT) in treating colonic cancer in a preclinical study. Photofrin, a porphyrin mixture, and pheophorbide a (Ph a), a bacteriochlorin, were tested on HT29 human colonic tumor cells in culture and xenografted into athymic mice. Their pharmacokinetics were investigated in vitro, and the PDT efficacy at increasing concentrations was determined with proliferative, cytotoxic and apoptotic assessments. The in vivo distribution and pharmacokinetics of these dyes (30 mg/kg, intraperitoneal) were investigated on HT29 tumor-bearing nude mice. The inhibition of tumor growth after a single 100 J/cm2 PDT session was measured by the changes in tumor volume and by histological analysis of tumor necrosis. PDT inhibited HT29 cell growth in culture. The cell photodamage occurred since the time the concentrations of Ph a and Photofrin reached 5.10(-7) M (or 0.3 microg/mL) and 10 microg/mL, respectively. A photosensitizer dose-dependent DNA fragmentation was observed linked to a cleavage of poly(ADP-ribose) polymerase and associated with an increased expression of mutant-type p53 protein. PDT induced a 3-week delay in tumor growth in vivo. The tumor injury was corroborated by histological observation of necrosis 48 h after treatment, with a correlated loss of specific enzyme expression in most of the tumor cells. In conclusion, PDT has the ability to destroy human colonic tumor cells in vitro and in vivo. This tumoricidal effect is likely associated with a p53-independent apoptosis, as HT29 cells express only mutated p53. The current study suggests a preferential use of Photofrin in PDT of colonic cancer because it should be more effective in vivo than Ph a as a consequence of better tumor uptake.