In vivo PET monitoring of an experimental model of peritoneal carcinogenesis

Introduction: Peritoneal carcinogenesis is present in 80% of ovarian carcinoma patients at the time of diagnosis and occurs also in gastric, colon, bladder and pancreas cancer patients[1]. It is characterized by neoplastic masses and an inflammatory infiltrate in which tumour associated-macrophages are present. These cells release a vast number of factors that promote tumourigenesis, remodelling of extracellular matrix and neo-angiogenesis[2]. Pre-clinical studies of peritoneal carcinogenesis have been hampered by the inaccessibility of the peritoneal cavity. In this work we evaluated the feasibility of PET imaging to monitor non invasively the extent of peritoneal carcinogenesis in living animals[3] and consequentially the role of macrophages in the peritoneal carcinosis evaluating an active therapy on Immune System. Methods: We established a murine model of peritoneal carcinogenesis by injecting i.p. in syngeneic hosts murine adenocarcinoma cell line (TS/A). Animals were treated with clodronate encapsulated in liposomes or empty liposomes as control to deplete macrophages. For animal-PET studies, mice were injected i.v. with 116.74±7.39 mCi of [18F]FDG. PET acquisition started at 60 minutes after tracer injection and lasted for 30 minutes (6 frames of 5 minutes). For the images quantification, tumour regions of interest (ROIs) were automatically generated by thresholding 3D PET studies using ImageJ software. To investigate regional differences in tumour uptake and dispersion in the same individual through time and in comparison with animals of different groups, an automatic analysis was performed in MATLAB 7 software. Briefly, each PET acquisition was masked with the respective 3D ROIs previously generated and tumour average (± SD) uptake values, minimal, maximal values, number of pixel involved in the region were calculated. After death, peritoneal liquid was analyzed by FACS (F4/80 and CD11) in order to characterize the tumour infiltration and to assess the efficiency of the depletion. Results: PET-[18F]FDG revealed the onset and the progression of masses in the abdominal cavity. Moreover the intensity of the radioactive signal correlated with the size of lesions. Animals treated with clodronate developed less neoplastic masses whereas animals no treated showed an higher PET signal due also to inflammation. By quantifying the 3D images obtained by PET by masking tumor regions we verified that information gathered through PET scans were in keeping with the net weight of tumor lesions for the clodronate treated group. We obtained an efficiently depletion of peritoneal macrophages. Conclusions: Our results indicated PET imaging represents an accurate PET in vivo preclinical technique to visualize and monitor peritoneal carcinosis. Preliminary results suggest that clodronate might exert a positive effect on peritoneal lesions producing a significant regression of neoplastic lesions or associated inflammatory process.