A Tumor-mimic Model for Evaluating the Accuracy of HIFU Preclinical Studies: An In Vivo Study

To date, the efficacy of ablative technologies such as HIFU for the treatment of liver tumors in humans has been studied in animal models without tumors or in small animals like rats and rabbits with established tumors. Because of the small size of these animals, the lesion produced by HIFU devices has to be small. Thus, the local and systemic effects of the treatment as encountered in humans cannot be studied. The purpose of this study was to use an in vivo tumor-mimic model to evaluate the accuracy of HIFU ablation in the liver in preclinical studies. Tumor mimics were created in in vivo porcine livers by injecting a 1-cc warm mixture of agarose, cellulose, glycerol and methylene blue, which formed 1-cm hyperechoic discrete lesions on sonograms. Three studies were carried out: (i) in vitro experiments were conducted to study the acoustical proprieties of the tumor mimics; (ii) in vivo experiments were conducted in 10 pigs to evaluate the tolerance of the tumor mimics when injected in the liver; (iii) ultrasound- guided HIFU ablation was performed in 10 pigs to demonstrate that it is possible to treat a predetermined zone accurately. It was shown that the acoustical properties of tumor mimics are visible in sonograms and do not modify the shape and dimensions of HIFU lesions. The local and biological tolerance of tumor mimics was excellent. In addition, it was demonstrated that the average difference between the predetermined location of the HIFU ablation and the actual coagulated area was 32%. Therefore, this tumor mimic can be used to teach HIFU ablation before starting clinical studies, especially if the ultrasound device is to be used manually, as the one presented in this study was.