Background and Aim: Sicne in many dosimetry calculations, the water and soft tissue phantoms are used, this study aimed to investigate the difference of these two phantoms with a phantom consisted of realistic liver materials in proton therapy for liver cancer.
Methods & Materials: Three phantoms with different materials of water, soft tissue and realistic liver materials were used for the study. A spherical tumor with 2 cm radius was considered in the liver. The Spread-out Bragg Peaks (SOBPs) were measured to cover the complete tumor for the three phantoms. Dose distribution and deposited dose ratio in tumor and surrounding organs were calculated using Monte Carlo N-Particle Extended (MCNPX) code. 
Ethical Considerations: The best proton energy interval to complete the coverage of tumor in the liver for phantoms with realistic and soft tissue materials was 90-120 MeV and for water phantom, it was 88-116 MeV. The shift of the Bragg peaks depth per energy in the water phantom mm relative to two other phantoms was about 4.5. The dose parameters were evaluated according to the International Commission on Radiation Units and Measurements (ICRU), and the results showed no any significant difference between them. The dose distribution in the tumor and surrounding organs showed that for all three phantoms, the dose distribution around the tumor was negligible.
Results: The use of soft tissue phantom has more acceptable results than water phantom in simulating treatment and can be replaced with realistic liver tissue. More realistic phantoms should be used in treatment plan. 
Conclusion: The use of soft tissue phantom has more acceptable results than water phantom in simulating treatment and can be replaced with realistic liver tissue. More realistic phantoms should be used in treatment plan.