Encrypted login | home

Program Information

Comparison of Radiobiological Models in Predicting TCP and NTCP for IMRT and VMAT Treatments

no image available
V Subramani

V Subramani1,2*, M Radhakrishnan2 , A Nambiraj2 , M Manavalan1 , S Chitra1 , M Venkataraman1 , (1) Apollo Cancer Institutes, Chennai, Tamil Nadu, (2) VIT University, Vellore, Tamil Nadu


SU-I-GPD-T-248 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall

Purpose: To compare the Tumour Control Probability (TCP) and Normal Tissue Complication Probability (NTCP) predicted using Poisson, Lyman-Kutcher-Burman (LKB) and Niemierko’s EUD based radiobiological models for Intensity Modulated Radiation Therapy (IMRT) and Volumetric Modulated Arc Therapy (VMAT) treatments.

Methods: A total of 32 patients are chosen from wide range of treatment sites which including brain, head & neck, thorax (Ca.Oesophagus), abdomen (Ca.Pancreas) and pelvis (Ca.Cervix, Ca.Rectum & Ca.Prostate) to study the sensitivity and effectiveness of radiobiological models in their TCP and NTCP prediction. For each patient IMRT and VMAT plans are optimized on set criterias. Dosimetric comparisons are made between the generated IMRT and VMAT treatment plans to quantify differences in dose coverage of the target and sparing of organs-at-risk (OARs). A matlab program code and a MS Excel file which accepts cumulative and differential DVH respectively are employed for calculations. The tumour-specific and tissue-specific parameters used in TCP and NTCP calculations respectively are taken from literature.

Results: The degrees of dose conformity and homogeneity of VMAT are better than IMRT plans. Paired t-test concludes that there exist statistical differences in the prediction of TCP between the Poisson and Niemierko’s EUD model for all treatment plans and techniques except Ca.Prostate case. The Poisson model overestimates TCP than EUD based model in brain and pelvis cases. In IMRT and VMAT both the LKB and EUD model calculates same NTCP for lungs, brainstem, opticnerve, lens, spinalcord, parotids and femoral heads among which lungs exhibit maximum NTCP because of its relatively low TD50/5. The LKB model predicted minimal NTCP for bladder, rectum, kidney, liver and heart but Niemierko’s EUD based model failed to predict the same.

Conclusion: The radiobiological models are used in calculating the TCP and NTCP for IMRT and VMAT treatment plans and their clinical correlation needs to be validated.

Contact Email: