Technical Note: Predicting dose distribution with replacing stopping power ratio for inter-fractional motion and intra-fractional motion during carbon ion radiotherapy with passive irradiation method for stage I lung cancer.

PURPOSE

We designed and evaluated a simple method for predicting the effects of intra-fractional and/or inter-fractional motion on dose distribution during carbon ion radiotherapy (CIRT) for solitary-lesion stage I lung cancer.

METHODS

The proposed method uses computed tomography (CT) images from treatment planning and intra-tumoral and/or inter-tumoral displacement. The predicted dose distribution (PDD) was calculated by replacing the current tumor region with the stopping power ratio (SPR) of the lung and replacing the moved tumor region with the SPR of the tumor. The actual dose distribution (ADD) was calculated without the replacement. Ten patients with solitary-lesion stage I lung cancer were retrospectively studied to evaluate the prediction method's accuracy. Four PDDs for intra-fractional motion (gate-in, exhalation, gate-out, inhalation phases during four-dimensional CT) and two PDDs for inter-fractional motion (CT images acquired 1-2 days before treatment) with bone- and tumor-matching methods were compared with each of six ADDs on each CT scan. Percentages of the planning/clinical target volumes (PTV/CTV) receiving >95% of the prescribed dose (V ) and of minimum doses covering 95% of the PTV/CTV (D ) were compared with dose volume histogram parameters.

RESULTS

The maximum tumor displacements occurred in the superior-inferior direction, with intra-fractional motion values of 3.75 and 8.97 mm for the superior and inferior directions, respectively, and inter-fractional values of 9.61 and 4.10 mm. The maximum average error for PTV V regarding intra-fractional motion was -0.43% for the gate-out phase and -0.63% for the inhalation phase. There were no significant differences for these parameters (P = 0.541, P = 0.571). Average errors for PTV and CTV V with inter-fractional motion with bone matching were 2.2% and 2.9%, respectively, with no significant differences (P = 0.387, P = 0.155).

CONCLUSIONS

The accuracy of the proposed method was good. Hence, it is feasible to use the proposed method during CIRT to predict dose distribution with respect to intra-fractional motion and/or inter-fractional motion of the tumor in patients with solitary-lesion stage I lung cancer.