Investigation of pulsed IMRT and VMAT for re-irradiation treatments: Dosimetric and delivery feasibilities

Many tumor cells demonstrate hyperradiosensitivity at doses below ∼50 cGy. Together with the increased normal tissue repair under low dose rate, the pulsed low dose rate radiotherapy (PLDR), which separates a daily fractional dose of 200 cGy into 10 pulses with 3 min interval between pulses (∼20 cGy/pulse and effective dose rate 6.7 cGy min^-1), potentially reduces late normal tissue toxicity while still providing significant tumor control for re-irradiation treatments. This work investigates the dosimetric and technical feasibilities of intensity modulated radiotherapy (IMRT) and volumetric modulated arc therapy (VMAT)-based PLDR treatments using Varian Linacs. Twenty one cases (12 real re-irradiation cases) including treatment sites of pancreas, prostate, pelvis, lung, head-and-neck, and breast were recruited for this study. The lowest machine operation dose rate (100 MU min^-1) was employed in the plan delivery. Ten-field step-and-shoot IMRT and dual-arc VMAT plans were generated using the Eclipse TPS with routine planning strategies. The dual-arc plans were delivered five times to achieve a 200 cGy daily dose (∼20 cGy arc^-1). The resulting plan quality was evaluated according to the heterogeneity and conformity indexes (HI and CI) of the planning target volume (PTV). The dosimetric feasibility of retaining the hyperradiosensitivity for PLDR was assessed based on the minimum and maximum dose in the target volume from each pulse. The delivery accuracy of VMAT and IMRT at the 100 MU min ^-1 machine operation dose rate was verified using a 2D diode array and ion chamber measurements. The delivery reproducibility was further investigated by analyzing the Dynalog files of repeated deliveries. A comparable plan quality was achieved by the IMRT (CI 1.10-1.38; HI 1.04-1.10) and the VMAT (CI 1.08-1.26; HI 1.05-1.10) techniques. The minimum/maximum PTV dose per pulse is 7.9 ± 5.1 cGy/33.7 ± 6.9 cGy for the IMRT and 12.3 ± 4.1 cGy/29.2 ± 4.7 cGy for the VMAT. Six out of the 186 IMRT pulses (fields) were found to exceed 50 cGy maximum PTV dose per pulse while the maximum PTV dose per pulse was within 40 cGy for all the VMAT pulses (arcs). However, for VMAT plans, the dosimetric quality of the entire treatment plan was less superior for the breast cases and large irregular targets. The gamma passing rates for both techniques at the 100 MU min^-1 dose rate were at least 94.1% (3%/3 mm) and the point dose measurements agreed with the planned values to within 2.2%. The average root mean square error of the leaf position was 0.93 ± 0.83 mm for IMRT and 0.53 ± 0.48 mm for VMAT based on the Dynalog file analysis. The RMS error of the leaf position was nearly identical for the repeated deliveries of the same plans. In general, both techniques are feasible for PLDR treatments. VMAT was more advantageous for PLDR with more uniform target dose per pulse, especially for centrally located tumors. However, for large, irregular and/or peripheral tumors, IMRT could produce more favorable PLDR plans. By taking the biological benefit of PLDR delivery and the dosimetric benefit of IMRT and VMAT, the proposed methods have a great potential for those previously-irradiated recurrent patients.