Dosimetric differences between scheduled and adapted plans generated from ethos adaptive radiotherapy for patients with prostate cancer

Abstract

Prostate is one of the organs that can easily move or change during the treatment fractions. Adaptive radiotherapy can help to reduce the uncertainty of interfraction and decrease the side effects produced from the radiation given to the patients. The objective of this study was to provide evidence of the efficiency of adaptive radiotherapy for prostate cancer patients by investigating the dosimetric differences between the scheduled and adapted plans generated from Ethos. The treatment data of 100 fractions of prostate cancer patients who had previously been treated on Ethos daily adaptive radiotherapy were collected. The treatment data in each fraction of both scheduled and adapted plans include PTVs D95%, PTV Dmin, PTV Dmax, OAR doses of bladder and rectum, homogeneity index (HI) and frequency of plan selection were compared. The PTVs consist of PTV1, PTV2, and PTV3 which have dose prescription at 48 Gy, 57.6 Gy, and 60 Gy, respectively. PTV doses were compared in each fraction between scheduled and adapted plans. The adapted plan with 77% of all the fractions has the value of PTV3 D95% closer to the reference plan than the scheduled plan. There were 79% of fractions that the adapted plan had higher PTV3 D95% for 0.2% ± 1.2% on average than the scheduled plan in our study. The adaptation significantly pushed the average of Dmin higher, lower the average of Dmax and HI in every patient. Except for one patient that the average of Dmax and HI index were higher than the scheduled plan. For bladder dose, there were 23 fractions that had values of V60Gy exceed the threshold, adaptation could lower the values for 0.71% ± 0.57% on average (p-value <0.001). For V40.8 Gy and V48.6 Gy, the adapted plan lowered the values for 0.08% ± 0.17% and 0.11% ± 0.20%, respectively (p-value <0.001). Overall, the adapted plan had the values less than the scheduled plan and both plans produced values below bladder constraints. Adaptation could reduce rectal dose of V20, V30, V40, V50, and V60 Gy for 1.47%, 4.83%, 5.70%, 12.09%, and 12.52%, respectively. Especially for V50 Gy that the value was higher than the rectal constraint in the scheduled plan but the adapted plan could lower it to within tolerance. In conclusion, the adapted plan produced better results and less variation in PTV doses, HI, and OAR doses, where it pushed the average of Dmin higher and lower the average of Dmax, HI, and OAR doses. In comparison with the scheduled plan, the adapted plan produced PTVs D95% closer to the reference plan for 66% of fractions. This showed significant improvements by the adaptation from Ethos. However, the higher dose of adapted plan over the reference plan might lead to creating some hot areas in target volume. Thus, a careful review by oncologists is required prior to dose delivery.