An overlap-volume-histogram based method for rectal dose prediction and automated treatment planning in the external beam prostate radiotherapy following hydrogel injection.

PURPOSE: Hydrogel injected between the rectum and prostate prior to radiotherapy provides a possible means of increased dose sparing to the rectum. Here the authors evaluate the overlap volume histogram (OVH) metric as a means to predict the rectal dose following hydrogel injection. Whether OVH predicted dose can serve as the dose objective or constraint for automated treatment planning was also investigated.
METHODS: Treatment planning was performed on 21 prostate cancer patients both pre- and posthydrogel injection, with five-field IMRT delivering 78 Gy to the planning target volume (PTV). The authors quantify the geometrical relationship between the rectum and the prostate PTV using an OVH metric which determines the fractional volume of the rectum that is within a specified distance of the PTV. For an OVH distance the authors selected, L(20), the PTV expansion distance at which 20% of the rectum overlaps. The authors calculated the rectal dose, D(20), received by 20% of the rectum volume on the dose volume histogram. Linear regression was used to examine the correlation between the L(20) and D(20), and between ΔL(20) and ΔD(20) (i.e., the change of L(20) and D(20) posthydrogel injection). Additionally, rectal dose D(15), D(25), D(35), D(50), and bladder dose D(15) were predicted from the OVH (L(15), L(25), L(35), L(50), for rectum and L(15) for bladder) by the L(x)-D(x) linear regression. The predicted doses were applied to the objectives for automated treatment planning of ten plans from five patients. Automatically generated plans were compared with plans manually generated on trial-and-error basis.
RESULTS: The rectal L(20) was increased and dose D(20) decreased due to the enlarged separation of rectum caused by the hydrogel injection. Linear regression showed an inverse linear correlation between L(20) and D(20), and between ΔL(20) and ΔD(20) (r(2) = 0.77, 0.60, respectively; p < 0.0001). The increase in rectal sparing (ΔD(20)) is only weakly correlated with the volume of injected hydrogel (r(2) = 0.17; p = 0.07), indicating OVH is a more predictive indicator of rectal sparing than the volume of hydrogel itself. Application of the predicted rectum and bladder doses to automated planning produced acceptable treatment plans, with rectal dose reduced for eight of ten plans.
CONCLUSIONS: The OVH metric can predict the rectal dose in the external beam prostate radiotherapy for patients with hydrogel injection. The predicted doses can be applied to the objectives of optimization in automated treatment planning to produce acceptable treatment plans.