The goal of this work was to develop a more practical technique to meet the requirements of the Radiation Therapy Oncology Group (RTOG) 0933 protocol for sparing the hippocampus while irradiating the whole brain. Whole-brain radiation therapy (WBRT) has been linked to a decrease in neurocognitive function and increase in memory loss leading to lower quality of life for patients. Hippocampus avoidance (HA) WBRT may help to mitigate these issues by delaying the decline in neurocognitive function. RTOG 0933 is a phase II clinical trial aimed at HA-WBRT for patients with metastatic brain lesions. Recently published results from this study showed significant benefit in preserving short-term memory function for these patients. Although the trial allows several methods of treatment, including tomotherapy, volumetric arc therapy, and linac-based IMRT, treatment at many cancer centers is limited to linear accelerator (linac) based step-and-shoot IMRT. The linac-based method suggested by RTOG involves nine gantry angles on a number of different planes. Our goal was to find a more practical solution for the planning and delivery of HA-WBRT according to the RTOG 0933 protocol.

A total of 14 patients were accrued to 0933 at our center. Patients receive magnetic resonance imaging, which is fused with the planning computed tomographic scan to aid in contouring the hippocampus. Immobilization of the patients is achieved using Mayo Mold and Thermoplast shell (CDR Systems, Calgary, AB). Treatment planning is performed using Pinnacle version 9 (Pinnacle3, Fitchburg, WI) and direct machine parameter optimization. Treatment is delivered using a Varian iX (Varian Medical Systems, Palo Alto, CA), 120-leaf multileaf collimator, 6 MV photon beams. Through an iterative process, we developed a technique using seven gantry angles and no couch rotations. The planning process was aided by the use of dose shaping rings and avoidance structures; the delivery time was further optimized by avoiding split fields and keeping the total number of monitor units and beam segments to the minimum required to meet RTOG 0933 dose constraints. The planning and dosimetry were centrally reviewed for the first five patients according to protocol requirement.

We were able to achieve our goal by developing a technique that requires only seven gantry angles and no couch movements. All of our plans met the protocol requirements and were accepted into the study. We assessed the average delivery time for four patients comparing our technique with the technique proposed by the RTOG. An average time saving of 523 seconds was achieved with the seven gantry angle technique. There was also a modest saving of 102 total monitor units with our technique.

The technique developed in our center is a practical technique for HA-WBRT. Our process gave reproducible results in the patients we accrued to RTOG 0933 and should allow many more centers to be able to deliver HA-WBRT. Should HA become a standard of care this technique is easier to adopt.