PO24 Presentation Time: 7:40 AM

Purpose

To quantify the impact of intraoperative MRI-guidance on implant quality for patients receiving brachytherapy (BT) for cervical cancer by calculating dose to the HR-CTV and OARs at three stages of the implant procedure.

Materials and Methods

Three patients enrolled on an IRB-approved prospective clinical trial were treated with EBRT followed by two fractions of pulsed-dose rate (PDR) brachytherapy. For each patient, one of the two PDR implants was performed in a dedicated MRI-Operating Room (MRI-OR) using 3.0 T MRI-guidance (Discovery MR750w, GE Healthcare, Waukesha, WI). An Advanced Gynecological Applicator (AGA) was used for each of the implants. Straight or straight and oblique needle trajectories were used. 3D T2-weighted MRI scans were acquired at three time points during the procedure: (1) following placement of the intracavitary components of the hybrid applicator (initial); (2) following initial needle selection and advancement (intermediate); and (3) following final modifications of implant position (final). The first scan was used to inform needle selection and estimate the appropriate advancement depth for each needle. The second scan was used to determine if modifications to the implant were necessary, including advancement of additional needles, removal of previously advanced needles, or modification of advancement depth. The final scan was used to generate a clinical plan for patient treatment. The titanium obturators used to advance the needles were replaced with C4 positive-contrast imaging markers (C4 Imaging, Houston, TX) prior to each scan to facilitate needle digitization during treatment planning. Retrospective plans were generated on the initial and intermediate scans for each of the three patients and compared to the final plans used for treatment. AGA solid applicator models were registered to the initial and intermediate MRI scans. Needles present at the time of scanning were manually digitized in the Oncentra Treatment Planning System (Elekta, Stockholm, Sweden). Clinical treatment plans were optimized to meet EMBRACE dose constraints. For a given patient, the initial and intermediate plans were optimized to escalate dose to the HR-CTV without exceeding the clinical doses delivered to the OARs. The dose-limiting structure varied based on the patient's anatomy and implant geometry. For each patient, the HR-CTV D90 and bladder, rectum, and sigmoid D2cc doses were compared between the initial, intermediate, and final plans.

Results

One needle was advanced during the first patient's implant. Activity was loaded into the tandem, ovoids, needle, and vaginal caps. The dose to the HR-CTV increased by 4.6 Gy between the initial and intermediate plans and 3.0 Gy between the intermediate and final plan used for treatment. The final dose to the HR-CTV was 10.2% greater than the initial HR-CTV dose. Five needles were advanced for the second patient. The dose to the HR-CTV increased by 3.8 Gy between the initial and intermediate implants and 0.96 Gy between the intermediate and final plan used for treatment. The final dose to the HR-CTV was 5.8% greater than the initial HR-CTV dose. Eight needles were advanced for the third patient. Dose to the HR-CTV increased by 8.3 Gy between the initial and intermediate implants and 5.7 Gy between the intermediate and final plans. The final dose to the HR-CTV was 19.7% greater than the initial HR-CTV dose. None of the OAR doses in the initial or intermediate plans exceeded the OAR doses delivered clinically.

Conclusions

Performing interstitial implants using an AGA under MRI-guidance improves implant quality, defined by dose escalation to the HR-CTV and maintenance of OAR doses.