Cervical Brachytherapy Research Articles

GSOR09 Presentation Time: 5:40 PM: AI Based Sentiment Analysis of Online Discussions Related to Cervical Brachytherapy

GSOR09 Presentation Time: 5:40 PM: AI Based Sentiment Analysis of Online Discussions Related to Cervical Brachytherapy

GSOR05 Presentation Time: 11:50 AM: Improving Plan Quality in Cervical Brachytherapy Using Simplified Knowledge-Based Planning for Direction Modulated Brachytherapy Tandem Applicator

GSOR05 Presentation Time: 11:50 AM: Improving Plan Quality in Cervical Brachytherapy Using Simplified Knowledge-Based Planning for Direction Modulated Brachytherapy Tandem Applicator

PO0223: How Hot is Too Hot? Assessing the Role of the V200 and Other Dosimetric Parameters in the Evolution of Complications Following HDR Cervical Brachytherapy

PO0223: How Hot is Too Hot? Assessing the Role of the V200 and Other Dosimetric Parameters in the Evolution of Complications Following HDR Cervical Brachytherapy

PO0222: Feasibility of 3-D Printed Prototypes of Indwelling Intrauterine Tube for Tandem-Based Cervical Brachytherapy

PO0222: Feasibility of 3-D Printed Prototypes of Indwelling Intrauterine Tube for Tandem-Based Cervical Brachytherapy

1574: Deep learning nnU-Net for auto-contouring in cervical brachytherapy, an African country’s experience

1574: Deep learning nnU-Net for auto-contouring in cervical brachytherapy, an African country’s experience

660: Improving cervical brachytherapy by better defining the contouring required for the rectum and bowel

660: Improving cervical brachytherapy by better defining the contouring required for the rectum and bowel

Neural network dose prediction for cervical brachytherapy: Overcoming data scarcity for applicator-specific models.

3D neural network dose predictions are useful for automating brachytherapy (BT) treatment planning for cervical cancer. Cervical BT can be delivered with numerous applicators, which necessitates developing models that generalize to multiple applicator types. The variability and scarcity of data for any given applicator type poses challenges for deep learning. The goal of this work was to compare three methods of neural network training-a single model trained on all applicator data, fine-tuning the combined model to each applicator, and individual (IDV) applicator models-to determine the optimal method for dose prediction. Models were produced for four applicator types-tandem-and-ovoid (T&O), T&O with 1-7 needles (T&ON), tandem-and-ring (T&R) and T&R with 1-4 needles (T&RN). First, the combined model was trained on 859 treatment plans from 266 cervical cancer patients treated from 2010 onwards. The train/validation/test split was 70%/16%/14%, with approximately 49%/10%/19%/22% T&O/T&ON/T&R/T&RN in each dataset. Inputs included four channels for anatomical masks (high-risk clinical target volume [HRCTV], bladder, rectum, and sigmoid), a mask indicating dwell position locations, and applicator channels for each applicator component. Applicator channels were created by mapping the 3D dose for a single dwell position to each dwell position and summing over each applicator component with uniform dwell time weighting. A 3D Cascade U-Net, which consists of two U-Nets in sequence, and mean squared error loss function were used. The combined model was then fine-tuned to produce four applicator-specific models by freezing the first U-Net and encoding layers of the second and resuming training on applicator-specific data. Finally, four IDV models were trained using only data from each applicator type. Performance of these three model types was compared using the following metrics for the test set: mean error (ME, representing model bias) and mean absolute error (MAE) over all dose voxels and ME of clinical metrics (HRCTV D90% and D2cc of bladder, rectum, and sigmoid), averaged over all patients. A positive ME indicates the clinical dose was higher than predicted. 3D global gamma analysis with the prescription dose as reference value was performed. Dice similarity coefficients (DSC) were computed for each isodose volume. Fine-tuned and combined models showed better performance than IDV applicator training. Fine-tuning resulted in modest improvements in about half the metrics, compared to the combined model, while the remainder were mostly unchanged. Fine-tuned MAE=3.98%/2.69%/5.36%/3.80% for T&O/T&R/T&ON/T&RN, and ME over all voxels=-0.08%/-0.89%/-0.59%/1.42%. ME D2cc were bladder=-0.77%/1.00%/-0.66%/-1.53%, rectum=1.11%/-0.22%/-0.29%/-3.37%, sigmoid=-0.47%/-0.06%/-2.37%/-1.40%, and ME D90=2.6%/-4.4%/4.8%/0.0%. Gamma pass rates (3%/3mm) were 86%/91%/83%/89%. Mean DSCs were 0.92%/0.92%/0.88%/0.91% for isodoses ≤150% of prescription. 3D BT dose was accurately predicted for all applicator types, as indicated by the low MAE and MEs, high gamma scores and high DSCs. Training on all treatment data overcomes challenges with data scarcity in each applicator type, resulting in superior performance than can be achieved by training on IDV applicators alone. This could presumably be explained by the fact that the larger, more diverse dataset allows the neural network to learn underlying trends and characteristics in dose that are common to all treatment applicators. Accurate, applicator-specific dose predictions could enable automated, knowledge-based planning for any cervical brachytherapy treatment.

Generalizability of deep learning in organ-at-risk segmentation: A transfer learning study in cervical brachytherapy

PurposeDeep learning can automate delineation in radiation therapy, reducing time and variability. Yet, its efficacy varies across different institutions, scanners, or settings, emphasizing the need for adaptable and robust models in clinical environments. Our study demonstrates the effectiveness of the transfer learning (TL) approach in enhancing the generalizability of deep learning models for auto-segmentation of organs-at-risk (OARs) in cervical brachytherapy. MethodsA pre-trained model was developed using 120 scans with ring and tandem applicator on a 3T magnetic resonance (MR) scanner (RT3). Four OARs were segmented and evaluated. Segmentation performance was evaluated by Volumetric Dice Similarity Coefficient (vDSC), 95 % Hausdorff Distance (HD95), surface DSC, and Added Path Length (APL). The model was fine-tuned on three out-of-distribution target groups. Pre- and post-TL outcomes, and influence of number of fine-tuning scans, were compared. A model trained with one group (Single) and a model trained with all four groups (Mixed) were evaluated on both seen and unseen data distributions. ResultsTL enhanced segmentation accuracy across target groups, matching the pre-trained model’s performance. The first five fine-tuning scans led to the most noticeable improvements, with performance plateauing with more data. TL outperformed training-from-scratch given the same training data. The Mixed model performed similarly to the Single model on RT3 scans but demonstrated superior performance on unseen data. ConclusionsTL can improve a model’s generalizability for OAR segmentation in MR-guided cervical brachytherapy, requiring less fine-tuning data and reduced training time. These results provide a foundation for developing adaptable models to accommodate clinical settings.

Dosimetric Comparison of Inverse Planning Simulated Annealing and Manual Optimization for Intracavitary Cervix Brachytherapy

Background/Aims: Different optimization methods in brachytherapy treatment planning is used. The aim of this study is to evaluate dosimetric differences between manual optimization (MO) and inverse planning simulated annealing (IPSA) planning techniques commonly used in brachytherapy of cervical cancer. Methods: Fifteen cervical cancer patients were included in this study. Nucletron standard tandem-ovoid (TO) applicators were used for treatment. High-risk clinical tumor volume (HR-CTV), bladder, rectum, and sigmoid contouring were performed according to GEC-ESTRO recommendations. Two plans were created for each patient using IPSA and MO techniques. While a dose of 700 cGy was prescribed to the target volume during the planning phase, an effort was made to protect the organs at risk in the best way possible. IPSA and MO planning techniques were compared via dose volume histogram (DVH). Results: There was no significant difference between HR-CTV and CI values for MO and IPSA techniques. There was a significant difference between IPSA and MO techniques for the 2cm3 volume of the rectum (p= 0.002). It was observed that the bladder was better protected by the IPSA technique. There was a 6.26% dose difference between IPSA and MO for the bladder. A significant difference was found between IPSA and MO techniques for the 2cm3 volume of the sigmoid (p= 0.002). The IPSA technique was superior to the MO technique in terms of time. Conclusions: The IPSA technique was superior to the MO technique in terms of protecting organs at risk (OARs). IPSA provides a faster and higher quality plan in cervical brachytherapy.

Safe dose escalation and reduction of the fraction number of uterine cervical brachytherapy using a gel spacer in the rectovaginal and vesicouterine septum: A planning study

PURPOSETo evaluate the possibility of dose escalation and reduction of fraction number in cervical brachytherapy using a gel spacer. MATERIAL AND METHODSTwenty patients with uterine cervical cancer treated with image-guided adaptive brachytherapy (IGABT) were selected. Hyaluronic acid gel injection (HGI) was performed in the rectovaginal and vesicouterine septum for 10 patients. The other ten patients were not with HGI. Both groups were treated with IGABT involving tandem/ovoid or cylindrical applicators along with additional interstitial needles. Dose distributions approved by radiation oncologists were retrospectively analyzed, and a dose summation of 45 Gy/25 of external beam radiation therapy and IGABT was performed. Dose constraints for D2cc of bladder, rectum, and sigmoid were 80, 70, and 70 Gy, respectively. Equivalent dose in 2-Gy fractions calculations used α/β = 10 Gy for high-risk clinical target volume (CTVHR) D90 and α/β = 3 Gy for organs at risks (OARs). As a planning study, dose distribution rescaling was conducted to deliver as much dose to CTVHR D90 as possible within the dose constraint limitation for OARs when IGABT was performed for four, three, and two fractions in both groups. RESULTSThe median CTVHR D90 was >80 Gy in the non-HGI group and >85 Gy in the HGI group for virtual two and three fractions. Rectum D2cc was significantly lower in the HGI group for three fractions (p < 0.01). CONCLUSIONSIn the HGI group, adequate dose delivery to CTVHR could be achieved with a reduced IGABT fraction number while meeting the dose constraints of OARs.

Feasibility of Using a Single Planning MRI for Two-Insertion Intracavitary Brachytherapy for Cervical Cancer

MRI is an essential tool in treatment planning for cervical cancer patients undergoing definitive chemoradiation involving brachytherapy. However, due access to MRI for many patients may be limited. We sought to examine the feasibility and quality of treatment planning for organs at risk (OAR) and target volumes of a single treatment planning MRI for a planned two-insertion HDR intracavitary cervical brachytherapy. Eighteen patients with cervical cancer who underwent MRI-based treatment planning for intracavitary brachytherapy were selected for analysis. Patients underwent two intracavitary applicator insertions using a tandem and ring applicator (TnR_1 and TnR_2) with both CT (for digitization only) and MRI following each insertion and two fractions delivered with each insertion (total 7 Gy x 4fx). MRI-based TnR_2 plans were retrospectively replanned with the following procedure: 1) register MRI from TnR_1 to TnR_2 CT, 2) transfer cervix contour (Cx_simul) and plan from TnR_1 to TnR_2 CT, simulating blinding from TnR_2 MRI, 3) crop Cx_simul from TnR_2 OARs, 4) adjust plan only to meet OAR constraints or clinical doses if clinical plan (TnR_2 original MRI plan summed with TnR_1 and EBRT doses) violates constraints. Coverage changes (∆_D90%) in actual MRI defined TnR_2 cervix (Cx_real) was compared to clinical plans and evaluated against our clinical target goal: >85 Gy EQD2 (EBRT+BT). Differences between Cx_simul and Cx_real contours were evaluated with Dice Similarity Coefficient (DSC). Changes in bladder, rectum, rectosigmoid, and bowel D2cc was evaluated. Procedure time was recorded to determine time added by MRI (time between CT and final MRI timestamps; n = 29). Plan modification to meet OAR constraints was required for 5/18 plans. Cx_real target coverage from clinical to simulated plans decreased in 7/18 plans (∆_D90% = 1.7±6%). Composite EQD2 was <85 Gy in 2/18 cases, with ∆_D90% of -1.3% (clinical EQD2<85 Gy for this case also) and -7.3%, which was due to inconsistent cervix shape between insertions 1 and 2. Cervix DSC values between Cx_simul and Cx_real ranged from 0.68-0.88 (mean = 0.78 ± 0.06). Bladder, rectum, rectosigmoid, and bowel D2cc increased on average 4 ± 12%, 3 ± 9%, 5 ± 10%, and 5 ± 17%, respectively. Including planning MRI increased procedure time by 41.8 ± 10.4 min. The proposed method of utilizing TnR_1 MRI contours for TnR_2 CT planning reduces procedure time and provides reasonable target coverage for most cases. OAR assessment varies between imaging modality which likely represents variability in OAR position rather than treatment planning modality. Next steps are to increase patients studied, and implement this technique prospectively while waiting for TnR_2 MRI completion and compare the clinical TnR_2 MRI plan with the TnR_2 CT plan created with Cx_simul. This will provide further confidence in validity of the approach and ensure plan quality is maintained prior to full implementation.

Comparison of three analgesic regimens in women undergoing cervical brachytherapy

PURPOSEEffective periprocedural analgesia is an important aspect of cervical brachytherapy delivery, with implications for patient comfort and attendance for subsequent fractions. We compared the efficacy and safety of three analgesic modalities: intravenous patient-controlled analgesia (IV-PCA), continuous epidural infusion (CEI) and programmed-intermittent epidural bolus with patient-controlled epidural analgesia (PIEB-PCEA). METHODS AND MATERIALSNinety-seven brachytherapy episodes involving 36 patients between July 2016 and June 2019 in a single tertiary center were retrospectively reviewed. Episodes were divided into two key phases: Phase 1 (while applicator remained in situ) and Phase 2 (following applicator removal until discharge or 4 h). For the primary endpoint, pain scores were retrieved and analyzed by analgesic modality with respect to median score and an internally defined “unacceptable” pain experience (>20% of scores being ≥4/10; i.e., moderate or greater). Total nonepidural oral morphine equivalent dose (OMED) and toxicity/complication events were reported as secondary endpoints. RESULTSIn Phase 1, there was a significantly higher median pain score (p < 0.001) and more episodes with unacceptable pain scores (46%) in the IV-PCA group compared with either epidural modality (6–14%; p < 0.001).In Phase 2, we observed a greater median pain score (p = 0.007) and higher proportion of patient episodes with unacceptable pain scores (38%) in the CEI group compared with both the IV-PCA (13%) and PIEB-PCEA (14%) groups (p = 0.001).There was a significant difference in median OMED used throughout all phases across the PIEB-PCEA (0 mg), IV-PCA (70 mg), and CEI (15 mg) groups (p < 0.001). CONCLUSIONSPIEB-PCEA is safe and offers superior analgesia compared to IV-PCA or CEI for pain control after applicator placement in cervical brachytherapy.

Improving plan quality in cervical brachytherapy using a simple knowledge-based prediction tool for OAR dose (D2cm3)

PurposeToxicity from cervical brachytherapy has been demonstrated to correlate with the D2cm3 of the bladder, rectum, and bowel. This suggests a simplified version of knowledge-based planning investigating the relationship of the overlap distance for 2cm3 and the D2cm3 from planning may be possible. This work demonstrates the feasibility of simple knowledge-based planning to predict the D2cm3, detect suboptimal plans, and improve plan quality. Methods and materialsThe overlap volume histogram (OVH) method was used to determine the distance for 2cm3 of overlap between the OAR and CTV_HR. Linear plots modeled the OAR D2cm3 and 2cm3 overlap distance. Two datasets of 20 patients (plans from 43 insertions in each dataset) were used to create two independent models, and the performance of each model was compared using cross-validation. Doses were scaled to ensure consistent CTV_HR D90 values. The predicted D2cm3 is entered as the maximum constraint in the inverse planning algorithm. ResultsMean bladder D2cm3 decreased by 2.9% for the models from each dataset, mean rectal D2cm3 decreased 14.9% for the model from dataset 1 and 6.0% for the model from dataset 2, mean sigmoid D2cm3 decreased 10.7% for the model from dataset 1 and 6.1% for the model from dataset 2, mean bowel D2cm3 decreased 4.1% for the model from dataset 1 but no statistically significant difference was observed for the model from dataset 2. ConclusionsA simplified knowledge-based planning method was used to predict D2cm3 and was able to automate optimization of brachytherapy plans for locally advanced cervical cancer.

PO-2142 Dosimetric uncertainties due to inter-fraction organ at risk movement in HDR cervical brachytherapy.

PO-2142 Dosimetric uncertainties due to inter-fraction organ at risk movement in HDR cervical brachytherapy.

PD-0408 Effects of simulated systematic applicator displacement in intra-cavitary HDR cervical brachytherapy

PD-0408 Effects of simulated systematic applicator displacement in intra-cavitary HDR cervical brachytherapy

Self-configuring nnU-Net for automatic delineation of the organs at risk and target in high-dose rate cervical brachytherapy, a low/middle-income country's experience.

The high-dose rate (HDR) brachytherapy treatment planning workflow for cervical cancer is a labor-intensive, time-consuming, and expertise-driven process. These issues are amplified in low/middle-income countries with large deficits in experienced healthcare professionals. Automation has the ability to substantially reduce bottlenecks in the planning process but often require a high level of expertise to develop. To implement the out of the box self-configuring nnU-Net package for the auto-segmentation of the organs at risk (OARs) and high-risk CTV (HR CTV) for Ring-Tandem (R-T) HDR cervical brachytherapy treatment planning. The computed tomography (CT) scans of 100 previously treated patients were used to train and test three different nnU-Net configurations (2D, 3DFR, and 3DCasc). The performance of the models was evaluated by calculating the Sørensen-dice similarity coefficient, Hausdorff distance (HD), 95th percentile Hausdorff distance, mean surface distance (MSD), and precision score for 20 test patients. The dosimetric accuracy between the manual and predicted contours was assessed by looking at the various dose volume histogram (DVH) parameters and volume differences. Three different radiation oncologists (ROs) scored the predicted bladder, rectum, and HR CTV contours generated by the best performing model. The manual contouring, prediction, and editing times were recorded. The mean DSC, HD, HD95, MSD and precision scores for our best performing model (3DFR) were 0.92/7.5mm/3.0mm/ 0.8mm/0.91 for the bladder, 0.84/13.8mm/5.3mm/1.4mm/0.84 for the rectum, and 0.81/8.5mm/6.0mm/2.2mm/0.80 for the HR CTV. Mean dose differences (D2cc/90% ) and volume differences were 0.08Gy/1.3 cm3 for the bladder, 0.02Gy/0.7 cm3 for the rectum, and 0.33Gy/1.5 cm3 for the HR CTV. On average, 65% of the generated contours were clinically acceptable, 33% requiring minor edits, 2% required major edits, and no contours were rejected. Average manual contouring time was 14.0 min, while the average prediction and editing times were 1.6 and 2.1 min, respectively. Our best performing model (3DFR) provided fast accurate auto generated OARs and HR CTV contours with a large clinical acceptance rate.

GSOR11 Presentation Time: 9:55 AM: 3D Printing Enhanced GYN Brachytherapy: From Novice to Expert in One Print with the TARGIT-FX

<h3>Purpose</h3> It is well established in cervical brachytherapy that the addition of interstitial (IS) needles to intra-cavitary (IC) applicators can significantly enhance dosimetry by improving target coverage without increasing normal tissues doses. Accurate placement of interstitial needles requires significant technical skill and imaging guidance proficiency, thereby limiting the benefits to a subset of practitioners and patients. Available commercial combined IC/IS applicators restrict the available needle positions to often unsatisfactory locations for optimized dosimetry. We have developed and further improved our supplemental templates that attach to the tandem applicator and guides needles to optimized positions for different tumor topologies. This new TARGIT-FX (Tandem Anchored Radially Guiding Interstitial Template - Flexible eXtended) is 3D printed from biologically compatible materials, can be steam sterilized and provides multiple degrees of flexibility to simplify and facilitate interstitial needle insertion. <h3>Materials/Methods</h3> Our previously described IS needle guide (TARGIT), which facilitates placement of complementary needles in a tandem and ovoid implant, was redesigned with the purported goal of added flexibility and ease of use. The tandem attached small template was extended all the way to the vaginal introitus to allow needle insertion from outside the patient. The number of needle position options increased from four to nine to better cover a wider range of clinical situations: three paired width options ((2.0, 1.5 and 1.25 cm width) at the level of point A and three individual posterior locations (figure, insert E). The new design has been used clinically for 14 different patient implants During the procedure, the tandem is first inserted (Figure 1A), the TARGIT-FX is then slid over in position and secured over the tandem (Fig. 1B-C); after the ovoids are placed, needles can be inserted from outside the patient in the selected locations as clinically required (Fig. 1 D). Figure 1E show all the needle potential positions. The procedure time and relevant dosimetric parameters were tracked, as well as usability. <h3>Results</h3> The procedure time with the TARGIT-FX was decreased to 24 min from a benchmark of 29 min with the older TARGIT design (average over 68 implants). The V100% remained similar with 92.2% as compared to the 90.2% TARGIT benchmark, while organs at risk remained within constraints. The main goals of flexibility were achieved as follows: the larger number of needle positions enabled greater customization and live changes during the procedure based on the patient anatomy as visualized on live transrectal ultrasound. The ease of use, in particular the ability to reliably insert and position needles from outside the patient, enabled our medical residents under supervision to confidently perform the entire combined IS/IC implants and even first fraction implants. In addition, the new design enabled in one particular case the addition of a needle after the initial procedure, following the planning CT, as the needle could be inserted from outside with the patient lying on the CT table in scanning position. The reliability and consistency of needle placement with the new design may further increase speed since ultrasound imaging during the procedure may be reduced or omitted. <h3>Conclusion</h3> The 3D printed TARGIT-FX enables simple, fast and reliable interstitial needle insertion for combined tandem and ovoid and IS implants. The flexibility of the design enables new or less experienced users to improve the degrees of freedom and dosimetry of their combined IC/IS implants.

GPP08 Presentation Time: 9:50 AM: From Learning Outcomes to Game Design: Basic Brickwork in Virtual Reality Simulation of Cervical Brachytherapy Procedures

<h3>Purpose</h3> Even before the pandemic, clinical training opportunity for brachytherapy (BT) procedures has been decreasing. In locally advanced cervix cancer (CC), this has been linked to declining BT utilization and poorer clinical outcomes. This work describes the culmination of our early exploration into how to augment resident training in CCBT, distilling and integrating educational keys for a first-in-kind immersive, hands-on, and interaction-enabled virtual reality (VR) simulator for CCBT applicator insertion procedures. <h3>Materials and Methods</h3> Informal needs assessment in consultation with frontline experts and novice stakeholders identified the basic competencies thought to be required to successfully execute a CCBT applicator insertion procedure according to local standards. There was early recognition that, even before a curricular intervention could be designed, a versatile training tool would be needed. After reviewing international guidelines relating to CCBT and BT training, procedure-based learning outcomes were articulated, then selected and simplified for programming a basic goal-oriented simulation (sim) in VR. A clinical vignette of a typical CC case was crafted, with specific narrative stops amenable to building knowledge, skills, and behaviors. Emphasis on the "why" and the "how" of certain BT steps was built in. Procedural milestones were generalizable to trainees with varying degrees of clinical BT experience, with some steps or ‘interactions' made evaluable (go, no go) in this linear design. Gamification expanded options for cross-linking and reinforcing learning concepts. To the extent feasible, realism was incorporated into the physical representation of sim objects, such as CT-scanned surgical instruments and an intracavitary/interstitial BT applicator (Elekta, Netherlands; with permission), extracted as user-configurable additions to the 3D immersive operating room (OR) environment coded on Unreal engine v. 2.46 (Epic Games, US). <h3>Results</h3> A working prototype was generated for a single-player VR sim, broadly mimicking the procedural steps for actual CCBT applicator insertion. Despite an estimated runtime of 20 minutes, depending on user proficiency with CCBT and with VR technology (headset, hand controllers), the sim can start/stop at user discretion, and be shown on-screen to non-playing observers in a classroom setting. The game can also be run iteratively offline, enabling self-paced study of CCBT procedural steps in sequence, repeated execution for kinesthetic learning, and easy focus on recurring problem areas. The user starts the sim in OR, with a computer-generated patient draped and BT instrument tray at the ready. The case vignette and step-by-step instructions trigger the user to complete specific procedural tasks in the correct order; sim will not advance otherwise. Hints are given when needed. Pop-up multiple choice questions, in between BT sim interactions, are formative, drawing on basic recall, understanding, and application, to strengthen learning of core concepts in CCBT. Further contextual information is given at end of sim, to better situate procedural learnings with post-procedure processes (e.g. applicator verification and BT planning). <h3>Conclusions</h3> A novel VRsim tool, inspired by educational design principles, was created as proof-of-concept for guided performance-oriented learning of CCBT procedural aspects. Broader range and complexity of simulated procedures is planned, with higher level of user interaction and realism. Prospective evaluation of face (realism) and content (usefulness) validity will inform on its potential, as part of comprehensive CCBT instructional programming.

Saturday, June 18, 20229:30 AM - 10:30 AM GPP06 Presentation Time: 9:30 AM: Image-Guided Cervical Cancer Brachytherapy: Summary of Consensus Recommendations from the Society for Abdominal Radiology and American Brachytherapy Society

<h3>Purpose</h3> To summarize recommendations regarding imaging guidance for cervical cancer brachytherapy implant placement from the Society of Abdominal Radiology (SAR) and American Brachytherapy Society (ABS). <h3>Materials and Methods</h3> A comprehensive literature search was conducted on PubMed for image-guided cervical cancer brachytherapy (BT) using radiography, ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI). Original research studies were included for evaluation if they included sensitivity and specificity of tumor detection, complication rates, or additional quantitative imaging results directly related to brachytherapy, such as geometric distortion measurements. The literature review was evaluated by an expert panel of members from the Society for Abdominal Radiology (SAR) and American Brachytherapy Society (ABS). These data were analyzed to determine an overall level of evidence for each recommendation and the subsequent strength of each claim. <h3>Results</h3> Three-dimensional (3D) imaging is recommended for treatment planning whenever feasible due to the rate of toxicities ≥ Grade 3 decreasing from between 0-13.5% with 2D planning to 0-5.8% with 3D planning. US guidance is highly recommended for tandem selection and insertion to reduce uterine perforation rates from 2.3-34%/insertion without US-guidance to 0-1.4%/insertion with US. US-guided interstitial needle placement is feasible when post-implant 3D imaging is available for treatment planning. Post-implant CT is recommended over orthogonal radiography to reduce genitourinary and gastrointestinal complications and evaluate proper applicator positioning. Tumor visualization is optimal on T2-weighted (T2w) MRI, which also provides improved specificity for assessment of parametrial invasion. CT-based treatment planning should refer to prior MRI for high risk-clinical target volume (HR-CTV) contouring, when available. Treatment planning on MRI is highly recommended for HR-CTV delineation, and post-applicator imaging should include at minimum both a 2D sagittal T2w and 3D axial T2w fast spin echo-based sequence. <h3>Conclusions</h3> Imaging-guidance for cervical brachytherapy implant placement is now standard-of-care. 3D imaging with MRI is highly recommended for treatment planning purposes, although CT-based planning also results in fewer complications as compared to 2D planning techniques. Ultrasound is useful for applicator placement, but application to treatment planning is limited.

GSOR2 Presentation Time: 9:05 AM: Implementation of Hybrid Intracavitary and Interstitial Brachytherapy for Cervical Cancer in a Freestanding Clinic

<h3>Purpose</h3> Radiation therapy for locally-advanced cervical cancer consists of external beam radiation therapy and brachytherapy. While the recent introduction of image-guided brachytherapy and hybrid intracavitary and interstitial brachytherapy treatments (HBT) may potentially improve clinical outcomes, current literature describes the procedure requiring significant resources (such as general anesthesia, conscious sedation, and MRI-based treatment planning). With increases in treatment time and patient movements, there is greater potential for positional changes of the applicator and neighboring organs at risk (OAR), which can alter dosimetry. Furthermore, these resources might not be universally available to all radiation oncology centers that perform intracavitary cervical brachytherapy, potentially limiting the widespread implementation of HBT. Here we describe our experience with CT-guided interstitial needle placement in a typical freestanding outpatient setting, with an emphasis on minimization of patient movement and treatment duration. <h3>Materials and Methods</h3> All patients that were treated with HBT from June 2018 to May 2020 were included in this series. Initially, patients were taken to the operating room, for an exam under anesthesia and Smit sleeve placement by the radiation oncologist. The patients were determined to need interstitial brachytherapy either by exam under anesthesia, due to anatomy and tumor extent, or after the first brachytherapy treatment, due to dosimetry. All tandem, ovoid and needle insertions and HDR treatments were performed in a freestanding clinic, on the CT simulator table. The patients were treated using minimal sedation (oral lorazepam and oxycodone/acetaminophen). An indwelling urinary catheter was inserted into the bladder. The tandem and ovoids (Elekta) were inserted in the patient, and a vaginal balloon was substituted for packing. Multiple CT scans were performed while advancing the interstitial needles, the bladder was filled with 180 cc of diluted contrast, and a final CT scan was taken for planning purposes. The patients were shifted from the CT table to a surgical bed (single shift), and then transported to the treatment room while treatment planning was performed using Oncentra (Elekta) treatment planning software. After the treatment plan was approved, the bladder was refilled to 180 cc and treatment was delivered using a microSelectron HDR afterloader (Elekta). The applicator was removed and the patient was examined for bleeding. If any vaginal bleeding occurred, vaginal packing was inserted with pressure applied. Time stamps and clinical data were collected during each procedure. <h3>Results</h3> Sixty-three patients underwent placement of a total of 244 interstitial implants. The median time for the entire procedure, from the time of patient entry into the CT simulation room to exit of the HDR treatment room, was 70.0 minutes (mean 70.3, range 54-100). The median time after tandem and ovoid insertion to complete interstitial needle insertion was 9.0 minutes (mean 9.8, range 4-24), using median 2 needles (mean 1.93, range 1-3) and median 3 CT scans (mean 2.63, range 1-4). Bleeding occurred in 22.1% (54/244) of implants, which resolved after median 3 minutes (mean 3.33, range 1-8) of vaginal packing with pressure. All patients (100%) required only one total shift, from the CT table to the mobile surgical bed. <h3>Conclusions</h3> In our series, the interstitial insertion portion of the HBT added 9 minutes, with an overall procedure time of 70 minutes (including insertion, planning, treatment and removal). There was an acceptable bleeding rate which quickly resolved with vaginal packing and pressure. Combining intracavitary and interstitial implants using a hybrid applicator appears to be feasible, efficient and safe in an outpatient freestanding clinic. Further studies aimed at evaluating long-term toxicities and tumor control are warranted.