KEY POINTS
- This retrospective single-center study evaluated 32 cervical cancer interstitial brachytherapy cases treated using a Syed–Neblett transperineal template between 2015 and 2021. Half the procedures incorporated iterative intraoperative MRI guidance (n=16), while the remaining 16 used conventional implantation without intraoperative MRI-based adjustments. Investigators retrospectively simulated optimized catheter placement using postimplant images to approximate a proposed post-template-and-applicator placement (post-TAPP) imaging workflow.
- The optimization pipeline used centroidal Voronoi tessellation (CVT) to distribute interstitial catheters across anatomically suitable template positions. Three-dimensional target and OAR contours were projected onto the template plane, with a 2-mm safety margin around excluded OAR regions. The tandem and six immediately adjacent catheter positions were kept fixed, while additional catheters were redistributed within eligible template holes. The algorithm required a mean 12.3 seconds per case (range 3.2–31.8 seconds).
- For each patient, two new HDR plans were generated: one using CVT-optimized catheter positions and another using the clinically implanted catheter configuration. Both were replanned by the same experienced brachytherapy physicist using 24 Gy in three fractions (8 Gy per fraction), with cumulative EQD2 evaluated against EMBRACE-II target and OAR criteria. All reconstructed plans satisfied the evaluated bladder, rectum, sigmoid, and bowel dose limits.
- In patients treated without intraoperative MRI guidance, CVT optimization increased compliance with CTVHR D90 ≥85 Gy EQD2 from 62.5% (10/16) to 93.8% (15/16), with a nominal p=0.033 that did not meet the study's multiplicity-adjusted significance threshold. Compliance with residual GTV D98 ≥90 Gy also increased from 50.0% to 75.0% (p=0.144). Five previously inadequate clinical catheter arrangements achieved the CTVHR D90 requirement after virtual optimization.
- In procedures already incorporating iterative MRI-guided catheter adjustments, the dosimetric benefit was considerably smaller. CTVHR D90 ≥85 Gy compliance remained 93.8% (15/16) with both clinical and CVT configurations, while residual GTV D98 compliance increased only from 75.0% to 81.3% (p=0.669). This suggests that intraoperative MRI-guided adjustments had already addressed many of the suboptimal catheter arrangements that automated planning sought to correct.
- The most consistent improvement was a reduction in unnecessary catheters. Compared with the reconstructed clinical plans, CVT reduced the mean number of inserted but unused catheters by 3.0 in the MRI-guided group and 2.8 in the non-MRI-guided group (both p<0.001). The corresponding reductions in the proportion of unused catheters were 21.8 and 20.6 percentage points, respectively. Importantly, these changes occurred without significant increases in CTVHR V150% or V200%, indicating no measurable worsening of high-dose volumes.
- The potential workflow benefit remains unproven. Intraoperative MRI-guided procedures required a mean two catheter-adjustment iterations, with total MRI-related time averaging 55.5 minutes and each additional adjustment adding approximately 14 ± 2 minutes. However, the optimization was retrospective, and no actual reduction in procedural time, needle trauma, or complications was demonstrated. The algorithm also assumes straight catheter trajectories despite observed tip displacements of approximately 4–6 mm, and two cases achieved better target coverage with clinically placed catheters than with the CVT-generated configuration.
CLINICAL TAKEAWAY
Geometry-based catheter optimization may improve the efficiency of interstitial cervical brachytherapy by reducing unnecessary needle insertions while maintaining target coverage and organ-at-risk sparing. Its greatest potential appears to be in procedures without iterative intraoperative MRI guidance, but the benefits remain dosimetric and computational until prospective studies demonstrate improved procedural efficiency, safety, or clinical outcomes.