Multi-institutional evaluation of direction-modulated brachytherapy tandems for cervical cancer

Background: Direction-Modulated Brachytherapy (DMBT) offers a means of generating anisotropic radiation beams to create conformal dose distributions when in combination with image-guided adaptive brachytherapy (IGABT) workflow. The purpose of this study was to integrate prototypical DMBT tandem applicators of varying physical dimensions within a commercial treatment planning system (TPS) and perform a comprehensive planning study to evaluate and quantify its performance. Methods: A multi-institutional cohort of 110 clinically delivered high-dose-rate (HDR) cervical cancer brachytherapy plans were retrospectively re-planned with nine iterations of the DMBT six-groove tandem model incorporated within the TPS, of varying thicknesses (4–8 mm), for a total of 990 replans. Original plans consisted of both intracavitary and intracavitary–interstitial (77:33) cases with an average prescription dose/fraction and high-risk clinical target volume (CTVHR) of 607.73 ± 113.43 cGy and 26.96 ± 14.95 [range 6.7–69.58] cm3, respectively. A novel inverse optimization process was implemented, specifically for DMBT, such that the lowest possible organ-at-risk (OAR) D2cc doses were sought while keeping equivalent target coverage (ΔCTVHR-D90 to within ±1%) as well as retaining the original pear-shape dose distributions. Results: Significant improvements in plan quality were achieved by all nine DMBT tandem models. Irrespective of the model, on average, about 45–60 cGy/plan reductions in D2cc across all OARs were achieved. In terms of total EQD2, over the course of a standard 4–5 fraction protocol, D2cc showed significant reductions of −6.29 ± 4.37 Gy (p < 0.003), −3.82 ± 2.06 Gy (p < 0.0001), and −4.85 ± 3.01 Gy (p < 0.001) for the bladder, rectum, and sigmoid, respectively, for the thickest DMBT model. Conclusions: We have successfully performed a treatment planning study integrating nine unique DMBT six-groove tandem models into a commercial TPS. All models were able to achieve significant dose reductions to OARs, without compromising target coverage, across plans from multiple institutions with varying applicators and clinical (planning) practice. The results indicate that DMBT technology has promising potential in the treatment of cervical cancer.