Influence of dwell time homogeneity error weight parameter on treatment plan quality in inverse optimized HDR cervix brachytherapy using SagiPlan

Abstract

Purpose Restricting the gradients of dwell times between adjacent dwell positions can potentially be beneficial in reducing the probability of unwanted hot/cold spots occurring, if the planned applicators/anatomy relative positions change before or during treatment. This constraint, however, may degrade plan quality. This study, for the first time, aims to quantify the impact of modulation restriction on plan quality indices in inverse optimization for cervix high-dose-rate (HDR) brachytherapy using the BEBIG SagiPlan treatment planning system. Material and methods Ten cervical cancer patient plans were optimized for treatment with a BEBIG SagiNova 60 Co HDR afterloader using the min/max inverse planning method, with dwell time homogeneity error weight (DTHEW) parameter values of 0 to 10. Dwell time homogeneity and gradients as well as various plan quality indices were analyzed. Results For DTHEW = 0, min/max-based optimization yielded higher HR-CTV D 90 values than the variance-based option ( p < 0.001) and was therefore selected for this study. Averaging over all patients, selecting non-zero DTHEWs resulted in a general increase in dwell time homogeneity and decrease in mean and maximum adjacent dwell time gradients, especially between DTHEWs of 0 and 1. For DTHEW > 1, an increase of this parameter did not always result in more homogeneous dwell times or reduced gradients in individual patients. There was a negative correlation between DTHEW and both HR-CTV D 90 and V 100 ( p < 0.001, r = –0.91). Increasing DTHEW also negatively affected conformity index ( p < 0.001, r = –0.99). Changes in rectum and sigmoid colon D 2cc were insignificant. There was a strong positive relationship between bladder D 2cc and DTHEW ( p < 0.001, r = 0.99). Conclusions Assuming a static geometry, statistically significant degradation of plan quality can result from restricting the dwell time homogeneity in min/max-based optimization of cervix HDR brachytherapy plans using SagiPlan. Therefore, setting DTHEW to zero is indicated for the type of patient plans considered in this study.

Does inverse planning improve plan quality in interstitial high-dose-rate breast brachytherapy?

“…The role of controlling neighboring dwell times in quality of treatment plan is controversial. In a recent study, a negative correlation was found between modulation restriction, and both HR-CTV D 90 and V 100 in cervix BT as well as an increasing restriction, negatively affected the conformity index [36]. According to the results of Mavroidis et al [37], dwell time regularization technique in HIPO introduced a minor improvement in the effectiveness of the optimized prostate HDR dose distributions.…”

Optimization in treatment planning of high dose‐rate brachytherapy — Review and analysis of mathematical models

“…[47] and [50] have a parameter between zero and one to control the DTMR, where zero is the least restrictive choice, and this parameter value is suggested 47 to be in the range 0.1–0.2. A DTMR is implemented in the treatment planning system SagiPlan but it is suggested 51 that the DTMR parameter should be zero, corresponding to no restriction.…”

Comprehensive methodology for commissioning modern 3D-image-based treatment planning systems for high dose rate gynaecological brachytherapy: A review

“…Sensitivity analysis of inverse planning parameters: Investigates whether various effective parameters (e.g., the optimization run-time) affect the quality of inverse planning routines [5,32,[60][61][62]. Selection of suitable parameters [32].…”

“…They offer DT/DW manual adjustment, graphical optimization, geometrical optimization, and inverse planning approaches, such as inverse planning by simulated annealing (IPSA) [77] and hybrid inverse treatment planning and optimization (HIPO) [78]. But before using the optimization algorithms, the physicist should evaluate these routines to ensure that they work correctly and robustly and assess the sensitivity of the resulting plans to the available parameter settings [3,11,14,15,17,18,58,62]. Also, appropriate training of users is required [5].…”