Comparison of single and multiple dwell position methods in MammoSite high dose rate (HDR) brachytherapy planning

The purpose of this study is to dosimetrically compare two plans generated using single dwell position method (SDPM) and multiple dwell position methods (MDPM) in MammoSite high dose rate (HDR) brachytherapy planning for 19 breast cancer patients. In computed tomography (CT) image-based HDR planning, a surface optimization technique was used in both methods. Following dosimetric parameters were compared for fraction 1 plans: %PTV_EVAL (planning target volume for plan evaluation) coverage, dose homogeneity index (DHI), dose con-formal index (COIN), maximum dose to skin and ipsilateral lung, and breast tissue volume receiving 150% (V150[cc]) and 200% (V200[cc]) of the prescribed dose. In addition, a plan was retrospectively generated for each fraction 2-10 to simulate the clinical situation where the fraction 1 plan was used for fractions 2-10 without modification. In order to create nine derived plans for each method and for each of the 19 patients, the catheter location and contours of target and critical structures were defined on the CT images acquired prior to each fraction 2-10, while using the same dwell-time distribution as used for fraction 1 (original plan). Interfraction dose variations were evaluated for 19 patients by comparing the derived nine plans (each for fractions 2-10) with the original plan (fraction 1) using the same dosimetric parameters used for fraction 1 plan comparison. For the fraction 1 plan comparison, the MDPM resulted in slightly increased %PTV_EVAL coverage, COIN, V150[cc] and V200[cc] values by an average of 1.2%, 0.025, 0.5 cc and 0.7cc, respectively, while slightly decreased DHI, maximum skin and ipsilateral lung dose by an average of 0.003, 3.2 cGy and 5.8 cGy, respectively. For the inter-fraction dose variation comparison, the SDPM resulted in slightly smaller variations in %PTV_EVAL coverage, DHI, maximum skin dose and V150[cc] values by an average of 0.4%, 0.0005, 0.5 cGy and 0.2 cc, respectively, while slightly higher average variations in COIN, maximum ipsilateral lung dose and V200[cc] values by 0.0028, 0.2 cGy and 0.2 cc, respectively. All differences were too small to be clinically significant. Compared to the MDPM, the SDPM combined with a surface optimization technique can generate a clinically comparable fraction 1 treatment plan with a similar interfraction dose variation if a single source is carefully positioned at the center of the balloon catheter.