Using Skin Dose Parameters Predicts Low Skin Toxicity in a Phase 2 Trial of Multiple Dwell Position Balloon-Based Brachytherapy for Partial Breast Irradiation

Abstract

Purpose/Objective(s): Preliminary trials with balloon-based brachytherapy for partial breast irradiation showed a 50% rate of grade 2 or higher skin toxicity, and a substantial number of catheters had to be removed prior to treatment because of a balloon-skin distance <5 mm. These trials employed a “single dwell position” HDR technique. The objectives of this trial were: (a) to prospectively determine if a multiple dwell position dose delivery method can decrease skin dose and subsequent toxicity over an in silico modeled single dwell position technique while achieving similar or superior PTV coverage, and (b) to evaluate whether specific skin parameters could be safely used in lieu of skin-balloon distance alone for predicting toxicity and treatment eligibility. Materials/Methods: A single-arm phase II study using a Simon two-stage design was performed on 28 women with stage 0, I, or II (<3 cm) breast cancer. Each patient was treated using a multiple dwell position Ir-192 plan with a balloon-based applicator with a prescription of 3400 cGy in 10 fractions delivered BID. Initial entry required a minimum skin-balloon distance 7 mm. Plans for single dwell position treatment were also run for comparison of dosimetric parameters. The primary endpoint was grade 2 or higher skin toxicity. Acute and late toxicity was assessed during treatment and at 1, 3, 6, and 12 months. Based on the minimal toxicity seen after treating the first 16 patients, additional patients were treated irrespective of skin-balloon distance as long as the maximum dose to 1 mm thick skin tissue was <130%. Results: Compared to the phantom single dwell plans, multiple dwell position planning yielded equivalent to slightly superior PTV coverage as measured by median V90 (100% vs 97%, p < 0.001), V95 (99% vs 93%, p < 0.001), and V100 (96% vs 88%, p < 0.001) but had slightly higher median V150 (39% vs 29%, p < 0.001) and V200 (10% vs 4%, p < 0.001), as well as a lower median dose homogeneity index (0.6 vs 0.7, p Z 0.0046). Maximum point dose to skin was decreased by multiple dwell planning for all skin depths (1 mm, 2 mm, and 3 mm; p < 0.001). The most common grade 1 toxicities were erythema (16 pts), fibrosis (9 pts), fatigue (8 pts), mild pain (8 pts), hyperpigmentation (5 pts), and telangiectasias (4 pts). There have been 2 patients (7%) who developed grade 2 toxicity (both transient erythema), and no patients have experienced grade 3, 4, or 5 toxicity. Conclusions: Multiple dwell position planning for balloon-based brachytherapy results in lower skin doses with equal to superior PTV coverage and an overall low rate of skin toxicity. Our data suggest that limiting the maximum point dose to <130% to 1 mm thick skin is achievable and results in minimal toxicity independent of skin-balloon distance. This parameter may therefore be a more clinically relevant selection criterion. Author Disclosure: S.K. Nath: None. Z. Chen: None. B.P. Rowe: None. R. Blitzblau: None. S. Aneja: None. B.J. Grube: None. N. Horowitz: None. J.B. Weidhaas: None.