Outcomes of and Treatment Planning Considerations for A Novel Technique Delivering Proton Pencil-Beam Scanning Radiation to Women With Metal-Containing Tissue Expanders Undergoing Post-Mastectomy Radiation

Abstract

Breast reconstructive tissue expanders (TEs) contain metallic ports (MCPs) comprised of high “Z” materials. This presents a challenge for proton pencil-beam scanning (PBS) planning given potential dose attenuation and concern for port movement. We hypothesize that a novel technique utilizing hybrid single- and multi-field optimizations (SFO, MFO) for delivering post-mastectomy radiation (PMRT) using proton PBS for patients with TEs and MCPs will be robust and uphold the dosimetric benefits of proton therapy. A protocol utilizing a hybrid SFO/MFO technique was prospectively developed. Plans were robustly optimized using a Monte Carlo algorithm. A CTV_eval structure including chest wall (CW) and regional nodal (RNI) targets and excluding boost plans and the TE metallic components plus a 1 cm margin was evaluated for target coverage, robustness, and dose homogeneity. Thirteen women with TEs completed PMRT between 2/12/19 and 1/2/20. Median age was 46 years (range, 37-60), 67% were white. Median body-mass index (BMI) was 24 (range, 19-36). 75% of patients had left-sided disease. A 2 or 3-field SFO technique was used superior and inferior to the MCP, while a 3 or 4-field MFO technique was used at the level of the MCP. At the level of the MCP, beams were blocked to deliver dose anterior/posterior and lateral to the port without going through it. A port-to-CW distance of 1cm was required to deliver dose posteriorly. Anterior, anterior oblique, and lateral beam angles were utilized. Anterior beams were given with a couch kick to enter under the breast and deliver dose behind the expander. Patients underwent daily image-guidance to ensure the port remained within a 5mm planning volume (PRV). The median total RT dose to CTV_eval was 50.4Gy (45.0-50.4) delivered in 1.8Gy/fraction; 5 (38%) of plans utilized a boost. Median ipsilateral lung V20 was 20% (2.0-29); median average heart dose was 1.7Gy (0.5-3.5). Median 100% CTV_eval coverage was 95% (90-95) with a median homogeneity index of 4.2(1.9-18). OSLD readings anterior to the port were performed on 5 patients and correlated to the dose measurements in the TPS. When evaluating plan robustness, median worst-case 95% CTV_eval coverage was 95% (90-97%). There was one ≥ G2 acute toxicity of G4 skin necrosis in a patient with very thin tissue flaps which did not occur immediately anterior to the port but required expander removal. At a median follow up of 3.0 months, 2 patients had successful permanent reconstructions after RT completion. Delivering PMRT with PBS for women with metal-containing TEs using a novel hybrid SFO/MFO technique is feasible, robust, achieves excellent dose distributions to targets and OARs, and did not lead to unexpected toxicities. Additional dosimetric, physics and set up techniques are required to safely deploy this treatment in the clinic. This is the largest series to date describing the delivery of PMRT to women with MCPs, and suggests that MCPs may not preclude treatment with PBS.