SBRT of large tumors and tumors located near organs at risk in the thoracic cavity

Sammanfattning: Stereotactic Body Radiotherapy (SBRT) is a non-invasive ablative radiation therapy developed at Karolinska University Hospital in the early 90-s. Its basic principles are highly focused inhomogeneous dose load with rapid fall of in the periphery of the target, delivered in few fractions with high-doses per fraction. Due to the ability to spare the surrounding healthy structures, the technique allows for deployment of very high radiation doses in the target, yielding high rates of local control to a price of a limited toxicity. However, high fractional doses of SBRT may cause excessive toxicity in case of proximity of a radiation sensitive organ. Likewise, the volume of damaged healthy tissue increases with the volume of the target, which is important in treatment of large tumors. Dose planning in SBRT is a balance between a dose sufficient for tumor control and compliance with dose restrictions for surrounding tissues. The purpose of this thesis is to elucidate key situations of this important trade-off from a radiological and clinical point of view, and contribute to useful guidelines in the SBRT practice. Paper I, was a single institution retrospective analysis of 52 patients treated for apical lung tumors in proximity of plexus brachialis. A method for contouring of plexus brachialis was established, and the delineation of the organ was performed retrospectively. Clinical data on radiation induced brachial plexopathy was collected. NTCP-modelling was performed based on different dose-volume parameters, showing a high predictive ability for adverse events. The study provided insights into the feasibility of anatomical delineation, and dose constrictions for plexus brachialis. Paper II, was a retrospective single institutional study of 164 patients with tumors larger than 70 cc treated with at least 8 Gy x 5, located in thorax and abdomen. 70 cc corresponds roughly to a spherical structure of 5.1 cm. Local control and toxicity were primary and secondary aims. Minimal dose to the GTV and histology were predictive for local control in multivariate analyses. The best local control was for renal cell carcinoma while worst for colorectal carcinoma. Seven of the ten patient that suffered from potentially lethal adverse events had tumor located close to central structures in the thorax. The study showed that, when possible to deploy a sufficient irradiation dose without compromising organs at risk, the SBRT might be a good treatment option for large tumors. Papers III and IV, were analyses of patients with tumors located near central bronchial structures in the thorax. Paper III is a multicenter non-randomized phase II-trial, involving 65 patients with tumors within 1 cm from trachea, main and lobar bronchi, treated with 7Gy x 8. Ten patients experienced possible lethal adverse event, including eights cases of hemoptysis. Distance to main bronchi and dose to main bronchi and distal trachea were the risk factor for bronchial bleeding. A hypothesis was put forward that tumors located more than 1 cm from distal trachea/main bronchus, and given that the main bronchus/trachea receive less that 70-80 Gy EQD2, are relatively safe to treat with the study fractionation. In paper IV, the cohort was expanded with a retrospective multicenter cohort of patients treated in the same way for tumors within 2 cm from the central bronchial tree. In total, there were 232 patients, with 30 possible treatment related deaths. The concept of the most central sensitive bronchi in the hypothesis described above needs to be expanded to also include the intermediate bronchus. Tumor compression of any part of the PBT was associated with treatment related death. The study provided important material and insights for further dose-volume modulations of individual radiation sensitivity of each part of the central bronchial tree.