ESTRO 38 Abstract book

S868 ESTRO 38

Purpose or Objective Cranio Spinal Irradiation (CSI) is a promising indication for pencil beam scanning (PBS) proton therapy. In this study we present the clinical and dosimetric evaluation of CSI treatments delivered at PSI, with special focus on tumour recurrences. Material and Methods 20 CSI patients, treated between 2004 and 2017, have been included in this study. 17/20 were positioned prone (Gantry1), and 3 supine (Gantry 2). The primary tumors were: medulloblastoma (10pts), PNET (4pts), anaplastic ependymoma (3pts), choroid plexus carcinoma (1pt) and others (2pts). 3 of 20 patients received CSI for local failure after a first course of local fractionated RT, thus requiring accurate sparing of pre-irradiated brain tissue. Daily imaging was performed based on orthogonal topograms acquired once along the full length of the spine before delivery, and a single positioning correction vector for the entire CSI PTV applied. Dose delivery on the other hand consists of a number of patched subfields. As such, inaccuracies of couch movements between sub-fields during delivery might result in under- or over-dosage. For all cases we analysed: i) planning approach; ii) target coverage (V95, D98) and OAR sparing; iii) dose averaged LET distributions and iv) robustness (range errors and set- up scaling for fractionation). Accuracy of prone (G1) vs. supine (G2) setup was evaluated. During follow-up, 6 patients (30%) presented with tumor recurrences, of which 5 (25%) could be considered high risk patients at diagnosis - 3 with leptomeningeal dissemination at the time of CSI, 1 with a recurrent PNET at the time of irradiation and one case where the tumor was an “in-field” relapse of an anaplastic ependymoma after re-irradiation. For all 6 relapsing patients, a detailed localization of the recurrence and any spatial relationship with field patching was performed. Results Clinical dosimetric data are summarized in Table 1. Supine positioning resulted in lower systematic and random errors as compared to prone (systematic residual errors < 0.1mm for supine as compared to 0.5mm for prone; random errors in PA direction were reduced from 1.7mm for prone to 1.1mm for supine). For the relapsing patients, the recurrences did not overlap with patch lines and were not correlated with setup or range error. Indeed, all recurrences were either located in the high risk PTV or were leptomeningeal dissemination. We observed two vascular radiation induced toxicities: one with Moya-Moya syndrome and one with multiple cerebral cavernomas. These observed vascular toxicities were not related to high-LET areas. Conclusion PBS CSI treatments can be safely delivered, as the plans are robust considering range and setup uncertainties. The pattern of recurrences was not correlated with potential dose inaccuracies in the patch lines or lack of robustness, and no correlation of toxicities to enhanced LET have been observed.

Conclusion Our study showed that the mean EQD2 to the hair follicles lower than 15.1 Gy seemed to be associated with prevention of permanent alopecia. More cases are needed to verify the efficacy and more accurate threshold to prevent permanent alopecia for pediatric patients who receive multi-drug CTx and cranial irradiation. EP-1610 Cranio Spinal Axis irradiations using Pencil Beam Scanning: the PSI experience D. Siewert 1 , F. Belosi 1 , R. Gonzalo Gleyzes 1 , L. Mikroutsikos 1 , D. Correia 2 , A. Pica 1 , F. Albertini 1 , A.J. Lomax 1 , D.C. Weber 1 , A. Bolsi 1 1 Paul Scherrer Institute, Center for Proton Therapy, Villigen, Switzerland ; 2 Bern University Hospital, Department of Radiation Oncology- University of Bern, Bern, Switzerland