ESTRO 38 Abstract book

S266 ESTRO 38

Conclusion Prediction models based on T2- and DW-MRI radiomics can be used for adequate non-invasive detection of patients with rectal cancer that will achieve a (near-)complete response after CRT. These radiomics models, however, do not outperform a previously reported prediction model based on two volumetric and two ADC parameters. OC-0511 Organ Preservation with Image Guided and Adaptive Brachytherapy for Patients with Rectal Cancer A. Garant 1 , S. Magnan 2 , S. Devic 3 , A. Martin 2 , M. Boutros 4 , C. Vasilevsky 4 , S. Ferland 5 , A. Bujold 6 , S. Desgroseilliers 7 , H. Sebajang 8 , C. Richard 9 , T. Vuong 10 1 UT Southwestern Medical Center, Radiation Oncology, Dallas, USA ; 2 Laval University, Radiation Oncology, Quebec City, Canada ; 3 McGill University, Medical Physics, Montreal, Canada ; 4 Jewish General Hospital, Colorectal Surgery, Montreal, Canada; 5 Gatineau Hospital, Radiation Oncology, Gatineau, Canada ; 6 Maisonneuve-Rosemont Hospital, Radiation Oncology, Montreal, Canada ; 7 Pierre-Boucher Hospital, Surgery, Longueuil, Canada ; 8 University of Montreal Hospital Center, Surgery, Montreal, Canada ; 9 University of Montreal Hospital Center, Colorectal Surgery, Montreal, Canada ; 10 Jewish General Hospital, Radiation Oncology, Montreal, Canada Purpose or Objective Organ preservation or non-operative management (NOM) of rectal cancer is of growing interest among colorectal experts. Image guided and adaptive endorectal brachytherapy (IG-AEBT) is a radiation dose escalation modality: we explored its role as a tumor-response adaptive therapy in elderly patients unfit for surgery and patients refusing surgery. Material and Methods In this prospective registry study, patients with rectal cancer who were ineligible for surgery began treatment with pelvic external beam radiotherapy (EBRT) to a dose of 40 Gy in 16 fractions; in patients refusing surgery, EBRT was prescribed at a dose of 45- 50 Gy in 25 fractions with concurrent 5-FU. This was followed by three weekly IG- AEBT boosts of 10 Gy to the residual clinical target volume, for a total of 30 Gy in 3 weekly fractions. Complete clinical response (cCR) was the primary endpoint. Results A total of 118 patients were included; the median age was 81 years. With a median follow-up of 2.5 years for living patients, the proportion of cCR was 86.4%, the tumor regrowth proportion was 12.8%, and the cumulative incidence of local relapse was 4.3% at 1 year and 16.4% at 2 years. When considering responders and non-responders, the 2-year local control was 72.8%. The overall survival at 2 years was 66.1%. Acute rectal grade 1-2 toxicity included all patients: 25.5% of patients had late rectal bleeding requiring iron replacement, or blood transfusions and/or argon plasma therapy. Conclusion The results of our registry study exploring IG-AEBT as a boost modality for an elderly medically unfit patient population with unselected tumors reveal that a high proportion of patients achieved cCR with a reasonable toxicity profile. Consequently, it is our belief that this technology will contribute to the challenging NOM paradigm of rectal cancer. OC-0512 Radiochemotherapy and hyperthermia in locally advanced rectal cancer - A prospective phase II trial C. Gani 1 , U. Lamprecht 1 , M. Bitzer 2 , J. Gellermann 3 , O. Voigt 1 , A. Ziegler 4 , M. Moll 1 , A. Königsrainer 5 , D. Zips 1 1 University Hospital Tübingen, Radiation Oncology, Tübingen, Germany ; 2 University Hospital Tübingen,

Gastroenterology, Tübingen, Germany; 3 Zentrum für Strahlentherapie und Radioonkologie, Radiation Oncology, Berlin, Germany ; 4 Städtische Kliniken Esslingen am Neckar, Oncology, Esslingen am Neckar, Germany ; 5 University Hospital Tübingen, Surgery, Tübingen, Germany Purpose or Objective Despite excellent local control rates after preoperative radiochemotherapy and total mesorectal excision there is growing interest in local treatment escalation strategies which could increase pathological complete response rates and therefore qualify more patients for non- operative management. Therefore, the goal of the present study was to evaluate tumor regression after preoperative radiochemotherapy combined with deep regional hyperthermia in locally advanced rectal cancer. Material and Methods A total of 78 patients with locally advanced rectal cancer, UICC stage II or III, were enrolled in this prospective phase 2 study. Radiotherapy consisted of 50.4 Gy and concomitant chemotherapy with 5-Fluorouracil (1000mg/m 2 CVI) during the first and fifth week of treatment. Deep regional hyperthermia was applied biweekly with a target of at least eight treatments. Tumor regression was graded according to the Dworak classification with "Dworak 4" corresponding to a pathological complete response. The primary endpoint was the pathological complete response rate, a good response (GR) was defined as Dworak grades three and four. Results All patients completed radiotherapy as prescribed. Sixty of 78 patients (77%) completed eight or more hyperthermia treatments, all patients underwent surgery. Five patients experienced grade three or higher hyperthermia related toxicity (Three with claustrophobia, two with bolus pressure). Dworak grades 1/2/3 and 4 were 11.5% / 38.5% / 34.6% and 15.4%. A GR to treatment was achieved by 50% of all patients. Patients with a GR had significantly higher cumulative equivalent minutes at 43 degrees (CEMT43) compared with patients with Dworak grades 1 and 2 (7.22 vs 4.47, p=0.012). Conclusion Our study showed a very favorable toxicity profile of deep regional hyperthermia and a promising proportion of patients with a high degree of tumor regression. The higher CEMT43 in good responders suggests a causal relationship between temperature and tumor regression. We will further investigate the potential of hyperthermia to maximize pathological complete response rates in the ongoing multicentric CAO/ARO/AIO-16 organ preservation trial. (Parts of the study results were presented at the "European Society of Hyperthermia in Oncology" Meeting in 2018) OC-0513 Cone-beam CT intensity correction using a generative adversarial network and unpaired training C. Kurz 1,2,3 , M. Maspero 3 , M.H.F. Savenije 3 , G. Landry 2 , F. Kamp 1 , M. Li 1 , K. Parodi 2 , C. Belka 1,4 , C.A.T. Van den Berg 3 1 University Hospital- LMU Munich, Department of Radiation Oncology, Munich, Germany ; 2 Faculty of Physics- Ludwig-Maximilians-Universität München, Department of Medical Physics, Munich, Germany ; 3 Universitair Medisch Centrum Utrecht, Center for Image Sciences, Utrecht, The Netherlands; 4 DKTK, German Cancer Consortium, Munich, Germany Proffered Papers: PH 9: Proffered paper: Artificial intelligence and novel imaging approaches