Abstract Book

S143

ESTRO 37

4 Leiden University Medical Center LUMC, Department of Gastroenterology, Leiden, The Netherlands Purpose or Objective The HERBERT study was a high-dose-rate endorectal brachytherapy (HDREBT) dose escalation study in elderly patients with rectal cancer who were unfit for surgery. The primary results showed promising response rates of 60%, but severe late toxicity was not uncommon and occurred in approximately one third of patients, illustrating the need for further analyses. The current analyses focus on the correlation of clinical and dosimetric parameters with severe late clinical and The HERBERT study was performed from 2007 to 2013 in inoperable rectal cancer patients. Patients received 13x3 Gy EBRT followed by three weekly applications HDREBT in different dose levels of 5 to 8 Gy per fraction, which was prescribed to the circumferential margin of the HDREBT clinical target volume (CTV). The CTV, contralateral rectal wall and anal canal were delineated on the planning-CT with applicator in situ acquired prior to the first brachytherapy application. Dose volume histogram parameters were collected for CTV coverage (physical dose/fraction): CTV D98 and CTV D90 and for hotspots in the CTV, contralateral wall and the anal canal: D2cc, D5cc and D10cc. In addition, the volumes of isodose lines corresponding to a cumulative total dose of 60 Gy and 75 Gy (EQD2 α/β=3) were also acquired. Severe late proctitis (CTCAEv3 grade ≥3) and endoscopic toxicity at the tumor site (categorised as 1. erythema or scarring, 2. superficial ulcer and 3. deep ulcer) were evaluated. Patients with a complete or partial response were included in the analyses with censoring at time of local progression. Univariable analyses were performed using Spearman correlation, Mann Whitney-U and Krukshal-Wallis-H test. Logistic regression was used for multivariable analyses. Results Thirty-five of 38 patients completed treatment and were included in the current analyses. Grade 3 or more late proctitis occurred in ten patients and was only associated with CTV D90 (no correlation with patient and tumor characteristics; or dosimetric parameters of the contralateral wall). When correcting for CTV volume in a multivariable analyses CTV D90 remained significant (OR 3.1, 95%CI; 1.1-8.5). The endoscopic toxicity at the tumor site was correlated with circumference at baseline, CTV volume, CTV D2cc, and the volumes correlating to a cumulative totaldose of 60 Gy and 75 Gy (Table 1 and Figure 1). endoscopic proctitis. Material and Methods

Montreal, Canada 13 Jewish General Hospital- McGill University, Colo-rectal surgery, Montreal, Canada 14 Jewish General Hospital- McGill University, Medical Oncology Department oncology, Montreal, Canada Purpose or Objective In North America, the standard of care for locally advanced rectal cancer is neoadjuvant chemo-radiation therapy (CRT) prior to total mesorectal surgery (TME) followed by adjuvant chemotherapy (CT). This results in a median time of 4.5-5.5 months from the start of local therapy to the start of systemic treatment.The present study is proposing to evaluate a modified strategy to shorten the local disease treatment time to 2.5-3.5 months with targeted high dose rectal brachytherapy (HDRBT) to allow for optimization of systemic therapy with FOLFOX CT given either in neoadjuvant or adjuvant setting. Material and Methods The primary endpoint is patient compliance to CT. A total of 180 patients were randomized using a 2:1 ratio to arm A: Pre-operative six cycles of FOLFOX CT followed by Imaged guided HDRBT to deliver 26 Gy in 4 fractions, then TME and 6 cycles of adjuvant CT; or arm B. Pre- operative HDRBT, TME and 12 cycles of adjuvant FOLFOX CT. The primary outcome was compliance rate to the first 6 cycles of CT.This study was approved by the Institutional Review Board of each of the five participating centers and written informed consent was obtained for eligible pts. Results 180 pts without metastases were recruited with 120 pts treated in arm A (neoadjuvant FOLFOX) and 60 patients in arm B (adjuvant FOLFOX). The CT compliance rate of 78.1 % for arm A compared to 51.9% in arm B P =.(p = 0.0008)..Grade 3 & 4 toxicity rate during chemotherapy was simllar with 29% for arm A and 31% for arm B. One patient in arm A died during CT (Grade 5). HDREBT was feasible in all patients but 4 pts. One patient refused HDRBT after randomization (arm B) and 3 patients on arm A had local tumor progression and were not deemed for HDRBT and all received a short course of external beam radiation therapy. Two patients refused surgery after neoadjuvant treatment. On an intention to treat basis, the pT0 rate was 35.2% for arm A and 32.1 % for arm B while the pT0N0 rate was respectively 28.7% and 22.4% (p= 0.72073). With a median follow up of 37 months (range: 11-84), the secondary outcomes for arms A and B at 3 years were respectively, local recurrence rate was 3% and 5% (p=0.9893), Disease free survival 80 % and 76%, (p=0.6020), the cancer specific survival was 94% and 92% (p=0.8386) and overall survival rate 94% and 85% The safety and improved compliance to neoadjuvant CT is confirmed in this multi-institutional randomized phase II study using HDREBT as a neoadjuvant RT modality for rectal cancer.The good oncological outcomes obtained in this patient population at the time of this reporting are encouraging. . OC-0280 Proctitis after brachytherapy for rectal cancer: clinical and dosimetric factors - The HERBERT study E.C. Rijkmans 1 , R.A. Nout 1 , E.M. Kerkhof 1 , A. Cats 2 , B. Van Triest 3 , A. Inderson 4 , R.P.J. Van den Ende 1 , M.S. Laman 1 , M. Ketelaars 1 , C.A.M. Marijnen 1 1 Leiden University Medical Center LUMC, Department of Radiotherapy, Leiden, The Netherlands 2 The Netherlands Cancer Institute, Department of Gastroenterology, Amsterdam, The Netherlands 3 The Netherlands Cancer Institute, Department of Radiotherapy, Amsterdam, The Netherlands (p=0.5327). Conclusion