ESTRO 38 Abstract book

S470 ESTRO 38

clinical practice to improve the current motion mitigation techniques such as margins reduction (SH mode), respiratory gating (SL mode) or tracking (VC mode). Indications still have to be refined and patients well- selected according to appropriate clinical situation. PO-0890 Abscopal effects in metastasized cancer patients treated with PD-1 inhibition and radiation therapy M. Trommer 1 , S.Y. Yeo 2 , T. Persigehl 2 , A. Bunck 2 , M. Schlaak 3 , H. Grüll 2 , S. Theurich 4 , M. Von Bergwelt 4 , J.M. Herter 1 , E. Celik 1 , S. Marnitz 1 , C. Baues 1 1 University Hospital of Cologne, Department of Radiation Oncology, Köln, Germany ; 2 University Hospital of Cologne, Department of Diagnostic and Interventional Radiology, Köln, Germany ; 3 Ludwig Maximilian University, Department of Dermatology and Allergology, Munich, Germany ; 4 Ludwig Maximilian University, Department III of Internal Medicine- Hematology and Oncology, Munich, Germany Purpose or Objective Immune-checkpoint inhibitors targeting the programmed death receptor 1 (PD-1) have shown promising results in the fight against several malignancies. Radiation therapy (RT) induces the release of immunogenic targets and the secretion of damage-associated molecules. Subsequent systemic anti-tumor reactions can lead to a regression of non-irradiated lesions, which is called abscopal effect. The aim of this study was to evaluate abscopal effects based on radiological imaging of irradiated cancer patients receiving simultaneously anti-PD1 antibodies. Material and Methods We analyzed 48 patients treated with PD-1 inhibitors and radiotherapy between August 2014 and December 2017. Inclusion criteria were at least one distant metastatic site outside V10% (volume of tissue receiving at least 10% of the prescribed irradiation dose) and start of RT after at least one month of anti-PD1 therapy (pembrolizumab 2 mg/kg every three weeks or nivolumab 3 mg/kg every two weeks) or within one month after the last application. Abscopal effect was defined as shrinkage of one or more non-irradiated lesions outside V10% with no additional treatment. To distinguish the abscopal effect from the systemic effects of immunotherapy alone, radiological imaging had to be performed twice before RT and at least once after completion of RT. In every radiological image taken during radio-immunotherapy, the largest diameter of the lesions was measured. A size change of more than 20 % was considered as significant decrease or enlargement. Results 11 patients met the inclusion criteria. We observed lesion shrinkage outside V10% in four out of 11 analyzed cases, one patient demonstrated as case example. In five patients, all lesions progressed during the analyzed period, whereas two patients showed already reduced or stable lesions after immunotherapy alone. Nivolumab was given in four cases and pembrolizumab in seven cases. Patients were irradiated with a mean of 50,67 Gy total dose (median 50 Gy) in the analyzed time period. Seven patients were treated stereotactically using Cyberknife ® (20 Gy / 65 %). One out of those received three Cyberknife ® treatments, one patient additionally received hypo-fractionated treatment (3 Gy single dose (SD)), two patients additionally hypo-fractionated and normo- fractionated treatment (3 Gy SD and 1,8 or 2 Gy SD). Two patients had only hypo-fractionated (3 Gy SD) and two only normo-fractioned (2 Gy SD) radiotherapy. Three out of four patients showing AE were treated stereotactically, all AE patients received a high total dose (mean: 77,6 Gy, median: 78 Gy). Conclusion We demonstrated that four out of 11 patients who were considered for this analysis showed a regression of lesions

holds (SPBH) (Figure1).

All the patients underwent one coaching session, one simulation session and then 2 dynamic MRI sessions. Each MRI acquisition lasted for 6 minutes with a first acquisition in SP/SPBH followed by those with MANIV. Patients’ objective tolerance was assessed with oxymetric monitoring (SpO2, etCO2). Subjective tolerance was assessed through questionnaires fulfilled after each session. The motion of the tumour, or the nipple, was tracked and expressed in position, amplitude, period and plateau within each MRI (intra-session analysis) and between MRI (inter-session analysis). Results Five lung cancer patients, 5 liver cancer patients (61-83 years old) and 10 left breast cancer patients (49-77 years old) were included (Table 1).

MANIV was well tolerated without any interruption, nor oxymetric changes. Main reasons of discomfort were the facial mask (47%) and the arms position (29%), before MANIV itself (18%). In cohort A, the BR in VC and SH were all more stable than in SP in intra- and inter-session analysis. Switching from VC to SH led to a motion amplitude reduction of 0.6mm to 9mm proportional to the BR increase (1.5 to 3.4-fold increase). The greatest effect was observed in a liver tumour with a spontaneous median amplitude of 28.7mm (IQR 12.8mm) reduced to 17.6mm (IQR 4.8mm) with SH. In cohort B, the repeated breath-holds in SL were as stable as in SPBH in terms of range within a same plateau (0.7mm in SPBH and 0.9mm in SL) or position between plateaus (0.2mm of median variation for both). Plateaus in SL lasted 16,7sec (IQR=4.6) and 20sec in SPBH. Conclusion This trial demonstrates that MANIV can be used safely in unsedated patients. MANIV may thus be proposed in