Abstract Book

S57

ESTRO 37

hypertension. No patients received prior local treatment. Thirteen patients (31%) presented a grade ≥ 3 acute adverse event of which liver biology perturbation was the most encountered (20,9%). Three patients (10%) experienced a decline in CP at 3 months post-SBRT. There was no grade ≥ 3 chronic adverse event reported. The 18- and 24-month local control rates were 98% (95% CI, 85%-99%) and 94% (95% CI, 79%-99%), respectively. The 18- and 24-month overall survival rates were 72% (95% CI, 56%-83%) and 69% (95% CI, 53%-81%), respectively. Median survival was 3.5 years. Conclusion Local control and overall survival after SBRT for untreated solitary HCC were excellent despite candidates being unfit for resection and ablation treatments. SBRT should be considered, as bridging to transplant or as definitive therapy for those ineligible for transplant. It should be considered in multidisciplinary committees alongside other localized standard treatments including radiofrequency ablation or transarterial chemoemboli- zation, depending on patients’ characteristics, and advantages and disadvantages of each technique. PV-0106 Liver-directed CCRT with high-dose radiation is effective in BCLC-C hepatocellular carcinoma H.K. Byun 1 , H.J. Kim 1 , J. Seong 1 1 Yonsei Cancer Center, Radiation oncology, SEOUL, Korea Republic of Purpose or Objective Barcelona Clinic Liver Cancer stage C hepatocellular carcinoma (BCLC-C HCC) remains a great clinical challenge due to its dismal prognosis. The aim of this study was to evaluate whether increased radiation dose could improve clinical outcomes when applying liver- directed concurrent chemoradiotherapy (CCRT) to patients with locally advanced BCLC-C HCC. Material and Methods From 2005 to 2016, 637 patients who received liver- directed CCRT for BCLC-C HCC were analyzed. Among them, 466 (73%) had vascular invasion and 225 (35.3%) had multiple lesions. The median tumor size was 10 cm (range, 1-22 cm). Patients were excluded if they had distant metastases or belonged to Child-Pugh class C. Patients were divided according to the biological equivalent dose (BED) of ≥ 72 Gy (range, 72.0-140.0 Gy; n=101) and < 72 Gy (range, 41.4-71.9 Gy; n=536) groups because ROC curve and Maxstat analysis indicated 72 Gy as an optimal cut-off dose predicting local control. Clinical outcomes and toxicities were compared between two groups. Additionally, propensity score matching was performed to reduce bias arising from the retrospective nature of this study. Results The median overall survival (OS) was 16.3 months for all patients with a median follow-up of 12.4 months (range, 0.9-158.8 months). In multivariate analysis, BED as either a continuous variable (per 1 Gy increased; p=0.001) or dichotomized variable (≥ 72 Gy vs. < 72 Gy; p=0.002) was the strongest predictor of improved local control. In addition, BED was a significant factor of survival along with other known prognostic factors such as Child-Pugh class, the size of tumors, the extent of portal vein tumor thrombus, and baseline AFP level. The ≥ 72 Gy group had better local failure-free survival (LFFS) (median, not reached vs. 88.6 months; p < 0.001) and OS (median, 39.7 vs. 14.5 months; p < 0.001) than < 72 Gy group. After propensity score matching, all patient characteristics were balanced between two groups, and the ≥ 72 Gy group still showed improved LFFS (p=0.008) and OS (p=0.002). The outfield failure-free survival and distant metastasis-free survival showed no differences between two groups. Twenty of 101 patients (19.8%) in the ≥ 72 Gy group and 64 of 536 patients (11.9%) in the < 72 Gy group converted to surgery. Patients who underwent conversion

Conclusion V 15 and VS 10

are crucial risk predictors for RIHT (≥1 and

≥2). And the cut-off points of V 15 may provide some clue s for risk stratification. Our NTCP model and nomogram based on V 15 plus pre-CP was validated by an independent cohort. NTCP models or nomograms can help clinical doctors obtain tailored constraints for each patient. It should be noted that the CP-B patients were greater susceptibility to the development of RIHT after SBRT. PV-0105 Stereotactic body radiotherapy treatment for hepatocellular carcinoma: A phase II study J. Durand-Labrunie 1 , H. Jarraya 2 , E. Boleslawski 3 , S. Cattan 4 , T. Lacornerie 5 , D. Peiffert 6 , X. Mirabel 1 1 Centre Oscar Lambret, Département de radiothérapie, Lille, France 2 Centre Oscar Lambret, Département d'imagerie médicale, Lille, France 3 Hôpital Claude Huriez- CHRU de Lille, Chirurgie digestive et transplantation, Lille, France 4 Hôpital Claude Huriez- CHRU de Lille, Maladies de l'appareil digestif, Lille, France 5 Centre Oscar Lambret, Département de physique médicale, Lille, France 6 Centre Alexis Vautrin, Département de radiothérapie, Nancy, France Purpose or Objective Liver transplantation is the standard definitive treatment for non-metastatic hepatocellular carcinoma (HCC). However, less than 5% of patients are ultimately candidates due to frequent comorbidities and graft shortage. Traditionally, radiotherapy played a limited role in the treatment of HCC due to the risk of radiation- induced liver disease. Development of stereotactic body radiotherapy (SBRT) allows delivery of high-dose conformal radiotherapy with limited liver toxicity. The aim of this study was to evaluate the efficiency and safety of SBRT as a local treatment for inoperable HCC. Material and Methods A prospective phase II trial of SBRT for patients with unique HCC was conducted from 2009 to 2014 in three French institutions. Eligibility criteria include d a single HCC lesion; without extra hepatic extension; unsuitable for standard loco regional therapies; with a tumor size from 1 to 6 cm. For patients with cirrhosis, only Child-Pugh (CP) A patients were enrolled. The SBRT dose was 45 Gy in 3 fractions delivered in 8 to 10 days. Primary end point was local control at 18 months, defined as a non-progressive disease of irradiated HCC by RECIST (Response Evaluation Criteria in Solid Tumors). Results A total of 44 patients were evaluable. Patients were 43 to 91 years of age (median 72 year). The median tumor size was 28 mm (range, 10-60 mm). Median follow-up was 3.0 years (range, 0.32 months to 4.3 years). All 44 patients had cirrhosis, 38 (88%) were CP grade A, and 5 (12%) grade B. 5 patients (11%) had portal vein involvement and 17 patients (39%) had a portal and VS 10