ESTRO 2020 Abstract book

S505 ESTRO 2020

therapy, but also other complications such as other malignancies and chronic obstructive pulmonary disease, and we haven't offered radiotherapy to all cancer patients of active smokers since 2009. Now, we evaluate the outcome of definitive radiation therapy for early stage glottis carcinoma before and after prohibiting smoking. Material and Methods Between 2005 and 2016, 103 patients (male/female =98/5) with early glottis carcinoma (T1N0/T2N0 = 79/24) treated by conventional radiotherapy were reviewed by independent two radiation-oncologists. The median age was 70 year-old (ranging from 46 to 86). The pathology was all squamous cell carcinoma. The prescribed dose was 60 to 70 Gy (median EQD 2 66Gy). The energy of X-ray was 4MV (until 2010) or 6MV (since 2011). Concurrent chemotherapy or molecular target therapy was done in 36 patients. From 2009 we checked that patients cessation of smoking only by medical interviews, and we confirmed it using smokerlyzer® (Bedfont Scientific Ltd. Kent. UK), which detects expiratory carbon monoxide, to monitor their smoking cessation more strictly from 2014 to 2016. The number of patients at diagnosis years was 48 patients from 2005 to 2008, 31 patients from 2009 to 2013, and 24 patients from 2014 to 2016, respectively. We defined each groups, patients from 2005 to 2008 as “freely cigarette”, from 2009 to 2013 as “incomplete cessation”, and from 2014 to 2016 as “complete cessation”. We analyzed local relapse rate as primary endpoint, and over all survival (OS) and disease free survival (DFS) as secondary endpoints. Results The median follow-up period was 53.1 months (range, from 1.9 to 141.0 months). Local control rate at 3 years was 83%, and 3-year local control rates before and after prohibiting smoking were 76% (n=48) and 91%(n=55), respectively. The 2-year and 5-year relapse rate of all patients were 15.6% and 18.8%. The 2-year relapse rate of complete cessation group was 5.3%, and tended to be lower than incomplete cessation group (13.7%) and freely cigarette group (21.2%) (p=0.24). There was no difference in the local control rate between with and without chemotherapy or molecular target therapy. In multivariate analyses, relapse rate and DFS of complete cessation group were significantly better than freely cigarette group (HR, 0.63 and 0.26 respectively, p <0.05).

Conclusion We confirmed the significant differences of the local control and survival between the complete smoking cessation patients and freely smoking patients by monitoring expiratory carbon monoxide. We think that the monitoring expiratory carbon monoxide is suitable for clinical working because it can supply answers in real-time and considering the biological effects such as tissue hypoxia. PO-0854 Prediction of sensorineural hearing loss in patients undergoing proton therapy for skull base tumors N. Roos 1 , F. Pfiffner 2 , M. Walser 1 , A. Pica 1 , D. Leiser 1 , U. Kliebsch 1 , S. Von Felten 3 , D.C. Weber 1 , D. Veraguth 2 , B. Bachtiary 1 1 Paul Scherrer Institute, Center for Proton Therapy, Villigen PSI, Switzerland ; 2 University Hospital Zurich, Departement of Otorhinolaryngology- Head and Neck Surgery, Zurich, Switzerland ; 3 University Zurich, Epidemiology- Biostatistics and Prevention Institute, Zurich, Switzerland Purpose or Objective Sensorineural hearing loss is a common complication after radiotherapy in patients with skull base tumors and has a significant impact on the quality of life. Proton therapy (PT) can contribute to reducing the dose to the cochlea and consequently the risk of hearing loss, in particular when a high dose of radiation to the tumor is necessary. Although there are certain guidelines for the threshold dose to the cochlea, in the daily clinical routine an accurate estimation of hearing loss after proton therapy is difficult due to additional intrinsic and extrinsic factors. In order to better estimate the hearing loss to be expected, we retrospectively determined the effect of proton beam irradiation on the hearing function in patients who had skull base tumors. Material and Methods This retrospective study include 51 patients (median age, 48.5 years) with chordoma (n=24), chondrosarcoma (n=15), head and neck tumors (n=9) and meningioma (n=3). There were 21 male and 30 female patients and all underwent PT. All patients had a Pure-Tone Audiometry before and after PT. As a measure for hearing loss the Pure-Tone Average (PTA) of 500Hz, 1.000Hz, 2.000Hz and 4.000Hz was calculated for both ears (n=102) separately. A hierarchical linear mixed-effects model was chosen to simultaneously account for co-variables. Results The median tumor dose was 72 Gy(RBE), (range, 52.2-77.8) and the median dose to the cochlea was 33.9 Gy(RBE) (range, 0.03 – 76.7). At baseline, 23 (22.5%) ears had sensorineural, 4 (3.9%) ears conductive and 7 (6.9%) ears mixed hearing loss. In 20 ears only the air conductivity was measured and the loss of hearing could be caused by obstruction or damage to the outer, middle and/or inner ear. The median PTA before PT (PTA.bl) was 15 dB (IQR 10-25) and the median PTA after PT was 26.25 dB (IQR 11.3 - 45). The median time interval between the end of PT and first audiometry was 279 days. Some patients were followed over a longer period of time with multiple hearing tests. In multivariate analysis, PTA.bl (p=<0.01), mean dose to cochlea (p=<0.01), time since treatment (p=0.01) and age (p=0.02), were independent prognostic factors for