ESTRO 38 Abstract book

S375 ESTRO 38

Livi 1 1 Azienda Ospedaliero-Universitaria Careggi, Radiation Oncology, Florence, Italy Purpose or Objective The definition of primary tumor clinical target volume (CTV) for HNSCC has been recently reviewed in international guidelines. The aim of our work was to assess the impact of two different delineation approaches (anatomical vs geometric expansion) on the clinical outcome of patients with oropharyngeal cancer (OPC). Material and Methods We performed a retrospective analysis of consecutive patients treated at our institution for OPC between January 2010 and April 2018. In 2014, our local policy to delineate high risk (HR) CTV for both primary (CTV-p) and nodal (CTV-n) GTVs shifted from an anatomical expansion (AE) to a geometric expansion (GE) modality, described as follows: - AE: CTV-p consisted of the primary GTV plus the whole oropharyngeal anatomical compartment, edited for air cavities and anatomical barriers; CTV-n consisted of the nodal GTVs plus the corresponding lymph node level (delineated according to Gregoire’s classification), including also the level immediately adjacent to it - GE: CTV-p consisted of the primary GTV plus an isotropic 5 mm margin, manually edited for air cavities and anatomical barriers; CTV-n consisted of the nodal GTVs plus an isotropic 7 mm margin, manually edited to take into account anatomical barriers (i.e, sternocleidomastoid muscle, if not infiltrated). Of note, an additional 5 mm margin to the CTV-p was not added. All patients received IMRT with 3 dose levels: the high, intermediate and low risk CTVs received 70, 60 and 50 Gy or 69.9, 59.4 and 52.8 Gy in 35 or 33 fractions, respectively, depending on whether a sequential or SIB technique was employed. A 3-5 PTV margin was added to CTV’s (3 mm in case of daily image guidance). HPV status was not routinely available until 2015. Locoregional control (LRC) was defined as the time from the last treatment day to local or nodal failure or both. Progression-free survival (PFS) was defined as the time from the last treatment day to disease progression or death from any cause. Overall survival (OS) was defined as the time from OPC diagnosis to death from any cause. Median LRC, PFS and OS were calculated. The relative estimates of LRC and PFS at 24 months were estimated by the Kaplan-Meier method. Results A total of 116 patients were included (table 1). In terms of HR CTV expansion modality, 43 (37.1%) were treated with AE and 63 (62.9%) with GE, respectively. One patient in the AE cohort was lost to follow-up after first evaluation, therefore 115 patients are evaluable for outcome analysis. At a median follow-up of 26 months (range: 1-86), 87 patients were alive (75.6%). The median LRC, PFS and OS for the whole group were 24 (95% CI: 25.3- 33.8), 24 (95% CI: 24.6-33.2) and 31 (95% CI: 30.7-38.6) months, respectively. There was no difference between the AE and GE cohorts in terms of 2-year LRC (86.9% vs 94.3%; p= 0.28) (figure 1) and PFS (68.8% vs 80.9%, p=0.56).

(penetration/aspiration) at 6 months after treatment, scored with VF (PAS6). From all baseline features, a set of candidate variables was selected based on plausible relevance in relation to PAS6. Average doses (D mean ) to SWOARs were collected. Additionally, univariable analysis with correction for cross correlation of candidate variables was performed. Multivariable logistic regression with stepwise forward elimination and internal validation, using bootstrapping, was performed to correct for optimism of the model for PAS6. Results The prevalence of PAS6 was 43% (73 patients). The best performing model included 4 independent prognostic factors for PAS6: PAS at baseline, T-stage, N-stage and D mean to the supraglottic larynx . Figure 1. After internal validation using bootstrapping the model performance (Nagelkerke R 2 : 0,398; ROC-AUC: 0.818) and calibration were good.

Conclusion Patients with more advanced disease are at the highest risk of penetration/aspiration after (CH)RT. This risk increases substantially in case of penetration/aspiration observed at baseline VF. Moreover, increasing dose to supraglottic larynx contributes to further escalation of the risk of penetration/aspiration 6 months after (CH)RT. PO-0731 Transition from anatomical to geometric expansion modality for high-risk CTV in oropharyngeal cancer P. Bonomo 1 , I. Desideri 1 , C. Becherini 1 , L. Visani 1 , V. Salvestrini 1 , M. Mariotti 1 , P. Garlatti 1 , L. Dominici 1 , L.