6th ICHNO Abstract Book

6th ICHNO 6 th ICHNO Conference International Conference on innovative approaches in Head and Neck Oncology 16 – 18 March 2017 Barcelona, Spain __________________________________________________________________________________________ Institute of Oncology, Department of Pathology, Gliwice, Poland page 17

stratified by trial and adjusted on gender, age, T-stage, N-stage, radiotherapy fractionation schedule (standard, altered), p16 (positive, negative) and smoking status (never/former, current). Both prognostic and predictive effects were estimated for progression-free survival (PFS, primary endpoint) and overall survival (OS). Only prognostic effect was estimated for locoregional control and OS after first failure. Results Data and tumor tissue from 815 patients enrolled in 4 trials (DAHANCA 6-7, RTOG 9003, ARTSCAN, RTOG 0129) was available for analysis: 350 (43%) p16 negative and 465 (57%) p16 positive. Patients with p16-pos tumors were significantly more likely to be younger (mean: 56 vs. 59 years, p=0.0002), have better performance (PS=0: 74% vs. 50%, p<0.0001), smaller tumors (T1-2: 46% versus 33%, p<0.0001), more advanced N-stage (N+: 87% versus 76%, p<0.0001) and were more often never or former smokers (77% versus 36%, p<0.0001) compared with the p16-neg subgroup. p16-status had a significant prognostic effect, with better PFS and OS in p16 positive patients: Hazard Ratio (HR)=0.42 [95% Confidence Interval (CI): 0.34;0.51] with a 37% absolute increase at 5 years for PFS, 0.40 [0.32;0.49] with a 39% absolute increase at 5 years for OS. Locoregional control and OS after first failure were also significantly better in p16 positive patients with HR of 0.31 [0.22;0.44] and 0.64 [0.47;0.87], respectively. Smoking also independently influenced outcome and never/former smokers had better prognosis than current smokers with HR of 0.61 [0.50;0.75] and 0.58 [0.47;0.72] for PFS and OS, respectively. Heterogeneity between trials was observed for PFS (p=0.03), and OS (p=0.02). Use of random effect model did not change conclusions. There was no statistical significant interaction between HPV status and fractionation schedule (p=0.45 for PFS and p=0.58, for OS) and as such no predictive impact of HPV could be demonstrated. Conclusion The significant prognostic effect of HPV in OPC was confirmed in this pooled analysis based on individual patient data, but HPV status was not found to be predictive of outcome after altered fractionated radiotherapy. Supported by PAIR-VADS. PD-029 Causes-of-death underestimate the burden of head and neck cancers in France (EPICORL study) Y. Pointreau 1 , L. Sagaon Teyssier 2 , L. Geoffrois 3 , M. Bec 4 , C. Godard 4 , C. Even 5 , L. Lévy-Bachelot 4 , F. Huguet 6 , S. Témam 5 , M. Schwarzinger 2 1 Centre Jean Bernard, Oncology, Le Mans, France 2 THEN Translational Health Economics Network, Public health, Paris, France 3 Institut de Cancérologie de Lorraine - Alexis Vautrin, Medical oncology, Vandoeuvre Les Nancy, France 4 MSD France, Market access, Courbevoie, France 5 Institut Gustave Roussy, Head & Neck Surgical & Medical Oncology, Villejuif, France 6 Tenon Hospital, Radiation Oncology, Paris, France Purpose or Objective : Patients with head and neck (H&N) cancer carry the highest risk of secondary primary cancers. Determining the underlying cause of death is conflicting in presence of multiple primary cancer sites, and the burden of H&N cancers may be underestimated by causes-of-death statistics. Material and Methods : Using the French National Hospital Discharge (PMSI) database, we identified all adult patients residing in metropolitan France and diagnosed with squamous cell carcinoma at hospital (ICD-10: C00-C06; C09-C14; C30.0; C31; C32) in 2008-2012. Among patients who died at hospital, we considered advanced H&N cancer (stage III/IV at diagnosis or disease progression) as a cause of death. We recorded competing causes of death from

Purpose or Objective Circulating free HPV DNA (cfHPV DNA) could be found in patients with HPV-related HNSCC. Amount of cfHPV DNA may be related to treatment respond. If the response to treatment is reflected in cfHPV DNA detection, it may become a marker of treatment results. cfHPV DNA estimation after radiotherapy (RT) or radiochemotherapy in the relation to treatment results has been presented. Material and Methods Between 2012 and 2016 blood from 540 consecutive patients with HNSCC before definitive RT/CHRT was collected. If HPV-positive sample was identified, serial blood collections were taken after treatment and during follow-up. cfHPV DNA assessment was correlated with treatment results. cfHPV DNA complete remission (cfHPVrem) was defined as not detectable cfHPV DNA after RT/CHRT. cfHPV DNA recurrence (cfHPVrec) was defined as a detection of cfHPV DNA in previously cured patients during follow-up. Results In 71/540 (13%) of patients cfHPV DNA was found. At the end of treatment in all 70(98%) cured patients cfHPVrem was found. In 1 uncured patient cfDNA was still detectable. During follow up cfHPVrec was found in 9(13%) patients. Despite of no evidence of locoregional recurrence of disease on physical examination nor on imaging diagnostic (TK or MRI) PET scanning was additionally performed in these patients. PET revealed locoregional recurrence and metastatic disease in 5(7%) and 4(6%) patients respectively. Conclusion In patients with HPV-related HNSCC the presence of cfHPV DNA in blood reflect active cancer. Serial estimation of cfHPV DNA in cured patients may help to diagnose early treatment failure because cfHPVrec strongly suggest disease recurrence. PD-028 Prognostic and predictive impact of HPV status in oropharyngeal cancer: the MARCH-HPV project B. Lacas 1 , P. Lassen 2 , A. Trotti 3 , B. Zackrisson 4 , Q. Zhang 5 , J.P. Pignon 1 , J. Overgaard 2 , P. Blanchard 6 1 Gustave Roussy, Ligue Nationale Contre le Cancer meta- analysis platform - Department of Biostatistics and Epidemiology,Villejuif,France 2 Aarhus University Hospital, Department of Experimental Clinical Oncology, Aarhus, Denmark 3 Moffitt Cancer Center, Department of Radiation Oncology,Tampa-Florida,USA 4 Umeå University, Department of Radiation Sciences - Oncology,Umeå,Sweden 5 NRG Oncology, Statistics and Data Management Center, Philadelphia,Pennsylvania, USA 6 Gustave Roussy, Department of Radiotherapy, Villejuif, France Purpose or Objective HPV-induced oropharyngeal cancers (OPC) have a better prognosis compared to classical squamous cell carcinoma of the head and neck (HNSCC). Whether HPV status is associated with a higher sensitivity to treatment modifications such as altered fractionation radiotherapy remains controversial. Material and Methods The MARCH-HPV project is based on the first update of the Meta-Analysis of Radiotherapy in HNSCC (MARCH), which included 33 trials and 11833 patients (Blanchard et al. Rad Oncol; 111 (Suppl 1): 32). HPV status was determined according to p16 immunohistochemistry. The HPV analysis was restricted to patients with OPC and performed using a Cox model