ESTRO 36 Abstract Book

S598 ESTRO 36 _______________________________________________________________________________________________

Material and Methods A total of 110 HNSCC patients undergoing chemotherapy with Cetuximab were retrospectively analyzed. Patients were treated in our Institution between February 2007 and May 2016. Performance Status (PS) evaluated with the Eastern Cooperative Oncology Group (ECOG) scale. Relevant comorbidity were evaluated with the Charlson Comorbidity Index (CCI). Skin toxicity was evaluated according to Common Terminology Criteria for Adverse Events (CTCAE) version 4.03. Results The median age was 67 years (range 32-84 years); 83 (75.5%) patients were male. Ninety (82%) were smokers, most of them had a smoking history > 20 pack/years. PS was scored ≤ 1 in 100 (90.9%) patients. The median CCI was 6 (range 1- 13). The most represented head and neck subsites in the study were the oropharynx (27%) and the larynx (32%). Fourty-three (39.1%) patients underwent a curative treatment with a concomitant CTX-RT treatment, while 67 (60.9%) received a CTX-containing chemotherapy regimen. In patients undergoing a curative treatment, the median number of CTX cycles administered amounted to 6 (range 2-8) ; median RT dose was 67Gy (range 66-70). Acute mucositis G≤2 and G3 was observed in 24 (55,8%) and 18 (41,9%) patients respectively. Actinic dermatitis G≤2 was observed in 25 (58,1%) patients, while G3 and G4 dermatitis was present in 17 (39,5%) and in 1 (2.3%) patient, respectively. In the advanced disease subgroup, CTX was administered as first- line CT in 39 cases (53.6%) as a triplet combination (cis/carboplatin, 5-FU, CTX), and in 21 (26.8%) as a 2-drugs regimen (carboplatin / CTX); 7 patients (19.6%) received CTX monotherapy. Incidence of acneiform rash G1-G2 was 37.3%, while G3 was 3.0%(25 and 2 patients, respectively); incidence of hypomagnesemia G1-G2 was 4.5% (3 patients), no G3 toxicity was reported. Infusion reactions G2 was reported in 2 (3.0%) patients. Conclusion Cetuximab is a mainstay in the radical and palliative treatment in HNSCC patients but its toxicity in our serie is not negligible, though in line with data from literature. Accurate patients selections and toxicity management during treatment are mandatory to ensure safety and clinical benefit. EP-1088 EGFR as blood biomarker improves prognostic value of nomogram for survival in laryngeal carcinoma F.W.R. Wesseling 1 , G. Feliciani 1 , C. Oberije 1 , M.P. Van de Waarenburg 2 , C.G. Schalkwijk 2 , P. Lambin 1 , F.J. Hoebers 1 1 Maastricht Radiation Oncology MAASTRO clinic- GROW - School for Oncology and Developmental Biology- Maastricht University Medical Centre, Department of Radiation Oncology, Maastricht, The Netherlands 2 Laboratory for Metabolism and Vascular Medicine- Maastricht University, Department of Internal Medicine, Maastricht, The Netherlands Purpose or Objective To improve the prognostic value of a previously validated nomogram for laryngeal carcinoma derived from clinical parameters 1 by adding blood biomarker results for hypoxia, inflammation, tumor load and cell growth. Material and Methods For 50 patients treated for laryngeal carcinoma at our clinic by radiotherapy with curative intent, blood samples were stored in a blood biobank ( ClinicalTrials.gov: NCT01084785) before start of treatment. Eight biomarkers representing hypoxia, inflammation, tumor load and cell growth were selected and measured with commercially available kits: IL6, IL8, CEA, c-kit, E-cad, EGFR, MMP-9 and osteopontin. Primary outcome was overall survival (OS). Blood biomarker results and a

prognostic score, based on our previously developed nomogram were entered in a Cox regression analysis using least absolute shrinkage and selection operator (LASSO) variable selection procedure with 10-fold cross validation. Significant biomarkers were added to the nomogram score to investigate whether the prognostic value would be improved. Concordance index (C-index) was used to evaluate the performance of the model. Results Clinical staging was: St. I: 18%, St. II: 20%, St. III: 34% and St. IV: 28%. Most patients were treated with accelerated locoregional radiotherapy (68%) or local radiotherapy (16%). After a median follow up of 56 months, 2 and 5 year OS was 75% and 58%. LASSO procedure selected EGFR and the nomogram score as variables significantly associated with OS. Spearman ρ coefficient and visual inspection showed no correlation between EGFR and nomogram score (ρ=0.29). After validating the existing nomogram with this new cohort, its performance was comparable to the published data 1 (C-index = 0.68). By performing multivariate Cox regression, the addition of EGFR to the nomogram improved its concordance index to 0.73 as shown in table 1. Interestingly we found that high-risk patients had lower EGFR levels in blood samples compared to low-risk patients. This is opposite to results obtained from tissue samples as reported in literature. 2

Conclusion EGFR as blood biomarker is an independent predictor for OS of patients with laryngeal carcinoma. OS curves for patients in high vs low EGFR group are visible in figure 1. Based on this small cohort the prognostic value of our existing nomogram was improved, but an independent cohort would be needed for external validation. Further research is needed to correlate EGFR blood values with EGFR expression in tumor samples.

References: 1. Egelmeer AGTM et al. Development and validation of a nomogram for prediction of survival and local control in laryngeal carcinoma patients treated with radiotherapy alone: A cohort study based on 994 patients. Radiother Oncol . 2011;100(1):108.