1021P - Analysis of genome alterations in head and neck squamous cell carcinoma to identify potential prognosis biomarkers or targetable genetic aberrations

Date 29 September 2014
Event ESMO 2014
Session Poster Display session
Topics Head and Neck Cancers
Translational Research
Presenter Benjamin Linot
Citation Annals of Oncology (2014) 25 (suppl_4): iv340-iv356. 10.1093/annonc/mdu340
Authors B. Linot1, O. Capitain1, I. Valo2, A. Septans3, P. Augereau1, A. Renoult4, S. Fronteau4, A. Morel4
  • 1Medical Oncology, Institut de cancérologie de l'ouest, 49000 - Angers/FR
  • 2Pathology Department, Institut de cancérologie de l'ouest, 49000 - Angers/FR
  • 3Statistics And Data Management, Institut de Cancérologie de l'Ouest, Angers/FR
  • 4Pharmacology Department, Institut de cancérologie de l'ouest, 49000 - Angers/FR

Abstract

Aim

It is crucial to delineate the aberrant growth signaling pathways in head and neck squamous cell carcinoma (HNSCC) and develop specific targeted therapies to improve the treatment outcome.

Methods

Formalin-fixed, paraffin-embedded (FFPE) surgically resected tumor samples from 50 patients with previously untreated resectable HNSCC were evaluated. All patients benefited from complete surgery with post-operative radiotherapy. We used targeted, massively exome sequencing (Life Technology, Ion Torrent) for 30 cancer relevant genes. Mutations were modeled bioinformatically using the CHASM prediction algorithm to retain single nucleotide variants with low CHASM scores consistent with oncogenic driver characteristics. We assessed by immunohistochemistry (IHC) the expression of p16, PTEN, cMet and Ki67. Univariate and multivariate analyses were performed to evaluate predictive variables (protein expression and integrated genomic alterations) for disease-free (DFS) and overall survival (OS).

Results

The median age was 57.7 years (range: 35-84) with tobacco smoking and alcohol consumption in 90%. The median DFS and OS were respectively 13.8 months [8.8; 16.5] and 20.5 months [CI 95%: 15.4; 27.3] with 8.2% [CI 95%: 2.2%; 19.3%] as 5-year OS rate (p > 0.05). PTEN, cMet, p16 expression and Ki67 assessed by IHC were not independent factors for poor DFS or OS in multivariate analysis in this small sample, but classically PS > 0 (p = 0.0102) and stage > T1 (p = 0.0073) were identified as such. Over 10 significantly mutated genes (CHASM p-value <0.01) were identified including TP53, ERBB2, PDGFRA, MET and some of PI3K family members such as PIK3CA (24%). Potential genomic alteration drivers in TP53, ERBB2 and IGF1R decreased significantly DFS and OS in multivariate analysis (p < 0.05).

Conclusions

HNSCC is a heterogeneous group and the understanding of biomarkers is not as well defined as in other tumor types. We identified multiple mutation drivers including potential targetable genes such as ERBB2 and IGF1R who decrease OS and PFS. Initiation of clinical trials adapted to individual genomic alterations will be necessary to drive therapeutic targeting.

Disclosure

All authors have declared no conflicts of interest.