Abstract 4743: Analysis of a protein expression biomarkers in HPV negative SCCHN

Abstract

Abstract Background: Human papillomavirus negative (HPV) (-) squamous cell head and neck cancer (SCCHN) is associated with a poor prognosis and decreased response to treatment. We sought to study the correlation between protein-based biomarkers related to DNA repair, mitotic regulators and apoptosis in a tissue microarray (TMA) of HPV(-) SCCHN. Methods: TMAs were constructed with SCCHN samples from patients treated at Fox Chase Cancer Center from 1990-2002. The TMA contained 85 p16 -surgical specimens from predominantly oral cavity tumors; 72 received adjuvant radiation. Slides were stained by a modified indirect immunofluorescence method. Sections were incubated with antibodies against ERCC1 (HPA029773, Sigma), Aurora A, (Bethyl Laboratory, IHC-00062), pAuroraA ( Bethyl Laboratory, IHC-00067), survivin (Novus Biological, NB100-56168) and a wide-spectrum screening mouse cytokeratin ab (Dako M3515). Prolong Gold mounting medium (P36931; Molecular Probes) containing 4,6-Diamidino-2-phenylindole (DAPI) defined nuclei. A tumor mask (TM) image was created from the cytokeratin stain of each histospot. ERCC1 levels were measured using fluorescent IHC on the HistoRx PM-2000 image analysis platform and data analyzed using AQUA algorithms. We assessed relationships between markers using Spearman's correlation, explored associations between markers and overall survival using Kaplan-Meier curves, and tested significance using the t-statistic. Parallel analysis of The Cancer Genome Atlas (TCGA) resources was used to establish whether correlative relationships observed at the protein level reflected commensurate changes in mRNA expression, in an independent study cohort. For this purpose, mRNA expression values were imported from the (provisional) TCGA data set encompassing 295 SCCHN samples, and Spearman's correlation coefficients were calculated. Results: Correlation coefficients indicated a significant positive association between nuclear expression of the following markers: ERCC1 and AuroraA (p<0.0001); ERCC1 and pAurora (p=0.0027); Aurora A and pAurora (p<0.0001); survivin and Aurora A (p=0.0064); and survivin and ERCC1 (p=0.0084). Review of mRNA levels of Aurora A and survivin in the TCGA database indicated a highly significant correlation between these two markers (r=0.59, p=0.0002). This striking correlation did not represent common amplification of a chromosomal locus, given distinct chromosomal locations of Aurora A (20q13), and survivin (17q25). The relation of elevated expression of the biomarkers to prognosis was analyzed, considering biomarkers alone and in combination. A particularly important role for survivin was established, independent of other variables. Conclusions: ERCC1, Aurora A, and survivin displayed significantly correlated expression in HPV (-) SCCHN, supporting linked function in SCCHN. This works suggests a cluster of biomarkers worthy of future study. Citation Format: Ranee Mehra, Donghua Yang, Elizabeth Handorf, Ilya Serebriiskii, Eric Ross, Miriram Lango, John A. Ridge, Erica Golemis, Barbara Burtness. Analysis of a protein expression biomarkers in HPV negative SCCHN. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4743. doi:10.1158/1538-7445.AM2014-4743