Imaging, Diagnosis, Prognosis Validation of Genetic Sequence Variants as Prognostic Factors in Early-Stage Head and Neck Squamous Cell Cancer Survival Abul Kalam Azad 1 , Isabelle Bairati 6 , Elodie Samson 6 , Dangxiao Cheng 1 , Maryam Mirshams 1 , Xin Qiu 5 , Sevtap Savas 1 , John Waldron 3 , Changshu Wang 7 , David Goldstein 4 , Wei Xu 5 , Francois Meyer 6 , and Geoffrey Liu 1,2,8 Abstract Purpose: From the published literature, we identified 23 germ line sequence variants in 17 genes from hypothesis-generating studies that were associated with prognosis of head and neck cancer, including sequence variants of DNA repair (ERCC1, ERCC4, ERCC5, MSH2, XPA, ERCC2, XRCC1, XRCC3), DNA methylation (DNMT3B), cell cycle and proliferation (CCND1, TP53), xenobiotic metabolism (GSTM1, GSTT1, CYP2D6), metastatic -potential (MMP3), immunologic (CTLA4), and growth factor pathways (FGFR4). The purpose of this study was to validate the role of these 23 sequence variants for overall (OS) and disease-free survival (DFS) in a large, comprehensive, well-annotated data set of patients with head and neck cancer. Experimental Design: We genotyped these sequence variants in 531 patients with stage I and II radiation-treated head and neck cancer (originally recruited for an alpha-tocopherol/beta-carotene place- bo-controlled secondary prevention study), and analyzed using Cox proportional hazards models, stratified by treatment arm, adjusting for clinical prognostic factors. Results: Two OS associations were statistically significant for each variant allele when compared with the wild-type: CTLA4:A49G [rs231775; adjusted HR (aHR), 1.32 (1.1–1.6); P ¼ 0.01] and XRCC1:Arg339Gln [rs25487; aHR, 1.28 (1.05–1.57); P ¼ 0.02]. Both of these sequence variants had significant results in the opposite direction as prior published literature. Two DFS associations were of borderline significance in the same direction as prior literature: ERCC2:Lys751Gln [rs13181; aHR, 0.80 (0.6–1.0); P ¼ 0.05] and TP53: Arg72Pro [rs1042522; aHR, 1.28 (1.0–1.6); P ¼ 0.03], comparing number of variant alleles with reference of zero variants. Conclusions: None of the prognostic sequence variants previously published was validated for OS in our patients with early-stage radiation-treated head and neck cancer, though rs1381and rs1042522 had borderline significant association with DFS. Clin Cancer Res; 18(1); 196–206. Ó2011 AACR. Introduction Carcinomas of the oral cavity, pharynx, and larynx are referred to as head and neck cancers; together they account for 2% to 3% of all newly diagnosed cancers in North America (1, 2). Most head and neck cancers are squamous cell carcinomas (SCC) and they vary according to their degree of differentiation to well, moderate, and poor car- cinomas. Some anatomic subsites appear to have better prognosis, but this may be more related to earlier symptoms leading to diagnosis at an earlier stage. Between 33% and 50% of head and neck cancers are early diagnosed at stages I and II (3, 4). In the 1990s, vast majority of patients had a history of smoking and alcohol consumption that was considered to be a significant cofactor. However, increasing evidence has recently documented human papilloma virus (HPV) as a cause of specific subsets of head and neck SCCs (HNSCC; ref. 5). Treatment of early-stage head and neck cancers is usually based on a multidisciplinary approach including radiation therapy and/or surgery, depending on the extent and location of the tumor. However, results with such treatments have not been completely satisfactory due to high mortality and early recurrence of disease. For all stages combined, about 83% of persons with head and neck cancers survive 1 year after diagnosis. The 5-year and 10-year relative survival rates are 61% and 50%, respectively (2). Authors' Affiliations: 1 Applied Molecular Oncology, Departments of 2 Medicine, 3 Radiation Oncology, and 4 Otolaryngology, 5 Biostatistical Unit, Ontario Cancer Institute, Princess Margaret Hospital, University Health Network, University of Toronto, Toronto, Ontario; 6 Centre de Recherche en Cancerologie de l'Universite Laval, l'Universite Laval; 7 D epartement de Radio Oncologie, Centre Hospitalier Universitaire de Qu ebec, Quebec City, Quebec, Canada; and 8 Department of Environmen- tal Health, Harvard School of Public Health, Boston, Massachusetts Note: F. Meyer and G. Liu are co-senior authors. Corresponding Author: Geoffrey Liu, Department of Medical Oncology, Princess Margaret Hospital, Room 7-125, 610 University Avenue, Toronto, Ontario, Canada M5G 2M9. Phone: 416-946-4501, ext. 3428; Fax: 416- 946-6546; E-mail: geoffrey.liu@uhn.on.ca doi: 10.1158/1078-0432.CCR-11-1759 Ó2011 American Association for Cancer Research. Clinical Cancer Research Clin Cancer Res; 18(1) January 1, 2012 196 Downloaded from http://aacrjournals.org/clincancerres/article-pdf/18/1/196/2002397/196.pdf by guest on 13 June 2022