Radiotherapy (RT) is integral in the treatment of head and neck cancer (HNC). Tumors have varying response to RT. Quantifying gene expression and correlating it with outcome can identify genes that influence a tumor's response to radiation. Our goal is to identify the genes predictive of locoregional recurrence (LRR) in HNC patients treated with RT.
Patient data was abstracted from a prospectively maintained institutional Total Cancer Care (TCC) database and chart review. All microarray chips were normalized using iterative rank-order normalization method. A supervised Gene Set Enrichment Analysis (GSEA) was performed to determine whether a priori defined gene pathways shows statistically significant, concordant differences between groups of samples obtained before and after radiation therapy. 50 well-annotated hallmark gene sets were obtained from Molecular signatures database v.5.1 for GSEA analysis. Linear Models for Microarray Data (LIMMA) was used to identify individual genes with differential expression between the two groups. To explore survival-associated genes, Cox proportional hazard regression analysis was performed with time to local-regional recurrence data of patients who underwent radiotherapy. Further, Spearman's rank correlation was calculated between expression and survival fraction after 2Gy RT (SF2) of cancer cell lines in NCI-60 panel. Two-sided false discovery rate (FDR) adjusted p-value less than 0.05 is considered statistically significant.
A total of 108 patients were analyzed, of which 49 (45%) were sampled prior to and 59 (55%) after receiving RT. There were a total of 48 (44%) LRRs. Thirty eight genes were identified in common with a differential expression in NCI 60, prior receipt of RT, and cox regression analysis of LRR. Upregulation of 26 genes were associated with LRR, the majority of which (22/26; 85%) were significantly elevated in the samples that were previously radiated. Downregulation of 12 genes were associated with LRR, of which 8 (67%) were significantly lower in samples that were previously radiated.
We identify a number of genes with alteration in expression, which may be associated with resistance to radiotherapy in HNC. Further validation is necessary in other patient cohorts as well as in vitro studies.
Clinical trial identification
Legal entity responsible for the study
Jimmy J. Caudell
Moffitt Cancer Center
All authors have declared no conflicts of interest.