A two-stage genome-wide association study of radiation-induced acute toxicity in head and neck cancer

Abstract

Background: Most head and neck cancer (HNC) patients receive radiotherapy (RT) and develop toxicities. This genome-wide association study (GWAS) was designed to identify single nucleotide polymorphisms (SNPs) associated with common acute radiation-induced toxicities (RITs) in an HNC cohort.

Methods: A two-stage GWAS was performed in 1279 HNC patients treated with RT and prospectively scored for mucositis, xerostomia, sticky saliva, and dysphagia. The area under the curve (AUC) was used to estimate the average load of toxicity during RT. At the discovery study, multivariate linear regression was used in 957 patients, and the top-ranking SNPs were tested in 322 independent replication cohort. Next, the discovery and the replication studies were meta-analyzed.

Results: A region on 5q21.3 containing 16 SNPs showed genome-wide (GW) significance association at P-value < 5.0 × 10^-8 with patient-rated acute xerostomia in the discovery study. The top signal was rs35542 with an adjusted effect size of 0.17*A (95% CI 0.12 to 0.23; P-value < = 3.78 × 10^-9). The genome wide significant SNPs were located within three genes (EFNA5, FBXL17, and FER). In-silico functional analysis showed these genes may be involved in DNA damage response and co-expressed in minor salivary glands. We found 428 suggestive SNPs (P-value < 1.0 × 10^-5) for other toxicities, taken to the replication study. Eleven of them showed a nominal association (P-value < 0.05).

Conclusions: This GWAS suggested novel SNPs for patient-rated acute xerostomia in HNC patients. If validated, these SNPs and their related functional pathways could lead to a predictive assay to identify sensitive patients to radiation, which may eventually allow a more individualized RT treatment.

Keywords: GWAS; Head and neck cancer; Radiation-induced toxicity; Radiogenomics.

Fig. 1

The dispersion of AUCs and STAT scores of acute radiotoxicity endpoints in HNC patients by discovery and replication cohorts. The X-axis shows the endpoints, and Y-axes show the measured value of the AUC and STAT score of endpoints. The Brown color represents the discovery study, and the grey color represents the replication study. The black border shows physician-rated endpoints, and the orange border shows patient-rated endpoints. The lowest line represents the minimum (Q0 or 0th percentile): the top line represents the maximum (Q4 or 100th percentile) data point excluding any outliers; The middle line represents the median (Q2 or 50th percentile); the box represents the interquartile range (IQR) which is the distance between the first quartile (Q1 or 25th percentile; that is the median of the lower half of the dataset) and the third quartile (Q3 or 75th percentile that is the median of the upper half of the dataset). Table 2 shows the comparison between discovery and replication studies for the significant difference in outcomes’ distribution

Fig. 2

Genome-wide association findings for AUC for patient-rated toxicity xerostomia in HNC patients. 2A. Manhattan plot: The X-axis shows the location in the genome. The Y-axis shows − log10 P-values for the association of each of the tested SNPs with the outcome. The red line shows the threshold for genome-wide significance (P < 5 × 10–8), and the blue line shows the suggestive threshold (P < 1 × 10–5). 2B. Quantile–quantile (QQ) plot comparing the distribution of observed P-values (test statistics) from discovery study to the distribution of expected P-value based on the theoretical probability distribution, inflation of plot to upper part suggest inflation of test statistics due to the possibility of population substructure or type 1 error (small sample size bias). The Y-axis shows observed − log10 P-values, and the X-axis shows the expected − log10 P-values. Each SNP is plotted as a dark blue dot, and the red line indicates a null hypothesis of no true association. Deviation from the expected P-value distribution is evident only in the tail area, with a lambda of 1.001, suggesting that population stratification was adequately controlled. 2C. Locuszoom plot of the associated region on chromosome 5. The blue circle (query variant) points to the top SNP (rs35542). Points representing nearby SNPs are color-coded according to linkage disequilibrium r2 value as indicated in the legends. The X-axis shows the genomic coordinates chromosome 5. The Y1 axis shows − log10 P-values for each of the SNPs in the genome. The Y2 axis shows the combined recombination rate which is estimated from the international HapMap project

Fig. 3

Single and multi-gene expression visualization across all tissues, obtained from GTEx. 3A: Single expression visualization of EFNA5 gene, as the nearest gene, is shown. The Y-axis shows the expression density of EFNA5 measured as linear count of transcript per million (TPM), and the X-axis shows the tissues sorted as a decrease in the median of TPM. Box plots are shown as median and 25th and 75th percentiles; points are displayed as outliers if they are above or below 1.5 times the interquartile range. 3B: A heat map of multi-gene expression visualization for EFNA5, FBXL17, and FER genes is shown. The genes and tissues are clustered using hierarchical clustering. The expression level of the gene per tissue is color-coded according to the linear count of TPM as indicated by the legend. The red box shows sub-clusters of high expression levels of the three genes in secretory tissues, including the minor salivary gland