Seven-CpG-based prognostic signature coupled with gene expression predicts survival of oral squamous cell carcinoma

Abstract

Background: DNA methylation has started a recent revolution in genomics biology by identifying key biomarkers for multiple cancers, including oral squamous cell carcinoma (OSCC), the most common head and neck squamous cell carcinoma.

Methods: A multi-stage screening strategy was used to identify DNA-methylation-based signatures for OSCC prognosis. We used The Cancer Genome Atlas (TCGA) data as training set which were validated in two independent datasets from Gene Expression Omnibus (GEO). The correlation between DNA methylation and corresponding gene expression and the prognostic value of the gene expression were explored as well.

Results: The seven DNA methylation CpG sites were identified which were significantly associated with OSCC overall survival. Prognostic signature, a weighted linear combination of the seven CpG sites, successfully distinguished the overall survival of OSCC patients and had a moderate predictive ability for survival [training set: hazard ratio (HR) = 3.23, P = 5.52 × 10^-10, area under the curve (AUC) = 0.76; validation set 1: HR = 2.79, P = 0.010, AUC = 0.67; validation set 2: HR = 3.69, P = 0.011, AUC = 0.66]. Stratification analysis by human papillomavirus status, clinical stage, age, gender, smoking status, and grade retained statistical significance. Expression of genes corresponding to candidate CpG sites (AJAP1, SHANK2, FOXA2, MT1A, ZNF570, HOXC4, and HOXB4) was also significantly associated with patient's survival. Signature integrating of DNA methylation, gene expression, and clinical information showed a superior ability for prognostic prediction (AUC = 0.78).

Conclusion: Prognostic signature integrated of DNA methylation, gene expression, and clinical information provides a better prognostic prediction value for OSCC patients than that with clinical information only.

Keywords: Gene expression; Methylation; Oral squamous cell carcinoma; Overall survival; Prognostic signature.

Fig. 1

Flow chart indicating study design. We identified candidate CpG sites from 32 paired OSCC and adjacent non-tumor tissues by methylation 450k assay in the discovery set. Then, we excluded a large proportion of CpG sites that were unrelated to survival and developed prognostic scores by SIS. The seven-CpG-based classifier was validated in two independent datasets. Relationships between methylation and gene expression were also analyzed in the training dataset

Fig. 2

Construction of the seven-CpG-based classifier. a Circos plot of epigenome-wide DNA methylation CpG sites. Results are presented as P values ordered by genomic position, including paired t test of the discovery set (green and red symbols) and univariable Cox regression analysis of the training set (orange and blue symbols). b Volcano plot comparing CpG methylation for OSCC tumor and non-tumor tissues. A total of 1490 CpG sites had an absolute value of differential methylation of > 0.4 and a paired t test P value of < 1 × 10⁻⁷ (blue dots). c Heatmap showing methylation of 15 CpG sites in tumor tissues and adjacent non-tumor tissues. d Coefficients of CpG sites calculated by univariate Cox regression and sure independence screening (SIS). After SIS selection, seven probes remained non-zero coefficients

Fig. 3

Prognostic signature and OSCC patient survival. Left panels show Kaplan-Meier survival analyses of patients, which are categorized into low-risk and high-risk groups using a cutoff value of 0.02, for the a training set, b validation set 1, and c validation set 2. P values were calculated using log-rank test, and HR indicates hazard ratio. Right panels show time-dependent ROC curves of different months used to evaluate patient survival, with risk score using the nearest neighbor method

Fig. 4

Subgroup and stratification analysis of the seven-CpG-based signature. Subgroup analysis for HPV+ cases (a) and HPV− cases (b) in the imputed combined dataset. c Kaplan-Meier curves plotting overall survival of the combined three datasets for respective prognostic score categories. d Subgroup analysis with clinical stage of the combined training set and validation set 2

Fig. 5

Association between gene expression and methylation. Left panels show correlation of a AJAP1, b SHANK2, c FOXA2, d MT1A, e ZNF570, f HOXC4, and g HOXB4 expression (X-axis) with methylation (Y-axis). Right panels show Kaplan-Meier survival plots of gene expression from the TCGA cohort. HR indicates hazard ratio. Correlation coefficients and hypothesis tests were based on Spearman rank correlation tests. Patients were categorized into high-risk and low-risk groups by an optimum cutoff point according to the highest χ ² value. h ROC curves for expression of the seven genes (left) and combinations of different types of data (right), including clinical characteristics (Clin), gene expression (Exp), and methylation (Methy)

Fig. 6

Mediation analysis for methylation prognostic signature through mRNA expression. a Diagram of mediation model. b Methylation signature from the seven CpG sites was treated as “exposure”; mediator was the linear combination of the corresponding seven genes’ expression level (score_expression) (Overall model). Total prognostic effect in hazard ratio (HR) were described as direct effect (HR_direct), indirect effect (HR_indirect), corresponding 95% confidence interval (95% CI), and the proportion of effect mediated (M%). Further, sensitivity analyses were performed by excluding each gene from score_expression, respectively, which retained statistical significance for mediation effect