Integrative DNA methylome analysis of pan-cancer biomarkers in cancer discordant monozygotic twin-pairs

Abstract

Background: A key focus in cancer research is the discovery of biomarkers that accurately diagnose early lesions in non-invasive tissues. Several studies have identified malignancy-associated DNA methylation changes in blood, yet no general cancer biomarker has been identified to date. Here, we explore the potential of blood DNA methylation as a biomarker of pan-cancer (cancer of multiple different origins) in 41 female cancer discordant monozygotic (MZ) twin-pairs sampled before or after diagnosis using the Illumina HumanMethylation450 BeadChip.

Results: We analysed epigenome-wide DNA methylation profiles in 41 cancer discordant MZ twin-pairs with affected individuals diagnosed with tumours at different single primary sites: the breast, cervix, colon, endometrium, thyroid gland, skin (melanoma), ovary, and pancreas. No significant global differences in whole blood DNA methylation profiles were observed. Epigenome-wide analyses identified one novel pan-cancer differentially methylated position at false discovery rate (FDR) threshold of 10 % (cg02444695, P = 1.8 × 10(-7)) in an intergenic region 70 kb upstream of the SASH1 tumour suppressor gene, and three suggestive signals in COL11A2, AXL, and LINC00340. Replication of the four top-ranked signals in an independent sample of nine cancer-discordant MZ twin-pairs showed a similar direction of association at COL11A2, AXL, and LINC00340, and significantly greater methylation discordance at AXL compared to 480 healthy concordant MZ twin-pairs. The effects at cg02444695 (near SASH1), COL11A2, and LINC00340 were the most promising in biomarker potential because the DNA methylation differences were found to pre-exist in samples obtained prior to diagnosis and were limited to a 5-year period before diagnosis. Gene expression follow-up at the top-ranked signals in 283 healthy individuals showed correlation between blood methylation and gene expression in lymphoblastoid cell lines at PRL, and in the skin tissue at AXL. A significant enrichment of differential DNA methylation was observed in enhancer regions (P = 0.03).

Conclusions: We identified DNA methylation signatures in blood associated with pan-cancer, at or near SASH1, COL11A2, AXL, and LINC00340. Three of these signals were present up to 5 years prior to cancer diagnosis, highlighting the potential clinical utility of whole blood DNA methylation analysis in cancer surveillance.

Keywords: Biomarker; Cancer; DNA methylation; Discordant monozygotic twins; Epigenetics; Twin study.

Fig. 1

Diagnostic characteristics and global methylation profiles of 41 cancer-discordant MZ twin-pairs. a Number of cases for each primary location of cancer, where a blood sample was obtained before (white) and after (black) cancer diagnosis. b Dendrogram of the unadjusted global methylation profiles. Annotation bars denote each individual’s cancer status, type of cancer (identical for both twins in a pair), and family identifier (identical for both twins in a pair). c Pair-wise correlation in DNA methylation profiles shows greater similarity within MZ pairs, compared to pairs of unrelated individuals, either paired at random or at within affection status

Fig. 2

Pan-cancer epigenome-wide results in 41 discordant MZ twin-pairs. a Manhattan plot of the epigenome-wide association results in 41 pan-cancer-discordant MZ twin-pairs, where each point represents the observed −log₁₀ P value at a CpG-site. b Direction of association at the top-ranked signal cg02444695, near SASH1. Results are plotted using normalised unadjusted beta values of cancer-affected individuals (left) and healthy individuals (right). The lines connect co-twins in twin-pairs and indicate a consistent direction of the effect with an average of 0.7 % within twin-pairs with a range of 0.9 to 3.0 %. The three suggestive probes are included in Additional file 1. c Pan-cancer DMR at TIMM44. Results are plotted using adjusted DNA methylation values at each CpG site in the DMR for individuals affected by cancer (red) and healthy co-twins (blue). Smooth (LOESS) lines with standard error are plotted for both groups. The CpG site driving the signal is at chr19:8,008,850 (hg19)

Fig. 3

Differential methylation with respect to time of cancer diagnosis. a Unadjusted DNA methylation values at cg02444695 (near SASH1) in affected individuals (red) and healthy co-twins (blue), shown with respect to time of diagnosis (years) with smooth (LOESS) lines fitted for both groups. The orange vertical line represents the time of diagnosis. b Methylation differences within twin pairs at the four top-ranked DMPs and cg04533633 (at COX7C). Each smooth (LOESS) line represents the methylation difference (affected − unaffected twin) at an individual probe (see legend)

Fig. 4

Functional follow-up of top-ranked pan-cancer DMPs. Adjusted whole blood DNA methylation profiles compared to adjusted gene expression levels for (a) cg21046959 in blood and ILMN_1809352 (PRL) in LCLs, and (b) cg27094856 in blood and ILMN_1701877 (AXL) in skin. Data are obtained from 283 healthy middle-aged females and lines represent the least squares regression fit. c Enrichment analysis of genomic annotation categories within the 500 top-ranked cancer DMPs. The bars indicate the difference in proportion of DMPs compared to the remainder of probes used in the study in the corresponding genomic annotation class. Nominally significant results were obtained for the ‘enhancer’ category (P = 0.03)