Detection of promoter hypermethylation in serum samples of cancer patients by methylation-specific polymerase chain reaction for tumour suppressor genes including RUNX3

The purpose was to validate the use of RUNX3 as a potential biomarker for detection of cancer in serum samples and to determine its sensitivity alone and in combination with p16, RASSF1A and CDH1 using methylation-specific polymerase chain reaction (MSP). We examined the promoter methylation status of RUNX3, p16, RASSF1A and CDH1 by MSP using the serum of 70 metastatic breast, non-small cell lung, gastric, pancreatic, colorectal or hepatocellular carcinomas. The DNA from 10 healthy serum controls was used to determine the specificity of methylation. According to our results, promoter hyper-methylation of RUNX3 was detected in the serum of 44 patients comprising breast 9/19 (47%), non-small cell lung 11/20 (55%), gastric 4/4 (100%), pancreatic 2/2 (100%), colorectal 11/17 (65%) and liver 7/8 (88%) carcinomas. Comparative figures for the other genes were as follows: p16 - 39/70 (7/19, 10/20, 2/4, 0/2, 12/17, 8/8); RASSF1A - 24/70 (8/19, 6/20, 1/4, 1/2, 4/17, 4/8); CDH1 - 10/70 (0/19, 4/20, 1/4, 1/2, 3/17, 1/8). Using a panel of four genes, hypermethylation of one or more genes was found in 62/70 samples (15/19, 19/20, 4/4, 2/2, 14/17, 8/8). A panel of three genes omitting RUNX3 detected hyper-methylation in only 50/70 samples. No methylation was detected in the 10 healthy serum controls. Thus, RUNX3 can be detected in the serum of a high proportion of advanced cancers. This suggests that serum hypermethylation of RUNX3 is at least as, or possibly more sensitive a marker, than other tumor suppressor genes currently under investigation. Inclusion of RUNX3 in gene panels can potentially increase the sensitivity of such panels for serum diagnosis of malignancies and warrants further study.