To screen candidate methylation markers for early detection of breast cancer and to explore the relationship between methylation and gene expression, we performed methylated-CpG island recovery assay (MIRA) combined with CpG island array on 61982 CpG sites across 4162 genes in 10 cancerous and 10 non-cancerous breast tissues. Direct bisulfite sequencing and combined bisulfite restriction analysis (COBRA) were carried out in independent cancerous and non-cancerous samples. Gene expression was analyzed by microarrays and validated using RT-PCR. We detected 70 significantly hypermethylated genes in breast cancer tissues, including many novel hypermethylated genes such as ITGA4, NFIX, OTX2 and FGF12. Direct bisulfite sequencing showed widespread methylation occurring in intragenic regions of the WT1, PAX6 and ITGA4 genes and in the promoter region of the OTX2 gene in breast cancer tissues. COBRA assay confirmed that the WT1, OTX2 and PAX6 genes were hypermethylated in breast cancer tissues. Clustering analysis of the gene expression of 70 significantly hypermethylated genes revealed that most hypermethylated genes in breast cancer were not expressed in breast tissues. RT-PCR assay confirmed that WT1 and PITX2 were only weakly expressed in the breast cancer tissues and were not expressed in most non-cancerous breast tissues. OTX2 and PAX6 were not expressed in either breast cancer or non-cancerous tissues. In conclusion, these results will expand our knowledge of hypermethylated genes and methylation sites for early detection of breast cancer and deepen our understanding of the relationship between methylation and gene expression. The MIRA approach can screen candidate methylated genes for further clinical validation more effectively than gene expression microarray-based strategy.