As a logical step toward developing this methodology, we targeted 2 gene promoters, TLX3 (T-cell leukemia homeobox 3)  and FOXE3 (forkhead box E3), which displayed the largest differential methylation between pediatric ALL patients and nonleukemic controls in validation experiments (8).
Additionally, we prepared reference materials for the promoter regions of TLX3 and FOXE3 by PCR amplification of FOXE3 and TLX3 promoter regions using human genomic DNA (Sigma) followed by in vitro methylation using M.
Three different schemes, all based on detection by MALDI-TOF mass spectrometry, were used for evaluation of the methylation level of the promoter regions of TLX3 and FOXE3.
Converted combined TLX3 and FOXE3 reference materials (5 [micro]L, 20 000 copies/[micro]L) were added to 37.
For patient samples, 12 fragments were measured for FOXE3 covering 19 CpG sites and 13 fragments for TLX3 covering 26 CpG sites.
50, implying that a very high proportion of the cell population present in the bone marrow aspirates contained hypermethylated CpG loci at FOXE3 and TLX3 promoters.
Seven of the 10 TLX3 false-negative values and 4 of the 9 FOXE3 values were close to the 0.
20 for either FOXE3 or TLX3 achieved high clinical specificity and sensitivity for ALL diagnosis, indicating that reliable detection of even low levels of methylated DNA may be diagnostic.
We designed a multiplex assay that combined TLX3 and FOXE3 in a single MALDI-TOF analysis for this work, since we envisioned that interrogation of multiple markers might be required for patient sub-typing.
The multiplex EpiTYPER technique analyzed fewer CpG sites then singleplex analysis owing to indistinguishable fragments (by mass) produced between the FOXE3 and TLX3 amplicons after cleavage.
We were able to correctly diagnose 18 of 20 pediatric B-cell ALL patients that lacked common genetic lesions using EpiTYPER interrogation of FOXE3 and TLX3 gene promoters and EF threshold >0.
10] Human genes: TLX3, T-cell leukemia homeobox 3; FOXE3, forkhead box E3.