A DNA-Based RQ-PCR Screening Assay for the Identification of Cases with RUNX1 Copy Number Changes in Childhood B Cell Precursor Acute Lymphoblastic Leukemia (BCP-ALL).

Abstract

An increased RUNX1 copy number is a common finding in childhood B cell precursor acute lymphoblastic leukemia (BCP-ALL). In the vast majority of cases it reflects the presence of additional copies of chromosome 21, which in turn is a typical feature of hyperdiploid BCP-ALL. In a much smaller proportion of cases, however, it results from rearrangements that specifically multiply the region 21q22 (intra-chromosomal RUNX1 multiplication; ICRM). This distinct genetic marker, also known as AML1 amplification, designates a specific form of BCP-ALL with a pronounced risk of relapse that is usually encountered in older children. Noteworthy, less than 100 cases with this abnormality have been reported so far worldwide. The apparent rarity, however, results in part also from the fact that at present an ICRM can only be detected with fluorescence in situ hybridization (FISH) and a systematic FISH screening is only conducted in very few treatment trials. To overcome this diagnostic obstacle, we developed a DNA-based real-time polymerase chain reaction (RQ-PCR) screening assay. It is based on the comparative quantification of three regions within RUNX1 at 21q22, PRSS7 at 21q21.1 (as an intra-chromosomal control) and BBS1 at 11q13.2 (as an inter-chromosomal control). The assay was set up and evaluated with DNA from cases with two (normal controls), three (Down syndrome patients) and four (hyperdiploid ALL) chromosomes 21 and put to test on samples from 13 Austrian cases with a previously FISH-verified ICRM. The number of additional RUNX1 copies in these samples was determined to range from 4 to approximately 8. Subsequent screening of 150 ALL samples from the German BFM-ALL trial identified altogether 37 cases with an increased RUNX1 copy number, but also one with a PRSS7 and two with RUNX1 deletions. In the vast majority of cases the respective PCR results were in good accordance with those suggested by DNA-index, cytogenetic or FISH analyses. Discrepant results were double-checked with RQ-PCR analyses utilizing primers that encompassed additional appropriate DNA sequences. The selective increase of RUNX1 copies in three instances discerned specifically samples with a ICRM (+3, +3 and +7 copies, respectively). A concordant overrepresentation of both RUNX1 and PRSS7 copies, on the other hand, defined cases with one to approximately five extra copies of chromosome 21, which included those with hyperdiploidy as well as one case each with a constitutional trisomy 21 and an ETV6/RUNX1 gene fusion. The results of our analyses prove that, provided the investigated material contains a sufficient number of blast cells, such a DNA-based RQ-PCR screening technique can reliably delineate and discriminate between both forms of RUNX1 overrepresentations. Since this approach is extremely well suited for the fast and efficient retro- and prospective analyses of a large number of cases we foresee that it will become the preferred method for the identification of such cases in childhood ALL treatment trials.