FISH improves risk stratification in acute leukemia by identifying KMT2A abnormal copy number and rearrangements

This study simultaneously identified and described KMT2A-r and KMT2A CNV alterations in AL using FISH analysis. KMT2Awhich is 92 kb in size, is located on chromosome 11, band 23 (11q23) and includes at least 37 exons13. The accurate detection of KMT2A-r remains a challenge in clinical practice because of the complex mechanisms involved, including reciprocal translocations, complex chromosomal rearrangements, gene internal duplications, deletions, inversions on chromosome 11q, insertion of chromatin material into KMT2Aand KMT2A insertions into other chromosomes or vice versa14. It has been reported that the occurrence of a 3′-KMT2A deletion is associated with KMT2A-r15th.

Currently, the methods routinely used for detecting KMT2A-r in AL patients in China include CBA and RT-PCR. However, FISH analysis using a dual-color BAP is used to further confirm the existence of KMT2A-r when these two primary methods yield inconsistent results.

In the current study, KMT2A FISH was performed for samples from every patient in the AL cohort, as were CBA and RT-PCR analyses. Through a combination of the three methods, a total of 23 cases were identified as being KMT2A-r, with eight being positive using all three methods. Methods of FG detection, karyotype analysis, and FISH analysis yielded positive results in 15, 16, and 20 cases, respectively (Table 2). Conventional karyotype analysis, which can reveal structural and numerical abnormalities of chromosomes, may miss some instances of KMT2A-r, probably owing to poor metaphase division of leukemic cells and poor detection of subtle chromosomal changes caused by complex or cryptic abnormalities6. Use of the LSI KMT2A BAP allows for the recognition of complex rearrangements, notably cryptic insertions of KMT2A segments into other chromosomes or the disruption of KMT2A by the insertion of other chromosomal segments, which improves the detection rate of KMT2A-r16. Three cases in our cohort (Patients 1, 2, and 3) were confirmed to have cryptic KMT2A insertions after a second review of karyotype analysis, which was based on the metaphase FISH analysis of KMT2A. This illustrates the need for FISH analysis using the KMT2A probe as an important auxiliary method to the CBA of AL patients. Furthermore, beyond the typical separation of signals in FISH detection of KMT2A, atypical FISH signal patterns are worth studying in clinical practice, especially changes in the number and intensity of orange and green signals. When an unusual KMT2A signal pattern is observed, metaphase FISH should be implemented to further clarify the mechanism of complex rearrangement of KMT2A. Several cases of insertions involving band 11q23 into chromosomes 2, 4, 5, 6, 9, and 10 and the X-chromosome have been previously reported16,17,18,19,20.

Compared to cytogenetic analysis, RT-PCR is a more sensitive and rapid technique. Furthermore, commercially available multiplex RT-PCR kits that can reveal KMT2A-r with frequently involved partner genes have been used worldwide. However, these kits do not contain primer sets for detecting rare or unknown partner genes, which may result in missed diagnosis of some patients with KMT2A-r. The RT-PCR panel used in the present study may not have included primers to detect the FG of KMT2A-r caused by t(2;11), t(5;11), and t(11;17), which was demonstrated by the karyotype analysis of Patients 9, 10, 12, and 13. However, the remaining sample volumes were insufficient to further confirm the presence of this FG by RNA sequencing. Additionally, fusion transcripts cannot be found in cases with such widespread aberrations as t(4;11) (Patient 14), t(10;11) (Patient 8), or t(11;19) (Patient 11). We speculated that there are several potential causes of the negative RT-PCR result. First, in addition to the major breakpoint cluster region (BCR) of the KMT2A gene (KMT2A exons 8–14), the other BCR is novel and minor (KMT2A intron 21–23), which was verified by Meyer et al. in 20195. The commercial multiplex RT-PCR kits are not always compatible and may only cover the major BCR, so if KMT2A-r occurred in minor BCR or other special mechanisms, including in the cases of 11q deletion, inversion, or three-way translocation, which can all cause KMT2A-r breakpoint outside of the major BCR. In these situations, the results were both beyond the range of the panel designed for the commercial kit. Second, in addition to AFF1, MLLT10, MLLT1and ELLdifferent partner genes can be involved in 4q, 10p, and 19p, including SEPT11 and ARGBP2 (4q), ABI1 and NEBL (10p), and ASAH3, EEN and MYO1F (19p)4. However, these different partner genes all demonstrate the same t(4;11), t(10;11), and t(11;19) in the karyotype, which cannot be clearly distinguished by CBA. Patients 4 and 6 showed the FG of KMT2ASEPT6 and KMT2AMLLT10 by RT-PCR without any aberration involving the rearrangement of KMT2A in FISH and karyotype analyses. Similar cases have been previously described by other groups, and mechanisms that involve either a fragment of KMT2A being inserted elsewhere into the genome or a fragment of a locus being inserted proximally to KMT2A have been identified6,16. Both circumstances create a copy-neutral KMT2A-fusion oncogene that is not detectable by BAP FISH testing. In addition, KMT2A-PTDs that are undetectable using currently available FISH probes or conventional karyotype analysis were identified in one case of the current study (Patient 5) and confirmed by RT-PCR. Therefore, the combined use of FISH, CBA, and multiplex RT-PCR analyzes can improve the detection of KMT2A-r, particularly in unusual, complex, or cryptic chromosomal rearrangements that are frequently observed in AL.

A total of 17 patients with KMT2A CNV, including 15 with AML and two with ALL, were identified in the current cohort. Moreover, 10 of 15 (66.7%) AML cases with KMT2A CNV detected by FISH were CK, which was higher than that for the group with normal FISH results (66.7% vs. 10.6%; Table 4). Amplification of KMT2A could be defined as the presence of at least two extra gene copies as seven of 10 patients with AML and CK demonstrated ≥ 4 copies of KMT2A. Tang et al. reported that AML/MDS with KMT2A amplification is associated with a CK and high frequency of TP53 mutations8. It has been demonstrated by Zatkova et al. that patients with AML and CK, including 11q/KMT2A amplification, respond poorly to therapy and have a poor prognosis with an extremely short overall survival9. Therefore, the application of KMT2A FISH may have an additional benefit of shedding light on the identification of CK, especially when conventional karyotype analysis fails. The sample from Patient 27 of the current study failed in CBA, but FISH analysis revealed five copies of KMT2A. Three months after AML diagnosis, the patient had not achieved remission and died of infectious shock. Of the 10 AML cases with CK, including KMT2A amplification, recurrent chromosomal aberrations of 5q deletions were observed in six cases, which was similar to previous findings of 5q deletions being the most frequently associated with 11q amplifications in AML with CK9,21.

The mechanisms underlying the KMT2A-r and KMT2A CNV pathogenesis differ in that some KMT2A-r are driver abnormalities that require very few cooperate mutations to induce tumorigenesis22while KMT2A CNV is easily accompanied by CK, meaning it must act in coordination with other genetic abnormalities to cause leukemia. Thus, detection and distinction of these two types of pathogenesis are important for development and management of treatment protocols. Several studies have the potential use of KMT2A inhibitors as promising targeted therapies for KMT2A-r leukemia23. However, the effect on KMT2A CNV was not yet known at the time of this study.

Our study has several limitations. First, the commercial multiplex RT-PCR kits did not disclose primers sequences for trademark restrictions, which hindered us in distinguishing whether there were some defects in their primer design leading to the false negative in our cohort. Second, because of the insufficient sample and unaffordable cost, the cases with discrepant results from the three detection techniques cannot be further clarified using the complex mechanism of KMT2A-r by taking long distance inverse PCR (LDI-PCR) or next generation sequencing (NGS) technologies, such as RNA sequencing, which were both previously considered as robust ways to identify KMT2A-r.

Despite these limitations, we have shown that using multiplex nested RT-PCR, karyotype analysis, and FISH techniques in combination can improve the detection rate of KMT2A-r. We think the results show that KMT2A FISH detection should become a routine component of diagnostic and prognostic workups to identify cryptic KMT2A-r in AL patients. Furthermore, analysis may reveal a set of different KMT2A CNVs, which is highly associated with poor prognosis in AL, especially for patients in which there is a failure to obtain adequate levels of evaluable metaphase cells in cytogenetic analysis.

Leave a Reply

%d bloggers like this: