The issue of clonal heterogeneity in leukaemia has diverse implications for disease monitoring, targeted therapy design and understanding the biology of disease relapse. In this study, we aimed to perform single cell analysis to improve understanding of clonal evolution in T-ALL. We used fluorescence in situ hybridisation (FISH) and phylogenetic analysis of diagnostic T-ALL samples to infer the order of acquisition of genetic events and define the early and likely founder events in this subtype of leukaemia. Screening of paired diagnostic and xenograft material from two STIL-TAL1 positive cases allowed analysis of clonal heterogeneity and evolution in leukaemia initiating cells.


DNA from seven diagnostic T-ALL patients aged 1 – 25 years (five STIL-TAL1 fusion positive, one TLX1 translocation positive, one TLX3 translocation positive) was screened using the Affymetrix® SNP 6.0 platform (analysed using CNAG version 3.3.0 software) and standard Sanger Sequencing to detect genome wide copy number alterations (CNA), mutations in NOTCH1 (exons 26, 27, 34) FBXW7 (exons 9,10), PTEN (exon 7) and IL7R (exon 6) and to define the genomic DNA sequence of the STIL-TAL1 fusion in relevant cases. Genomic localisation of CNA was used to design patient-specific FISH probes for the STIL-TAL1 fusion and losses of CDKN2A, PTEN, LEF1, SYNCRIP (marker of 6q deletion), MYC (marker of chromosome 8 gain) and ETV6 (marker of 12p loss). TLX1 status was assessed using a commercial Cytocell FISH probe and TLX3 status using an in-house probe set. Interphase FISH scoring of one to two hundred methanol-acetic acid fixed nuclei per patient with four - five differentially labelled FISH probes allowed phylogenetic analysis and reconstruction of the evolution of the malignant clone. In two of the STIL-TAL1 cases, we completed xenograft transplants using NOD/ShiSCID/IL2Rgamma null (NOG) mice. Mutation screening, copy number analysis and FISH were performed on xenograft spleen DNA and methanol-acetic acid fixed spleen nuclei.


Construction of phylogenetic trees for each patient based on single cell four - five colour FISH analysis demonstrated the presence of two - four abnormal sub-clones per patient. Both linear clonal progression and branching sub-clonal patterns were observed. STIL-TAL1 fusion, TLX1 and TLX3 translocations were placed at the root of each phylogenetic tree as they were present in all cells with an abnormal FISH pattern signal; the majority (but not all) cells also demonstrated loss of CDKN2A. Copy number losses of PTEN and LEF1 as well as losses of 12p and 6q and gain of chromosome 8 occurred at a lower level than CDKN2A loss. Sequencing of the STIL-TAL1 fusion sequence in the two paired diagnosis/xenograft cases (A and B) demonstrated that the fusion sequence was conserved in the xenografts. Case A harboured a NOTCH1 PEST domain truncating mutation detected in the xenograft. Sequencing of the diagnostic sample was suggestive of the presence of the same mutation at low level. At diagnosis, case B was found to have a PTEN exon 7 truncating mutation which was not detectable by Sanger Sequencing in the xenograft. In case B, the dominant diagnostic clone assessed by FISH had the STIL-TAL1 fusion plus losses of CDKN2A, 4p and 6q, compared to the dominant clone in the paired xenograft, which harboured the STIL-TAL1 fusion and loss of CDKN2A but retained both copies of 4p and 6q. This dominant xenograft clone was present at low level in the diagnostic sample.


These results support the hypothesis that, in accordance with other types of leukaemia, intra or inter-chromosomal rearrangements (STIL-TAL1 fusion, TLX1 and TLX3 translocation) are early events. As the STIL-TAL1 fusion was present in all cells with an abnormal FISH signal pattern in all five cases and the fusion sequence was conserved in the xenograft transplants, this provides evidence to support the hypothesis that it is a founder event in the development of T-ALL. CDKN2A loss also occurred early in the evolution of the leukaemogenic clone, in line with its key role in T-ALL. Assessment of the order of acquisition of genetic events in T-ALL and their stability as evolution of the malignant clone unfolds are key factors to determine rational treatment targets for molecularly targeted therapies. Future work will incorporate mutations and gene fusions in to the single cell phylogenetic analysis using a multiplex Q-PCR approach.


Ferrando:Pfizer: Research Funding; BMS: Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.