AUTHORS
Petri Polonen, PhD, Huimei Wei, PhD2, Yiping Fan, PhD3, Shaohua Lei, PhD4, Qingsong Gao, PhD5, Yiming Wu6, Ti-Cheng Chang, PhD, MSc7, Emily Backhaus8, So Hyeon Kim9, Haley Newman, MD, MSCE10, Lahari Uppuluri, PhD11, Sandi Radko-Juettner, PhD12, Gail Budhu, MD13, Yantao Chen14, Alejandro Gutierrez, MD15, Rahul S Bhansali, MD16, Gerd A. Blobel, MD, PhD17, Niroshan Nadarajah, MSc18, Torsten Haferlach, MD, PhD19, Marketa Zaliova, MD, PhD20, Jan Trka, MD, PhD21, Lu Wang22, Hiroto Inaba, MD. PHD23, Ilaria Iacobucci, PhD24, Stanley Pounds, PhD25, Gang Wu, PhD26, Lindsey Elizabeth Montefiori, PhD27, David T. Teachey, MD28, Charles G. Mullighan, MBBS (Hons), MSc, MD29
AFFILIATIONS
1 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
2 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
3 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
4 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
5 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
6 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
7 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
8 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
9 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
10 Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, United States
11 Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, United States
12 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
13 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
14 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
15 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
16 Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, United States, Hospital of the University of Pennsylvania, Philadelphia, United States
17 Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, United States, University of Pennsylvania, Philadelphia, United States
18 MLL Munich Leukemia Laboratory, Munich, Germany
19 MLL Munich Leukemia Laboratory, Munich, Germany
20 Charles University, Prague, Czech Republic
21 Charles University, Prague, Czech Republic
22 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
23 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
24 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
25 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
26 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
27 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
28 Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, United States, University of Pennsylvania, Philadelphia, United States
29 St. Jude Children’s Research Hospital, Memphis, Tennessee, United States
ABSTRACT
Background: Molecular subtypes of T-lineage acute lymphoblastic leukemia (T-ALL) are defined by lineage-specific transcriptional programs and recurrent coding and noncoding drivers. However, the genetic alterations that increase risk of relapse remain incompletely resolved. We investigated recurrent LYL1-activating alterations, their regulatory mechanism, and their association with relapse.
Methods: We used TALLForest, an integrated whole-transcriptome (WTS) and whole-genome sequencing (WGS) classifier, to identify and annotate candidate LYL1-associated structural and copy-number alterations in patients enrolled in Children's Oncology Group AALL0434 (N=1309) and the AALL1231 (N=489) cohort. Candidate enhancer gains were evaluated using PacBio long-read sequencing, EPIC v2, ATAC-seq, ChIP-seq, motif analysis, sequence-based regulatory prediction, and luciferase reporter assays with patient-derived duplicated sequences. Associations with relapse were tested using competing-risk regression. Clonal evolution was examined in paired diagnosis-relapse samples with a LYL1 enhancer gain. Enhancer occupancy was examined in Loucy cells and paired diagnosis-relapse samples.
Results: Integrated TALLForest analysis identified previously unrecognized duplications of an accessible enhancer upstream of LYL1 in 14/1309 patients in AALL0434 and 4/489 patients in AALL1231 in the TAL1 DP-like subtype. LYL1 enhancer gains and enhancer-hijacking rearrangements involving TCRD/TCRB and LYL1 were mutually exclusive with TAL1 and TAL2 alterations, but frequently co-occurred with TCRD/TCRB::LMO2 rearrangements and LEF1 deletions. The enhancer was accessible in normal CD34+ progenitors and ETP-like T-ALL, but inactive in most TAL1/LMO2-driven TAL1 DP-like leukemias lacking a LYL1 alteration, supporting alteration-dependent activation.
Long-read sequencing confirmed 46-87 bp tandem enhancer duplications in all five tested cases, with a shared 42-bp interval. DNA methylation profiling showed reduced regional methylation across the LYL1 locus in LYL1-altered cases (N=9) compared with TAL1 DP-like wild-type (N=75) cases (P=0.0069). The patient-derived duplicated sequence increased reporter activity 1.84-fold relative to the nonduplicated reference sequence (95% CI, 1.50–2.27; P=0.003).The minimal recurrently duplicated interval overlapped ATAC-seq accessible chromatin and contained predicted E-box, GATA:SCL, and ETS-family motifs. AlphaGenome identified E-box and ETS-family sites as the sequences most influential for predicted ATAC-seq accessibility.
ChIP-seq in Loucy cells and paired diagnosis- and relapse-derived patient xenografts demonstrated focal co-occupancy of LMO2, LYL1, and the bridging factor LDB1 at the LYL1 enhancer, together with RUNX1 and ERG enrichment. In the paired diagnosis-relapse case, the LYL1 enhancer gain, TCRB::LMO2 rearrangement, LEF1deletion, and CDKN2A deletion persisted at relapse, whereas a subclone carrying FBXW7 R465H, NRAS G13D, and a PTEN indel expanded. Compared with diagnosis, relapse showed increased enhancer accessibility, increased LYL1expression, and stronger LMO2, LYL1, and LDB1 occupancy at the enhancer. Together, these findings support selection for a self-reinforcing LMO2-LYL1 regulatory circuit in which an LMO2-LDB1 complex activates the LYL1enhancer, leading to increased LYL1 expression and incorporation of LYL1 into the enhancer-bound complex, thereby reinforcing enhancer activity and sustaining LYL1 transcription.
Exploratory pooled analysis of both cohorts identified LYL1 and PTEN co-alterations in 10 patients, five of whom relapsed within one year (1-year cumulative incidence of relapse, 50%). A cohort-stratified Fine–Gray model provided preliminary evidence of a positive LYL1×PTEN interaction on the subdistribution hazard scale (interaction sHR, 10.23; 95% CI, 1.09–95.96; P=0.042). In contrast, LYL1 alterations without PTEN alterations did not show evidence of increased relapse incidence.
Conclusions: LYL1-activating enhancer duplications and enhancer-hijacking rearrangements define a genetically distinct T-ALL subgroup characterized by frequent LMO2 rearrangements and LEF1 deletions. Orthogonal genomic, epigenomic, and functional evidence supports an LMO2-LYL1-LDB1 positive-feedback circuit operating through the LYL1 enhancer and becoming reinforced during relapse evolution. Concurrent LYL1 and PTEN alterations may identify patients at increased risk of early relapse.
DISCLOSURES
Petri Polonen, PhD: No relevant conflicts of interest to declare.