Comparison of large structural variants between cth positive and cth negative T-ALL
In the next step, we searched for differences in the frequency and spectrum of structural aberrations exceeding 5 Mbp between the cth pos patients and cth neg individuals. Overall, the SNP array detected 552 events (0–64; 3.19 per case), including 199 losses, 212 gains, and 141 copy-neutral losses of heterozygosity (cnLOH). CNAs were present in all the cth pos T-ALL patients and in half of the cth neg patients (n = 95; 55%). The number of CNAs per patient was eight times greater in the first group than in the second group [median number of CNAs was 19.2, range: 4–64; range: 2.2, range: 0–17].
A total of 360 segmental alterations, including whole-chromosome trisomy (n = 60), monosomy (n = 2), segmental duplication (n = 63), segmental deletion (n = 141), segmental loss of heterozygosity (n = 94), and loss of heterozygosity of whole chromosomes (n = 3), were detected in 163 cth neg T-ALL patients. In total, 49 patients had only one large segmental abnormality exceeding 5 Mbp. A total of 192 segmental CNAs, including whole-chromosome trisomies (n = 1), monosomies (n = 2), duplications (n = 13), deletions (n = 21), and segmental loss of heterozygosity (n = 1), were observed in 10 cth pos T-ALL cohort. The number of CNAs ranged from 8–38 segmental CNAs, with an average of 72.29 (± 55.35) Mb.
We further investigated whether cth pos T-ALL patients exhibit a common pattern of structural abnormalities or whether they resemble any specific nonsyndromic T-ALL subgroup. For this purpose, we performed unsupervised hierarchical clustering analysis using information about whole-genome CNAs exceeding 5 Mbp in 173 leukemia samples. As shown in Fig. 2A, all NBS T-ALL patients grouped within Cluster 3, which may reflect their intrinsic genetic resemblance. The remaining cth pos non-NBS patients clustered either in Cluster 3 (two samples) or in Cluster 2 (three samples).
Overall, there was a significant difference in the distribution of large segmental aberrations, both deletions and duplications, between the cth pos and cth neg patients, which are marked in black in the inner circle in Fig. 2B. Segmental duplications involving chromosomal regions 1q, 5p, 8q, 14q, 18p, 18q, 19q, and 22q were more frequent in cth pos T-ALL than in cth neg patients. Similarly, segmental deletions affecting 1q, 5q, 7q, 8p, 12p, and 13q were significantly more common in T-ALL patients with cth.
We also investigated the pattern of large structural CNAs according to T-cell immunophenotype (Supplementary Figure S1) and found that duplication of chromosome 8 was significantly common among pre-T-ALL patients. Although ETP-ALL was underrepresented in our study group, we observed a frequent incidence of deletions involving chromosomes 5q, 6q, and 11q in T-ALL samples representing this leukemic subtype.
As expected, the most recurrent copy number losses (n = 37; 22.15%) in nonsyndromic T-ALL were found on the short arm of chromosome 9 and encompassed the common 9p22.2p21.1 (17262395_28275543) region, including the SH3GL2, MLLT3, IFNA1, MTAP, CDKN2A/B, and TEK genes. Another frequent structural deletion (n = 20; 11.97%) was identified on the long arm of chromosome 5q and involved the 5q33.3q35.1 (157861361_172589313) region; this deletion included the EBF1, PTTG1, RANBP17, TLX3, and NPM1 genes. Moreover, we also observed recurrent deletions on the long arm of chromosome 6 (n = 19; 11.37%) encompassing the 6q14.1q16.3 (79658603_104885348) region, leading to loss of the CASP8AP2, BACH2, MAP3K7, EPHA7, and CCNC genes. Other less common deletions were observed on the short arm of chromosome 12p13.2p12.1 (11256830_22116241) (n = 12; 7.18%), including ETV6, BCL2L14, CDKN1B, ATF7IP, GUCY2C, EPS8 and 1p36.33p36.22 (854277_10485206) (n = 11; 6.58%), encompassing the DVL1, MIB2, PRDM16, RPL22, CAMTA1, ERRFI1, ENO1, and PIK3CD genes. The most recurrent duplications were observed on the long arm of chromosome 4q32.1q35.2 (159407548_190957473) (n = 7; 4.2%), which included FBXW7, NPY1R, ING2, SORBS2, FAT1, and DUX4; on the long arm of chromosome 8q23.3q24.3 (116593872_146295771) (n = 6; 3.6%), which included TNFRSF11, ENPP2, and RNF139; and on the long arm of chromosome 9q33.1q34.3 (118514469_141020389) (n = 6; 3.6%), which included TSC1, GFI1B, PAEP, and MIR126.
Interestingly, the most characteristic structural aberrations for NBS were those affecting chromosome 8, such as loss of the short arm and duplication of the long arm. In two patients, we cytogenetically confirmed the presence of isochromosome 8 [i(8)(q10)]. Segmental LOH involving the NBN gene at 8q21 was present in the entire group of NBS patients. Apart from the short arm of chromosome 8, the most frequent deletion was located at the 12p13.31p12.1 (9910041_23561624) region and involved GABARAPL1, KLRK1, ETV6, BCL2L14, CDKN1B, ATF7IP, GUCY2C, and EPS8 (n = 4). The most recurrent duplications in syndromic T-ALL occurred on the short arm chromosome 5p15.33p12 (113577_42844121) region and encompassed IL7R, PRLR, LIFR, PDCD6, TRIO, DAB2, and SEPP1. In two NBS patients, we also noted coincidence of CNAs involving chromosomes 7 and 14, suggesting reciprocal translocations, which could not be confirmed due to the lack of biological sample for cytogenetic studies.
We also investigated the frequency of UPDs, which are relatively rare events in T-ALL (< 1%) (18), in contrast to pediatric BCP-ALL, in which UPDs occur in 5–10% of patients (19). Loss of heterozygosity involving chromosomes 8, 11, and 18 was identified in a single cth neg patients with low hyperdiploid karyotypes. However, common segmental LOH affects the short arm of chromosome 9p24.3p13.3 (192128_21894495), with a concomitant deletion occurring prior to sUPD9p in all instances (20). The presence of multiple regions of homozygosity (ROH) exceeding ≥ 10% of the genome using a threshold of 3–5 Mb or larger was identified in one NBS patient (M38). According to ACMG guidelines (1), a possible first- or second-degree relationship between the patient’s parents may be suspected, but this was not reported by the care givers. (Supplementary Figure S2).
Profile of microaberrations in T-ALL patients with the chromothripsis pattern
The total number of microdeletions and microduplications in the study group was 2953 in 173 patients (17.17/per sample). The median number of microCNAs did not significantly differ between cth pos and cth neg individuals [n = 13 (range: 8-20.5) and n = 17 (range: 11-21.50), respectively]. In both groups, the most recurrent CNAs affected genes encoding lymphocyte T receptors (TRA, TRD, TRB, and TRG) and immunoglobulin heavy chain (IGH) (Supplementary Table S4). Moreover, there was no specific pattern of microaberrations in the cth pos T-ALL patients (Table 1).
In contrast, cth neg T-ALL patients presented recurrent microdeletions in the following genes: LEF1, HOXA11 (TLX1), FLI1, CEBPE, RB1, SORBS2, IFNA1, PTEN, NDRG2, MMP14, CTCF, NF1, DCC, STIL, CDKN2A, CDKN2B, FIP1L1, MLLT3 (AF9), ANG, MTAP, and ST13P4. The most frequent microduplications involved MYB, AHI1, TSC1, IGL@, IL3RA, L1CAM, AIFM1, GPC3, PHF6, ZRSR2, and SEPT6 (Supplementary Table S4). Analysis of the frequency of microalterations according to gene function revealed no significant differences between the cth pos and cth neg T-ALL patients, except for cell cycle regulators (Supplementary Figure S3). This difference resulted from the common incidence of microdeletions affecting CDKN2A, CDKN2B, and STIL in cth neg T-ALL patients (263 deletions in 118 patients).