Introduction

T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive malignancy of the thymocytes caused by the accumulation of genomic lesions that affect the development of T cells. It mainly affects children, commonly boys, accounting for 15% of pediatric ALL and 20% of adult ALL [1,2,3]. Historically, T-ALL has been associated with poor outcome; however, with more contemporary therapy, outcomes for T-ALL and B-cell ALL become similar [1, 4].

Although there are validated prognostic factors in T-ALL, e.g., response to treatment including induction remission status and minimal residual disease (MRD) levels, there is controversy regarding the prognostic significance of the other factors including early T-cell precursor (ETP) phenotype and recurrent genetic aberrations [1, 4].

The ETP subtype, occurring in 15% of childhood T-ALL cases, was first identified within pediatric T-ALL cases based on its unique immunophenotypic and genetic features [1, 4, 5]. The hypothesis regarding the cell of origin and its leukemogenic pathways is still debated. Although having characteristics of T-cell lineage commitment, these cells continue to have the potential for myeloid/dendritic cell differentiation. Therefore, it is considered as a stem cell leukemia at the crossroads of the lymphoid and myeloid fates [4, 6].

As such, ETP-ALL immunophenotype was defined as follows: (1) absent (< 5% positive cells) CD1a and CD8 expression; (2) absent or dim (< 75% positive cells) CD5 expression; and (3) expression (> 2 5% positive cells) of 1 or more myeloid (CD11b, CD13, CD33, CD117) or stem cell (CD34, HLA-DR) markers [1, 5, 6]. Initially, ETP status was associated with high rates of induction failure, early relapse, and poor overall survival. However, in more recent reports, ETP status has lacked prognostic significance; therefore, the prognostic impact of ETP remains a subject of debate [1, 4].

While several genetic aberrations associated with T-ALL pathogenesis had been identified [5, 6], few of them have been reproducibly associated with prognosis, and none is prospectively used in risk stratification [4]. There are four major mechanisms known to cause aberrant expression of transcription factors in T-ALL including the following: (1) chromosomal translocations involving one of the TCR genes, (2) chromosomal rearrangements with other regulatory sequences, (3) duplication/amplification of the transcription factor, and (4) mutations or small insertions generating novel regulatory sequences acting as enhancers [3].

Translocations and inversions involving the TCR genes are now recognized as the oncogenic hallmark of T-ALL. TCR translocations occur in 35% of cases, as ascertained by FISH and PCR analyses, with almost 50% going undetected by chromosome banding analysis. The TCR aberrations include the TCR alpha/delta chain (TCRA/D) which is located at 14q11.2 (in 17% of T-ALL cases), the TCR beta chain (TCRβ) which is located at 7q34, and the TCR gamma chain (TCRδ) which is located at 7p14 [7].

Upon characterizing TCR rearrangements, about 30 partner oncogenes have been identified and proved to be important for T-ALL oncogenesis [7,8,9]. The overexpression of TAL1 (chromosome 1p32) and LMO1/LMO2 (11p13) expression are caused by rearrangements to the TCRA/D chain in 3–9% of pediatric T-ALL patients, while overexpression of HOX11 (TLX1) (10q24) is caused by rearrangements to the promoters of the TCRβ or TCRA/D chains in 30% of adult T-ALL [7, 10].

Current treatment of pediatric T-ALL consists of high-intensity combination chemotherapy which results in a very high overall survival (OS) [11]. Unfortunately, this treatment comes with significant side effects that should not be underestimated [12, 13]. Moreover, the occurrence of relapse is a challenging matter, as it is observed in up to 20% of cases and is often refractory to chemotherapeutics [14]. In addition, the development of targeted therapy for T-ALL might be controversial, given that it was demonstrated that TCR-positive T-ALL cells can be targeted by mimicking thymic negative selection [13].

Accordingly, in this study, we aimed at investigating the characterization of ETP-ALL and TCRA/D-positive cases in pediatric T-ALL with special emphasis on the clinical features associated with those cases and their impact on the outcome of the patients.

Patients and methods

Patients

The present retrospective study included 63 newly diagnosed pediatric T-ALL patients who were diagnosed during the period from January 2016 to July 2020. All patients were presented to the Outpatient Clinic of the Paediatric Oncology Department, National Cancer Institute (NCI), Cairo University, Egypt.

Methods

The diagnosis of T-ALL was based on the morphological examination of the peripheral blood (PB) and bone marrow (BM) smears, cytochemistry, immunophenotyping by flow cytometry, conventional cytogenetics, and cerebrospinal fluid (CSF) analysis. In addition to the initial workup such as chest radiographs (X-ray or CT chest) and abdominal ultrasonography.

Flow cytometry analysis

Immunophenotyping was performed on either BM or PB samples using a wide panel of monoclonal antibodies (Abs) for initial lineage assignment, according to the standard techniques. Fluorescein-labelled mouse monoclonal antibodies (MPO, CD13, CD33, CD117, CD64, CD14, and CD11c) were used to identify myeloid lineage. B-lymphoid cells were identified by using CD19, CD10, CD20 cytoplasmic μ chain, kappa, Lambda, and cytCD22/cytCD79a), while T-lymphoid antigens were identified using CD3, CD2, CD5, CD7, CD4, CD8, TDT, and CD1a in addition to C45, CD34, and HLA-DR. The acquisition was done on Navios 6 colors flow cytometry (Beckman Coulter, Miami, FL, USA).

The diagnosis of T-phenotype was considered only upon the expression of the lineage-specific marker (surface/cytoplasmic CD3). Patients were classified into immunophenotypic subgroups: early T-cell precursor phenotype (ETP) (CD1a−, CD8−, CD5dim/−, positivity for ≥ 1 myeloid/stem cell-related markers, i.e., CD34, CD117, CD13, CD33 or HLADR), early T-phenotype (CytCD3+, CD1a−, SCD3−, CD4−, and CD8−), intermediate T-phenotype (CytCD3+, CD1a+, and/or co-expression of CD4 and CD8), and late T-phenotype (CytCD3+, SCD3+, CD4+, or CD8+).

Cytogenetic analysis

The pre-treatment diagnostic BM samples were subjected to conventional karyotyping using G-banded metaphase cells from unstimulated 24- and 48-h cultures following the standard techniques. In most of the cases, at least 20 metaphases were analyzed using an IKAROS imaging system (MetaSystems, Altlussheim, Germany). The karyotypes were interpreted using the International System for Human Cytogenetic Nomenclature (ISCN 2016 [15] and 2020 [16]).

For the detection of TCRA/D rearrangements, amplifications, and deletions, fluorescence in situ hybridization (FISH) was performed using XL TCRA/D (MetaSystems, Altlussheim, Germany) according to the manufacturer’s instructions. A minimum of 200 interphase nuclei and 10 metaphases were analyzed using a fluorescence microscope (AxioImager.Z1 mot, Carl Zeiss Ltd., Hertfordshir, UK) equipped with appropriate filter sets. Image capture and processing were performed using an ISIS imaging system (MetaSystems, Altlussheim, Germany).

Management of the patients

All children received total XV protocol (modified from St. Jude total XV protocol) [17]. Patients with T-ALL were classified into either standard or high-risk based on response to treatment assessed morphologically and by minimal residual disease levels measured by flowcytometry on day 15 of start of chemotherapy and at the end of induction therapy (day 42).

The treatment protocol consisted of three phases, induction of remission, consolidation, and maintenance [18]. The induction phase (6 weeks) included prednisone, vincristine, doxorubicin, L-asparaginase, cyclophosphamide, cytarabine, and mercaptopurines. Age-dependent triple intrathecal chemotherapy was given on days 1, 4, 8, 12, 15, and 22 regardless of the initial CNS status in all T-cell patients. Regarding the patients who did not achieve morphological CR at day 15 (≥ 5% BM blasts) or had minimal residual disease (MRD) ≥ 1% by flow cytometry, they received additional 3 doses of L-asparaginase, while the patients who achieved morphological CR at day 42 induction (< 5% BM blasts) or had MRD < 1% were considered a standard risk and received the consolidation phase. Patients who did not achieve morphological CR at day 42 induction (≥ 5% BM blasts) or had MRD ≥ 1% were considered high risk and received re-intensification therapy before the consolidation phase and were considered for allogeneic hematopoietic stem cell therapy (HSCT) if they have HLA-matched sibling donor. The consolidation therapy (8 weeks) consisted of 4 cycles of high-dose MTX (HDMTX) and 56 days of 6MP followed by maintenance treatment duration for 120 weeks for girls and 146 weeks for boys (supp. 1).

Statistical analysis

Data management and analysis were performed using SPSS, version 22 (IBM, Armonk, NY, USA). Qualitative data were presented as numbers and percentages, while the quantitative data were presented as median and interquartile ranges (IQR) according to the appropriate normality test. The comparison between groups was performed using chi-square test and/or Fisher exact test which is appropriate. The Mann–Whitney test was used for comparing numerical variables between two groups. Survival analysis was done using the Kaplan–Meier test, and comparison between survival curves was done using the log-rank test. Overall survival (OS) was calculated from the date of diagnosis until the date of last follow-up or death due to any cause. Disease-free survival (DFS) was calculated from the date of CR until the date of relapse, death during complete remission, or second malignant neoplasms. All tests of hypotheses had been conducted at the alpha level of 0.05, with a 95% confidence interval.

Results

Patient’s characteristics

The current study included 63 pediatric patients with de novo T-ALL with a median age of 7 (range of 1 to 18) years old. Males represented 69.8% (44/63), and females were 30.2% (19/63). Seventeen patients (26.9%) had intermediate T-phenotype, while 26 (41.3%) had late T-phenotype. The median of modal chromosome number (MCN) was 46 (range: 41–91); 11/63 patients (17.5%) had hyperdiploidy, while 6/63 (9.5%) had hypodipiody.

Abnormal cytogenetics were detected in 68.3% (43/63) T-ALL patients. Structural changes were more common than numerical ones with 62.8% (27/43) and 20.9% (9/43) of abnormal cases, respectively, while 16.3% (7/43) of abnormal cases displayed both structural and numerical anomalies. Baseline demographic, clinical, and laboratory characteristics are shown in Table 1.

Table 1 Clinical and laboratory characteristics of the study cohort
Full size table

Frequency of ETP and TCR aberrations in pediatric T-ALL

The ETP was observed in 7/63 (11.1%) of the patients. Upon screening by FISH, 4 types of TCRA/D aberrations were found among the 63 pediatric patients: (I) TCRA/D translocations in 20/63 (31.7%) patients, (II) TCRA/D amplifications in 11/63 (17.5%) patients, (III) TCRA/D deletions in 3/63 (4.8%) patients, and (IV) collective TCRA/D abnormalities, including the previously mentioned 3 aberrations, in 30/63 (47.6%) patients (Fig. 1).

Fig. 1
figure 1

Frequency of TCR/D, TCR-β, and ETP alterations in pediatric T-ALL cases

Full size image

TCRA/D translocation was the sole structural abnormality in 15/20 (71%) patients, while 5/20 (29%) harbored additional structural and/or numerical aberrations. The identified partners were chromosome 8 (n = 2.10%), chromosome 11 (n = 2.10%), and chromosome 6 and chromosome 12 (one case each), while 14 cases were of unidentified partner chromosome. Of note, 3/20 (15%) patients had concurrent TCRA/D amplification, while 1 patient had TCRA/D deletion.

Out of 11 cases with TCRA/D amplification, 7 cases (63.6%) gained one or more copy of chromosome 14 in hyperdiploid karyotypes, while 4 cases (36.3%) showed TCRA/D duplications. Notably, all cases with TCRA/D deletions had monosomy 14 in hypodiploidy karyotypes. Aberrations involving TCR-β were found in 13/63 patients (20.6%); 9/13 (69.2%) showed TCR-β structural abnormalities of which 3 (23%) cases with t(1;7)(p32;q34), while 4/13 (30.8%) gained extra-copy of TCR-β in hyperdiploid karyotypes.

Association of ETP with ALL patients’ characteristics

Based on the immunophenotype, 7/63 (11.1%) patients were classified as ETP and 56/63 (88.9%) as non-ETP. Patients with ETP were significantly older than the other T-ALL patients (P = 0.013). Initial WBC count less than 50,000/µl was seen significantly in 71.4% of ETP compared to 20.3% of non-ETP (P = 0.001). The ETP patients showed significantly lower percentage of PB blast cells compared to non-ETP patients [50% (0–85%) vs 85% (0–99%); respectively, P = 0.037]. Of note, all patients (100%) with ETP-ALL had lymph node enlargement at presentation, but the relation was not statistically significant.

Regarding cytogenetics, patients with ETP-ALL were more likely to have hyperdiploid karyotype compared to non-ETP patients (57.1% vs 12.5%, respectively, P = 0.009) with a nonsignificant trend to have complex karyotypes (42.9% vs 12.5%, respectively, P = 0.073, Table 2).

Table 2 Clinical features, laboratory characteristics, and outcome of ALL in relation to ETP subtype
Full size table

Interestingly, there was 85.7% (6/7) of the ETP-ALL patients showed increased incidence of total TCRA/D abnormalities compared to 42.9% (24/56) in non-ETP patients, P = 0.047. Similarly, there was a significant association between ETP-ALL and TCRA/D amplifications (P = 0.014), as TCRA/D amplifications were present in 57.1% (4/7) of ETP patients compared to 12.5% (7/56) in non-ETP patients (Fig. 2).

Fig. 2
figure 2

Frequency of TCR aberrations at the initial diagnosis according to the immunophenotype subgroups of T-ALL patients. Early-T phenotypes are double negative for CD4 and CD8; intermediate-T phenotypes were double positive for CD4 and CD8. Late T-phenotype: single positive for either CD4 or CD8

Full size image

Association between TCR aberrations and ALL patients’ characteristics

Among patients presenting with bleeding, 71.4% were positive to TCRA/D translocations (P = 0.028). Patients with TCRA/D amplification had lower WBC count (median-range): 52 (3–546) vs 240 (2–967), P = 0.02), lower percentage of BM blast cells (median-range): 80 (38–92) vs 90 (0–99), P = 0.05), compared to TCRA/D amplification-negative cases. Additionally, patients with TCRA/D translocations were positively associated with aberrant myeloid markers (P = 0.06); had additional chromosomal abnormalities (54.5% vs 23.1%, P = 0.036); and, subsequently, presented in complex karyotypes (45.5% vs 9.6%, P = 0.003) compared to those who did not show TCRA/D amplification (Table 3).

Table 3 Clinical features, laboratory characteristics, and outcome in relation to TCRA/D translocation and TCRA/D amplification
Full size table

Notably, there was a highly statistically significant relation between TCR-copy number variation groups (CNV) and MCN, where positive cases to TCRA/D amplification had higher MCN with a median of 50 (range: 46–91) chromosomes (P < 0.01). While all cases with TCRA/D deletion cases had MCN lower than 46 with a median of 45 (range: 41–45) chromosomes (P < 0.001, supp. 2). There was a significant association between TCRA/D-abnormalities and hyperdiploidy, as 10/11 (90.9%) of the patients with hyperdiploidy showed TCRA/D abnormalities (P = 0.006).

On the other hand, patients with TCR-β aberrations had lower WBC count (median-range):75 (2–883) vs 269 (2–967), P = 0.002), lower risk of CNS involvement at presentation (53.8% vs 94%, P = 0.003), positive association with CD7 expression (84.6% vs 51%, P = 0.05) compared to the ALL patients without TCR-β aberrations (Table 3).

Impact of ETP and TCR aberrations on the T-ALL patients’ response to treatment

Minimal residual disease (MRD) was performed by flow cytometry for 46 patients on day 15, and for 42 patients on day 42 according to the adequacy of the samples and the presence of leukemia-associated immunophenotypes (Table 1).

Patients with ETP-ALL showed a nonsignificant inferior clearance rate of leukemic cells compared to the other T-ALL patients, as all patients had detectable MRD (≥ 0.01%) after 15 days of treatment (Table 2). Only two patients had MRD ≥ 1% at the end of induction of whom one patient had ETP-ALL. However, due to the low number of patients, no statistical analysis could be done.

Eleven patients died during induction, 7 were from treatment-related causes (TRM), while 4 were from disease-related causes. Two of the seven patients with ETP (28.6%) died in induction from disease-related cause, while 9/56 patients with non-ETP (16%) died mainly due to TRM.

The morphological CR of the assessed ALL patients was significantly associated with the presence of TCR-β aberrations (P = 0.028), while a nonsignificant association between TCRA/D amplification and the achievement of the morphological CR was noted (P = 0.095, Fig. 3). In addition, patients with TCR-β abnormalities were more likely to have negative MRD on day 42 compared to TCR-β-negative patients (80% vs 29.7%, P = 0.047, Tables 3 and 4).

Fig. 3
figure 3

A–F showing the association of ETP and TCR abnormalities with the patients’ complete remission, while G–L showing the association of ETP and TCR abnormalities with the patients’ morphological complete remission. The colored parts represent the percent of patients who were positive for the ETP or TCR abnormalities

Full size image
Table 4 Clinical features, laboratory characteristics, and outcome in relation to total TCR-B aberrations and TCRA/D abnormality
Full size table

Impact of ETP and TCR aberrations on the clinical outcome of the assessed ALL patients

With a median follow-up of 22 (range: 1–86) months, eleven patients (17.5%) died during the induction, 8 cases (12.7%) relapsed after achieving CR of which 1 (14.3%), and 2 (64.2%) patients were positive to ETP and TCR abnormalities, respectively. The total number of deaths at the end of the study was 27 cases (42.9%). Characteristics of the relapsed pediatric T-ALL patients are illustrated in supp. 3.

Notably, 5/13 (38.5%) of patients with TCR-β aberrations died before the end of the induction compared to 6/50 (12%) of the negative cases (P = 0.04). The present study showed that there was no significant impact of TCR aberrations or ETP on the DFS or the OS of the assessed T-ALL patients (P > 0.05, Fig. 4).

Fig. 4
figure 4

Survival of pediatric T-ALL according to TCR aberrations status and ETP. A, B, C, D, I, and K Disease-free survival (DFS) and E, F, G, H, J, and L overall survival (OS) of different TCR aberrations

Full size image

Discussion

Acute lymphoblastic leukemia is the most common childhood malignancy. Although first described in 2009 by Coustan-Smith et al. [19], ETP-ALL was included, as a provisional entity only in the 2016 revision of the “WHO classification of tumors of hematopoietic and lymphoid tissues” [20] recognizing its fascinating scenario of biological heterogeneity [4]. Nevertheless, the distinction of ETP-ALL is still challenging. The “partial CD5 expression” criterion had a negative impact on the correct identification of ETP-ALL cases, as ETP patients with high CD5 expression showed a comparable survival to non-ETP T-ALL. Consequently, Zuubier et al. [21] proposed refined immunophenotypic criteria by excluding CD5 expression and adding negativity for CD4, categorizing the cases with elevated CD5 expression as “near ETP” [4, 22, 23].

In the current series, we analyzed the frequency, clinical, and prognostic features of pediatric ETP and TCRA/D-ALL-positive cases compared to the other T-ALL cases. Based on the immunophenotypic criteria proposed by recent studies [5, 6, 22,23,24,25], our data showed that ETP-ALL represented 11.1% of T-ALL patients which goes in agreement with the previously reported frequency in pediatric T-ALL [18, 21,22,23]. In contrast, a higher frequency (17%) was detected by Burns et al. [1], which was comparable to that observed in adult patients, due to the inclusion of patients ranged from 1 to 21 years in their series compared to 1–18 years in our cohort.

In concordance with previous studies [1, 5, 15, 19, 21,22,23, 26], our series demonstrated that there were no significant differences between the ETP and non-ETP patients except that ETP patients were older, presented with lower WBC count, and had lower percentage of PB blast cells than non-ETP patients. In contrast to these findings, Jain et al. [24] reported higher BM blast count and CNS involvement at presentation, while Puglianini et al. [6] reported an association of ETP with younger age. Moreover, several studies [5, 6, 21, 24] reported the frequent occurrence of ETP in males compared to female patients.

In the present study, most of our ETP-ALL patients (85.7%) had chromosomal aberrations and were less likely to have diploid karyotype (28.6%). In addition, ETP-ALL patients had a significant association with hyperdiploid karyotype, TCRA/D gene amplification, and a nonsignificant trend to have complex karyotypes. These findings were consistent with the other published series [4, 6, 19, 22, 24, 25], which reported that the higher likelihood of ETP-ALL patients harboring clonal cytogenetic lesions, with highly variable non-diploid karyotypes, and lower frequency of classical T-ALL translocations than the other T-ALL patients. In the same context, Coustan-Smith et al. [19] reported that ETP-ALL patients had significantly more DNA copy number abnormalities (genomic gains or losses) than other T-ALL patients. This high genomic instability may provide an explanation for the higher incidence of hyperdiploid karyotype and TCRA/D gene amplification in ETP phenotype compared to non-ETP patients in our series.

There was a nonsignificant association between ETP-ALL patients and positive MRD (> 0.1) after induction therapy; in addition, it was noted that all patients with ETP-ALL had detectable MRD on day 15. This goes in agreement with previous studies [1, 5, 6, 19, 22,23,24], which reported a lower incidence of clearance of leukemic cells after the first phase of induction therapy in patients with ETP-ALL compared to other T-ALL.

In line with Morita et al. [22], our study showed that patients with ETP-ALL tend to have worse OS than non-ETP patients, with no significant impact on patients’ DFS. In concordance with Wood et al. [23], no relapses occurred in patients with ETP-ALL later than 12-month post-diagnosis.

There was a discrepancy in long-term outcomes of ETP-ALL patients in different studies compared to non-ETP patients. The original early studies [1, 6, 19] reported that patients with ETP-ALL had a significantly worse OS and DFS, while the recent studies [4, 5, 23, 25,26,27] reported that ETP-ALL was not associated with a poor outcome, despite the higher incidence of a positive MRD at the end of induction. Such discrepancy may be explained by the limited numbers of pediatric patients enrolled in the early studies, while the recent studies were based on larger number of children. In addition, this was also likely due to the treatment intensification of the MRD-positive patients that abolished the negative prognostic impact of ETP-ALL [23].

In line with the previous studies [19], the frequency of TCRA/D aberrations in pediatric patients with T-ALL was 47.6% upon screening by FISH, which was higher than that previously reported by Kim et al. [28].

In agreement with the other series [8, 9], the incidence of TCR rearrangements was 20%. Interestingly, patients with TCRA/D amplification tend to have lower WBC count and lower percentage of BM blast cells and were positively associated with aberrant myeloid markers. In addition, patients with TCRA/D amplification had hyperdiploid karyotype and presented in complex karyotypes when compared to TCRA/D amplification-negative patients. These associations are similar to that found in patients with ETP-ALL which may explain the strong association of TCRA/D amplification in patients with ETP-ALL. In line with Kim et al. [28] who detected two or more instances of TCR clonality in most children with T-ALL, patients with TCRA/D amplification had concurrent TCR-β aberrations.

Moreover, the current study demonstrated that patients with TCR-β aberrations showed a morphological CR as well as negative MRD on day 42 compared to TCR-β-negative patients. These data are in agreement with Dutta et al. [29] who concluded a better outcome in patients with TCR-β selection. Also, Brüggemann et al. [30] found that clonal TCR-β gene rearrangements are excellent targets for RQ-PCR detection of MRD in T-ALL patients.

High percentage of pediatric and young adult patients achieves early MRD negativity; more than 80% of T-ALL patients remain MRD positive at the end of induction as demonstrated in the AIEOP-BFM-ALL 2000 study [31].

In agreement with previously published papers [8, 9], the present study showed that cytogenetic features do not play a prognostic role in T-ALL, nor do types of TCR aberrations as there was no significant difference between TCRA/D aberrations positive and negative cases regarding the DFS or OS.

Conclusion

In summary, there was a nonsignificant trend of ETP-positive cases to have lower outcome but mainly due to increased mortalities. TCR-β aberrations were significantly associated with negative MRD at the end of induction compared to TCR-β-negative patients. No significant association was found between TCR aberrations and DFS and OS. However, the main limitation of the present study was the small number of the ETP-ALL-positive patients in relation to the other compared non-ETP-ALL patients which affects the reliability of the results. Therefore, further studies with a larger number of patients are highly required for better validation of the data. This will allow for proper integrating the genetic and the clinico-biological data of the T-ALL patients, translating them into better risk stratification, and subsequently improving the outcomes of T-ALL patients.