Frequency and Early Prognostic Significance of Gene-Class Fusions Identified By RNA-Seq in Children and Adolescent with T-Cell Acute Lymphoblastic Leukemia Enrolled on AIEOP-BFM ALL 2017 Study

Introduction

T-cell acute lymphoblastic leukemia (T-ALL) has been historically associated with inferior outcome compared to B-cell precursor ALL with slower treatment response and lack of recurrent prognostic cytogenetic alterations. With the introduction of next-generation sequencing analyses, a deeper characterization of childhood ALL has been achieved. However, if for B-ALL the identification of fusion-class genes has clinical implication, the frequency and relevance of gene fusions in T-ALL still remain poor. Here we describe the incidence of gene-class fusions in the pediatric cohort of T-ALL patients treated according to the AIEOP-BFM ALL 2017 protocol and explore the association with presenting patient characteristics and early treatment response.

Methods

From Dec 1^st 2021 to Mar 31^st 2024, in AIEOP Italian centers 132 consecutive patients with T-ALL have been enrolled in the trial AIEOP-BFM ALL 2017 Study. According to material availability, Whole transcriptome RNA-seq was performed at disease onset in 128 (97%) patients (Universal RNA-seq kit, Tecan with paired-end sequencing on Nextseq2000). The analysis was performed with 4 independent pipelines with the purpose to detect fusion genes (Dragen RNA, Star, Fusion Catcher, Arriba) and results were evaluable in 123 (96%) patients (5 cases were excluded due to poor library quality).

Results

Among 123 evaluable patients, at least one fusion gene was identified in 51 (41.5%) patients and validated through real-time PCR. Most common gene-class fusions were STIL::TAL1 (n=10, 8.1%), KMT2A-rearranged (KMT2Ar, n=9, 7.3%), ABL-class (n=8, 6.5%) and MLLT10-class (n=8, 6.5%). Other fusions were identified in 16 (13.0%) patients, with ETV6-class in 4 patients (3.3%) and CDK6-class in 2 patients (1.6%), in addition to single non recurrent fusion/cases. All gene-class fusions were mutually exclusive.

Patient characteristics at diagnosis were evaluated in association with the presence of gene-class fusions. ABL-class and MLLT10-class fusions were more common in females and patients with MLLT10-class fusion tended to be older (age ≥ 10 years). Patients with STIL::TAL1 fusion were more likely to have hyperleukocytosis (white blood cell counts ≥ 100 × 10⁹/L) and central nervous system disease (CNS3).

We investigated whether the presence of gene-class fusion was associated with treatment response in term of prednisone poor response (PPR), high level of bone marrow disease (≥10% of blasts) measured by flow-cytometry (FCM-MRD) at induction day 15, end of induction (EOI) PCR-MRD, end of consolidation (EOC) PCR-MRD and final risk group classification. Patients carrying a KMT2Ar were characterized by higher proportion of PPR, FCM-MRD with ≥10% of blasts and higher EOI MRD levels: of note, approximately half (44.4%) of them had EOI MRD ≥5x10^-3. The presence of MLLT10-class fusion was associated with high EOI MRD, with 50% of patients having levels ≥5x10^-3, as well as for STIL::TAL1 patients (40% of patients with ≥5x10^-3). No association was found at EOC between PCR-MRD levels and gene-class fusions. Interestingly, all patients carrying KMT2Ar except one were classified as HR while a lower proportion of HR patients was found among patients with ABL-class fusion. Indeed, the presence of ABL-class fusion was not associated with any unfavorable early treatment responses.

Conclusion

Our study showed a high incidence of gene-class fusions in children and adolescents with newly diagnosed T-ALL. We found that KMT2Ar, STIL::TAL1 and MLLT10-class fusions are associated with high risk features in a prospective cohort of uniformly treated patients with pediatric T-ALL. On the contrary, patients carrying ABL-class fusions did not show adverse characteristics of early treatment response. Despite a longer follow-up is required, prognostic relevant gene-class fusions might integrate current risk stratification and identify patients who might qualify for early interventional studies or targeted therapies.