An artificial intelligence-assisted clinical framework to facilitate diagnostics and translational discovery in hematologic neoplasia.

BACKGROUND: The increasing volume and intricacy of sequencing data, along with other clinical and diagnostic data, like drug responses and measurable residual disease, creates challenges for efficient clinical comprehension and interpretation. Using paediatric B-cell precursor acute lymphoblastic leukaemia (BCP-ALL) as a use case, we present an artificial intelligence (AI)-assisted clinical framework clinALL that integrates genomic and clinical data into a user-friendly interface to support routine diagnostics and reveal translational insights for hematologic neoplasia. METHODS: We performed targeted RNA sequencing in 1365 cases with haematological neoplasms, primarily paediatric B-cell precursor acute lymphoblastic leukaemia (BCP-ALL) from the AIEOP-BFM ALL study. We carried out fluorescence in situ hybridization (FISH), karyotyping and arrayCGH as part of the routine diagnostics. The analysis results of these assays as well as additional clinical information were integrated into an interactive web interface using Bokeh, where the main graph is based on Uniform Manifold Approximation and Projection (UMAP) analysis of the gene expression data. At the backend of the clinALL, we built both shallow machine learning models and a deep neural network using Scikit-learn and PyTorch respectively. FINDINGS: By applying clinALL, 78% of undetermined patients under the current diagnostic protocol were stratified, and ambiguous cases were investigated. Translational insights were discovered, including IKZF1plus status dependent subpopulations of BCR::ABL1 positive patients, and a subpopulation within ETV6::RUNX1 positive patients that has a high relapse frequency. Our best machine learning models, LDA and PASNET-like neural network models, achieve F1 scores above 97% in predicting patients' subgroups. INTERPRETATION: An AI-assisted clinical framework that integrates both genomic and clinical data can take full advantage of the available data, improve point-of-care decision-making and reveal clinically relevant insights promptly. Such a lightweight and easily transferable framework works for both whole transcriptome data as well as the cost-effective targeted RNA-seq, enabling efficient and equitable delivery of personalized medicine in small clinics in developing countries. FUNDING: German Ministry of Education and Research (BMBF), German Research Foundation (DFG) and Foundation for Polish Science.

The kinetics of blast clearance are associated with copy number alterations in childhood B-cell acute lymphoblastic leukemia.

We analyzed the pattern of whole-genome copy number alterations (CNAs) and their association with the kinetics of blast clearance during the induction treatment among 195 pediatric patients with B-cell precursor acute lymphoblastic leukemia (BCP-ALL) who displayed intermediate or high levels of minimal residual disease (MRD). Using unsupervised hierarchical clustering of CNAs > 5 Mbp, we dissected three clusters of leukemic samples with distinct kinetics of blast clearance [A - early slow responders (n=105), B - patients with persistent leukemia (n=24), C - fast responders with the low but detectable disease at the end of induction (n=66)] that corresponded with the patients' clinical features, the microdeletion profile,the presence of gene fusions and patients survival. Low incidence of large CNAs and chromosomal numerical aberrations occurred in cluster A which included ALL samples showing recurrent microdeletions within the genes encoding transcription factors (i.e., IKZF1, PAX5, ETV6, and ERG), DNA repair genes (XRCC3 and TOX), or harboring chromothriptic pattern of CNAs. Low hyperdiploid karyotype with trisomy 8 or hypodiploidy was predominantly observed in cluster B. Whereas cluster C included almost exclusively high-hyperdiploid ALL samples with concomitant mutations in RAS pathway genes. The pattern of CNAs influences the kinetics of leukemic cell clearance and selected aberrations affecting DNA repair genes may contribute to BCP-ALL chemoresistance.

Case report: Successful allogeneic stem cell transplantation in a child with novel GATA2 defect associated B-cell acute lymphoblastic leukemia.

GATA-binding protein 2 (GATA2) is a transcription factor responsible for the regulation of blood cell proliferation, differentiation, and maintenance in hematopoietic stem cells. Here, we describe successful bone marrow transplantation in a carrier of a novel GATA2 pathogenic variant who was diagnosed with immunodeficiency a few years after completion of B-cell precursor acute lymphoblastic leukemia (BCP-ALL) treatment. At the age of 4 years, the patient was diagnosed with and treated for BCP-ALL. Antileukemic therapy was complicated by pulmonary cryptococcosis. Two years after completion of the maintenance therapy, the child was consulted by an immunologist because of recurrent respiratory tract infections and an episode of sepsis. Flow cytometry revealed deep monocytopenia, lymphopenia, absence of B lymphocytes, considerably reduced NK cells, poor thymic T lymphocyte production, minor defects in T cell maturation, and absence of TCRγδ+ T cells. The presence of the likely pathogenic, heterozygous missense variant within exon 5 of GATA2 (NM_032638.5: c.1047T>G, Cys349Trp) was identified in the proband and confirmed in the father of the patient, who underwent allogeneic hematopoietic stem cell transplantation (HSCT) from a matched unrelated donor due to myelodysplastic syndrome with excess blasts at the age of 22 years. An allogeneic hematopoietic stem cell transplantation with a reduced toxicity conditioning protocol was performed using a matched sibling donor. Pre-transplant conditioning included fludarabine (5 × 30 mg/m2), treosulfan (3 × 14 g/m2), and thiotepa (10 mg/kg). Complete donor chimerism was achieved on post-transplant day 17. During the 12 months of the posttransplant observation period, she remained free from symptoms of acute or chronic graft-versus-host disease, and immunosuppressive treatment was therefore stopped. This is the second reported case of BCP-ALL in a patient with GATA2 deficiency, and the first successfully treated with a reduced-toxicity conditioning HSCT protocol. The co-occurrence of lymphoid malignancies and primary immunodeficiencies points to the role of genetic counseling and family screening for possible cancer predisposition syndromes prior to the selection of related HSCT donors.