Deep Learning-Based Blood Abnormalities Detection as a Tool for VEXAS Syndrome Screening.

INTRODUCTION: VEXAS is a syndrome described in 2020, caused by mutations of the UBA1 gene, and displaying a large pleomorphic array of clinical and hematological features. Nevertheless, these criteria lack significance to discriminate VEXAS from other inflammatory conditions at the screening step. This work hence first focused on singling out dysplastic features indicative of the syndrome among peripheral blood (PB) polymorphonuclears (PMN). A deep learning algorithm is then proposed for automatic detection of these features. METHODS: A multicentric dataset, comprising 9514 annotated PMN images was gathered, including UBA1 mutated VEXAS (n = 25), UBA1 wildtype myelodysplastic (n = 14), and UBA1 wildtype cytopenic patients (n = 25). Statistical analysis on a subset of patients was performed to screen for significant abnormalities. Detection of these features on PB was then automated with a convolutional neural network (CNN) for multilabel classification. RESULTS: Significant differences were observed in the proportions of PMNs with pseudo-Pelger, nuclear spikes, vacuoles, and hypogranularity between patients with VEXAS and both cytopenic and myelodysplastic controls. Automatic detection of these abnormalities yielded AUCs in the range [0.85-0.97] and a F1-score of 0.70 on the test set. A VEXAS screening score was proposed, leveraging the model outputs and predicting the UBA1 mutational status with 0.82 sensitivity and 0.71 specificity on the test patients. CONCLUSION: This study suggests that computer-assisted analysis of PB smears, focusing on suspected VEXAS cases, can provide valuable insights for determining which patients should undergo molecular testing. The presented deep learning approach can help hematologists direct their suspicions before initiating further analyses.

Evaluation of a machine-learning model based on laboratory parameters for the prediction of acute leukaemia subtypes: a multicentre model development and validation study in France.

BACKGROUND: Acute leukaemias are life-threatening haematological cancers characterised by the infiltration of transformed immature haematopoietic cells in the blood and bone marrow. Prompt and accurate diagnosis of the three main acute leukaemia subtypes (ie acute lymphocytic leukaemia [ALL], acute myeloid leukaemia [AML], and acute promyelocytic leukaemia [APL]) is of utmost importance to guide initial treatment and prevent early mortality but requires cytological expertise that is not always available. We aimed to benchmark different machine-learning strategies using a custom variable selection algorithm to propose an extreme gradient boosting model to predict leukaemia subtypes on the basis of routine laboratory parameters. METHODS: This multicentre model development and validation study was conducted with data from six independent French university hospital databases. Patients aged 18 years or older diagnosed with AML, APL, or ALL in any one of these six hospital databases between March 1, 2012, and Dec 31, 2021, were recruited. 22 routine parameters were collected at the time of initial disease evaluation; variables with more than 25% of missing values in two datasets were not used for model training, leading to the final inclusion of 19 parameters. The performances of the final model were evaluated on internal testing and external validation sets with area under the receiver operating characteristic curves (AUCs), and clinically relevant cutoffs were chosen to guide clinical decision making. The final tool, Artificial Intelligence Prediction of Acute Leukemia (AI-PAL), was developed from this model. FINDINGS: 1410 patients diagnosed with AML, APL, or ALL were included. Data quality control showed few missing values for each cohort, with the exception of uric acid and lactate dehydrogenase for the cohort from Hôpital Cochin. 679 patients from Hôpital Lyon Sud and Centre Hospitalier Universitaire de Clermont-Ferrand were split into the training (n=477) and internal testing (n=202) sets. 731 patients from the four other cohorts were used for external validation. Overall AUCs across all validation cohorts were 0·97 (95% CI 0·95-0·99) for APL, 0·90 (0·83-0·97) for ALL, and 0·89 (0·82-0·95) for AML. Cutoffs were then established on the overall cohort of 1410 patients to guide clinical decisions. Confident cutoffs showed two (0·14%) wrong predictions for ALL, four (0·28%) wrong predictions for APL, and three (0·21%) wrong predictions for AML. Use of the overall cutoff greatly reduced the number of missing predictions; diagnosis was proposed for 1375 (97·5%) of 1410 patients for each category, with only a slight increase in wrong predictions. The final model evaluation across both the internal testing and external validation sets showed accuracy of 99·5% for ALL diagnosis, 98·8% for AML diagnosis, and 99·7% for APL diagnosis in the confident model and accuracy of 87·9% for ALL diagnosis, 86·3% for AML diagnosis, and 96·1% for APL diagnosis in the overall model. INTERPRETATION: AI-PAL allowed for accurate diagnosis of the three main acute leukaemia subtypes. Based on ten simple laboratory parameters, its broad availability could help guide initial therapies in a context where cytological expertise is lacking, such as in low-income countries. FUNDING: None.

Phase I study of zoledronic acid combined with escalated doses of interleukine-2 for early in vivo generation of Vγ9Vδ2 T-cells after haploidentical stem cell transplant with posttransplant cyclophosphamide.

The presence of donor Vγ9Vδ2 T-cells after haploidentical hematopoietic stem cell transplant (h-HSCT) has been associated with improved disease-free survival. These cells kill tumor cells in a non-MHC restricted manner, do not induce graft-versus-host disease (GVHD), and can be generated by stimulation with zoledronic acid (ZA) in combination with interleukin-2 (IL-2). This monocentric phase I, open-label, dose-escalating study (ClinicalTrials.gov: NCT03862833) aimed at evaluating the safety and possibility to generate Vγ9Vδ2 T-cells early after h-HSCT. It applied a standard 3 + 3 protocol to determine the maximum tolerated dose (MTD) of increasing low-doses of IL-2 (5 days [d] per week, 4 weeks) in combination with a single dose of ZA, starting both the first Monday after d + 15 posttransplant. Vγ9Vδ2 T-cell monitoring was performed by multiparameter flow cytometry on blood samples and compared with a control cohort of h-HSCT recipients. Twenty-six patients were included between April 2019 and September 2022, 16 of whom being ultimately treated and seven being controls who received h-HSCT only. At the three dose levels tested, 1, 0, and 1 dose-limiting toxicities were observed. MTD was not reached. A significantly higher number of Vγ9Vδ2 T-cells was observed during IL-2 treatment compared with controls. In conclusion, early in vivo generation of Vγ9Vδ2 T-cells is feasible after h-HSCT by using a combination of ZA and repeated IL-2 infusions. This study paves the way to a future phase 2 study, with the hope to document lesser posttransplant relapse with this particular adaptive immunotherapy.

[Can academic structures improve access to CAR-T cells?] / Les structures académiques peuvent-elles permettre une amélioration de l'accessibilité aux CAR-T cells ?

In France, hospital cell therapy units have not been authorised to routinely produce chimeric antigen receptor T lymphocytes (CAR-T cells), which would then be referred to as academic CAR-T cells. CAR-T cells are classified as advanced therapy medicinal products and correspond to genetically modified T lymphocytes ex vivo. The CAR-T cell production process is complex and requires scientific and technical expertise to meet the acceptance criteria of the pharmaceutical quality system. The most commonly used method for genetically modifying T lymphocytes is viral transduction (lentiviral or retroviral), which requires prior access to a batch of good manufacturing practice (GMP) grade viral vector. Because of its cost, this reagent is the main limiting factor for developing CAR-T cells. A CAR-T cell produced by an industrial company is expensive (around €350,000 per injection) and the time taken by the manufacturer to make it available to the clinician can vary from three to five weeks. By meeting the economic and ecological challenges, can academic structures improve access to CAR-T cells? In this article, we present the elements necessary for the feasibility of setting up CAR-T cell production in an academic structure.

Added value of molecular karyotype in childhood acute lymphoblastic leukemia.

Background: Thanks to an improved therapeutic regimen in childhood B-cell precursor acute lymphoblastic leukemia (BCP-ALL), 5 year-overall survival now exceeds 90%. Unfortunately, the 25% of children who relapse have an initial poor prognosis, potentially driven by pre-existing or emerging molecular anomalies. The latter are initially and essentially identified by cytogenetics. However, some subtle alterations are not visible through karyotyping. Methods: Single nucleotide polymorphisms (SNP) array is an alternative way of chromosomal analysis allowing for a more in-depth evaluation of chromosomal modifications such as the assessment of copy number alterations (CNA) and loss of heterozygosity (LOH). This method was applied here in retrospective diagnosis/relapse paired samples from seven children with BCP-ALL and in a prospective cohort of 38 newly diagnosed childhood cases. Results: In the matched study, compared to the initial karyotype, SNP array analysis reclassified two patients as poor prognosis cases. Modulation during relapse was seen for 4 CNA and 0.9 LOH. In the prospective study, SNP reclassified the 10 patients with intermediate karyotype as 7 good prognosis and 3 poor prognosis. Ultimately, in all the children tested, SNP array allowed to identify additional anomalies compared to conventional karyotype, refine its prognostic value and identify some druggable anomalies that could be used for precision medicine. Overall, the anomalies detected could be segregated in four groups respectively involved in B-cell development, cell proliferation, transcription and molecular pathways. Conclusion: SNP therefore appears to be a method of choice in the integrated diagnosis of BCP ALL, especially for patients initially classified as intermediate prognosis. This complementary method of both cytogenetics and high throughput sequencing allows to obtain further classified information and can be useful in case of failure of these techniques.

Assessment of monocytic-myeloid-derived suppressive cells (M-MDSC) before and after allogeneic hematopoietic stem cell transplantation in acute leukemia patients.

In this monocentric prospective study, the influence on long-term outcomes of peripheral blood levels of monocytic-myeloid-derived suppressive cells (M-MDSC) was investigated in 56 patients with acute leukemia (myeloid n = 47; lymphoid n = 9) before and after (Days+60/+90) allogeneic hematopoietic stem cell transplantation (Allo-HSCT). A risk of relapse was found to be associated with a level of pregraft M-MDSC above 1.4% by ROC curve analysis. In multivariate analysis, this threshold retained a strong statistical significance (HR: 5.94 [2.09-16.87], p = 0.001). Considering only the group of patients who were in complete remission prior to Allo-HSCT (n = 44), a significant prediction of relapse was found to be associated, in multivariate analysis, with a level of pregraft M-MDSC above 1.4% (HR: 55.01 [14.95-202.37], p < 0.001) together with pregraft-positive measurable -residual disease (MRD) (HR: 11.04 [1.89-64.67], p = 0.008). A poorer OS (HR: 6.05 [1.24-29.59], p = 0.026) and disease-free survival (HR: 6.52 [1.41-30.19], p = 0.016) were also associated with higher levels of pregraft M-MDSC. Remarkably, no relapse occurred in patients with pregraft-negative MRD and ≤1.4% of M-MDSC (vs. a 3-year relapse rate of 60% for others, p = 0.004). Patients developing grade 3-4 acute graft-versus-host-disease (GVHD, median occurrence: day+30 posttransplant) showed significantly higher levels of M-MDSC% at days +60 and +90, suggesting a possible amplification of these immunosuppressive cells as a reaction to GVHD. In conclusion, this prospective study demonstrates a negative impact of higher proportions of peripheral M-MDSC before Allo-HSCT in leukemic patients. This paves the way to potential therapeutic intervention to decrease M-MDSC levels before Allo-HSCT and thus perhaps the incidence of relapse in such patients.

Tocilizumab in combination with a standard induction chemotherapy in acute myeloid leukaemia patients (TOCILAM study): a single-centre, single-arm, phase 1 trial.

Background: In acute myeloid leukaemia (AML), interleukin-6 (IL-6) promotes chemo-resistance and its levels correlate with poor prognosis. IL-6 blockade may represent a promising therapeutic strategy. We aimed to test, tocilizumab, an anti-IL-6 receptor (R) monoclonal antibody in combination with standard intensive AML induction chemotherapy. Methods: This investigator-initiated single-centre phase 1 trial was conducted at Nantes University Hospital in France. According to a continual reassessment method, three escalating doses were tested of intravenous (IV) tocilizumab (4, 6, and 8 mg/kg) administered at day (d) 8 of a standard AML induction chemotherapy (IV idarubicine 8 mg/m2 d1 to d5 + IV cytarabine 100 mg/m2 d1 to d7). All adults (aged ≥ 18 years) with an Eastern Cooperative Oncology Group performance status of 0-2 and with a newly diagnosed (excluding patients with a favourable risk according to ELN-2017 classification if <60 year-old) or a relapsed/refractory AML were eligible. The primary objective was to determine the maximum tolerated dose of tocilizumab to administrate with a standard intensive AML induction. Safety outcomes were continuously monitored for at each participant contact. This trial is registered with ClinicalTrials.gov, NCT04547062. Findings: Between Dec 29, 2020 and Dec 1, 2022, 12 patients were enrolled, of whom 75% had an ELN-2017 high-risk profile, and were treated with tocilizumab- two patients at 4 mg/kg, two at 6 mg/kg and eight at 8 mg/kg of tocilizumab. No dose-limiting toxicity related to tocilizumab was documented. There were nine serious adverse events, none of which were related to tocilizumab, and there was no treatment-related deaths. MTD was thus not reached. Two deaths occurred during induction. In the remaining ten evaluable patients, nine responded to treatment. Interpretation: The combination of tocilizumab with standard AML intensive induction appears to be safe and resulting responses are encouraging. A dose of 8 mg/kg of tocilizumab given at day 8 of induction could be used for further phase 2/3 studies. Funding: The Leucémie Espoir Atlantique Famille (LEAF)-"Tous avec Fabien" association.

Prothrombin consumption as an indicator of hemorrhagic phenotype in mild platelet function disorders.

BACKGROUND: The bleeding risk of patients with mild platelet function disorders is difficult to assess and their phenotype remains ill-explored. AIM: This study was designed to establish a comprehensive biological phenotype of patients with mild platelet function disorders. METHODS: Twenty patients were included with persistent abnormal light transmission aggregometry (LTA). The ISTH bleeding assessment tool (ISTH-BAT) was assessed to identify laboratory analyses associated with an abnormal hemorrhagic score. RESULTS: The majority of patients had defects that might affect Gαi protein signaling pathways or minor abnormalities. No LTA nor flow cytometry parameters were associated with an above-normal hemorrhagic score. However, prothrombin consumption, which corresponds to the ratio of serum residual factor II to plasma residual factor II, was significantly higher (p = .006) in the abnormal ISTH-BAT group (mean = 14%, SD = 6) compared with the normal ISTH-BAT group (mean = 8%, SD 4). Prothrombin consumption was significantly associated with ISTH-BAT score (r = .5287, IC 95% 0.0986-0.7924, p = .0165). CONCLUSION: In this group of patients, there was an association between a pathological bleeding score and increased prothrombin consumption. This test could be used as an additional indicator of platelet function abnormality liable to be related to bleeding risk.

Two Ways of Targeting a CD19 Positive Relapse of Acute Lymphoblastic Leukaemia after Anti-CD19 CAR-T Cells.

BACKGROUND: Therapeutic options for CD19+ relapses after anti-CD19 CAR-T cells are still debated; second infusion of anti-CD19 CAR-T cells, therapeutic antibodies, or targeted therapies can be discussed. Here, we explore the immunophenotyping and lysis sensitivity of CD19+ ALL relapse after anti-CD19 CAR-T cells and propose different therapeutic options for such a high-risk disease. METHODS: Cells from successive B-ALL relapses from one patient were collected. A broad immunophenotype analysis was performed. 51Cr cytotoxic assays, and long-term killing assays were conducted using T-cell effectors that are capable of cytotoxicity through three recognition pathways: antibody-dependent cell-mediated cytotoxicity (ADCC), anti-CD19 CAR-T, and TCR. RESULTS: Previously targeted antigen expression, even if maintained, decreased in relapses, and new targetable antigens appeared. Cytotoxic assays showed that ALL relapses remained sensitive to lysis mediated either by ADCC, CAR-T, or TCR, even if the lysis kinetics were different depending on the effector used. We also identified an immunosuppressive monocytic population in the last relapse sample that may have led to low persistence of CAR-T. CONCLUSION: CD19+ relapses of ALL remain sensitive to cell lysis mediated by T-cell effectors. In case of ALL relapses after immunotherapy, a large immunophenotype will make new therapies possible for controlling such high risk ALL.

Peripheral levels of monocytic myeloid-derived suppressive cells before and after first induction predict relapse and survivals in AML patients.

Myeloid Derived Suppressive Cells (MDSC) are capable to suppress innate and adaptive immune responses, thus favouring solid cancer progression. However, little is known about the role of MDSC in acute myeloid leukaemia (AML). In this monocentric prospective study, 73 adult AML patients, eligible for first-line intensive chemotherapy, were included with the aim to study the influence on long-term outcomes of peripheral blood (PB) levels of monocytic (M) MDSC (M-MDSC) assessed by flow cytometry. A percentage of peripheral M-MDSC higher than 0.55% of leukocytes at diagnosis and a decrease of M-MDSC% after induction came out both as independent negative prognostic factors for leukaemia-free and overall survival.

Biological, clinical features and modelling of heterozygous variants of glycoprotein Ib platelet subunit alpha (GP1BA) and glycoprotein Ib platelet subunit beta (GP1BB) genes responsible for constitutional thrombocytopenia.

Constitutional thrombocytopenias are rare disorders, often difficult to discriminate from acquired thrombocytopenias. More than 80 genes have been described as being at the origin of these diseases. Among them, several variants of the glycoprotein Ib platelet subunit alpha (GP1BA) and glycoprotein Ib platelet subunit beta (GP1BB) genes, coding for the GpIb-IX-V glycoprotein complex, have been reported in the literature. The study reported here aimed at describing newly identified monoallelic anomalies affecting the GP1BA and GP1BB genes on a clinical, biological and molecular level. In a cohort of nine patients with macrothrombocytopenia, eight heterozygous variants of the GP1BA or GP1BB genes were identified. Five of them had never been described in the heterozygous state. Computer modelling disclosed structure/function relationships of these five variants.