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.

[Can academic structures improve access to 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.

Outcomes after allogeneic hematopoietic stem cell transplantation for adults with primary mediastinal B cell lymphoma: a SFGM-TC and LYSA study

Background: Despite therapeutic progress, 10 to 30% of adult patients with primary mediastinal B cell lymphoma (PMBCL) are primary refractory or experience early relapse (R/R). Allogeneic stem cell transplantation (allo-HSCT) thus remains a potentially curative option in this setting.Material and Methods: In this multicenter retrospective study, the outcomes of 33 French and Belgian adult patients allo-transplanted for R/R PMBCL between January 1999 and December 2018, were examined.Results: At allo-HSCT time, patients had received a median of 3 treatment lines, 50% of them were in complete response, 40% in partial response and 10% had a progressive disease. Forty-two percent of the donors were siblings and 39% matched related. The median follow-up for alive patients was 78 months (3.5-157). Considering the whole cohort, 2-year overall survival (OS), progression free survival (PFS) and graft-versus-host disease-free/relapse-free survival (GRFS) were 48% (95%CI: 33-70), 47% (95%CI: 33-68) and 38.5% (95%CI: 25-60) respectively. Cumulative incidence of relapse and non-relapse mortality rates were respectively 34% (95%CI: 18-50) and 18% (95%CI: 7-34). Disease status at transplant was the only factor predicting survivals, patients with progressive disease showing significant lower 2-year PFS (HR: 6.12, 95%CI: 1.32-28.31, p = 0.02) and OS (HR: 7.04, 95%CI: 1.52-32.75, p = 0.013). A plateau was observed for OS and PFS after 4 years with 10 patients alive after this date, suggesting that almost one third of the patients effectively salvaged and undergoing allo-SCT could be cured.Conclusion: This study indicates that allo-HSCT is a valid therapeutic option for R/R PMBCL, providing durable remissions.

Extensive myelitis with eosinophilic meningitis after Chimeric antigen receptor T cells therapy

Immune effector cell‐associated neurotoxicity syndrome (ICANS) is a frequent adverse event after Chimeric antigen receptor T cells (CAR‐T cells). A patient treated with anti‐CD19 CAR‐T cells for a refractory mantle cell lymphoma presented at Day 8 post‐infusion with extensive myelitis. Unusual eosinophilia was disclosed in the patient's cerebrospinal fluid. After treatment with methylprednisolone and siltuximab, a decrease in clinical symptoms and magnetic resonance imaging lesions were obtained. This unprecedented presentation of eosinophilic meningitis after CAR‐T cells therapy highlights the need for a better understanding of the physiopathology of ICANS, especially to identify potentially targetable pathways.

Immune checkpoint inhibitors for the treatment of myeloma: novel investigational options

INTRODUCTION

Multiple myeloma (MM) is still considered incurable and the outcome of patients with triple-class refractory remains very poor. Immunotherapy is considered as a standard of care for the treatment of MM. Among immunotherapeutic approaches, the PD-1/PD-L1 axis is an attractive target because PD-L1 is highly expressed in most myeloma plasma cells. While many types of cancer benefit from checkpoint inhibitor treatment, their relevance in multiple myeloma needs to be defined.

AREAS COVERED

The authors evaluate the published data regarding the mechanism of action, safety profile, and clinical efficacy of the immune checkpoint inhibitors (ICI) for the treatment of multiple myeloma.

EXPERT OPINION

The use of ICI monotherapy does not offer any clinical benefit in myeloma patients. In combination with immunomodulatory drugs (IMID), ICI failed to demonstrate clinical benefit and were associated with increased toxicity. Given the toxicities of these treatments, predictive markers would be useful to select patients who would benefit most. Clinical studies are necessary to evaluate the safety and efficacy of checkpoint inhibitors in combination with other standards of care such as proteasome inhibitors and monoclonal antibodies. The combination of anti-PD-1 with T-cell engager (TCE) or CAR-T cells seems theoretically attractive and should be explored in clinical trials.

RAS mutation leading to acquired resistance to dabrafenib and trametinib therapy in a multiple myeloma patient harboring BRAF mutation

Multiple myeloma (MM) is still considered incurable and new therapeutic approaches are therefore needed. Deep‐sequencing analysis revealed the presence of BRAF mutations in up to 15% of patients. The clinical experience of BRAF‐targeted therapy in myeloma patients harboring BRAF mutation is still limited. We here report the case of a patient with penta‐refractory (bortezomib, lenalidomide, carfilzomib, pomalidomide, and daratumumab) MM with extramedullary BRAF‐mutated disease that achieved clinical response to dual BRAF and MEK inhibition. At the time of disease progression, gene sequencing analysis of the tumor at the time of progression demonstrated a clonal evolution with emergence of a NRAS mutation and persistence of BRAF and TP53 mutations. Backtracking of the NRAS mutation was performed by digital polymerase chain reaction on the baseline biopsy and identified the pre‐existence of the NRAS at a subclonal level. This observation is the first report of acquired NRAS mutation leading to resistance to dual BRAF/MEK inhibitors in MM. These data suggest that a systematic search for RAS mutations using highly sensitive techniques should be performed before considering targeted therapy in relapsed myeloma with BRAF mutation.