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.

Comparative diagnostic accuracy between simplified and original flow cytometric gating strategies for peripheral blood neutrophil myeloperoxidase expression in ruling out myelodysplastic syndromes.

BACKGROUND: Flow cytometric analysis of peripheral blood neutrophil myeloperoxidase expression is accurate in ruling out myelodyplastic syndromes (MDS) but might not be suitable for implementation in busy clinical laboratories. We aimed to simplify the original gating strategy and examine its accuracy. METHODS: Using the individual data from 62 consecutive participants enrolled in a prospective validation study, we assessed the agreement in intra-individual robust coefficient of variation (RCV) of peripheral blood neutrophil myeloperoxidase expression and compared diagnostic accuracy between the simplified and original gating strategies. RESULTS: Cytomorphological evaluation of bone marrow aspirate confirmed MDS in 23 patients (prevalence, 37%), unconfirmed MDS in 32 patients (52%), and was uninterpretable in 7 patients (11%). Median intra-individual RCV for simplified and original gating strategies were 30.7% (range, 24.7-54.4) and 30.6% (range, 24.7-54.1), with intra-class correlation coefficient quantifying absolute agreement equal to 1.00 (95% confidence interval [CI], 0.99 to 1.00). The areas under the receiver operating characteristic (ROC) curves were 0.93 (95% CI, 0.82-0.98) and 0.92 (95% CI, 0.82-0.98), respectively (P = .32). Using simplified or original gating strategy, intra-individual RCV values lower than a pre-specified threshold of 30.0% ruled out MDS for 35% (19 of 55) patients, with both sensitivity and negative predictive value estimates of 100%. CONCLUSIONS: The simplified gating strategy performs as well as the original one for ruling out MDS and has the potential to save time and reduce resource utilization. Yet, prospective validation of the simplified gating strategy is warranted before its adoption in routine. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT03363399 (First posted on December 6, 2017).

Flow cytometry lyophilised-reagent tube for quantifying peripheral blood neutrophil myeloperoxidase expression in myelodysplastic syndromes (MPO-MDS-Develop): protocol for a diagnostic accuracy study.

INTRODUCTION: Suspicion of myelodysplastic syndromes (MDS) is the most common reason for bone marrow aspirate in elderly patients. Peripheral blood neutrophil myeloperoxidase expression quantified by flow cytometric analysis might rule out MDS for up to 35% of patients referred for suspected disease, without requiring bone marrow aspiration. Yet laboratory-developed liquid antibody cocktails have practical limitations, because of lack of standardisation and poor stability. This research project aims to estimate the level of agreement and comparative accuracy between a single-use flow cytometry tube of lyophilised reagents (BD Lyotube Stain 468) and its laboratory-developed liquid reagent counterpart in quantifying peripheral blood neutrophil myeloperoxidase expression, among adult patients referred for suspected MDS. METHODS AND ANALYSIS: The MPO-MDS-Develop project is a cross-sectional diagnostic accuracy study of two index tests by comparison with a reference standard in consecutive unselected adult patients conducted at a single university hospital. Flow cytometry analysis of peripheral blood samples will be performed by independent operators blinded to the reference diagnosis, using either Lyotube Stain 468 or laboratory-developed liquid reagent cocktail. The reference diagnosis of MDS will be established by cytomorphological evaluation of bone marrow aspirate by two independent haematopathologists blinded to the index test results. Morphologic assessment will be complemented by bone marrow flow cytometric score, karyotype and targeted next-generation sequencing panel of 43 genes, where relevant. The target sample size is 103 patients. ETHICS AND DISSEMINATION: An institutional review board (Comité de Protection des Personnes Sud Est III, Lyon, France) approved the protocol prior to study initiation (reference number: 2020-028-B). Participants will be recruited using an opt-out approach. Efforts will be made to release the primary results within 6 months of study completion. TRIAL REGISTRATION NUMBER: NCT04399018.

AML with inv(16)/t(16;16) and high-risk cytogenetic abnormalities: atypical features and unfavorable outcome.

OBJECTIVES: Acute myeloid leukemia (AML) with inv(16)/t(16;16) is among the most frequent AML subtypes. It is recognized by the detection of the CBFB-MYH11 fusion which confers a favorable prognosis, irrespective of the presence of secondary cytogenetic abnormalities. However, the effect of additional genetic anomalies on the behavior of inv(16) AML is debatable. Recent case reports describe an unfavorable prognosis for those patients, characterized by early relapse and death. In this study, we present a series of patients with CBFB-MYH11 fusion and high-risk rearrangements to increase knowledge about this potentially distinct subgroup. METHODS: All cases with inv(16)/ t(16;16) and one or more high risk abnormalities were reviewed at two tertiary healthcare centers between years 2006 and 2020 in terms of demographics, biological and clinical data. RESULTS: Among the total 1447 and 1283 AML cases, the frequency was found to be 0,2% and 0.3%. Clinical data could be retrieved for 5 patients. Detected high-risk abnormalities included TP53 and 5q deletion, complex and monosomal karyotype. The median age was 67 years, with a majority of females (M:F = 1:1.5). Two out of 5 patients presented with therapy related AML, with short latency periods. All patients presented with thrombocytopenia and/or leukocytopenia. Bone marrow aspirates revealed atypical morphology and the detection of rare CBFB-MYH11 fusion transcripts. All 5 patients died, with a short mean overall survival of 5.8 months. DISCUSSION AND CONCLUSION: Our series suggests that the presence of high risk abnormalities confers distinct biological features and poor prognosis to inv(16) AML.

CYCLON and NPM1 Cooperate within an Oncogenic Network Predictive of R-CHOP Response in DLBCL.

R-CHOP immuno-chemotherapy significantly improved clinical management of diffuse large B-cell lymphoma (DLBCL). However, 30-40% of DLBCL patients still present a refractory disease or relapse. Most of the prognostic markers identified to date fail to accurately stratify high-risk DLBCL patients. We have previously shown that the nuclear protein CYCLON is associated with DLBCL disease progression and resistance to anti-CD20 immunotherapy in preclinical models. We also recently reported that it also represents a potent predictor of refractory disease and relapse in a retrospective DLBCL cohort. However, only sparse data are available to predict the potential biological role of CYCLON and how it might exert its adverse effects on lymphoma cells. Here, we characterized the protein interaction network of CYCLON, connecting this protein to the nucleolus, RNA processing, MYC signaling and cell cycle progression. Among this network, NPM1, a nucleolar multi-functional protein frequently deregulated in cancer, emerged as another potential target related to treatment resistance in DLBCL. Immunohistochemistry evaluation of CYCLON and NPM1 revealed that their co-expression is strongly related to inferior prognosis in DLBCL. More specifically, alternative sub-cellular localizations of the proteins (extra-nucleolar CYCLON and pan-cellular NPM1) represent independent predictive factors specifically associated to R-CHOP refractory DLBCL patients, which could allow them to be orientated towards risk-adapted or novel targeted therapies.

c-MYC and p53 expression highlight starry-sky pattern as a favourable prognostic feature in R-CHOP-treated diffuse large B-cell lymphoma.

Diffuse large B-cell lymphoma (DLBCL) is a clinically heterogeneous entity, in which the first-line treatment currently consists of an immuno-chemotherapy regimen (R-CHOP). However, around 30% of patients will not respond or will relapse. Overexpression of c-MYC or p53 is frequently found in DLBCL, but an association with prognosis remains controversial, as for other biomarkers previously linked with DLBCL aggressivity (CD5, CD23, or BCL2). The aim of this study was to explore the expression of these biomarkers and their correlation with outcome, clinical, or pathological features in a DLBCL cohort. Immunohistochemical (c-MYC, p53, BCL2, CD5, and CD23), morphological ('starry-sky' pattern [SSP]), targeted gene panel sequencing by next-generation sequencing (NGS), and fluorescence in situ hybridisation analyses were performed on tissue microarray blocks for a retrospective cohort of 94 R-CHOP-treated de novo DLBCL. In univariate analyses, p53 overexpression (p53high ) was associated with unfavourable outcome (p = 0.04) and with c-MYC overexpression (p = 0.01), whereas c-MYC overexpression was linked with an SSP (p = 0.004), but only tended towards an inferior prognosis (p = 0.06). Presence of a starry-sky morphology was found to be correlated with better survival in p53high DLBCL (p = 0.03) and/or c-MYC-positive DLBCL (p = 0.002). Furthermore, NGS data revealed that these three variables were associated with somatic mutations (PIM1, TNFRSF14, FOXO1, and B2M) involved in B-cell proliferation, survival, metabolism, and immune signalling. Taken together, these results show that the SSP pattern seems to be a protective factor in high-risk DLBCL subgroups and highlight cell death as a built-in failsafe mechanism to control tumour growth.

Adverse outcome in follicular lymphoma is associated with MYC rearrangements but not MYC extra copies.

Follicular lymphomas (FLs) with MYC rearrangements (MYC-R) and extra copies of MYC (MYC-EC) are rare and the prognosis impact is uncertain. We conducted a retrospective study including 321 FL patients, among whom 259 (81%) had no 8q24 alterations and 62 (19%) were assigned to 8qAlt. Forty-five cases were classified as MYC-EC and six as MYC-R. MYC-R patients were significantly older (P = 0·008), had higher follicular lymphoma international prognostic index (FLIPI) stage (P = 0·05) and ß2-microglobulin (ß2m; P = 0·05). Among patients treated with immuno-chemotherapy, four presented a MYC-R and 25 a MYC-EC. Univariate analysis showed the absence of significant difference between MYC-EC and normal MYC (MYC-NL) regarding progression-free survival (PFS; HR1·3; 95% CI [0·4-1·6]) and specific overall survival (SOS; HR 1·6; 95% CI [0·4-5·7]). Those results were compared to data from the PRIMA trial. This confirmed that MYC-EC had no impact on PFS (P = 0·86) or SOS (P = 0·9). Conversely, MYC-R was associated with a trend to inferior outcome regarding PFS (HR : 6·1; 95% CI [2·2-17·1]; P = 0·00026), lymphoma-related death (SOS; HR 13·6; 95% CI [2·9-65]; P = 0·00014) and risk of transformation (transformation-free survival (TFS); HR 82·7; 95% CI [14·8-463·4]; P < 0·0001). In conclusion, MYC-EC has no prognostic impact in FL but MYC-R FL tended to be associated with an increased risk of transformation and poorer outcome.

Pathogenic Roles of S100A8 and S100A9 Proteins in Acute Myeloid and Lymphoid Leukemia: Clinical and Therapeutic Impacts.

Deregulations of the expression of the S100A8 and S100A9 genes and/or proteins, as well as changes in their plasma levels or their levels of secretion in the bone marrow microenvironment, are frequently observed in acute myeloblastic leukemias (AML) and acute lymphoblastic leukemias (ALL). These deregulations impact the prognosis of patients through various mechanisms of cellular or extracellular regulation of the viability of leukemic cells. In particular, S100A8 and S100A9 in monomeric, homodimeric, or heterodimeric forms are able to modulate the survival and the sensitivity to chemotherapy of leukemic clones through their action on the regulation of intracellular calcium, on oxidative stress, on the activation of apoptosis, and thanks to their implications, on cell death regulation by autophagy and pyroptosis. Moreover, biologic effects of S100A8/9 via both TLR4 and RAGE on hematopoietic stem cells contribute to the selection and expansion of leukemic clones by excretion of proinflammatory cytokines and/or immune regulation. Hence, the therapeutic targeting of S100A8 and S100A9 appears to be a promising way to improve treatment efficiency in acute leukemias.

Mitochondria in human acute myeloid leukemia cell lines have ultrastructural alterations linked to deregulation of their respiratory profiles.

Mitochondria not only are essential for cell metabolism and energy supply but are also engaged in calcium homeostasis and reactive oxygen species generation and play a key role in apoptosis. As a consequence, functional mitochondrial disorders are involved in many human cancers including acute myeloid leukemia (AML). However, very few data are available on the deregulation of their number and/or shape in leukemic cells, despite the evident link between ultrastructure and function. In this context, we analyzed the ultrastructural mitochondrial parameters (number per cell, mitochondria area, number of cristae/mitochondria, cristal thickness) in five leukemia cell lines (HEL, HL60, K562, KG1, and OCI-AML3) together with the functional assay of their respiratory profile. First, we describe significant differences in basal respiration, maximal respiration, ATP production, and spare respiratory capacity between our cell lines, confirming the various respiratory profiles among leukemia subtypes. Second, we highlight that these variations are obviously associated with significant interleukemia heterogeneity of the number and/or shape of mitochondria. For instance, KG1, characterized by the smallest number of mitochondria together with reduced cristal diameter, had a particularly deficient respiratory profile. In comparison, the HEL and K562 cell lines, both with high respiratory profiles, harbored the largest number of mitochondria/cells with high cristal diameters. Moreover, we report that K562, carrying the ASXL1 mutation, presents significant mitochondria-endoplasmic reticulum deficiency reflected by decreases in the numbers of matrix granules and mitochondria-associated endoplasmic reticulum membrane (MAM) and mitochondrial-derived vesicle (MDV) precursors, which are implicated in the regulatory pathways of cell mortality via the processes of mitophagy and calcium homeostasis. Contrarily, HL60 carried high levels of matrix granules and MAMs and had a higher sensitivity to drugs targeting mitochondria (rotenone/antimycin). We confirm the implication of ASXL1 mutation in this mitochondria dysregulation through the study of transcript expression (from 415 patients with public data) involved in three mitochondrial pathways: (1) endoplasmic reticulum-mitochondria contacts (MAMs), (2) matrix granule homeostasis, and (3) MDV precursor production. Our study offers new and original data on mitochondria structural alterations linked to deregulation of respiration profiles in AMLs and some genetic characteristics, suggesting that modifications of mitochondrial shape and/or number in leukemic cells participate in chemoresistance and could be a targeted mechanism to regulate their proliferative potential.

Pharmacogenetics may influence the impact of inflammation on voriconazole trough concentrations.

How pharmacogenetics modulates the inhibitory effects of inflammation on voriconazole trough concentration (Cmin) remains unknown. In 29 recipients of allogeneic hematopoietic stem cell transplantation retrospectively studied, both a genetic score (which aggregated CYP2C19 and CYP3A genotypes) and inflammation significantly influenced voriconazole Cmin (n = 260). A trend toward (p = 0.03) a greater impact of inflammation in patients with the highest genetic score (corresponding to ultra-rapid metabolizers) was observed. Further researches are needed to confirm these data.