From microscope to micropixels: A rapid review of artificial intelligence for the peripheral blood film.

Artificial intelligence (AI) and its application in classification of blood cells in the peripheral blood film is an evolving field in haematology. We performed a rapid review of the literature on AI and peripheral blood films, evaluating the condition studied, image datasets, machine learning models, training set size, testing set size and accuracy. A total of 283 studies were identified, encompassing 6 broad domains: malaria (n = 95), leukemia (n = 81), leukocytes (n = 72), mixed (n = 25), erythrocytes (n = 15) or Myelodysplastic syndrome (MDS) (n = 1). These publications have demonstrated high self-reported mean accuracy rates across various studies (95.5% for malaria, 96.0% for leukemia, 94.4% for leukocytes, 95.2% for mixed studies and 91.2% for erythrocytes), with an overall mean accuracy of 95.1%. Despite the high accuracy, the challenges toward real world translational usage of these AI trained models include the need for well-validated multicentre data, data standardisation, and studies on less common cell types and non-malarial blood-borne parasites.

Heavy/light chain assay in the monitoring of multiple myeloma.

Serum protein (SPE) and immunofixation electrophoresis (IFE) have been extensively validated for the routine use of identifying, characterising and quantifying monoclonal proteins. However, accurate quantitation of IgA monoclonal proteins can be difficult when they migrate in to the ß fraction, due to co-migration with transferrin and complement components. The heavy/light chain (HLC) immunoassay is an additional tool for measuring intact immunoglobulin monoclonal proteins. Therefore, we aimed to examine the clinical utility of the HLC assay for the disease monitoring of IgG and IgA multiple myeloma (MM) patients. A total of 177 samples from 30 MM patients (21 IgG and 9 IgA) were analysed retrospectively with median number of six follow up samples per patient (range 3-13). Serum free light chains (sFLC) and HLC were quantified using Freelite and Hevylite immunoassays. Details of M-protein concentration, ß-globulin levels, total immunoglobulin levels and disease treatment response were obtained from the laboratory and patient information system. Passing-Bablok regression analysis was performed to compare (i) M-protein quantification with involved HLC (iHLC) and (ii) total immunoglobulin with summated HLC pairs for each immunoglobulin type (e.g., IgGκ+IgGλ). For 127 IgG MM samples, IgG iHLC levels showed a good correlation with SPE quantification (iHLC y=0.96x+4.9; r=0.917) and summated HLC showed a good correlation with total IgG concentration (summated HLC y=0.94x+5.74; r=0.91). In total, 95/127 (75%) IgG MM follow-up samples had an abnormal HLC ratio and 122/127 (96%) had a positive SPE, probably due to the lower sensitivity of HLC assay in detecting clonality in patients with IgG MM. Consistent with this, one patient assigned a very good partial response by International Myeloma Working Group criteria would be assigned a complete response based on HLC measurements. For 50 IgA MM samples, 42/50 (84%) had an abnormal HLC ratio. Conversely, 50/50 (100%) of M-proteins showed ß fraction migration and were difficult to accurately quantify by SPE. Therefore, M-protein concentration and iHLC did not correlate as well in IgA MM (y=1.9x-8.4; r=0.8) compared to IgG MM. However, there was good correlation between total IgA and summated IgA HLC (IgAκ+IgAλ y=1.35x-0.33; r=0.95). Of the 8/50 (16%) IgA samples with a normal HLC ratio, 6/8 (75%) were consistent with the disease status being in complete remission. Interestingly, in one IgA MM patient, SPE and IFE were negative, but the serum FLC ratio and involved FLC were highly abnormal, consistent with the presence of light chain escape. Our data suggest HLC measurements could add value to the current disease monitoring of MM patients. In IgG MM patients, the M-protein level correlated well with HLC values. The HLC assay complements the serum FLC assay and is especially useful for monitoring of IgA MM patients who display M-proteins migrating in the ß region on SPE.

Effect of HbE heterozygosity on the measurement of HbA1c.

AIM: Measurement of HbA1c provides an excellent measure of glycaemic control for diabetic patients. However, haemoglobin (Hb) variants are known to interfere with HbA1c analysis. In our laboratory HbA1c measurement is performed by Variant II turbo 2.0. The aim of this study is to investigate the influence of HbE trait on HbA1c analysis. METHODS: Haemoglobin variants were identified by HbA1c analysis in 93 of 3522 samples sent to our laboratory in a period of 1 month. Haemoglobin analysis identified HbE trait in 81 of 93 samples. To determine the influence of HbE trait on HbA1c analysis by Variant II Tubo 2.0, boronate affinity high performance liquid chromatography (HPLC) method (Primus PDQ) was used as the comparison method. Two stage linear regression analysis, Bland Altman plot and Deming regression analysis were performed to analyse whether the presence of HbE trait produced a statistically significant difference in the results. The total allowable error for HbA1c by the Royal Australasian College of Pathologists (RCPA) external quality assurance is 5%. Hence clinically significant difference is more than 5% at the medical decision level of 6% and 9%. RESULTS: Statistically and clinically significant higher results were observed in Variant II Turbo 2.0 due to the presence of HbE trait. A positive bias of ∼10% was observed at the medical decision levels. CONCLUSION: Laboratories should be cautious when evaluating HbA1c results in the presence of haemoglobin variants.

Myeloid sarcoma: an unusual presentation of acute promyelocytic leukemia causing spinal cord compression.

Acute promyelocytic leukemia with concurrent myeloid sarcoma is a rare clinical event. Herein we describe a patient that presented with back pain and bilateral leg weakness caused by spinal cord compression due to extramedullary deposition of leukemic cells. Acute promyelocytic leukemia was suspected based on immunophenotypic findings of malignant cells in bone marrow aspirate. The diagnosis was confirmed by the presence of PML-RARα fusion copies. MRI showed multiple hyperintense changes on the vertebral bodies, together with intraspinal masses causing spinal cord compression. The patient immediately underwent radiotherapy, and was treated with all-trans retinoic acid and idarubicin. Reassessment MRI showed complete resolution of all intraspinal masses and the disappearance of most of the bony lesions. Post-treatment bone marrow aspirate showed complete hematological and molecular remission. The motor power of his legs fully recovered from 0/5 to 5/5; however, sensory loss below the T4 level persisted.

A case report of adult T-cell leukaemia/lymphoma.

Adult T-cell leukaemia/lymphoma (ATLL) is a rare T lymphoproliferative disorder which is aetiologically linked with human T-cell lymphotropic virus type-1 (HTLV-1). HTLV-1 is endemic in Japan, Caribbean and Africa. The highest incidence of ATLL is in Japan although sporadic cases have been reported elsewhere in the world. We describe a case of ATLL with an unusual presentation which we believe is the first reported case of ATLL in Malaysia based on our literature search. A 51-year-old Indian lady was referred to University Malaya Medical Centre for an incidental finding of lymphocytosis while being investigated for pallor and giddiness. Clinical examination revealed bilateral shotty cervical lymph nodes with no hepato-splenomegaly or skin lesions. Laboratory investigations showed absolute lymphocytosis (38 x 10(9)/L) with a mildly increased serum lactate dehydrogenase. The peripheral blood smear showed the presence of predominantly small to medium sized, non-flower lymphocytes. The bone marrow showed similar findings of prominent lymphocytosis. Immunophenotyping of the bone marrow mononuclear cells showed CD3+, CD4+, CD5+, CD7- and CD25+ which is characteristic of ATLL phenotype. HTLV-1 infection was confirmed by the presence of HTLV-1 proviral DNA in the tumor cells using conventional Polymerase Chain Reaction (PCR) and real-time PCR. Here, we discuss the pathogenesis and characteristics of ATLL as well as the detection of HTLV-1 by real time PCR.