{"title":"Hematological cytomorphology: Where we are","authors":"G. Zini","doi":"10.1111/ijlh.14330","DOIUrl":"10.1111/ijlh.14330","url":null,"abstract":"<p>The manuscript discusses the historical evolution of observing blood cell morphology under an optical microscope, from the earliest microscopes in the 17th century to the modern digital era, highlighting key advancements and contributions in the field. Blood has historically held symbolic importance in various cultures, with early medical observations dating back to Hippocrates and Galeno. The discovery of cells and subsequent advancements in microscopy by scientists like Hooke and van Leeuwenhoek paved the way for understanding blood cell morphology. Influential figures such as Hewson, Donné, and Ehrlich followed. Diagnostic cytology evolved from manual cell counting to the development of automated hematological systems. Automated complete blood counting came to support microscopic examination in diagnosing hematological disorders. Morphology is crucial in predicting disease outcomes and guiding treatment decisions, particularly hematological neoplasms. The introduction of flow cytometry and its integration with traditional morphological analysis and the new cytogenetic and molecular techniques revolutionized the classification and prognostication of hematologic disorders. Digital microscopy has emerged as a powerful tool in recent years, offering rapid acquisition and sharing of blood cell images. Integrating Artificial Intelligence with digital microscopy has further enhanced morphological analysis, improving diagnostic efficiency. We also discuss the prospects of AI in pre-classifying blood cells in bone marrow aspirate samples, potentially revolutionizing diagnostic pathways for hematologic diseases. Overall, the manuscript provides a comprehensive overview of the historical development, clinical significance and technological advancements in observing blood cell morphology, underscoring its continued relevance in modern hematology practice.</p>","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 5","pages":"789-794"},"PeriodicalIF":2.2,"publicationDate":"2024-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/ijlh.14330","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141428568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Stefanos Eskioglou, Loredana-Mariana Gheorghe, Nikolaos J. Tsagarakis, Ioulia Chaliori, Sofia Chaniotaki
{"title":"A case report of a Chediak-Higashi syndrome diagnosed by peripheral blood smear","authors":"Stefanos Eskioglou, Loredana-Mariana Gheorghe, Nikolaos J. Tsagarakis, Ioulia Chaliori, Sofia Chaniotaki","doi":"10.1111/ijlh.14329","DOIUrl":"10.1111/ijlh.14329","url":null,"abstract":"","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 6","pages":"994-996"},"PeriodicalIF":2.2,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141422286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Near-triploidy with four Philadelphia chromosomes in adult B-lymphoblastic leukemia with BCR::ABL1 fusion","authors":"Katsuya Yamamoto, Yuri Hirakawa, Sakuya Matsumoto, Kimikazu Yakushijin, Hironobu Minami","doi":"10.1111/ijlh.14327","DOIUrl":"10.1111/ijlh.14327","url":null,"abstract":"","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 6","pages":"988-990"},"PeriodicalIF":2.2,"publicationDate":"2024-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141422287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Homayemem Weli, John L. Frater, Gail Shimer, Stephen T. Oh, Cara Lunn Shirai
{"title":"Pseudo-lymphocytosis caused by circulating megakaryocyte fragments in a patient with post-essential thrombocythemia myelofibrosis","authors":"Homayemem Weli, John L. Frater, Gail Shimer, Stephen T. Oh, Cara Lunn Shirai","doi":"10.1111/ijlh.14328","DOIUrl":"10.1111/ijlh.14328","url":null,"abstract":"","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 6","pages":"991-993"},"PeriodicalIF":2.2,"publicationDate":"2024-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141319371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Negative expression of CD117 predicted inferior OS and PFS in acute promyelocytic leukemia","authors":"Hui Zeng, Jie He, Hai-Bo Dong, Min Zhou, Qian Zhang, Lan-Xin Chen, Cui-Ying Yuan, Ru-Ru Jiang, Jin-Wen Liu, Jian Ou-Yang, Yu Ben, Bing Chen","doi":"10.1111/ijlh.14326","DOIUrl":"10.1111/ijlh.14326","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>In recent years, the correlation between CD117 antigen and the prognosis of hematological malignancies has been demonstrated. However, there is limited literature on the clinical significance of CD117 antigen in acute promyelocytic leukemia (APL). The aim of this study was to retrospectively analyze the clinical features and prognostic significance of CD117 in APL.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>In this study, we retrospectively investigated the clinicopathological characteristics, outcome, and prognostic impact of negative CD117 expression (CD117<sup>−</sup>) in 169 APL patients treated with all-trans retinoic acid (ATRA) and arsenic trioxide (ATO) containing regimen.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>The median follow-up period was 63.0 months. CD117<sup>−</sup> was detected in 13 APL patients (7.7%). No significant differences were found in baseline characteristics between CD117+ and CD117<sup>−</sup> subgroups. However, compared to CD117+ APL, the incidence of early death (ED) was significantly higher in CD117<sup>−</sup> APL (<i>p</i> = 0.023). By multivariate analysis, CD117- was an independent adverse prognostic factor for overall survival (OS) and progression-free survival (PFS) (<i>p</i> = 0.022 and <i>p</i> = 0.014, respectively).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>To sum up, CD117<sup>−</sup> is associated with greater risk of ED and has the statistical power to predict inferior OS and PFS, this marker may be considered to build prognostic scores for risk-adapted therapeutic strategies in APL management.</p>\u0000 </section>\u0000 </div>","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 6","pages":"1052-1058"},"PeriodicalIF":2.2,"publicationDate":"2024-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/ijlh.14326","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141319370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Implementation of the recommended immunohistochemistry algorithm for classification of peripheral T-cell lymphoma, not otherwise specified into the prognostically significant GATA3 and TBX21 subtypes","authors":"Surabhi Jain, Aijaz Ahmad, Ambreen Jan, Ajay Gogia, Mukul Aggarwal, Ganesh Kumar Viswanathan, Trisha Mandal, Atul Sharma, Ranjit Sahoo, Mehar Chand Sharma, Sameer Bakhshi, Lalit Kumar, Saumyaranjan Mallick","doi":"10.1111/ijlh.14325","DOIUrl":"10.1111/ijlh.14325","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>Current molecular research has shown the several oncogenic pathways that give rise to the peripheral T-cell lymphoma, not otherwise defined (PTCL, NOS) subtypes, which alter prognosis and might have predictive value. This study was conducted to assess the immunohistochemistry (IHC) algorithm by Amador et al for the subtyping of PTCL, NOS and determine its applicability in relation to the clinicopathological profile.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>This study included 43 patients with PTCL, NOS diagnosis. Following the use of IHC for the transcription factors GATA3, TBX21, CCR4, and CXCR3, two pathologists subtyped the samples. Comprehensive clinicopathological correlation was carried out.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Applying the algorithm of Amador et al., cases were classified into GATA3 (20), TBX21 (15), and unclassified (8) subtypes. No significant association with clinical parameters of subtypes or CD4/ CD8 positivity was observed. Although a higher proportion of cases in the TBX21 subgroup showed a polymorphic population compared with the GATA3 subgroup, which had a monomorphic population, no significant <i>p</i>-value (0.111) was observed. Two Lennert lymphomas were classified into the GATA3 subgroup. Multivariate analysis showed no significant difference in overall survival (<i>p</i>-value = 0.105) and progression-free survival (<i>p</i>-value = 0.0509) between IHC-defined subtypes; trends indicate that overall survival and progression-free survival are worse in the GATA3 subgroup.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusion</h3>\u0000 \u0000 <p>Although the algorithm is reproducible, a proportion of cases remains unclassifiable and may require additional investigation and gene expression profiling. The GATA3 subgroup was found to have a monomorphic population with a poor overall prognosis and thus requires a larger sample size for validation.</p>\u0000 </section>\u0000 </div>","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 6","pages":"1059-1067"},"PeriodicalIF":2.2,"publicationDate":"2024-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141319369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Koen Jacobs, Alena Moerman, Karl Vandepoele, Tim Van den Abeele, Katrien De Mulder, Eva Steel, Maxim Clauwaert, Henk Louagie
{"title":"Variant-specific BCR::ABL1 quantification discrepancy in chronic myeloid leukemia","authors":"Koen Jacobs, Alena Moerman, Karl Vandepoele, Tim Van den Abeele, Katrien De Mulder, Eva Steel, Maxim Clauwaert, Henk Louagie","doi":"10.1111/ijlh.14320","DOIUrl":"10.1111/ijlh.14320","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>Accurate quantification of the <i>BCR::ABL1</i> fusion gene in whole blood is pivotal for the clinical management of chronic myeloid leukemia (CML) patients. The fusion protein encoded by <i>BCR::ABL1</i> can vary in size, depending on the <i>BCR</i> and/or <i>ABL1</i> gene breakpoint. The vast majority of CML patients have a p210 <i>BCR::ABL1</i> fusion gene (M-BCR), which can be attributed to the presence of either e14a2 (b3a2) or e13a2 (b2a2) mRNA transcript junctions.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>Twenty-five CML samples were analyzed in two different ISO15189-accredited centers that both use an Europe Against Cancer-based quantitative polymerase chain reaction (qPCR) protocol. Reanalysis of the sample set with transcript-specific standard curves and digital droplet PCR (ddPCR) were performed.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>qPCR quantification revealed a significant (up to 1 log) difference specifically for the e13a2 transcript variant in contrast to e14a2 transcripts (Hodges–Lehman 4.29; <i>p</i> < 0.001). Reanalysis of the sample set with transcript-specific standard curves abolishes the initial transcript-specific difference (Hodges–Lehman 0.003; <i>p</i> = 0.8192). Comparison of transcript-specific qPCR results of both centers with ddPCR, an absolute quantification method, showed a statically significant association, especially in the lower range, indicating the clinical utility of transcript-specific or absolute quantification methods.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusion</h3>\u0000 \u0000 <p>Our data show that differences between transcript-specific quantification might exist between centers, leading to potential clinical impact on the follow-up of CML patients. The use of transcript-specific standard curves for qPCR quantification, or absolute quantification, can significantly reduce these differences. Specific attention should be applied to the interpretation of quantification differences of CML patients that switch between diagnostic centers.</p>\u0000 </section>\u0000 </div>","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 5","pages":"910-917"},"PeriodicalIF":2.2,"publicationDate":"2024-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141260734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fei Gao, Shuang Chen, Jianwei Li, Zailin Yang, Cui Mao
{"title":"The critical role of phytohemagglutinin-stimulated cell cultures in the diagnosis of T-cell prolymphocytic leukemia: A case-based approach","authors":"Fei Gao, Shuang Chen, Jianwei Li, Zailin Yang, Cui Mao","doi":"10.1111/ijlh.14323","DOIUrl":"10.1111/ijlh.14323","url":null,"abstract":"<p>T-cell prolymphocytic leukemia (T-PLL) is a very rare subtype of the mature lymphocytic malignancies that typically occur in middle-aged and older individuals, accounting for approximately 2% of all mature T-cell lymphomas.<span><sup>1</sup></span> T-PLL is characterized by a poor median survival rate and distinctive cell morphological, immunophenotypic, and cytogenetic features. The major diagnostic criteria of the guidelines for T-PLL diagnosis<span><sup>2</sup></span> include: (1) 5 × 10<sup>9</sup>/L cells of the T-PLL phenotype in peripheral blood or bone marrow (BM); (2) T-cell clonality determined via PCR or flow cytometry (FCM); and, (3) an abnormal 14q32 or Xq28 karyotype or expression of <i>TCL1A/B</i> or <i>MTCP1</i>. Additionally, four minor criteria also contribute to the diagnostic framework: (1) abnormalities involving chromosome 11 (11q22.3; <i>ATM</i>); (2) abnormalities involving chromosome 8 such as idic(8)(p11), t(8;8), trisomy 8q; (3) abnormalities in chromosomes 5, 12, 13, 22, or a complex karyotype; and, (4) T-PLL-specific features (e.g., splenomegaly, effusions). A diagnosis of T-PLL is established if all three major criteria are met or if the first two such criteria and at least one minor criterion are met. Obviously, analysis of chromosomal abnormalities is essential for diagnosis of T-PLL. However, T-PLL poses a challenge in this regard given the low proliferation capacity of mature T lymphocytes, leading to a high failure rate of chromosomal karyotyping associated with conventional culture methods (short-term culture).</p><p>In this study, we retrospectively analyze and present five cases for whom we used a phytohemagglutinin (PHA)-stimulated culture method to increase the detection rate of chromosomal abnormalities in BM cells. This enhanced T-PLL diagnostic accuracy.</p><p>All five patients were males of median age 58 years (range, 42 to 83 years). Immunophenotypic abnormalities were observed in 50% ~ 90% of T cells of all patients (Table S1) in BM sample. Most exhibited the phenotype CD3<sup>+</sup>, CD2<sup>+</sup>, CD4<sup>+</sup>, CD5<sup>+</sup>, CD7<sup>+</sup>, CD56<sup>−</sup>, CD34<sup>−</sup>, CD117<sup>−</sup>, CD1a<sup>−</sup>, and CD52<sup>+</sup>. The most notable immunophenotypic difference was that for CD8: cases 1, 4, and 5 were of CD8<sup>part+</sup> status; case 2 CD8<sup>−</sup>; and case 3 CD8<sup>+</sup>. Subsequently, chromosomal karyotyping was performed on cultures grown with and without PHA (Table 1). Three of the five patients (cases 1, 2, and 3) exhibited only a 46,XY karyotype when cells were grown without PHA, but complex karyotypes, inv(14), trisomy 8q, and abnormalities in 11q when cells were grown with PHA. One patient (case 5) exhibited no metaphase cells when cells were grown without PHA, but a complex karyotype, abnormalities in 11q, and t (14;14) when cells were grown with PHA. Only one patient (case 4) yielded consistent chromosomal karyotypes when cells were grown with and wi","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 5","pages":"975-977"},"PeriodicalIF":2.2,"publicationDate":"2024-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/ijlh.14323","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141260585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xue-Shen Yan, Yu-Jiao Sun, Juan Du, Wen-Yan Niu, Han Qiao, Xiang-Cong Yin
{"title":"Effects of ferroptosis-related gene HSPB1 on acute myeloid leukemia","authors":"Xue-Shen Yan, Yu-Jiao Sun, Juan Du, Wen-Yan Niu, Han Qiao, Xiang-Cong Yin","doi":"10.1111/ijlh.14319","DOIUrl":"10.1111/ijlh.14319","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>The purpose of this study was to investigate the effects and potential mechanisms of ferroptosis-related gene heat shock protein beta-1 (HSPB1) on acute myeloid leukemia (AML).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>The RNA-seq and clinical data of AML samples were obtained from the Genomic Data Commons database, and the FerrDb database was used to screen the marker, drive and suppressor of ferroptosis. Besides, DESeq2 was applied for differential expression analysis on AML samples and screening for differentially expressed genes (DEGs). The screened DEGs were subjected to the intersection analysis with ferroptosis-related genes to identify the ferroptosis-related DEGs. Next, the functional pathways of ferroptosis-related DEGs were further be discussed by Gene Ontology as well as Kyoto Encyclopedia of Genes and Genomes enrichment analysis of DEGs. Additionally, lasso regression analysis was employed to determine the differential genes related to prognosis in patients with AML and the survival analysis was performed. Subsequently, quantitative real-time polymerase chain reaction and western blot assay were applied to detect the mRNA and protein expression levels of HSPB1 in normal/AML bone marrow tissues and human normal (HS-5)/AML (HL-60) bone marrow cells, respectively. Furthermore, HSPB1 was knocked down to assess the expression changes of glutathione peroxidase 4 and acyl-CoA synthetase long-chain family member 4. Ultimately, the viability and oxidative stress levels of HL-60 were analyzed by Cell Counting Kit-8 and biochemical detection.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>A total of 4986 DEGs were identified in AML samples, with 3324 up-regulated and 1662 down-regulated. The enrichment analysis illustrated that ferroptosis-related DEGs were significantly enriched in response to metal irons, oxidative stress, and other pathways. After lasso regression analysis, 17 feature genes related to the prognosis of patients with AML were obtained, with HSPB1 exhibiting a significant correlation. The reliability of our models was verified by Cox regression analysis and survival analysis of the hazard model. Furthermore, the outcomes of quantitative real-time polymerase chain reaction and western blot showed that mRNA and protein expression levels of HSPB1 were significantly increased in the AML Group and HL-60 cells. The knockdown of HSPB1 in HL-60 cells reduced the protein level of glutathione peroxidase 4, increased the protein level of acyl-CoA synthetase long-chain family member 4, decreased the cell viability, and aggravated oxidative stress.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusion</h3>\u0000 ","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 5","pages":"899-909"},"PeriodicalIF":2.2,"publicationDate":"2024-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141201589","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Construction of the prediction model for multiple myeloma based on machine learning","authors":"Jiangying Cai, Zhenhua Liu, Yingying Wang, Wanxia Yang, Zhipeng Sun, Chongge You","doi":"10.1111/ijlh.14324","DOIUrl":"10.1111/ijlh.14324","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Introduction</h3>\u0000 \u0000 <p>The global burden of multiple myeloma (MM) is increasing every year. Here, we have developed machine learning models to provide a reference for the early detection of MM.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>A total of 465 patients and 150 healthy controls were enrolled in this retrospective study. Based on the variable screening strategy of least absolute shrinkage and selection operator (LASSO), three prediction models, logistic regression (LR), support vector machine (SVM), and random forest (RF), were established combining complete blood count (CBC) and cell population data (CPD) parameters in the training set (210 cases), and were verified in the validation set (90 cases) and test set (165 cases). The performance of each model was analyzed using receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis (DCA). Accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and area under the ROC curve (AUC) were applied to evaluate the models. Delong test was used to compare the AUC of the models.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Six parameters including RBC (10<sup>12</sup>/L), RDW-CV (%), IG (%), NE-WZ, LY-WX, and LY-WZ were screened out by LASSO to construct the model. Among the three models, the AUC of RF model in the training set, validation set, and test set were 0.956, 0.892, and 0.875, which were higher than those of LR model (0.901, 0.849, and 0.858) and SVM model (0.929, 0.868, and 0.846). Delong test showed that there were significant differences among the models in the training set, no significant differences in the validation set, and significant differences only between SVM and RF models in the test set. The calibration curve and DCA showed that the three models had good validity and feasibility, and the RF model performed best.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusion</h3>\u0000 \u0000 <p>The proposed RF model may be a useful auxiliary tool for rapid screening of MM patients.</p>\u0000 </section>\u0000 </div>","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"46 5","pages":"918-926"},"PeriodicalIF":2.2,"publicationDate":"2024-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141186952","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}