Leukemia can be Effectively Early Predicted in Routine Physical Examination with the Assistance of Machine Learning Models.

Objectives: The diagnosis of leukemia relies very much on the results of bone marrow examinations, which is never generally performed in routine physical examination. In many rural areas even community hospitals and primary care clinics, the lack of hematological specialist and facility does not allow a definite diagnosis of leukemia. Thus, there will be a significant benefit if machine learning (ML) models could help early predict leukemia using preliminary blood test data in a routine physical examination in community hospitals to save time before a definite diagnosis. Methods: We collected the routine physical examination data of 1230 newly diagnosed leukemia patients and 1300 healthy people. We trained and tested 3 machine learning (ML) models including linear support vector machine (LSVM), random forest (RF), and XGboost models. We not only examined the accordance between model results and statistical analysis of the input data but also examined the consistency of model accuracy scores and relative importance order of model factors with regard to different input data sets and different model arguments to check the applicability of both the models and the input data. Results: Generally, the RF and XGboost models give more identical, consistent, and robust relative importance order of factors that is also accordant with the statistical analysis, while the LSVM gives much different and nonsense orders for different inputs. Results of the RF and XGboost models show that (1) generally, the models achieve accuracy scores above 0.9, indicating effective identification of leukemia, and (2) the top three factors that contribute most to the identification of leukemia include red blood cell (RBC), hematocrit (HCT), and white blood cell (WBC), while the other factors contribute relatively less. Conclusions: This study shows a feasible case example for early identification of leukemia using routine physical examination data with the assistance of ML models, which can be conveniently, cheaply, and widely applied in community hospitals or primary care clinics to save time before definite diagnosis; however, more studies are still needed to validate the applicability of more ML models to a larger variety of input data sets.

Collective outbreak of severe acute histoplasmosis in immunocompetent Chinese in South America: the clinical characteristics and continuous monitoring of serum cytokines/chemokines.

BACKGROUND: Acute histoplasmosis is a rare fungal disease in China. This study is aimed to summarize the clinical characteristics of the first large-scale outbreak of imported acute histoplasmosis in Chinese, so as to provide suggestions for clinical diagnosis and treatment. METHODS: We collected the symptoms, signs, laboratory examination and imaging data of 10 patients in so far the biggest outbreak of imported acute histoplasmosis in immunocompetent Chinese. Their clinical characteristics and time-varying cytokine/chemokine levels were analyzed, and rank correlation analysis between these markers was utilized to show their condition. RESULTS: The 10 patients of imported acute histoplasmosis were working without any respiratory protection in an abandoned mine tunnel in Guyana. The most common symptoms were fever and cough. Their chest CT imaging showed multiple nodular shadows in lungs. Laboratory examination showed that at admission the CRP, PCT, LDH, CysC, G-test, ß2-MG were all increased in at least 9 patients, and the CD4/CD8 was decreased to < 1 in all patients. Most cytokines/chemokines (other than IL-4, IL-12, INF-α, TNF-α) varied widely with patients and time, but their overall trend is higher at admission and decreasing gradually during hospitalization, especially for the IL-6, IL-8, IL-10 and IFN-γ. The LDH, CysC, G-test, ß2-MG, N/L, IL-6, IL-8, IL-10, IFN-γ, IL-27 are in positive associations to both CRP and PCT. CONCLUSIONS: The diagnosis of acute histoplasmosis needs a comprehensive analysis of epidemiological history, clinical symptoms and signs, and results of imaging, laboratory, microbiological and pathological examinations. Although none of the CRP, PCT, G-test, N/L, LDH, CysC, ß2-MG, IL-6, IL-8, IL-10, IFN-γ shows specificity in the diagnosis of acute histoplasmosis, there is possibility that the above factors might help in the inflammation and prognosis estimation. However, more studies and further investigation are still required for the verification.

The biomarkers of key miRNAs and gene targets associated with extranodal NK/T-cell lymphoma.

Gene expression profiling studies have shown the pathogenetic role of oncogenic pathways in extranodal natural killer/T-cell lymphoma (ENKL). In this study, we aimed to identify the microRNAs (miRNAs) playing potential roles in ENKL, and to evaluate the genes and biological pathways associated to them. Gene expression profiles of ENKL patients were acquired from the gene expression omnibus (GEO) database. Most differentially expressed (DE)-miRNAs were identified in ENKL patients using limma package. Gene targets of the DE-miRNAs were collected from online databases (miRDB, miRWalk, miRDIP, and TargetScan), and used in Gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) analyses on Database for annotation, visualization, and integrated discovery database, and then used in protein-protein interaction (PPI) analysis on STRING database. Hub genes of the PPI network were identified in cytoHubba, and were evaluated in Biological networks gene ontology. According to the series GSE31377 and GSE43958 from GEO database, four DE-miRNAs were screened out: hsa-miR-363-3p, hsa-miR-296-5p, hsa-miR-155-5p, and hsa-miR-221-3p. Totally 164 gene targets were collected from the online databases, and used in the GO and KEGG pathway analyses and PPI network analysis. Ten hub genes of the PPI network were identified: AURKA, TP53, CDK1, CDK2, CCNB1, PLK1, CUL1, ESR1, CDC20, and PIK3CA. Those hub genes, as well as their correlative pathways, may be of diagnostic or therapeutic potential for ENKL, but further clinical evidence is still expected.

Allogeneic Hematopoietic Stem Cell Transplantation in Extranodal Natural Killer/T-cell Lymphoma

Objective: Extranodal NK/T-cell lymphoma (ENKL) is aggressive and resistant to chemotherapy and radiotherapy. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a potentially curative treatment for high-risk lymphomas owing to its associated graft-versus-lymphoma (GVL) effect. However, its application to ENKL is limited. We aim to summarize the characteristics of allo-HSCT for ENKL and, more importantly, evaluate whether allo-HSCT could offer any benefits for ENKL. Materials and Methods: A systematic review and data analysis were performed to evaluate the performance of allo-HSCT in the treatment of ENKL using studies obtained from PubMed, Medline, and Embase from January 2000 to December 2019 in the English language. Results: A total of 136 cases from 17 eligible publications were included in this study. It was found that after allo-HSCT, with an average follow-up time of 34 months (range: 1-121 months), 37.5% (52) of 136 patients had acute graft-versus-host disease (GVHD) and 31.6% (43) had chronic GVHD. Furthermore, 35.3% (48) of the patients were reported to have relapsed, but 2 of those relapsed only locally and achieved complete remission (CR) again with additional irradiation, chemotherapy, and donor lymphocyte infusions for one and rapid tapering and discontinuation of cyclosporine for the other, earning more than one year of extra survival. Finally, of the 136 patients, 51.5% (70) died because of primary disease progression (42.9%), infection (20.0%), GVHD (11.4%), organ failure (7.1%), hemorrhage (4.3%), and other causes (not specified/unknown) (14.3%). Conclusion: Allo-HSCT may be a treatment option for advanced or relapsed/refractory ENKL, but its role still requires more rigorous future studies.