scholarly journals Artificial Neural Network Analysis of Gene Expression Data Predicted Non-Hodgkin Lymphoma Subtypes with High Accuracy

2021 ◽  
Vol 3 (3) ◽  
pp. 720-739
Author(s):  
Joaquim Carreras ◽  
Rifat Hamoudi

Predictive analytics using artificial intelligence is a useful tool in cancer research. A multilayer perceptron neural network used gene expression data to predict the lymphoma subtypes of 290 cases of non-Hodgkin lymphoma (GSE132929). The input layer included both the whole array of 20,863 genes and a cancer transcriptome panel of 1769 genes. The output layer was lymphoma subtypes, including follicular lymphoma, mantle cell lymphoma, diffuse large B-cell lymphoma, Burkitt lymphoma, and marginal zone lymphoma. The neural networks successfully classified the cases consistent with the lymphoma subtypes, with an area under the curve (AUC) that ranged from 0.87 to 0.99. The most relevant predictive genes were LCE2B, KNG1, IGHV7_81, TG, C6, FGB, ZNF750, CTSV, INGX, and COL4A6 for the whole set; and ARG1, MAGEA3, AKT2, IL1B, S100A7A, CLEC5A, WIF1, TREM1, DEFB1, and GAGE1 for the cancer panel. The characteristic predictive genes for each lymphoma subtypes were also identified with high accuracy (AUC = 0.95, incorrect predictions = 6.2%). Finally, the topmost relevant 30 genes of the whole set, which belonged to apoptosis, cell proliferation, metabolism, and antigen presentation pathways, not only predicted the lymphoma subtypes but also the overall survival of diffuse large B-cell lymphoma (series GSE10846, n = 414 cases), and most relevant cancer subtypes of The Cancer Genome Atlas (TCGA) consortium including carcinomas of breast, colorectal, lung, prostate, and gastric, melanoma, etc. (7441 cases). In conclusion, neural networks predicted the non-Hodgkin lymphoma subtypes with high accuracy, and the highlighted genes also predicted the survival of a pan-cancer series.

2019 ◽  
Vol 42 (3) ◽  
pp. 303-318 ◽  
Author(s):  
Julieta Afonso ◽  
Tatiana Pinto ◽  
Susana Simões-Sousa ◽  
Fernando Schmitt ◽  
Adhemar Longatto-Filho ◽  
...  

Blood ◽  
2005 ◽  
Vol 106 (2) ◽  
pp. 668-672 ◽  
Author(s):  
Andrea Altieri ◽  
Justo Lorenzo Bermejo ◽  
Kari Hemminki

Abstract Non-Hodgkin lymphoma (NHL) consists of a heterogeneous group of tumors. Population-based data on the familial risk for specific histopathologic subtypes have not been established. Such data are useful for clinical counseling and for searching tumor subtypes sharing common genetic pathways. We used the Swedish Family-Cancer Database to calculate standardized incidence ratios (SIRs) for histopathology-specific subtypes of NHL in 4455 offspring with NHL whose parents or siblings were affected with different types of lymphoproliferative malignancies. A familial history of NHL significantly increased the risk for NHL (SIRparent = 1.8; SIRsibling = 1.9) and for diffuse large B-cell lymphoma (SIRparent = 2.3), follicular lymphoma (SIRsibling = 2.3), and B-cell lymphoma not otherwise specified (NOS) (SIRsibling = 3.4). For a parental history of histopathology-specific concordant cancer, the risks were significantly increased for diffuse large B-cell lymphoma (SIR = 11.8), follicular NHL (SIR = 6.1), plasma cell myeloma (SIR = 2.5), and chronic lymphocytic leukemia (SIR = 5.9). Familial clusters for NHL seemed stronger in females and in siblings. Our study provides the first quantification of the familial risks for NHL by histopathology. The present findings give evidence for a strong familial association of NHL, with little differences in the magnitude of risks for various histopathologic subtypes. The patterns of risks in parents and siblings support the hypothesis of an autosomal-dominant component for diffuse large B-cell NHL and a recessive one for follicular NHL. (Blood. 2005;106:668-672)


Blood ◽  
2013 ◽  
Vol 122 (21) ◽  
pp. 1775-1775
Author(s):  
Hideaki Nitta ◽  
Yasuhito Terui ◽  
Masahiro Yokoyama ◽  
Noriko Nishimura ◽  
Kyoko Ueda ◽  
...  

Abstract Background In the rituximab era, there are several studies that have reported the risk factors for central nervous system (CNS) involvement in non-Hodgkin lymphoma, but the same factors emerge, such as high international prognostic index (IPI) score, >1 extranodal site, elevated lactate dehydrogenase (LDH) level, poor performance status (PS), advanced stage, bone marrow involvement. Macrophages are an important component of the tumor microenvironment and the immune response to malignancy. Recently, elevated peripheral blood monocyte counts have been shown to be an independent marker associated with poor prognosis in patients with both non-Hodgkin and Hodgkin lymphoma. Patients and methods We reviewed data from a total of 1238 lymphoma patients(1185 non-Hodgkin lymphoma, 53 Hodgkin lymphoma) at our institution between February 2005 and May 2013. Of these, 42 patients (3.4%) developed CNS complications during the clinical course. Thirty patients out of these 42 (71.4%) were diagnosed with diffuse large B-cell lymphoma (DLBCL). Therefore, we focused on DLBCL. In this study, we retrospectively analyzed data from a total of 557 DLBCL patients, 30 patients (5.4%) who developed CNS involvement and 527 patients with DLBCL but without CNS involvement. This study was approved by the Institutional Review Board of the Cancer Institute Hospital of the Japanese Foundation for Cancer Research. The clinical features of all 557 DLBCL patients, including 30 patients with CNS involvement, are summarized in Table 1. CNS involvement was defined by the presence of at least one histologically confirmed CNS involvement; neuroimaging findings compatible with CNS involvement with lymphoma, in conjunction with consistent clinical presentation; and the absence of other clinically feasible diagnosis or positive cerebrospinal fluid (CSF) (lymphoma cells detected by cytology). The absolute monocyte counts (AMC) and monocyte ratio were derived from pre-treatment complete blood counts. Pathological studies Immunohistochemical analysis was carried out using mAbs against CD68 at our institution. Results The incidence of CNS involvement was 5.4%, 1.3% having CNS involvement at diagnosis with DLBCL. Intriguingly, absolute monocyte counts (AMC) ≥0.6 (×109/L) at diagnosis were significantly frequent in 30 DLBCL patients (p=0.0420) with CNS involvement, compared with in 527 DLBCL patients without CNS involvement. Furthermore, the monocyte ratio ≥8% in peripheral blood at diagnosis was significantly frequent in 30 DLBCL patients (p=0.0325) with CNS involvement, compared with in 527 DLBCL patients without CNS involvement. DLBCL patients with CNS involvement showed age ≤60 years, stage III-IV, IPI score ≥3, and PS ≥2, elevated soluble IL-2 receptor levels was significantly frequent, compared with in DLBCL patients without CNS involvement. Neither gender, elevated LDH level, white blood cell counts (WBC) differed significantly in the two groups. With regard to pathological immunohistochemistry, the numbers of CD68 positive cells in or around lymphoma samples did not differ in the 14 DLBCL patients with CNS involvement that we were able to analyze, compared with DLBCL patients without CNS involvement. CNS involvement free survival rate in DLBCL patients was significantly lower in AMC ≥0.6 (×109/L) and/or the monocyte ratio ≥8% (Log-rank test, P=0.0102) in peripheral blood at diagnosis, compared with in AMC less than 0.6 (×109/L) and the monocyte ratio less than 8%. Conclusions These results suggest that in DLBCL patients, AMC and monocyte ratios in peripheral blood at diagnosis are closely correlated with the risk of eventual CNS involvement. AMC and monocyte ratios in peripheral blood at diagnosis in DLBCL patients could be a useful prognostic marker for the risk of CNS involvement during the clinical course. Disclosures: Yokoyama: Chugai Pharmaceutical CO., LTD.: Consultancy. Nishimura:Chugai Pharmaceutical CO., LTD.: Consultancy.


2002 ◽  
Vol 93 (11) ◽  
pp. 1207-1212 ◽  
Author(s):  
Tatsuya Ando ◽  
Miyuki Suguro ◽  
Taizo Hanai ◽  
Takeshi Kobayashi ◽  
Hiroyuki Honda ◽  
...  

Blood ◽  
2005 ◽  
Vol 106 (11) ◽  
pp. 233-233 ◽  
Author(s):  
Janina Salzburg ◽  
Birgit Burkhardt ◽  
Olga Wachowski ◽  
Martin Zimmermann ◽  
Reza Parwaresch ◽  
...  

Abstract We evaluated the prevalence and clinical significance of CNS involvement in childhood and adolescence Non-Hodgkin Lymphoma (NHL). Between 10/86 and 12/02, 2,086 eligible patients (pts) were registered in the subsequent multicenter trials NHL-BFM86, −90, −95. Median follow up was 6.5 years (0.3–17.7 years). Initial staging included examination of cerebrospinal fluid (CSF) and cranial CT or MRI. CNS involvement was diagnosed in case of CSF blasts, and/or intracerebral mass (ICM), and/or cranial nerve palsy (CNP), not caused by an extradural mass. Epidural NHL without any of the above criteria was not considered as CNS disease. CNS positive (pos) pts with lymphoblastic lymphoma (LBL) received an 8-drug induction, consolidation, re-intensification, and maintenance up to 2 years. CNS therapy included dexamethason, methotrexate (MTX) 5 g/m2 i.v., 13 dosis of intrathecal (i.th.) MTX, and cranial radiotherapy (CRT). CNS pos pts with non-LBL received six 5-day courses based upon vincristine, vindesine, dexamethason, oxazophorins, cytarabine, etoposide, doxorubicin, MTX 5 g/m2 i.v., and intraventricularely or i.th. applied chemotherapy. CRT was omitted since study BFM90, except for pts with anaplastic large cell lymphoma (ALCL). 111 of the 2,086 analyzed NHL pts were initially diagnosed as CNS pos. 1,933 pts were CNS negative (neg) and in 42 pts the CNS status was questionable or not evaluable due to incomplete diagnostics. Prevalence and outcome of CNS pos pts according to NHL subtypes were as follows. In the total group, the probability of event free survival at 5 years (pEFS) was 63 ± 5% for CNS pos pts compared to 81 ± 1% for CNS neg pts with stage III/IV NHL (n=1,323) (p< 0.0001). In LBL pts pEFS was 81 ± 10% for CNS pos pts and 84 ± 2% for CNS neg pts with stage III/IV (n=359) (p=0.54), while in Burkitt/B-ALL pEFS was 60 ± 5 % for CNS pos pts versus 85 ± 1% for CNS neg pts with stage III/IV (n=599) (p<0.0001). For CNS pos Burkitt/B-ALL pts pEFS was 57 ± 7% for 57 pts with and was 67 ± 10% for 24 pts without bone marrow involvement (p=0.31). Total LBL (T-, pB-) Burkitt/B-ALL PMLBL* DLBL° ALCL Others *primary mediastinal large B-cell lymphoma, °diffuse large B-cell lymphoma Number of pts 2086 433 1003 40 222 215 173 CNS pos pts 111 16 81 0 4 5 5 Percentage 5,3% 3,7% 8,1% 0 1,8% 2,3% 2,9% Chracteristics and outcome of CNS pos pts CSF blasts +/ − others 81 13 60 0 1 4 3 ICM (without CSF blasts) 18 2 11 0 2 1 2 CNP 12 1 10 0 1 0 0 Death unrelated to tumor 6 0 6 0 0 0 0 Relapse/Nonresponse 30 2 24 0 0 2 2 CNS involved 18 1 15 0 0 2 0 In summary, CNS-disease was most frequent in pts with Burkitt/B-ALL, while it was rare in DLBL pts. In Burkitt/B-ALL, CNS pos pts had a worse outcome compared to CNS neg pts with advanced stage disease, while in LBL pts outcome was comparable for CNS pos and CNS neg pts.


Blood ◽  
2007 ◽  
Vol 110 (11) ◽  
pp. 4432-4432
Author(s):  
Luis D. Sumoza ◽  
Jeffrey L. Jorgensen ◽  
Ilia R. Sumoza

Abstract We report the first case of composite Nodular Sclerosis Hodgkin and Diffuse Large B-Cell Non-hodgkin’s lymphoma of the mediastinum. We present a case of an inmunocompetent patient operated on for a mediastinal tumor similar to a Thymoma, which the histological examination morphological, and immunophenotyping were performed and confirmed the existence of 2 independent, unrelated tumors. The pathology blocks submitted show a composite lymphoma, with components of both classical Hodgkin lymphoma (Hodgkin’s disease), nodular sclerosis type, grade 2 of 2 in all three blocks. One block also shows non-Hodgkin lymphoma, namely diffuse large B-cell lymphoma, with a distinct immunophenotype. The Hodgkin-Reed-Sternberg cells were positive for CD15, CD20, CD30, PAX-5 (weak/partial), and EBV (EBERs), and negative for CD45. In contrast, the large B-cells are positive for CD20, PAX-5 (strong), CD45, and CD30 (very focal), and negative for CD15 and EBERs. The large B-cell area had an increased mitotic rate. Taken together, these data indicate that these Hodgkin and the Non-Hodgkin’ lymphomas arose as a consequence of independent malignant transformation events.


2014 ◽  
Vol 2014 ◽  
pp. 1-3 ◽  
Author(s):  
Christopher D. Jaeger ◽  
Kelly L. McAlvany ◽  
Shannon N. Zingula ◽  
Stephen A. Kramer ◽  
Candace F. Granberg

Lymphoma may affect the ureter in cases of retroperitoneal involvement. We present a case of an adolescent male found to have non-Hodgkin lymphoma initially presenting as ureteral stricture evident on imaging. He was treated and responded to multiagent chemotherapy with resolution of both the lymphoma and the ureteral stricture. Although rare, non-Hodgkin lymphoma should be included in the differential diagnosis of pediatric patients with noncalculous, idiopathic ureteral strictures.


2021 ◽  
Vol 20 (1) ◽  
Author(s):  
Atish Kizhakeyil ◽  
Nurmahirah Binte Mohammed Zaini ◽  
Zhi Sheng Poh ◽  
Brandon Han Siang Wong ◽  
Xinpeng Loh ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document