scholarly journals The differential diagnosis of IgG4-related disease based on machine learning

Author(s):  
Motohisa Yamamoto ◽  
Masanori Nojima ◽  
Ryuta Kamekura ◽  
Akiko Kuribara-Souta ◽  
Masaaki Uehara ◽  
...  

Abstract Introduction: To eliminate the disparity and maldistribution of physicians and medical specialty services, the development of diagnostic support for rare diseases using artificial intelligence is being promoted. Immunoglobulin G4 (IgG4)-related disease (IgG4-RD) is a rare disorder often requiring special knowledge and experience to diagnose. In this study, we investigated the possibility of differential diagnosis of IgG4-RD based on basic patient characteristics and blood test findings using machine learning. Methods Six-hundred and two patients with IgG4-RD and 212 patients with non-IgG4-RD that needed to be differentiated who visited the participating institutions were included in the study. Ten percent of the subjects were randomly excluded as a validation sample. Among the remaining cases, 80% were used as training samples, and the remaining 20% were used as test samples. Finally, validation was performed on the validation sample. The analysis was performed using a decision tree and a random forest model. Furthermore, a comparison was made between conditions with and without the serum IgG4 concentration. Accuracy was evaluated using the area under the receiver-operating characteristic (AUROC) curve. Results In diagnosing IgG4-RD, AUROC curve values of the decision tree and the random forest method were 0.905 and 0.970, respectively, when serum IgG4 levels were included in the analysis. Excluding serum IgG4 levels, the AUROC curve value of the analysis by the random forest method was 0.919. Conclusion Based on machine learning in a multicenter collaboration, with or without serum IgG4 data, basic patient characteristics and blood test findings alone were sufficient to differentiate IgG4-RD from non-IgG4-RD.

2020 ◽  
Author(s):  
L Schulte ◽  
F Arnold ◽  
F Siegel ◽  
J Backhus ◽  
L Perkhofer ◽  
...  

2015 ◽  
Vol 26 (4) ◽  
pp. 583-587 ◽  
Author(s):  
Ping Li ◽  
Hua Chen ◽  
Chuiwen Deng ◽  
Ziyan Wu ◽  
Wei Lin ◽  
...  

2010 ◽  
Vol 37 (7) ◽  
pp. 1380-1385 ◽  
Author(s):  
YASUFUMI MASAKI ◽  
SUSUMU SUGAI ◽  
HISANORI UMEHARA

Since the first report of serum IgG4 elevation in sclerosing pancreatitis in 2001, various systemic disorders have been reported to elevate IgG4, and many names have been proposed from the perspective of the systemic condition. Despite similarities in the organs damaged in IgG4-related Mikulicz’s disease and Sjögren’s syndrome, there are marked clinical and pathological differences between the 2 entities. The majority of cases diagnosed with autoimmune pancreatitis in Japan are IgG4-related sclerosing pancreatitis, and it should be recognized that this is distinct from the Western type. Diagnosis of IgG4-related disease is defined by both elevated serum IgG4 (> 1.35 g/l) and histopathological features, including lymphocyte and IgG4+ plasma cell infiltration (IgG4+ plasma cells/IgG+ plasma cells > 50% on a highly magnified slide checked at 5 points). Differential diagnosis from other distinct disorders is necessary: these include sarcoidosis, Castleman’s disease, Wegener’s granulomatosis, lymphoma, cancer, and other existing conditions. The Japanese IgG4 research group has begun multicenter prospective studies to improve diagnostic criteria and treatment strategies.


JHEP Reports ◽  
2020 ◽  
Vol 2 (4) ◽  
pp. 100135 ◽  
Author(s):  
Lucas A. Schulte ◽  
Frank Arnold ◽  
Florian Siegel ◽  
Johanna Backhus ◽  
Georg Beyer ◽  
...  

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 1224.1-1224
Author(s):  
I. Mizushima ◽  
T. Yamano ◽  
H. Kawahara ◽  
S. Hibino ◽  
R. Nishioka ◽  
...  

Background:Recently, the 2019 ACR/EULAR classification criteria for IgG4-related disease (IgG4-RD) were published mainly to identify more homogeneous subjects for inclusion in clinical trials and observational studies [1]. However, although their high specificity is presumed to be useful to differentiate IgG4-RD from various mimickers, their value in daily clinical practice needs to be evaluated.Objectives:This study aimed to clarify the usefulness of the 2019 ACR/EULAR classification criteria for IgG4-RD and characteristics of false-negative patients in daily clinical practice.Methods:We retrospectively reviewed the medical records of 162 patients with IgG4-RD and 130 consecutive non-IgG4-RD patients (mimickers) diagnosed by experts whose serum IgG4 levels were measured at a single center in Japan. Using the collected data, we calculated sensitivity, specificity, and fulfillment rates for the entry criteria, exclusion criteria, and threshold of inclusion criteria points. In addition, to clarify the characteristics of false-negative cases in IgG4-RD, we performed an intergroup comparison of their clinical features including disease-specific autoantibodies.Results:Both the patients with IgG4-RD and mimickers were relatively old (66 and 65 years) with male predominance (67% and 60%). The final diagnoses of mimickers mainly consisted of cancer, lymphoma, vasculitis, sarcoidosis, multicentric Castleman’s disease, and atherosclerotic or infectious aortic aneurysm. The classification criteria had a sensitivity of 72.8% and a specificity of 100%. Of the 44 false-negative cases, one did not fulfill the entry criteria, 20 fulfilled one exclusion criterion, and 27 did not achieve sufficient inclusion criteria points. Compared with the true-positive cases, the false-negative cases had significantly fewer affected organs, lower serum IgG4 levels, higher serum CH50 levels, and lower prevalence of salivary/lacrimal gland and renal parenchymal lesions. They were also less likely to have had biopsies (61% vs 97%). Of note, positivity of disease-specific autoantibodies including SSA/Ro antibody, ANCA, ds-DNA antibody, and ACPA was the most common exclusion criterion fulfilled in 18 patients, only 2 of whom were diagnosed with a specific autoimmune disease (rheumatoid arthritis) complicated by IgG4-RD. The remaining 16 patients had no specific clinical symptoms related to such autoantibodies. In addition, compared with IgG4-RD patients without disease-specific autoantibodies, the 18 patients with them had almost equal serum IgG4 and complement levels, number of affected organs, and histopathology and immunostaining scores despite higher serum IgG and CRP levels.Conclusion:The present study suggests that the 2019 ACR/EULAR classification criteria for IgG4-RD has excellent diagnostic specificity and moderate sensitivity in daily clinical practice. Positive disease-specific autoantibodies alone, which lowered the sensitivity in this study, may have little clinical significance concerning the diagnosis of IgG4-RD.References:[1]Wallace ZS et al. The 2019 American College of Rheumatology/European League Against Rheumatism classification criteria for IgG4-related disease. Ann Rheum Dis. 2020 Jan;79(1):77-87.Disclosure of Interests:None declared


2021 ◽  
Vol 36 (Supplement_1) ◽  
Author(s):  
Peifen Liang ◽  
Qianqian Han ◽  
Bo Liu ◽  
Qiongqiong Yang

Abstract Background and Aims Both antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) and IgG4-related disease (IgG4-RD) are multi-system inflammatory disorders. The coexistent of both diseases present the possibility of a new overlap syndrome which leads to different treatment and outcome. In this study, we aim to investigate the possibility and clinicopathologic clues to the diagnosis of this overlap syndrome. Method A case of a 67-year-old man in our hospital who exhibited the clinicopathologic characteristics of both AAV and IgG4-RD was presented. The serial alterations in serum parameters and treatment response of the case were followed up for the next 15 months. Then, a systematic literature review of the overlap syndrome was performed on PUBMED database from 1976 until January 2020. Results Forty-two patients fully met both AAV and IgG4-RD criteria in the literature. The median age was 60 years ranged from 31 to 73 years at diagnosis. Twenty-eight (66.7%) patients were men. Serum IgG4 concentration increased in 41 (97.6%) patients (median:395 mg/dl; range:177-876mg/dl). Forty-one patients (97.6%) tested positive for ANCA with 37 (90.2%) patients showing a specificity for MPO. Kidney histology of 23 (69.7%) patients presented pauci-immune necrotizing or crescentic glomerulonephritis and IgG4-relative tubulointerstitial nephritis. Glucocorticoids combined with cyclophosphamide therapy was commonly prescribed with a high remission rate within 3 months. Four common clinicopathologic features of the overlap syndrome were identified from the case and literature. Conclusion AAV may overlap with IgG4-RD while presenting atypical manifestations. Four common clinicopathologic characteristics could be used as specific clues to the diagnosis of overlap syndrome.


2012 ◽  
Vol 35 (1) ◽  
pp. 30-37 ◽  
Author(s):  
Motohisa YAMAMOTO ◽  
Norihiro NISHIMOTO ◽  
Tetsuya TABEYA ◽  
Yasuyoshi NAISHIRO ◽  
Keisuke ISHIGAMI ◽  
...  

Author(s):  
Jagadeesh Chandrasekaran ◽  
Phani Krishna Machiraju

Immunoglobulin G4 (IgG4)-related disease is a multi-organ, immune-mediated, fibro-inflammatory disorder characterized by tumefactive masses in the affected organs. Incidence and prevalence of IgG4-related disease (RD) are not clearly known and have slight male preponderance. It often involves multiple organs at the time of presentation or over the course of disease mimicking malignancy, Sjogren's syndrome, antineutrophil cytoplasmic antibodies associated vasculitis, infections. A thorough workup is needed to rule out these mimickers. A 33-year-old gentleman presented to us with history of progressive swelling in the right peri-orbital region for four years. On evaluation, abdominal imaging was notable for the sausage-shaped pancreas and hypoenchancing nodules in bilateral kidneys. Histological examination of right lacrimal gland revealed lymphoplasmacytic infiltrate and storiform fibrosis. Serum IgG4 levels were normal, and immunostaining was negative. A diagnosis of IgG4-RD was suggested because of multi-organ involvement, classical radiological and histopathological features. Awareness about IgG4-RD, an under-recognized entity is essential, as it is treatable, and early recognition may help in a favourable outcome. Appropriate use of clinicopathological, serological and imaging features in the right clinical context may help in accurate diagnosis. Elevated serum IgG4 levels and biopsy are not mandatory for the diagnosis.


Sign in / Sign up

Export Citation Format

Share Document