超声和免疫学指标的特征能否反映RA临床缓解的表型?

 

THU0121

DO THE IMMUNOLOGICAL AND
ULTRASOUND CHARACTERISTICS REFLECT THE CLINICAL REMISSION PHENOTYPE
IN PATIENTS WITH RHEUMATOID ARTHRITIS?

H. Gul1,*, M. J.
Isorna Porto1, F. Ponchel2, E. Hensor2, R. Wakefield2, P. Emery2

1Leeds Institute of Rheumatic and Musculoskeletal Medicine,
2Leeds institute of Rheumatology and Musculoskeletal
Medicine, University of Leeds, leeds, United Kingdom

Background: Rheumatoid arthritis (RA) is an immune-mediated disease and
primarily affects the synovium. The goal of treatment in RA is
remission (1), however we do not routinely assess immunological
markers or perform ultrasound imaging to define
this.

背景:  类风湿关节炎(RA)是一种免疫介导的疾病,主要影响滑膜。RA的治疗目标是缓解,但我们并没有惯常评估免疫标志物或用超声成像确定这一点。

Objectives: We aimed to explore the clinical, imaging and immunological
characteristics of RA patients in clinical remission
(DAS28<2.6).

目的: 本研究的目的旨在探讨RA临床缓解(DAS28 < 2.6)患者临床、影像学和免疫学特征。

Methods: A retrospective observational study was performed, using
our inflammatory arthritis database. Patients were selected when
fulfilling the inclusion criteria of RA diagnosis (1987 or 2010
ACR/EULAR criteria) and DAS28CRP≤3.2. The lowest observation of DAS28CRP after
diagnosis and the availability of ultrasound data were prioritised.
Disease/remission duration and treatment modality were not
considered in this initial descriptive study. We collected data of:
Tender and swollen joint counts, inflammatory markers,
autoantibodies, patient-reported outcomes (PRO’s), ultrasound and x-ray findings.
Furthermore, T-cell subsets [naïve CD4+T-cells, Inflammation
related cells (IRC), and regulatory T-cells (Treg)] were measured
and analysed by advanced 8 colour flow-cytometry. Frequencies of
T-cells   were compared with
reference range values obtained from 120 healthy
controls.

方法: 我们用炎性关节炎的数据库进行一项回顾性观察研究。RA患者入选需满足1987年或2010年ACR/
EULAR诊断标准和DAS28CRP≤3.2。诊断后DAS28CRP最低值和超声资料的可用性按优先次序区分。疾病/缓解持续时间和治疗方式在最初的描述性研究并没有收集。我们收集的资料有:疼痛和肿胀关节数、炎症标志物、自身抗体、患者报告结果(PRO’s)、超声和X线结果。同时,由先进8色流式细胞术分析T细胞亚群 [幼稚CD4
+ T细胞,炎症相关细胞(IRC)和调节性T细胞(Treg)]。
T细胞的百分率与来自120名健康对照获得的参考值范围进行比较。

Results: We included 633 patients with a minimum
DAS28CRP≤3.2 (mean 1.85/SD
0.67).  The cohort was
predominantly female (68.7%) and the mean age was 57.6 years. Of
these, 513 were in DAS-defined remission. LDA was included as a
control. In the remission group, 324 patients had
ultrasound data available. Active synovitis, confirmed by positive
Power Doppler signal was present in 43.2%
(140/324) of patients. 76.2%
(247/324)  had evidence of grey
scale (GS) changes >1 in ≥1 joint. For 51 patients where T-cell subset analysis was
available, abnormal frequencies were observed in 11.8% of patients
for naïve CD4+T-cells, 27.5% for IRC and 54.9% for Treg in the
total cohort. For patients in remission, abnormal cell frequencies
were seen in 8.7% (4/46), 23.9% (11/46) and 50%
(23/46)  for naïve, IRC and
Treg's respectively. In LDA 65.3%
(47/72) had evidence of Power Doppler synovitis
and 86.1% (62/72) had GS changes
>1 in ≥1
joint. Abnormal cell frequencies were found in 40%
(2/5) of patients for naïve CD4 + T-cells, 60%
(3/5) for IRC and all had abnormal Treg
frequencies. The remission group had generally low levels of
disability, functional and life impairment as
expected.

结果: 本研究包括633例RA患者,DAS28CRP≤3.2(平均1.85/0.67
SD)。队列为以女性为主(68.7%),平均年龄为57.6岁。这些患者中,513例为疾病缓解。LDA(低疾病活动度)患者作为对照。在缓解组,324例患者有超声数据。正能量多普勒证实有活动性滑膜炎的患者有43.2%(140/324)。 在≥1关节中76.2%(247/324)证实有灰度(GS)变化>1。51例患者有T细胞亚群的资料。观察到11.8%的患者有异常的幼稚CD4+ T细胞百分率,27.5%的患者IRC百分率异常和54.9%Treg细胞百分率异常。临床缓解的患者中,异常T细胞百分率患者占比分别为幼稚8.7%(4/46),IRC
23.9%(11/46)和Treg
50%(23/46)。在LDA患者中,正能量多普勒证实有滑膜炎的患者有65.3%(47/72)和86.1%(62/72)的患者≥1关节GS变化>1。异常T细胞百分率患者占比分别为幼稚CD4 + T细胞40% (2/5),IRC
60%(3/5)和Treg
100%(5/5)。与预期的结果一致,缓解组较低水平的致残、功能和生活障碍。

Conclusions: Despite being in clinical/DAS remission or LDA state, a
substantial proportion of patients exhibit subclinical inflammation
on ultrasound (higher for LDA). T-cell abnormalities were present
in the majority. Current remission criteria are composite scores
and do not measure inflammation directly (1). Identification of
remission biomarkers could potentially help predict the ability to
sustain remission and facilitate treatment withdrawal strategies,
leading to a better understanding of what constitutes true
remission.

结论: 尽管处于临床/ DAS缓解或LDA状态,患者有相当比例表现出超声下亚临床炎症反应(LDA者更高)。 T细胞的异常也存在于大多数患者中。当前缓解标准是综合得分并没有直接测量炎症。缓解的生物标志物的鉴定可能有助于预测维持缓解和治疗停药的策略,从而更好地理解什么是真正的缓解。

References:

1. Saleem et al. Can remission be maintained with or
without further drug therapy in rheumatoid arthritis? Clinical and
experimental rheumatology. 2006 Nov-Dec; 24((6 Suppl
43)):S-33-36.

超声和免疫学指标的特征能否反映RA临床缓解的表型?[EULAR2015_THU0121]的更多相关文章

  1. 想搞机器学习,不会特征工程?你TM逗我那!

    原文:http://dataunion.org/20276.html 作者:JasonDing1354 引言 在之前学习机器学习技术中,很少关注特征工程(Feature Engineering),然而 ...

  2. 【读书笔记与思考】《python数据分析与挖掘实战》-张良均

    [读书笔记与思考]<python数据分析与挖掘实战>-张良均 最近看一些机器学习相关书籍,主要是为了拓宽视野.在阅读这本书前最吸引我的地方是实战篇,我通读全书后给我印象最深的还是实战篇.基 ...

  3. AI - 概念(Concepts)

    01 - AI.ML与DL的关系 从涵盖范围上来讲,人工智能(AI)大于机器学习(ML)大于深度学习(DL) 人工智能(AI):能够感知.推理.行动和适应的程序: 机器学习(ML):能够随着数据量的增 ...

  4. GIS-"地理空间大数据与AI的碰撞"学习笔记

    1.关系 人工智能>机器学习>神经网络>深度学习 2.机器学习-两个过程 训练/学习过程:样本数据.学习器.模型参数 测试/预测过程:预测.预测值 3.神经网络 机器学习模拟人脑神经 ...

  5. 关于fmri数据分析的两大类,四种方法

    关于fmri数据分析的两大类,四种方法: 数据驱动: tca:其实这种方法,主要是提取时间维的特征.如果用它来进行数据的分析,则必须要利用其他的数据方法,比如结合ICA. ica:作为pca的一般化实 ...

  6. 机器学习——LightGBM

    基础概念 LigthGBM是boosting集合模型中的新进成员,它和xgboost一样是对GBDT的高效实现,很多方面会比xgboost表现的更为优秀.原理上它和GBDT及xgboot类似,都采用损 ...

  7. 《Applying Deep Learning to Answer Selection: A Study And an Open Task》文章理解小结

    本篇论文是2015年的IBM watson团队的. 论文地址: 这是一篇关于QA问题的一篇论文: 相关论文讲解1.https://www.jianshu.com/p/48024e9f7bb22.htt ...

  8. Coursera课程《Machine Learning》学习笔记(week2)

    1 特征 1-1 什么是特征? 我的理解就是,用于描述某个样本点,以哪几个指标来评定,这些个指标就是特征.比方说对于一只鸟,我们评定的指标就可以是:(a)鸟的翅膀大还是小?(b)鸟喙长还是短?(c)鸟 ...

  9. 从零开始写一个Exporter

    前言 上一篇文章中已经给大家整体的介绍了开源监控系统Prometheus,其中Exporter作为整个系统的Agent端,通过HTTP接口暴露需要监控的数据.那么如何将用户指标通过Exporter的形 ...

  10. 选股公式blog+节选

    <大智慧软件选股_大智慧软件如何选股>——一般性操作 http://jingyan.baidu.com/article/fa4125acb2028d28ac70923e.html < ...

随机推荐

  1. Springboot整合thymeleaf报错whitelabel page

    1.SpringBootApplication未放在最外层 2.application.properties未配置spring.thymeleaf.check-template-location=tr ...

  2. python中调用C代码

    首先我们需要明晰为什么我们需要在python中调用C语言的代码,原因不外乎有二: 其一,python不擅长"大量运算"任务,而擅长于编写简单,"IO密集型作业" ...

  3. NOIP 口胡

    因为没准备啥东西 这两天口胡一下近年 NOIP 的题 大概会一道不落?没什么很寄的考点主要是 2021 T1 报数 打一个 \(O(\log n)\) 查询 \(n\) 中是否有 \(7\),打一个类 ...

  4. Apache手动安装教程及报错解决梳理

    操作参考教程:https://www.cnblogs.com/haw2106/p/9839655.html 下载地址(例):wget https://dlcdn.apache.org/httpd/ht ...

  5. 【大数据面试】Flink 03-窗口、时间语义和水印、ProcessFunction底层API

    三.窗口 1.窗口的介绍 (1)含义 将无限的流式数据切割为有限块处理,以便于聚合等操作 (2)图解 2.窗口的分类 (1)按性质分 Flink 支持三种划分窗口的方式,time.count和会话窗口 ...

  6. Spring IOC源码(二):IOC容器之 刷新前的准备

    1.源码解析 prepareRefresh()容器刷新refresh()的第一个方法,是容器刷新前的准备工作. 1 // 容器启动的开始时间 毫秒级 2 private long startupDat ...

  7. 李宏毅机器学习笔记:从0到写AI

    part1.基本介绍 1.机器学习的三个任务 一般情况下,我们在机器学习中有三个基本任务,分别是Regression Classification和Structured Regression是计算数值 ...

  8. 微服务系列之服务监控 Prometheus与Grafana

    1.为什么需要监控服务   监控服务的所属服务器硬件(如cpu,内存,磁盘I/O等)指标.服务本身的(如gc频率.线程池大小.锁争用情况.请求.响应.自定义业务指标),对于以前的小型单体服务来说,确实 ...

  9. STL map容器常用API

    map容器:键值和实值是分开的,排序规则按照键值排序 #define _CRT_SECURE_NO_WARNINGS #include<iostream> #include<map& ...

  10. CTFshow——funnyrsa2

    题目如下: 题目分析: 发现n很小,可以考虑yafu分解n,或者使用在线网站,例如:http://factordb.com/.即得p,q,r.因为常规rsa只有p和q,则phi = (p -1) * ...