Scholarlink

Background <p>The type I interferon-alpha (IFN-α) gene expression signature refers to the analysis of specific IFN-stimulated genes (ISG) as a surrogate measure of IFN-α pathway activity. It is widely used to assess IFN-α activity in patients with systemic lupus erythematosus (SLE), as well as to evaluate treatment response or serve as an index of disease activity. ISGs are often aggregated into a composite score that provides a quantitative measure. However, previous literature has employed differing combinations of ISG, and no clear consensus exists regarding which should be preferentially used. Our study aims to identify the combination of IGS that best correlates with serum IFN-α levels in patients with SLE.</p> Methods <p>A total of 252 patients with SLE were recruited in this cross-sectional study. IFN-α serum levels were measured using Simoa (Single Molecule Array) technique. Based on prior literature, 72 transcripts previously used in other studies in rheumatic musculoskeletal diseases were selected and analyzed by qPCR using Nanostring<sup>®</sup> technology. Fifty-one distinct IFN scores were computed using previously reported ISG combinations. Besides, LASSO (Least Absolute Shrinkage and Selection Operator) regression analysis was performed to identify the minimal and most parsimonious combination of ISGs associated with serum IFN-α levels.</p> Results <p>LASSO analysis identified a 13-ISG combination as the most parsimonious predictor of serum IFN-α levels. However, the IFN score constructed using these 13 genes showed lower correlation with serum IFN-α levels than previously reported IFN scores from the literature. This suggests that statistically derived ISG selection is inferior to mechanistically- or pathophysiologically-informed gene combinations. Three literature-derived ISG combinations—<i>IFI44L IFIT1 RSAD2 IFI27 ISG15 SIGLEC1</i>, <i>IFI27 IFI44 IFI44L RSAD2</i>, and <i>IFI44L IFI44 RSAD2 IFI27 IFI6</i>—demonstrated high IFN-α correlations, statistically indistinguishable from each other but significantly superior to the remaining 48 combinations.</p> Conclusions <p>This study provides evidence-based recommendations for ISG selection in constructing optimal IFN signatures to accurately predict serum IFN-α levels in SLE patients.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Defining the optimal interferon-stimulated genes panel to reflect serum interferon-α levels in patients with systemic lupus erythematosus

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月31日

查看全文
  • Marta Rodríguez-García,
  • Fuensanta Gómez-Bernal,
  • Juan Carlos Quevedo-Abeledo,
  • Beatriz Tejera-Segura,
  • Enrique García-Barrera,
  • Luisa María Villar,
  • Javier Gonzalo Ocejo-Vinyals,
  • Nerea Mota-Pérez,
  • Javier Llorca,
  • Raquel Largo,
  • Miguel Ángel González-Gay,
  • Iván Ferraz-Amaro

摘要

Background

The type I interferon-alpha (IFN-α) gene expression signature refers to the analysis of specific IFN-stimulated genes (ISG) as a surrogate measure of IFN-α pathway activity. It is widely used to assess IFN-α activity in patients with systemic lupus erythematosus (SLE), as well as to evaluate treatment response or serve as an index of disease activity. ISGs are often aggregated into a composite score that provides a quantitative measure. However, previous literature has employed differing combinations of ISG, and no clear consensus exists regarding which should be preferentially used. Our study aims to identify the combination of IGS that best correlates with serum IFN-α levels in patients with SLE.

Methods

A total of 252 patients with SLE were recruited in this cross-sectional study. IFN-α serum levels were measured using Simoa (Single Molecule Array) technique. Based on prior literature, 72 transcripts previously used in other studies in rheumatic musculoskeletal diseases were selected and analyzed by qPCR using Nanostring® technology. Fifty-one distinct IFN scores were computed using previously reported ISG combinations. Besides, LASSO (Least Absolute Shrinkage and Selection Operator) regression analysis was performed to identify the minimal and most parsimonious combination of ISGs associated with serum IFN-α levels.

Results

LASSO analysis identified a 13-ISG combination as the most parsimonious predictor of serum IFN-α levels. However, the IFN score constructed using these 13 genes showed lower correlation with serum IFN-α levels than previously reported IFN scores from the literature. This suggests that statistically derived ISG selection is inferior to mechanistically- or pathophysiologically-informed gene combinations. Three literature-derived ISG combinations—IFI44L IFIT1 RSAD2 IFI27 ISG15 SIGLEC1, IFI27 IFI44 IFI44L RSAD2, and IFI44L IFI44 RSAD2 IFI27 IFI6—demonstrated high IFN-α correlations, statistically indistinguishable from each other but significantly superior to the remaining 48 combinations.

Conclusions

This study provides evidence-based recommendations for ISG selection in constructing optimal IFN signatures to accurately predict serum IFN-α levels in SLE patients.