A cost-optimized 5-protein panel revolutionizes systemic lupus erythematosus diagnosis

Wenhua Lv, Zhenwei Shang, Chen Sun, Yuping Zou, Siyu Wei, Haiyan Chen, Junxian Tao, Hongsheng Tian, Yu Dong, Chen Zhang, Mingming Zhang, Hongchao Lv and Yongshuai Jiang

PLOS Computational Biology, 2026, vol. 22, issue 7, 1-16

Abstract: Early diagnosis of systemic lupus erythematosus (SLE) is hindered by a lack of reliable biomarkers. This study sought to identify and evaluate diagnostic plasma protein biomarkers for SLE. We analyzed plasma protein profiles, polygenic risk scores (PRS), and clinical data from 544 SLE cases and 48,036 controls in the UK Biobank. Using LASSO regression, we identified 35 high-confidence SLE-associated proteins and derived a protein risk score (ProtRS). The ProtRS model achieved exceptional diagnostic performance (AUC = 0.91), significantly outperforming models based on PRS or clinical factors alone. Notably, a cost-optimized 5-protein panel (TRIM21, SOD2, KLK3, IL15, ADIPOQ) retained high accuracy (AUC = 0.82) while reducing costs by ~87%. ProtRS also demonstrated the highest population attributable fraction (96.34%), underscoring its dominant contribution to SLE burden. This study establishes a protein-driven framework for early SLE detection, offering tiered diagnostic solutions to balance accuracy and cost. The findings underscore the translational potential of protein biomarkers in bridging theoretical research and clinical practice.Author summary: Systemic lupus erythematosus (SLE) is a severe autoimmune disease that primarily affects women, yet early diagnosis remains challenging due to a lack of reliable biomarkers. Current diagnostic approaches rely on clinical symptoms and autoantibody tests that suffer from poor sensitivity or specificity. In this study, we leveraged plasma protein data from over 48,000 individuals in the UK Biobank to identify protein biomarkers for SLE. Using a balanced case-control sampling strategy combined with machine learning, we identified 35 proteins that accurately distinguish SLE patients from healthy controls. Furthermore, we developed a cost-optimized panel of just five proteins (TRIM21, SOD2, KLK3, IL15, and ADIPOQ) that maintains high diagnostic accuracy while reducing testing costs by approximately 87%. This 5-protein panel can be readily implemented in routine clinical laboratories using existing immunoassay platforms, offering a practical and affordable solution for early SLE detection. Our findings provide a protein-driven framework that bridges computational discovery and clinical application, potentially improving outcomes for patients with this debilitating disease.

Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014513 (text/html)
https://journals.plos.org/ploscompbiol/article/fil ... 14513&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pcbi00:1014513

DOI: 10.1371/journal.pcbi.1014513

Access Statistics for this article

More articles in PLOS Computational Biology from Public Library of Science
Bibliographic data for series maintained by ploscompbiol ().