{
  "abstract": "Objectives Systemic Lupus Erythematosus (SLE) exhibits high clinical and molecular complexity that may reflect distinct disease subtypes. Currently, two biologics targeting B-cell differentiation and type I interferon pathways are approved, while others are used off-label. No evidence-based molecular criteria exist to guide biologic selection, resulting in trial-and-error approaches that adversely affect patient outcomes. Previous molecular stratification studies suggest shared pathogenic pathways across systemic autoimmune diseases (SADs) and highlight the potential of molecular signatures to inform therapeutic decisions. In this analysis we evaluated whether molecular stratification could improve treatment selection effectiveness.Methods Unsupervised matrix factorization was applied to PRECISESADS transcriptomes from over 1,500 individuals across seven SADs. Resulting factors were evaluated for immune pathway associations through Gene Set Enrichment Analysis. These pathway-associated factors enabled consensus clustering to identify robust molecular clusters termed pathotypes. In-silico drug simulations were conducted based on actionable molecular pathways for biologic targets commonly employed in autoimmune disease treatment, including BAFF, IFNAR, TNF, CD20, IL1R1, and others. Pathotype stratification was then applied to an independent SLE cohort (GSE88887) for validation.Results Molecular pathotypes exhibited higher in-silico response rates for certain therapeutic targets than clinical diagnoses alone, supporting biological relationships between dysregulated pathways and corresponding therapies. For example, BAFF inhibition showed high response proportions within interferon-associated and plasma cell-associated pathotype. All clinical phenotypes were represented across molecular pathotypes, each displaying trends consistent with overall pathotype patterns, confirming that biologics may be effectively used across SADs. Having validated the in-silico response rates in an independent SLE cohort, pathotype classification was successfully transferred.Conclusions Molecular pathotype stratification identifies patient subgroups with differential predicted responses to biologic therapies, offering superior predictive capacity compared to clinical diagnosis alone. The distribution of clinical phenotypes across molecular pathotypes while maintaining consistent response trends indicates that pathway-directed therapy may improve outcomes across SADs. Successful validation in an independent cohort demonstrates robustness and clinical potential. These findings support the future design of clinical trials guided by molecular pathotype selection that may improve clinical outcomes, representing a significant step toward precision medicine in SADs.",
  "authors": [
    {
      "affiliations": [
        "Pfizer,University of Granada,Junta de Andalucía Centre for Genomics and Oncological Research, Granada, Spain"
      ],
      "name": "Ana Brescia Zapata"
    },
    {
      "affiliations": [
        "Pfizer,University of Granada,Junta de Andalucía Centre for Genomics and Oncological Research, Granada, Spain"
      ],
      "name": "María Rivas Torrubia"
    },
    {
      "affiliations": [
        "Pfizer,University of Granada,Junta de Andalucía Centre for Genomics and Oncological Research, Granada, Spain"
      ],
      "name": "PRECISESADS Clinical Consortium"
    },
    {
      "affiliations": [
        "Division of Rheumatology, Department of Medicine Solna, Karolinska Institutet and Karolinska University Hospital, Stockholm, Sweden",
        "Department of Rheumatology, Faculty of Medicine and Health, Örebro University, Örebro, Sweden"
      ],
      "name": "Ioannis Parodis"
    },
    {
      "affiliations": [
        "Pfizer,University of Granada,Junta de Andalucía Centre for Genomics and Oncological Research, Granada, Spain",
        "Institute of Environmental Medicine, Karolinska Institute, Stockholm, Sweden"
      ],
      "name": "Marta E Alarcón Riquelme"
    },
    {
      "affiliations": [
        "Pfizer,University of Granada,Junta de Andalucía Centre for Genomics and Oncological Research, Granada, Spain",
        "Department of Genetics, Faculty of Science, University of Granada, Granada, Spain",
        "Bioinformatics Laboratory, Biotechnology Institute, Centro de Investigación Biomédica, PTS, Granada, Spain"
      ],
      "name": "Guillermo Barturen"
    }
  ],
  "title": "PO:08:221 Pathway-based molecular stratification identifies patient subgroups with differential in-silico response rates to biologic therapies in systemic lupus erythematosus",
  "uid": "1ffbc201-e137-50e7-acee-fd8256b8e011"
}