{
  "abstract": "Background Recent advancements in RNA-sequencing technologies have facilitated the identification of transcriptomic-based scores capable of predicting prognosis and therapeutic response in lung adenocarcinoma (LUAD). However, the prognostic utility of transcriptomic profiling specifically focused on innate and adaptive immune-response genes (IRGs) remains inadequately validated in this context. In this study, we developed and validated a novel transcriptomic immune gene signature that reliably predicts clinical outcomes in patients with LUAD.Methods Transcriptomic and proteomic data from 507 LUAD patients and 59 normal controls were obtained from TCGA. IRGs were retrieved from molecular signature database (MSigDB), and differentially expressed genes (DEGs) were identified using DESeq2 (adjusted P < 0.05, |Log2FC| > 2). Mutation profiles were analyzed with the maftools R package. DE-IRGs were subjected to LASSO regression and Cox models. An Immune Response Stratification Index (IRSI) was constructed from significant genes, and patients were divided into high- and low-IRSI groups based on the median scores. Survival outcomes (OS, PFS) were assessed by Kaplan-Meier and log-rank test. Immune infiltration was evaluated using CIBERSORT and xCell. External validation was performed using two GEO datasets (GSE30219, GSE50081) and a real-world clinicogenomic dataset (n=19).Results We identified 82 DE- IRGs between normal and tumor samples, LASSO and cox regression models revealed seven independent prognostic biomarkers: SIRPA, CX3CR1, F12, F2RL1, EREG, WFDC3, TRIM6, all significantly associated with survival (p<0.05). High IRSI (n= 365) showed worse OS, PFS (P<0.05) in all datasets, and showing a trend towards significance in our clinicogenomic real-world dataset with a P= 0.051. Mutational and proteomic analysis revealed different genomic profiles between the high- and low-IRSI groups. High-IRSI group showed significant mutations in: TP53, CSMD3, CSMD1 and PAPPA2 (P < 0.05), with loss of MTAP, CDKN2A, and CDKN2B (P<0.05). Protein expression analysis between the two groups revealed higher SERPINE1, ITGA2, and CCNB1 in the high-IRSI patients, with lower expression of CDKN1B, PDCD4, STAT3, and MTOR. Low-IRSI patients showed higher immune, stromal, and tumor microenvironment (TME) scores, with elevated CD8+, CD4+, and B cells (P<0.001). Higher IRSI levels were significantly (p<0.05) associated with more advanced disease, including T stage (T1-T4), N stage (N0-N3), and AJCC clinical stage (Stage I-IV).Conclusions We developed an immune response-based index that efficiently predicts survival outcomes and disease progression in patients with LUAD. A high-IRSI score was significantly correlated with worse clinical outcomes and advanced disease stage. Additional efforts are ongoing to explore the generalizability and robustness of our score.",
  "authors": [
    {
      "affiliations": [
        "The University of Alabama at Birmingham, Birmingham, AL, USA"
      ],
      "name": "Sanad Alhushki"
    },
    {
      "affiliations": [
        "Jordan University of Science and Technology, Amman, Jordan"
      ],
      "name": "Mohammad Khaled Alhushki"
    },
    {
      "affiliations": [
        "The University of Alabama at Birmingham, Birmingham, AL, USA"
      ],
      "name": "Juan Juncos"
    },
    {
      "affiliations": [
        "The University of Alabama at Birmingham, Birmingham, AL, USA"
      ],
      "name": "Amr Ismail"
    },
    {
      "affiliations": [
        "The University of Alabama at Birmingham, Birmingham, AL, USA"
      ],
      "name": "Ellen McNeeley"
    },
    {
      "affiliations": [
        "Oregon Health and Science University, Portland, OR, USA"
      ],
      "name": "Rajat Thawani"
    },
    {
      "affiliations": [
        "The University of Alabama at Birmingham, Birmingham, AL, USA"
      ],
      "name": "Yanis Boumber"
    },
    {
      "affiliations": [
        "UAB Medicine, Birmingham, AL, USA"
      ],
      "name": "Aakash Desai"
    }
  ],
  "title": "774 Integrating immune response gene profiles for prognostic modeling in lung adenocarcinoma",
  "uid": "3b93bdb4-012e-5474-bb93-d982c542798c"
}