ساخت اکانت کاربری رایگان

مقالات پذیرفته شده کنگره

  • صفحه نخست
  • مقالات پذیرفته شده اولین کنگره بین المللی ژنتیک سرطان
بازنشر

  • Cross-Cancer Transcriptomic Profiling Reveals Context-Dependent Conservation of Mesenchymal Programs Associated with Immune Checkpoint Inhibitor Resistance

  • Seyedeh Alaleh Jahanpour Shirvan,1,* Zahra Aliabadi,2 Fatemeh Darvish,3 Keyvan Mohebbi,4
    1. Department of Clinical Biochemistry, School of Medicine, Golestan University of Medical
    2. Department of Microbiology, School of Medicine, Golestan University of Medical Sciences, Gorgan, Iran
    3. Department of Biochemistry and Biophysics » School of medicine, Golestan University of Medical Sciences » Gorgan, Iran
    4. Department of Dentistry, School of Medicine, Golestan University of Dental Sciences, Gorgan, Iran


  • Introduction: Immune checkpoint inhibitors (ICIs) have transformed cancer therapy; however, only a subset of patients achieves meaningful and durable responses. Established biomarkers such as tumor mutational burden (TMB) incompletely capture the complex tumor-intrinsic and microenvironmental mechanisms underlying ICI resistance. Transcriptomic profiling may provide complementary biological information and identify resistance-associated programs that extend across tumor types. We investigated whether epithelial–mesenchymal transition and extracellular-matrix (EMT/ECM) transcriptional programs associated with ICI response are conserved across independent cancer cohorts and detectable in peripheral blood.
  • Methods: Three publicly available transcriptomic datasets representing independent ICI-treated cohorts were analyzed. GSE78220 served as the discovery cohort and included pretreatment tumor RNA-sequencing profiles from pembrolizumab-treated melanoma. After exclusion of an on-treatment specimen and patient-level consolidation of duplicate pretreatment biopsies, 26 patients were evaluated (14 responders and 12 non-responders). Complete and partial responses were classified as response, whereas progressive disease was classified as non-response. FPKM values were log2-transformed. Gene-level differential expression and prespecified transcriptional programs representing EMT/ECM, angiogenesis, interferon-γ response, cytotoxic T-cell activity, antigen presentation, and other tumor-microenvironmental processes were evaluated with multiple-testing correction. Cross-cancer evaluation was performed in GSE176307, an independent metastatic urothelial carcinoma cohort treated with immune checkpoint blockade. RNA-sequencing profiles were matched to clinical response metadata, yielding 87 evaluable patients (16 responders and 71 non-responders). The discovery EMT/ECM signature was applied without feature reselection. Discrimination was evaluated using receiver-operating-characteristic area under the curve (ROC-AUC), precision-recall AUC (PR-AUC), effect sizes, and bootstrap confidence intervals. TMB was assessed as an established genomic comparator. Exploratory cross-compartment analysis was additionally performed using GSE225620 whole-blood transcriptomes collected before and after neoadjuvant PD-1 blockade-based therapy in non-small-cell lung cancer.
  • Results: In melanoma, the EMT/ECM transcriptional program was significantly elevated in non-responders compared with responders (standardized mean score, 0.328 vs. −0.281; Cohen’s d=1.18; p=0.0059; false-discovery rate [FDR]=0.047). The inverse EMT/ECM score discriminated response with an ROC-AUC of 0.798 and PR-AUC of 0.849. Angiogenesis was also elevated in non-responders (0.352 vs. −0.302; Cohen’s d=1.01; p=0.020), although the association did not remain significant following multiple-testing correction (FDR=0.081). In the independent urothelial carcinoma cohort, the direction of the EMT/ECM association was conserved, with lower scores in responders than non-responders (−0.274 vs. 0.056), although the effect was attenuated and not statistically significant (Cohen’s d=−0.43; p=0.166). The EMT/ECM signature achieved an ROC-AUC of 0.621 (bootstrap 95% CI, approximately 0.45–0.78) and PR-AUC of 0.347. TMB demonstrated stronger discrimination (ROC-AUC=0.776; PR-AUC=0.519), and addition of EMT/ECM to TMB did not improve cross-validated discrimination over TMB alone. Exploratory peripheral-blood analysis demonstrated detectability of all 12 tumor-derived EMT/ECM signature genes. However, the aggregate score showed no significant overall change between pretreatment (n=17) and post-treatment (n=21) blood samples (−0.066 vs. 0.053; Cohen’s d=0.18; p=0.555).
  • Conclusion: EMT/ECM remodeling was strongly associated with anti-PD-1 resistance in melanoma, while the direction of this association was preserved but substantially attenuated in an independent urothelial carcinoma cohort. The signature was measurable in peripheral blood but was not uniformly modulated following therapy. Collectively, these findings suggest that mesenchymal and extracellular-matrix remodeling represents a biologically relevant but context-dependent component of ICI resistance. Our results further demonstrate that directional biological conservation across cancer types does not necessarily translate into universal predictive performance, emphasizing the importance of independent cross-cancer and cross-compartment validation in transcriptomic biomarker development for precision immuno-oncology.
  • Keywords: Immune checkpoint inhibitors;Transcriptomics;Immunotherapy resistance

به خانواده بزرگ کنسر ژنتیکس و ژنومیکس سرطان بپیوندید!

ایجاد حساب کاربری رایگان