TG468: A Text Graph Convolutional Network for Predicting Clinical Response to Immune Checkpoint Inhibitor Therapy

Immunotherapy has achieved significant success in tumor treatment. However, due to disease heterogeneity, only a fraction of patients respond well to immune checkpoint inhibitor (ICI) treatment. To address this issue, we developed a Text Graph Convolutional Network (Text GCN) model called TG468 for clinical response prediction, which uses the patient’s whole exome sequencing (WES) data across different cohorts to capture the molecular profile and heterogeneity of tumors. TG468 can effectively distinguish survival time for patients who received ICI therapy and outperforms single gene biomarkers and TMB, indicating its strong predictive ability for the clinical response of ICI therapy. Moreover, the prediction results obtained from TG468 allow for the identification of immune status differences among specific patient types in the TCGA dataset. This rationalizes the model prediction results. Overall, TG468 could be a useful tool for predicting clinical outcomes and the prognosis of patients treated with immunotherapy. This could further promote the application of ICI therapy in the clinic. ### Competing Interest Statement The authors have declared no competing interest. ### Funding Statement National Natural Science Foundation of China (T2225002, 82273855 to M.Z. and 82204278 to X.T.L), Lingang Laboratory (LG202102-01-02 to M.Z.), National Key Research and Development Program of China (2022YFC3400504 to M.Z.), China Postdoctoral Science Foundation (2022M720153 to X.T.L.). ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: There are seven datasets were used as the reference set. The Miao2019 cohort comprises renal clear cell carcinoma patients treated with anti-PD-1 drugs. [Miao, D., et al., Genomic correlates of response to immune checkpoint therapies in clear cell renal cell carcinoma. Science, 2018. 359(6377): p. 801-806.] The Hugo and Riaz cohorts consist of melanoma patients treated with anti-PD-1 drugs [Hugo, W., et al., Genomic and Transcriptomic Features of Response to Anti-PD-1 Therapy in Metastatic Melanoma. Cell, 2016. 165(1): p. 35-44.; Riaz, N., et al., Tumor and Microenvironment Evolution during Immunotherapy with Nivolumab. Cell, 2017. 171(4): p. 934-949 e16.]. The Miao2018 cohort comprises pan-cancer patients treated with either 1) anti-cytotoxic T lymphocyte-associated protein-4 (CTLA-4) drugs, 2) anti-PD-1 drugs, or 3) a combination of both anti-CTLA-4 and anti-PD-1 drugs [Miao, D., et al., Genomic correlates of response to immune checkpoint blockade in microsatellite-stable solid tumors. NAT GENET, 2018. 50(9): p. 1271-1281.]. The Rizvi cohort comprises non-small cell lung cancer (NSCLC) patients treated with anti-PD-1 drugs[Rizvi, N.A., et al., Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science, 2015. 348(6230): p. 124-8.]. The Snyder and Van Allen cohorts consist of melanoma patients treated with anti-CTLA-4 drugs [Snyder, A., et al., Genetic basis for clinical response to CTLA-4 blockade in melanoma. New Engl. J. Med., 2014. 371(23): p. 2189-2199.; Van Allen, E.M., et al., Genomic correlates of response to CTLA-4 blockade in metastatic melanoma. Science, 2015. 350(6257): p. 207-211.]. There are two datasets were designated as the test sets. The Hellmann cohort comprises non-small cell lung cancer patients treated with both anti-CTLA-4 and anti-PD-1 drugs [Hellmann, M.D., et al., Genomic Features of Response to Combination Immunotherapy in Patients with Advanced Non-Small-Cell Lung Cancer. Cancer Cell, 2018. 33(5): p. 843-852 e4.]. The Liu cohort comprises melanoma patients treated with anti-PD-1 drugs [Liu, D., et al., Integrative molecular and clinical modeling of clinical outcomes to PD1 blockade in patients with metastatic melanoma. Nat Med, 2019. 25(12): p. 1916-1927.]. We downloaded eight WES datasets and corresponding clinical information from the cBioPortal database (). The Riaz cohort was obtained from the original literature [Miao, D., et al., Genomic correlates of response to immune checkpoint blockade in microsatellite-stable solid tumors. NAT GENET, 2018. 50(9): p. 1271-1281.]. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes All data produced in the present study are available upon reasonable request to the authors