Integration of Histopathological Image Features and Multi-dimensional Omics Data in Predicting Molecular Features and Survival in Glioblastoma

Yeqian Huang ¹ Linyan Chen ² Zhiyuan Zhang ³ Yu Liu ¹ Leizhen Huang ¹ Yang Liu ¹ Pengcheng Liu ¹ Fengqin Song ¹ Zhengyong Li ^1* Zhenyu Zhang ^1*

• ¹ Department of Burn and Plastic Surgery, West China Hospital, Sichuan University, Chengdu, China
• ² Department of Biotherapy, Cancer Center and State Key Laboratory of Biotherapy, West China Hospital, Sichuan University, Chengdu, Sichuan Province, China
• ³ West China School of Medicine, West China Hospital, Sichuan University, Chengdu, Sichuan Province, China

Glioblastoma (GBM) is a highly malignant brain tumor characterized by complex molecular mechanisms. Histopathological images provide valuable morphological information of tumors. In this study, we assessed the predictive potential of quantitative histopathological image features (HIF) for molecular characteristics and overall survival (OS) in GBM patients by integrating HIF with multi-omics data. A total of 439 GBM patients with eligible histopathological images and corresponding genetic data from the Cancer Genome Atlas (TCGA) were included in our study. Initially, 550 image features were extracted, and machine learning algorithms were employed to identify molecular characteristics, with random forest (RF) models demonstrating the best predictive performance. Subsequently, we assigned the patients to the training and validation cohorts at a 1:1 ratio and constructed predictive models for OS based on HIF through RF, which achieved significant prognostic accuracy (5-year AUC = 0.829). Then, we established prognostic models based on single-omics, the integration of HIF and single-omics (HIF + genomics, HIF + transcriptomics, HIF + proteomics), and all features (multi-omics). The multi-omics model achieved the best prediction performance (1-, 3- and 5-year AUCs of 0.820, 0.926 and 0.878, respectively). Additionally, we enrolled tissue microarrays of 67 patients as an external validation set. Furthermore, three prognostic histopathological image features (PHIF) were identified using two machine learning algorithms, and prognosis-related gene modules were discovered through WGCNA. The results indicated a certain prognostic value of HIF, and the integrated multi-omics model may enhance the prognostic prediction of GBM with higher accuracy and robustness.