DDR1 is a Novel Biomarker and Potential Therapeutic Target for the Combination Treatment of Liver Hepatocellular Carcinoma

Abstract

Aim: This study aimed to investigate the role of discoidin domain receptor tyrosine kinase 1 (DDR1) in liver hepatocellular carcinoma (LIHC) and to evaluate its prognostic value on patient response to combination therapy.

Methods: In the current retrospective study, we examined the protein expression of DDR1 in various cancers by standard immunohistochemical (IHC) methods and evaluated its clinical significance in LIHC personalized treatment. Multiple online databases, including The Cancer Genome Atlas (TCGA), TIMER, GEO, ROC Plotter, and Genomics of Drug Sensitivity in Cancer (GDSC), were used.

Results: DDR1 protein expression was higher in LIHC than in other nine examined cancer types. Additionally, DDR1 exhibited higher expression levels in adjacent normal tissues compared to HBs-positive LIHC tissues. Analysis at single-cell resolution revealed that DDR1 was expressed primarily in epithelial cells but not in stromal and immune cells, and DDR1 expression was lower in HBs-positive LIHC cells in comparison with normal hepatocytes. Correlation of DDR1 upregulation and sorafenib resistance was observed in the patient cohort. Moreover, DDR1 expression positively correlated with the expression of inflammatory response-related genes, ECM-related genes, and collagen formation-related genes, but negatively correlated with the infiltration of CD8⁺ T cells, NK cells, and dendritic cells in LIHC.

Conclusions: Our findings suggest that DDR1 expression might be induced by collagen production-related cellular events involved in liver injury and repair, and that DDR1 overexpression might contribute to the resistance to LIHC targeted therapy and immunotherapy, highlighting DDR1 as a potential prognostic biomarker and therapeutic target.

Keywords: discoidin domain receptor tyrosine kinase 1; hepatitis B virus; immunotherapy; liver hepatocellular carcinoma; sorafenib.

Figure 1.

Genetic function and clinical significance of DDR1 expression in LIHC. (A) Protein expression of DDR1 across cancers. All values are presented as the means ± SDs. (B) Comparison of DDR1 protein expression between LIHC tissues and adjacent normal tissues. (C) The t-distributed stochastic neighbor embedding (t-SNE) plot of 88 448 single-cell transcriptomes in the HBsAg-positive LIHC dataset (GSE202642). Cell types were color-coded and annotated post hoc based on their transcriptional profile identities. (D) Distribution of DDR1 in various cell types. (E) Distribution of DDR1 across various cell types in the HBsAg-negative LIHC dataset (GSE166635). (F) Association between DDR1 expression and the response to sorafenib in liver cancer cell lines based on data from the GDSC2 dataset. (G) Kaplan‒Meier curves of OS for the DDR1-high and DDR1-low groups of HBsAg-positive and untreated patients with LIHC from the clinical cohort. (H) Kaplan‒Meier curves of OS for the DDR1-high and DDR1-low groups of HBsAg-positive and sorafenib-treated patients with LIHC from the clinical cohort. (I) Correlations between DDR1 expression and relevant parameters. (J) Protein expression of DDR1 in LIHC tissues from 6 patients treated with lenvatinib plus a PD-1 inhibitor-, including patients with progressive disease (PD) (n = 1), stable disease (SD) (n = 2), and a partial response (PR) (n = 3). (K) IHC staining of DDR1 in LIHC tissues from patients treated with lenvatinib plus a PD-1 inhibitor. LIHC: liver hepatocellular carcinoma; PRAD: prostate cancer; KIRC: kidney clear cell carcinoma; CHOL: bile duct cancer; PAAD: pancreatic cancer; COAD: colon cancer; STAD: stomach cancer; BLCA: bladder cancer; DLBC: large B-cell lymphoma; UVM: ocular melanoma; OS: overall survival; ***P < 0.001.