Immune analysis according to Lauren type for gastric cancer and its significance in individual treatment and prognostic prediction

Huiwen Lu¹Junqiao Yao²Chi Xue²Rui Huang²Baojun Huang^2*

• ¹Tianjin Union Medical Centre, Nankai University, Tianjin, China
• ²The First Affiliated Hospital of China Medical University, Shenyang, Liaoning Province, China

Background Numerous studies have proved that Lauren types are associated with the prognosis of gastric cancer patients. Whereas their associations with tumor immunity and significance in treatment remain unclear. Method Eligible GC patients who underwent curative resection at our institution were identified for this study. Tumor specimens were collected and processed with immunohistochemical staining. Additional GC data related to human subjects were collected from GEO public dataset. Further analysis was then performed using TCGA and GEO datasets. GC patients in public datasets were divided into two groups according to their Lauren types. Survival analysis was performed between subtypes. The differences in infiltrating immune cells, human leukocyte antigen (HLA) family, checkpoints, and apoptosis-regulated genes between groups were analyzed. Then, associations between Lauren types and clinicopathological features were analyzed. Results GC patients with diffuse type showed higher expression of CD3 (P=0.042), CD8 (P=0.025), and CD57 (P=0.020) then those with intestinal types. Among 300 GC patients in the GEO training set, patients with diffuse type showed a poorer prognosis than intestinal-type ones (OS: P<0.001; RFS: P=0.005). The diffuse subtype had more immune cells but was less functional than the intestinal subtype. Notably, checkpoints were highly expressed among diffuse-type patients. Intestinal patients had a higher positive rate of HER2 than diffuse ones. To find the hub genes, a three-gene-included risk model based on Lauren was constructed. The risk score was independently associated with survival of gastric cancer patients, regardless of OS and RFS. ROC analysis showed that this risk model had a good predictive ability. High-risk patients had more advanced T, M and pathological stage, indicating that those with the high risk presented more aggressive features. Immune analysis was consistent with Lauren type. Results from the TCGA validation group were consistent with the GEO training set. Conclusion Diffuse-type tumors exhibited greater immune cell abundance but reduced functional activity, contributing to poorer prognosis. These tumors also demonstrated potentially higher sensitivity to immunotherapy and chemotherapy compared to intestinal-type tumors. HER2-targeted therapy combined with chemoradiotherapy is strongly recommended for intestinal-type GC patients. These disparities are primarily attributable to upregulated LINC00702, C8orf88, and FILP1 in diffuse-type GC patients