Immunotherapy for advanced gastric cancer

Combined single cell and spatial transcriptome analysis reveals cellular heterogeneity of hedgehog pathway in gastric cancer

Gastric cancer (GC) is one of the most common and deadly malignancies in the world. Abnormal activation of hedgehog pathway is closely related to tumor development and progression. However, potential therapeutic targets for GC based on the hedgehog pathway have not been clearly identified. In the present study, we combined single-cell sequencing data and spatial transcriptomics to deeply investigate the role of hedgehog pathway in GC. Based on a comprehensive scoring algorithm, we found that fibroblasts from GC tumor tissues were characterized by a highly enriched hedgehog pathway. By analyzing the development process of fibroblasts, we found that CCND1 plays an important role at the end stage of fibroblast development, which may be related to the formation of tumor-associated fibroblasts. Based on spatial transcriptome data, we deeply mined the role of CCND1 in fibroblasts. We found that CCND1-negative and -positive fibroblasts have distinct characteristics. Based on bulk transcriptome data, we verified that highly infiltrating CCND1 + fibroblasts are a risk factor for GC patients and can influence the immune and chemotherapeutic efficacy of GC patients. Our study provides unique insights into GC and hedgehog pathways and also new directions for cancer treatment strategies.

Immunotherapy Based on Immune Checkpoint Molecules and Immune Checkpoint Inhibitors in Gastric Cancer-Narrative Review

Due to its rapid progression to advanced stages and highly metastatic properties, gastric cancer (GC) is one of the most aggressive malignancies and the fourth leading cause of cancer-related deaths worldwide. The metastatic process includes local invasion, metastasis initiation, migration with colonisation at distant sites, and evasion of the immune response. Tumour growth involves the activation of inhibitory signals associated with the immune response, also known as immune checkpoints, including PD-1/PD-L1 (programmed death 1/programmed death ligand 1), CTLA-4 (cytotoxic T cell antigen 4), TIGIT (T cell immunoreceptor with Ig and ITIM domains), and others. Immune checkpoint molecules (ICPMs) are proteins that modulate the innate and adaptive immune responses. While their expression is prominent on immune cells, mainly antigen-presenting cells (APC) and other types of cells, they are also expressed on tumour cells. The engagement of the receptor by the ligand is crucial for inhibiting or stimulating the immune cell, which is an extremely important aspect of cancer immunotherapy. This narrative review explores immunotherapy, focusing on ICPMs and immune checkpoint inhibitors in GC. We also summarise the current clinical trials that are evaluating ICPMs as a target for GC treatment.

Interpretable machine learning for predicting the response duration to Sintilimab plus chemotherapy in patients with advanced gastric or gastroesophageal junction cancer

Background

Sintilimab plus chemotherapy has proven effective as a combination immunotherapy for patients with advanced gastric and gastroesophageal junction adenocarcinoma (GC/GEJC). A multi-center study conducted in China revealed a median progression-free survival (PFS) of 7.1 months. However, the prediction of response duration to this immunotherapy has not been thoroughly investigated. Additionally, the potential of baseline laboratory features in predicting PFS remains largely unexplored. Therefore, we developed an interpretable machine learning (ML) framework, iPFS-SC, aimed at predicting PFS using baseline (pre-treatment) laboratory features and providing interpretations of the predictions.

Materials and methods

A cohort of 146 patients with advanced GC/GEJC, along with their baseline laboratory features, was included in the iPFS-SC framework. Through a forward feature selection process, predictive baseline features were identified, and four ML algorithms were developed to categorize PFS duration based on a threshold of 7.1 months. Furthermore, we employed explainable artificial intelligence (XAI) methodologies to elucidate the relationship between features and model predictions.

Results

The findings demonstrated that LightGBM achieved an accuracy of 0.70 in predicting PFS for advanced GC/GEJC patients. Furthermore, an F1-score of 0.77 was attained for identifying patients with PFS durations shorter than 7.1 months. Through the feature selection process, we identified 11 predictive features. Additionally, our framework facilitated the discovery of relationships between laboratory features and PFS.

Conclusion

A ML-based framework was developed to predict Sintilimab plus chemotherapy response duration with high accuracy. The suggested predictive features are easily accessible through routine laboratory tests. Furthermore, XAI techniques offer comprehensive explanations, both at the global and individual level, regarding PFS predictions. This framework enables patients to better understand their treatment plans, while clinicians can customize therapeutic approaches based on the explanations provided by the model.

Establishment of a prognostic model for gastric cancer patients who underwent radical gastrectomy using machine learning: a two-center study

Objective

Gastric cancer is a prevalent gastrointestinal malignancy worldwide. In this study, a prognostic model was developed for gastric cancer patients who underwent radical gastrectomy using machine learning, employing advanced computational techniques to investigate postoperative mortality risk factors in such patients.

Methods

Data of 295 patients with gastric cancer who underwent radical gastrectomy at the Department of General Surgery of Affiliated Hospital of Xuzhou Medical University (Xuzhou, China) between March 2016 and November 2019 were retrospectively analyzed as the training group. Additionally, 109 patients who underwent radical gastrectomy at the Department of General Surgery Affiliated to Jining First People's Hospital (Jining, China) were included for external validation. Four machine learning models, including logistic regression (LR), decision tree (DT), random forest (RF), and gradient boosting machine (GBM), were utilized. Model performance was assessed by comparing the area under the curve (AUC) for each model. An LR-based nomogram model was constructed to assess patients' clinical prognosis.

Results

Lasso regression identified eight associated factors: age, sex, maximum tumor diameter, nerve or vascular invasion, TNM stage, gastrectomy type, lymphocyte count, and carcinoembryonic antigen (CEA) level. The performance of these models was evaluated using the AUC. In the training group, the AUC values were 0.795, 0.759, 0.873, and 0.853 for LR, DT, RF, and GBM, respectively. In the validation group, the AUC values were 0.734, 0.708, 0.746, and 0.707 for LR, DT, RF, and GBM, respectively. The nomogram model, constructed based on LR, demonstrated excellent clinical prognostic evaluation capabilities.

Conclusion

Machine learning algorithms are robust performance assessment tools for evaluating the prognosis of gastric cancer patients who have undergone radical gastrectomy. The LR-based nomogram model can aid clinicians in making more reliable clinical decisions.

Spatial analysis of NOS2 and COX2 interaction with T-effector cells reveals immunosuppressive landscapes associated with poor outcome in ER- breast cancer patients

Multiple immunosuppressive mechanisms exist in the tumor microenvironment that drive poor outcomes and decrease treatment efficacy. The co-expression of NOS2 and COX2 is a strong predictor of poor prognosis in ER- breast cancer and other malignancies. Together, they generate pro-oncogenic signals that drive metastasis, drug resistance, cancer stemness, and immune suppression. Using an ER- breast cancer patient cohort, we found that the spatial expression patterns of NOS2 and COX2 with CD3+CD8+PD1- T effector (Teff) cells formed a tumor immune landscape that correlated with poor outcome. NOS2 was primarily associated with the tumor-immune interface, whereas COX2 was associated with immune desert regions of the tumor lacking Teff cells. A higher ratio of NOS2 or COX2 to Teff was highly correlated with poor outcomes. Spatial analysis revealed that regional clustering of NOS2 and COX2 was associated with stromal-restricted Teff, while only COX2 was predominant in immune deserts. Examination of other immunosuppressive elements, such as PDL1/PD1, Treg, B7H4, and IDO1, revealed that PDL1/PD1, Treg, and IDO1 were primarily associated with restricted Teff, whereas B7H4 and COX2 were found in tumor immune deserts. Regardless of the survival outcome, other leukocytes, such as CD4 T cells and macrophages, were primarily in stromal lymphoid aggregates. Finally, in a 4T1 model, COX2 inhibition led to a massive cell infiltration, thus validating the hypothesis that COX2 is an essential component of the Teff exclusion process and, thus, tumor evasion. Our study indicates that NOS2/COX2 expression plays a central role in tumor immunosuppression. Our findings indicate that new strategies combining clinically available NOS2/COX2 inhibitors with various forms of immune therapy may open a new avenue for the treatment of aggressive ER-breast cancers.