Immunotherapy have shown encouraging results and may improve the survival status and quality of high-risk NB patients [13–14]. However, the inconsistency of their exploratory study of tumor surface markers highlights the necessity of determining the ideal subgroup of NB immunotherapy, which is still a major challenge in immunotherapy. A monoclonal antibody named dinutuximab attached to GD2 on the surface of neuroblastoma cells has been performed as an immunotherapy drug for the neuroblastoma . Recently, various evidence has confirmed the potential of PD-1/PD-L1 in immunotherapy of neuroblastoma [28–29]. However, only a small percentage of patients benefit from immunotherapy. Therefore, it is important to identify those subgroups that can benefit from immunotherapy. In our research, we validated a method that can quantify the integrated tumor microenvironment in NB. Our results indicated that the ICI score was not only an effective prognostic feature, but also been used to identify NB patients that could benefit from immunotherapy.
As in previous studies, patients with specific active immunity have a significantly better prognosis, while high-risk NB patients with immune cell dysfunction increased immune resistance, resulting in tumor progression and unfavorable prognosis [30–31]. At first, we analyzed the ICI from 438 NB sample cohorts in two databases and divided NB into two different immune subtypes. The results displayed that the relative content of resting dendritic cells in all samples was nearly zero. ICI cluster A was marked by high ImmuneScore, naive B cells, T cells regulatory (Tregs), naive CD4 T cells and CD8 T cells. The samples in the ICI cluster B were represented with a significantly higher expression of StromalScore, T cells CD4 memory resting, M0 and M2 macrophages, and Mast cells resting. Although the Kaplan-Meier survival curves was not statistically significant.
Studies have shown that a number of cytokines, TME components and the host's immune response have a positive impact on the anti-tumor response and maintain a dynamic balance. These molecular changes during tumorigenesis may interfere with the function of infiltrating immune cells, resulting in the destruction of the dynamic balance between immune tolerance and activity .
We utilized the combination of ICI and immune associated gene expression as a new method for patient-specific customized treatment strategies. Then, the novelty immune-related genes based on the significant ICI gene clusters were used to explore its prognostic significance by integrating ICI gene clusters with survival information. The Kaplan-Meier analysis revealed that the samples in gene cluster A were significantly associated with improved outcome, while samples in remaining gene clusters had a poor prognosis (log rank test, p = 0.040; Fig. 2E). Patients the lowest expression of immune score and matrix score are grouped in ICI gene cluster B subtype, which are related to the immune cold phenotype. While patients with ICI gene cluster A and C subtypes showed higher inflammatory cell density and immune scores. Moreover, we found that the increased expression of M2 macrophage infiltration in ICI gene cluster C subtype was related to high stromal score, meaning an unfavorable prognosis . Compared with other subtypes, samples ICI gene cluster A subtype are associated with a good prognosis, which could benefit from immunotherapy.
There is an urgent need to quantify the ICI pattern of neuroblastoma because of the heterogeneity of the individual microenvironment. Similar individual models have been well established in colorectal cancer, head and neck squamous cell carcinoma and breast cancer to improve prediction accuracy [18, 33–34].
The potential "signature subtype" was established to quantify the ICI model in our study. The Kaplan-Meier analysis showed that the samples high ICI score subtype was significantly associated with better outcome, while samples in low ICI score subtype had an unfavorable OS (log rank test, p = 0.025). The most immune checkpoint associated targets and immune activity associated targets were overexpressed in the high ICI subgroup. It is inferred that compared with ICI NB subtypes, patients with high ICI NB subtypes were more likely to respond to immunotherapy. For example, PD-1/PD-L1 targeted immunotherapy was more likely to be effective because of the PD-L1 expression was more expressed in the high ICI NB subgroup. The genes of immunosuppressive glycosaminoglycan biosynthesis heparan sulfate signaling pathway were remarkably enriched in the low ICI score subgroup by GSEA analysis. Destroying the integrity of the extracellular matrix and basement membrane was the primary condition for the invasion and metastasis of malignant tumors. Microscopic damage on the extracellular matrix in the tumor microenvironment is a sign of tumor aggressive growth . Heparanase (HPSE), as an endogenous endoglycosidase, was highly expressed in high-risk neuroblastoma. Tumor cells secreted a large amount of heparanase to destroy the strong network structure of extracellular matrix and basement membrane, promoting the invasion and metastasis of inflammatory cells and tumor cells and related diseases Progress . In the high ICI group, Apoptosis, Cytokine receptor interaction, Toll like receptor and Leukocyte transendothelial migration signaling pathways were enriched.