AUTHOR=Yuan Feng , Cai Xiangming , Zhu Junhao , Yuan Lei , Wang Yingshuai , Tang Chao , Cong Zixiang , Ma Chiyuan 

TITLE=A Novel Immune Classification for Predicting Immunotherapy Responsiveness in Patients With Adamantinomatous Craniopharyngioma

JOURNAL=Frontiers in Neurology

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2021.704130

DOI=10.3389/fneur.2021.704130

ISSN=1664-2295

ABSTRACT=Adamantinomatous craniopharyngioma (ACP) is the most common tumor of the sellar region in children. Owing to the aggressive behavior of ACP, current therapeutic management of ACP is limited to either aggressive surgical resection or surgical debulking followed by external beam radiation. Cancer immunotherapy has profoundly changed the concept of cancer treatments. However, immunotherapy is rarely studied in patients with ACP. In this research, through analyzing transcriptome data from GSE60815 and GSE94349 databases. We summarized the immune profile of ACP and identified two novel immune subtypes, namely IR subtype and IG subtype. We found that the ACPs with IG subtype (34.78%, 8/23) was more likely to respond to immunotherapy than the ACPs with IR subtype (6.25%, 1/16) via Tumor immune dysfunction and exclusion (TIDE) method. Simultaneously, the enrichment analysis results indicated that the DEGs of IG subtype were chiefly involved in various inflammatory and immune responses. However, the DEGs of IR subtype were mainly involved in RNA splicing, RNA catabolic process and cell cycle. Next, the CIBERSORT results found that compared with normal pituitary group, the proportion of M0 and M2 macrophages was significantly higher in ACPs. And the proportion of M2 macrophage in the IG subtype was higher than that in the IR subtype. We also found that compared with IR subtype, the expression of immune checkpoint molecules (PD1, PDL1, PDL2, TIM3, CTLA4, Galectin9, LAG3 and CD86) significantly upregulated in IG subtype. The ssGSEA results demonstrated that compared with IR subtype, the biofunction of carcinogenesis in IG subtype were significantly enriched, such lymphocyte infiltration, mesenchymal phenotype, stemness maintain and tumorigenic cytokines. Finally, we constructed a WDR89-based model to predict immune classification of ACPs. ROC curves analysis showed that WDR89-based predictive model could provide a more reliable classification assessment in the training sets (AUC=0.971) and validation sets (AUC=0.929). This predictive model might provide a reliable classification assessment tool for clinician and aid treatment decision-making in the clinic.