Suitability assessment of CD24 targeted-therapy in the cancer patients with COVID-19: Preliminary results from pan-cancer.

Dear Editor,

The continuing coronavirus disease 2019 (COVID-19) pandemic has resulted in diverse implications at each stage of the cancer progression for individuals with a present or prior history of cancer. We read with great interest the recent paper published in the Journal of Infection by Afshar et al., who reported that the percentage of severe cases and deaths among COVID-19-infected cancer patients was higher than among COVID-19-infected patients without cancer, due to an abnormal immune function. Mounting data shows that the severe incidence among cancer patients infected with severe acute respiratory syndrome coronavirus 2 was elevated (SARS-CoV-2).

Glycosyl-phosphatidyl-inositol (GPI)-anchored glycoprotein CD24 was discovered as a phagocytic inhibitor (the "do not eat me" signal) that has a suppressive effect in tumor immunity through binding partners such as SIGLEC10, L1CAM, and P-selectin. Genetic ablation or pharmacological inhibition of CD24 led to macrophage-dependent suppression of tumor development in vivo and enhanced survival. These results resemble the evidence that soluble CD24 may reduce COVID-19-associated systemic immunopathology. More recently, a randomized, double-blind, placebo-controlled phase 3 research confirmed that CD24-Fc is typically well tolerated and promotes clinical recovery in hospitalized COVID-19 patients requiring oxygen support. However, the potential applicability of CD24-Fc in cancer patients infected with COVID-19 remains unclear.

We first analyzed CD24 expression in pancancer by using the TCGA/GTEx dataset. CD24 downregulation was detected in 6 malignancies relative to normal tissues, including adrenocortical carcinoma (ACC), head and neck squamous cell carcinoma (HNSC), kidney chromophobe (KICH), acute myeloid leukemia (LAML), skin cutaneous melanoma (SKCM), and thymoma (THYM). In contrast, 19 malignancies demonstrated enhanced CD24 expression in established tumors (Fig. 1 A).

Fig. 1

Characterization of pan-cancer CD24 expression. (A) CD24 expression in human normal and tumor tissues statistically evaluated by the Mann-Whitney test. (B) Using univariate Cox regression, CD24 expression was linked with the survival of six kinds of cancer. (C) Kaplan-Meier plot with log-rank test of LGG samples stratified by the median CD24 value. The OS or PFS curves were shown in the top and bottom panels, respectively. (D) Effect of CD24 on survival in KIRC. (E) Correlation between CD24 expression and tumor-infiltrating immune cell infiltration levels.

CD24 expression was linked with the prognosis of six malignancies, according to univariate Cox analysis (Fig. 1B). Using the median value as the cutoff in lower grade glioma (LGG) and kidney renal clear cell carcinoma (KIRC), respectively, overall survival (OS) and progression-free interval were considerably decreased in CD24-low tumours (Fig. 1C, D).

Next, the CIBERSORT algorithm was used to assess the context-specific functions of CD24 programming tumor-infiltrating immune cells (TIICs) in the tumor microenvironment (TME). CD24 is negatively connected with CD8+ T cells and M1 macrophages, whereas it is favorably correlated with CD4+ T cells in LGG. In KIRC, CD24 had a positive connection with M2 macrophages and a negative relationship with memory B cells, M1 macrophages and Tregs (Fig. 1E), indicating that CD24 has a differential immunological role as a driving factor in various malignancies.

The host inflammatory response, involving immune-cell hyperactivation and high amounts of circulating cytokines, has been demonstrated to explain severe COVID-19, resulting in a cytokine storm. Consistent with recent findings that COVID-19 were severely inflamed but had limited T cell responses, re-analysis of a single-cell RNA sequencing dataset (GSE158055) revealed substantially elevated CD24 expression in epithelial and B cells (Fig. 2 A). The inflammatory response was then assessed using the single-sample gene set enrichment analysis (ssGSEA) technique, which was based on the signature gene sets from the Molecular Signatures Database (MSigDB). CD24 expression exhibited either an anticorrelation or no significant connection with the ssGSEA score of LGG or KIRC. In contrast, the expression of CD24 was strongly correlated with inflammation in both BRCA and LIHC cohorts (Fig. 2B). These data indicated that CD24-Fc might have mitigated the systemic inflammation of LGG or KIRC.

Fig. 2

Potential application of CD24-Fc in LGG and KIRC. (A) Annotation refinement of cell subsets within lineages using gene expression data from the NCBI GEO database with accession number GSE158055. The bottom panel displayed the CD24 expression of several cell lineages. (B) The correlation between CD24 and the inflammation score in various types of cancer. (C) CD24-related PPI network assembled using StringDB. (D) KEGG pathway enrichment of the network. (E) Differences in the TIDE score between CD24-high and CD24-low groups in LGG and KIRC respectively.

To investigate the underlying mechanism, the protein-protein interaction (PPI) network of CD24 and its ligands was built using the STRING database (Fig. 2C). The functional enrichment analysis revealed that the network was significantly enriched with KEGG pathways associated with cells and molecules involved in local acute inflammatory response or tumorigenesis, such as cell adhesion molecules, proteoglycans in cancer, bacterial invasion of epithelial cells, platelet activation, and phagosome (Fig. 2D).

Finally, we used the Tumor Immune Dysfunction and Exclusion (TIDE) framework (tide.dfci.harvard.edu) to evaluate tumor immune dysfunction and exclusion in order to predict the impact of CD24 on immune checkpoint blockade (ICB) treatment. Consequently, there was no difference in ICB response prediction between the CD24-high and CD24-low groups in LGG and KIRC. This demonstrated that CD24 treatment may be administered to COVID-19-infected LGG or KIRC patients, either alone or in combination with ICB, without compromising the immunotherapy's efficacy.

In conclusion, owing to the effective implementation of CD24-Fc in clinical trails, our findings have contributed the understanding of the relationship between CD24 in COVID-19 and malignancies. We proposed that CD24-Fc should not be administered to COVID-19 individuals with the BRCA, CESC, LIHC, or MESO cancers, but that it could be appropriate for LGG and KIRC patients. We hoped that our results might hasten the therapies in cancer patients during the COVID-19 pandemic.

Declaration of competing interest

All the authors declare that there are no conflicts of interest.