Germline variants in UNC13D and AP3B1 are enriched in COVID-19 patients experiencing severe cytokine storms

Abstract

Critically ill coronavirus disease 2019 (COVID-19) is characterized by severe cytokine storms, a hyperinflammatory condition intimately related to the development of fatal outcomes. Why some individuals seem particularly vulnerable to severe cytokine storms is still unknown. Primary immunodeficiency (PID)-related genes are inherited factors that dysregulate host inflammatory responses to infection, especially hemophagocytic lymphohistiocytosis (HLH)-related genes, established as contributors to the development of excessive cytokine storms. We analyzed the association between PID gene variants with severe cytokine storms in COVID-19. We conducted whole-exome sequencing in 233 hospitalized COVID-19 patients and identified four PID gene (UNC13D, AP3B1, RNF168, DHX58) variants were significantly enriched in COVID-19 patients experiencing severe cytokine storms. The total percentage of COVID-19 patients with variants in UNC13D or AP3B1, two typical HLH genes, was dramatically higher in high-level cytokine group than in low-level group (33.3 vs. 5.7%, P < 0.001). Germline variants in UNC13D and AP3B1 were associated with the development of severe cytokine storms, fatal outcomes in COVID-19. These findings advance the understanding of individual susceptibility to severe cytokine storms and help optimize the current management of COVID-19.

Fig. 1. Germline variants were significantly enriched in COVID-19 patients with severe cytokine storms.

A The red, blue, and green lines represent the receiver operating characteristic (ROC) curves of IP-10, IL-Ra, or MCP-3 to discriminate COVID-19 patients with ICU admission or without ICU admission, respectively. The areas under the curve (AUC) and the optimal cutoff values set by Youden’s index are shown in the graph. B The bar chart shows the mutation percentage of the four genes in high- and low-level cytokine groups and in Chinese population from 1000 G. The statistically significant mutation burden was analyzed by the chi-square test and the Fisher’s test. *P < 0.05; **P < 0.01. C The stacked bar shows the total percentage of patients harboring UNC13D or AP3B1 variants. The patients were grouped according to the severity of cytokine storms, COVID-19 severity, ICU admission, and fatal outcomes. The statistical significance between the groups was assessed by the chi-square test. *P < 0.05; **P < 0.01; ***P < 0.001.