Dynamic Angiopoietin-2 Serum Level as Endothelial Damage Marker and Potential Therapeutic Target

To the Editor-in-Chief:

We read with great attention the study of Price et al.¹ The authors identified an interesting low vascular injury signature, which was significantly associated with lower platelet count and higher mortality. The data also suggested that this signature could help in identifying coronavirus disease 2019 (COVID-19) acute respiratory distress syndrome patients in whom vascular targeted therapies may be useful. The signature includes 22 proteins associated with vascular injury, platelet levels, and vascular function. Tie2 was also added, being an angiopoietin-1 and angiopoietin-2 (ANGPT2) receptor. Interestingly, this signature has an inverse correlation with the severity of patients' condition, with these proteins constituting in a way a kind of normalcy signature when normally expressed. ANGPT2 predictive power for prognosis and mortality was not tested in the reported cohort; however, the data showed that the ANGPT2 levels were higher in the plasma of patients with the low mean protein abundance cluster. We previously reported a large cohort of hospitalized patients with COVID-19.² A three-day ANGPT2 increase of at least twofold from baseline was strongly associated with in-hospital mortality, whereas 10-day angiopoietin-2 increase of at least twofold from baseline was significantly associated with nonresolving pulmonary condition by multivariate analysis in our study. Interestingly, the pathologic picture in the lungs from nine autopsies in our study was similar to that reported in the present article. In our study, we stained the lung samples for ANGPT2, Tie2, CD68, and CD34, whereas in the present study, lungs were stained for ANGPT2, CD61, phosphorylated mixed lineage kinase domain-like protein (pMLKL), and CD31. In both of the studies, diffuse alveolar damage in COVID-19, interstitial inflammation, dilated capillaries, and microthrombi were seen. A marked ANGPT2 staining was present in the areas of higher damage, in newly formed vessels in the alveolar walls, in pericytes, and in macrophages inside alveolar spaces. In our study, ANGPT2 colocalized with Tie2, which was similarly overexpressed, likely powerfully induced by elevated ANGPT2.

On the whole, these two studies^{1 , 2} address an extremely relevant topic (ie, the role of the angiogenic factor ANGPT2 in the pathogenesis of the inflammatory and vascular disruptive syndrome occurring in COVID-19). ANGPT2 is released from endothelial granules Weber Palade bodies in case of vascular inflammation. The endothelium can be activated by circulating inflammatory mediators, such as polymorphonuclear leukocyte or monocyte-derived enzymes. The aberrant dysregulation of ANGPT2 influences and promotes vascular hyperpermeability. ANGPT2 modulates the function of angiopoietin-1 (that promotes instead the stabilization of nascent blood vessels). ANGPT2 is able to disrupt this action and become predominant in the pathogenic mechanism. Furthermore, abnormally high ANGPT2 levels can promote vascular leakage, such as in acute respiratory distress syndrome conditions.^{3 , 4} Interestingly, in this study,¹ the authors indicated a novel feature in the ANGPT2-associated vascular injury (ie, induction of vascular cell death). It is possible that the low vascular signature may be associated with worse prognosis, as it highlights a condition of down-regulated control mechanisms. ANGPT2, both as circulating factor and at the lung level, represents a much more direct indicator for the pathogenic mechanism underlying COVID-19 acute respiratory distress syndrome and therefore is, in our view, a more powerful candidate prognostic factor. Not less important, two tests 3 days apart may be simpler and more affordable than the complex low vascular signature proposed.