Patients With Hereditary Angioedema Do Not Develop More Severe COVID-19 but SARS-CoV-2 Infection May Trigger Attacks: 66 Cases

Purpose Hereditary angioedema (HAE) is a rare genetic disease with hyperactivated contact and kallikrein-kinin systems leading to bradykinin (BK) release and edema. SARS-CoV-2 infection results in inammatory exacerbation. C1 inhibitor (C1-INH) deciency could aggravate clinical outcomes, with HAE patients at a greater risk of adverse outcomes of COVID-19, however, data are still limited. Our aim was to characterize the course and severity of COVID-19 in patients with HAE. Methods


Introduction
Hereditary angioedema (HAE) is a rare genetic disease with autosomal dominant inheritance characterized by localized and asymmetric episodes of subcutaneous and submucosal edema with a high impact on quality of life 1,2,3 . Unnecessary surgeries and death due to airway obstruction can occur 2,4 . Two types of HAE are recognized: HAE with de ciency and/or dysfunctional C1 inhibitor (HAE-C1-INH) and HAE with normal C1-INH (HAE-nC1-INH); and the estimated prevalence is 1:50,000 and 1:400.000, respectively 4,5 . Gene variants involved with HAE-C1-INH are identi ed in SERPING1 gene and variants in genes encoding for coagulation factor XII (F12), Plasminogen, Kininogen 1, Angiopoetin1, Myoferlin and, more recently, heparan sulfate (HS)-glucosamine 3-O-sulfotransferase 6 6 were associated with HAE-nC1-INH. There is a cluster of patients with unknown variants within the HAE-nC1-INH group 2 .
The mechanism involved in HAE is the activation of the contact and kallikrein-kinin systems, resulting in the release of bradykinin (BK), after the cleavage of high molecular weight kininogen (HK) by kallikrein.
Kallikrein is the active cleaved form of pre-kallikrein formed after cleavage by activated FXII. After release, BK or its derivative des-Arg9-bradykinin (DABK), binds to B2R or B1R, respectively, generating increased vascular permeability, which causes angioedema. C1-INH inhibits not only contact and kallikrein-kinin systems but also brinolytic, complement and coagulation pathways and its de ciency or dysfunction leads to increased BK release. In most HAE-nC1-INH, the same mechanism is observed 7,8 .
The clinical spectrum of COVID-19, the disease caused by SARS-CoV-2, varies widely, from asymptomatic or mild symptoms similar to common colds to severe respiratory failure with risk of death. Recognized risk groups for the severity of the infection are age over 60, hypertension, diabetes and obesity 9 . Several fatal complications have been observed in the course of the disease: heart, liver or kidney failure, among others 9,10 .
SARS-CoV-2 uses surface proteins to invade host cells. The virus' spike glycoprotein binds to the angiotensin-converting enzyme 2 (ACE2), which is highly expressed on the surface of respiratory cells, and this interaction is considered essential for SARS-CoV-2 to enter target cells. In some cells, the "spike" glycoprotein can bind to other surface proteins, such as TCD4 + cells 10,11 . The binding of SARS-CoV-2 causes negative regulation of ACE2, in uencing the regulation of the renin-angiotensin system, possibly related to changes in blood pressure and in ammation 10 .
Considering the action of ACE2 inhibitors in the kallikrein-kinin system and the observation that antiin ammatory treatments have limited e cacy in respiratory distress, the relationship between the activated contact and kallikrein-kinin systems and the pulmonary manifestations of COVID-19 is a strong possibility [12][13][14][15] . Clinical studies with small number of patients and severe COVID-19 treated with recombinant or plasma-derived C1 inhibitor, or even icatibant, showed improved clinical, laboratory and radiological parameters [12][13][14] , reinforcing the role of these systems 15 .
A patient with HAE and COVID-19 has been reported, which manifested a mild illness without triggering an attack of angioedema 16 . In addition, a patient with acquired non-histaminergic angioedema was reported with mild COVID-19 and without angioedema attack 17 . A recent casuistry showed mild disease in HAE patients 18 .
Considering the scarcity of data about hereditary angioedema with or without C1 inhibitor de ciency and COVID-19, we collected clinical characteristics of these patients in a wider population focusing the severity and evolution of the infection.

HAE Referral Centers in Latin
American countries were consulted about patients with SARS-CoV-2 infection. For inclusion in the study, there was no age limit or restriction to risk factors for COVID-19. The diagnosis of HAE was con rmed by typical clinical symptoms, biochemical tests evaluating the dosage and / or function of C1 inhibitor, family history and for patients with HAE-nC1-INH, whenever possible, F12 variants were evaluated. We registered the tests performed for the con rmation of SARS-CoV-2 infection: PCR, serology and / or rapid test. A questionnaire was distributed to the centers including age, sex, type of HAE, risk factors, variants for HAE-nC1-INH if available, prophylaxis for HAE, COVID-19 symptoms, occurrence of angioedema attacks and therapy used for treating each attack, need for hospitalization, period of symptomatology, evolution and complications. The project was submitted and approved by the Ethics Committee (CAAE: 40745220.0.1001.0082).
Comorbidities were not identi ed in 69.7% of the patients, obesity was present in 12.1%, diabetes in 6.0%, arterial hypertension 4.5%, neoplasms and other conditions in 7.7%. Median time of disease was 11 (IQR: 7-15) and 10 (IQR 3-15) days in patients with HAE-C1-INH and HAE-nC1-INH, respectively. Complete cure was observed in 61 patients (92.4%), pulmonary sequelae in 4 and death in one patient ( Table 1). The disease progression had no difference in relation to the sex or type of HAE (p = 0.803).
Angioedema attacks occurred in 45.5% of patients with HAE during SARS-CoV-2 infection, predominantly in HAE-C1-INH (26/53) comparing with HAE-nC1-INH (4/13), however, there was no signi cant difference (p>0.05). Attacks affected the following sites: face and tongue in 7/66 (10.6%); extremities in 12/66 (18.2%); abdomen in 7/66 (10.6%) and larynx in 4/66 (6.1%). Discriminating by sex, a signi cant association was evidenced between the groups (p = 0.030), with the attacks occurring mainly in women with HAE-C1-INH during COVID-19. Most of the patients who suffered attacks were not receiving prophylaxis and no association between the occurrence of attacks and the use of prophylaxis was observed (p = 0.648). Androgens was used for long term prophylaxis in 48.4% of the patients (32/66), among which 93.3% progressed to cure after suffering SARS-CoV2 infection with no sequelae. Only a 71year-old woman was reported as deceased within this group due to septic shock.
Treatment of the SARS-CoV-2 infection in these patients was based on analgesics and antipyretics in 53%, mainly paracetamol (acetaminophen) (42.4%), antibiotics in 22.7%, azithromycin being the most used (18.1%), followed by anti-in ammatories in 11%. Combined therapies, based on the aforementioned drugs, were utilized by 22.1% of the cases; and other therapies such as vitamins and ivermectin were prescribed in 7.6%.

Discussion
The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) rst emerged in Wuhan (Hubei Province, China), in December 2019, and COVID-19 pandemic has spread rapidly worldwide. Only a small percentage of patients develop the potentially lethal complications of acute respiratory distress syndrome (ARDS), disseminated intravascular coagulation (DIC), and multiorgan failure (MOF) 19,20 . After entering the host, SARS-CoV-2 elicits a series of innate and adaptive immune responses, which are responsible for viral clearance as well as in ammation. Emerging evidences suggest that complement is chronically activated in severe COVID-19 and plays a key role in critically ill patients, due to a dysregulated in ammation characteristic of severe COVID-19 21 . Therefore, we decided to evaluate HAE patients infected with SARS-CoV-2 in order to describe its in uence on their prognosis. Of interest, we compared with patients diagnosed with HAE-nC1-INH, considering that C1-INH is normal in this HAE type.
As a collaborative study, we were able to collect data from 66 patients and no differences were evidenced comparing patients with or without C1 inhibitor de ciency. This result was not expected since C1-INH is a major regulator of all three pathways of complement activation 22 . Through covalent bond formation with the complement components C1s, C1r, MASP1 and MASP2 and reversible binding to C3a, C1-INH attenuates the consequences of complement activation, including the generation of proin ammatory anaphylatoxins, especially C5a, and the formation of the membrane attack complex (MAC) that leads to cell lysis 22,23 . Coagulation and brinolytic pathways are also regulated by C1-INH and it is presumed that the lower C1-INH activity could predispose to more severe SARS-CoV-2 infection 24 . Moreover, C1-INH has been demonstrated to interact with components of the extracellular matrix, leading to the hypothesis that these components concentrate C1-INH at the local site of in ammation in order to regulate complement and contact systems 25 .
The similar course of disease in our patients compared to the general population, can be explained by some ndings. Females were the predominant sex in our group as it is described in most HAE casuistries 26-30 . Men have been described with higher severity of COVID-19 in comparison with women 31,32 . The median age observed by us was in the 30s and it may have in uenced the prognosis. Seven HAE patients were older than 60 years of age and this is a risk factor for severe COVID-19. Unfortunately, one 71-year-old patient died from pulmonary complications and multiorgan disfunction related to COVID-19, 28 days after the beginning of symptoms. No additional comorbidity was reported in this patient and she was under HAE prophylaxis with danazol. This death accounted for 1.5% of total cases, less than the mortality reported for general population in Latin American countries 33 . Furthermore, approximately 70% of our HAE patients had no comorbidities; however, 12% of them were obese. Additionally, low C4 serum levels found in HAE patients might prevent further complement activation and deleterious clinical effects derived from increased complement activity. All these factors could contribute to the better prognosis observed in our population. Different trials inhibiting complement activation, targeting either C3 or C5, have been proposed for critically ill patients 34  Androgens, which are able to induce C1-INH liver production, have been the most accessible therapy in Latin American countries; therefore, it is possible that prophylaxis improved the evolution or severity of COVID-19 in HAE patients. On the other hand, a role of a co-receptor for SARS-CoV-2 infection, transmembrane protease serine 2 (TMPRSS2), was described to be upregulated by androgens in a lungderived cell line model, but not con rmed in physiologic settings 35 . Although this might be a concern in male patients treated with androgens because of the possibility to in uence the outcome of COVID-19, none of our 13 male patients had severe clinical manifestations of SARS-CoV-2 infection.
Eight patients were treated with tranexamic acid and no thromboembolism was reported. One patient with HAE-nC1-INH and no variant identi ed was hospitalized due to high D-dimer values; however, this nding is described in HAE patients during attacks 36 and in COVID-19. This patient was not experimenting angioedema symptoms and evolved with no complications except for involvement of 25% of the lungs during the symptomatic phase.
Long term prophylaxis treatment, to reduce the severity and frequency of HAE episodes, includes infusion of C1-INH, which not only reduces kinin-kallikrein activation and BK production, but also might protect against lung injury by inhibiting the cytotoxic activity of extracellular histones 37 . This treatment may improve the outcome of HAE patients with SARS-Cov-2 infection, however, access in Latin America is still limited.
Treatment of acute episodes of HAE includes icatibant, a bradykinin receptor 2 antagonist, or ecallantide, a kallikrein antagonist both of which reduce BK production having a modulatory effect on the "bradykinin storm" described in the pathophysiology of COVID-19 38 . Some of our patients were treated with icatibant for acute episodes and the clinical response was adequate.
We evaluated a more representative number of patients with HAE C1-INH and HAE-nC1-INH infected with SARS-CoV-2. The expression of the COVID-19 was not different from the population without HAE, possibly due to the predominance of female and young patients as well as HAE therapy. Our ndings suggest that SARS-CoV-2 infection is a trigger for angioedema attacks, however, the prognosis was not in uenced, as previously observed by us and others in a much smaller casuistry 18, 39 . Fortunately, none of our children had MIS-C; and it is believed that pediatric COVID-19 has a more favorable prognosis 40