The Impact of SARS-CoV-2 Infection in Patients with Inborn Errors of Immunity: the Experience of the Italian Primary Immunodeficiencies Network (IPINet)

During the last 2 years, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has spread worldwide, causing about 5 million deaths. Clinical manifestations are very variable, going from a completely asymptomatic infection to multiorgan failure and death [1]. The main risk factors of worse outcome in the general population include older age, male sex, and pre-existing comorbidities. Recent studies suggest that the clinical course of the infection may not be different in patient with inborn errors of immunity (IEI) compared to the general population [2‐4]. However, since IEI are a heterogeneous group of disorders including more than 400 different clinical conditions [5], more studies are necessary to better define the outcome of the infection in the different forms. Specific immune defects may serve as an experimentum naturae suggesting the role of specific branches of the immunity in predisposing or even protecting from a worse outcome and a better clarification of the role of different branches of the immunity in the response to SARS-CoV-2 may pave the way for the development of novel targeted treatments. Many aspects of the infection still need to be clarified in this syndrome. These include the ability to clear the infection in different IEI and the ability to develop an effective immunological memory post-infection.

In this paper, we describe the clinical course and outcome in a cohort of IEI patients followed at centers of the of the Italian Primary Immunodeficiencies Network (IPINet) who experienced SARS-CoV-2 infection between March 2020 and April 2021. The aim of the study is to better define the clinical course of the infection and define risk factors for a worse outcome in a large cohort of patients.

Patients in follow-up at IPINet centers who experienced SARS-CoV-2 infection between March 2020 and April 2021 were included in the study. SARS-CoV-2 infection was confirmed by RT-PCR on the nasopharyngeal swab or through antibody testing. For each patient, data were collected retrospectively from the clinical records. The survey was sent to 16 Italian reference centers for adult (eight centers: Milano, Brescia, Padova, Ancona, Roma, Napoli, Bari, Cagliari) or pediatric (eight centers: Milano, Brescia, Padova, Bologna, Firenze, Roma, Napoli, Bari) IEI. Most of the cases have been already included in another study on the IPINet cohort focusing on incidence, outcome, and infection-fatality rate [6]. A total of 33 patients included in the previous study were not included in the current study since the referring center decided not to take part in the second study or because detailed information on the clinical course of the infection was missing. Moreover, compared to the previous study, the current study includes 16 novel patients identified after the conclusion of the previous study. These include patients n° 13, 18, 25, 38, 40–45, 58, 64, 86, and 99–101 in the supplemental table. In this study, we performed an in-depth evaluation including detailed information on the type of underlying IEI, sex, age, comorbidities, and ongoing therapies along with baseline immunological and lung function assessment. As for the baseline immunological assessment, information collected included immunoglobulin levels, number and percentage of white blood cells (WBC), neutrophils, lymphocytes, T lymphocytes (CD3 +), helper T cells (CD4 +), cytotoxic T cells (CD8 +), B cells (CD19 +), and NK cells (CD16 + CD56 +). As for the lung function, information collected included baseline SatO2% and pre-existing structural lung damage (atelectasis, bronchiectasis, interstitial lung disease, or nodules). To evaluate the duration of the viral shedding, the dates of the first positive and first negative nasopharyngeal swab were recorded. The patients were then stratified according to hospitalization. For patients not requiring hospitalization, information collected included body temperature, resting SatO2%, SatO2% after 6-min walking test (6MWT), and type and duration of the symptoms including cough, headache, sore throat, nasal obstruction, anosmia/ageusia, myalgia, nausea/vomiting, diarrhea, loss of appetite, conjunctivitis, dyspnea, increased expectorate, and hemoptoe. The average time to recovery from COVID-19 is highly variable and depends on age and pre-existing comorbidities and illness severity [7, 8]. In individuals with mild infection, recovery time is usually about 2 weeks, whereas in individuals with severe disease, longer time was observed [7, 8]. We evaluated the percentage and the clinical feature of patients with COVID symptoms lasting more than 3 weeks. In the group of patients requiring hospitalization, when available, we collected data about PCR, LDH, ferritin, IL-6, D-dimer, procalcitonin, platelet count, number and percentage of white blood cells (WBC), neutrophils, lymphocytes, T lymphocytes (CD3 +), helper T cells (CD4 +), cytotoxic T cells (CD8 +), B cells (CD19 +), NK cells (CD16 + CD56 +), SatO2%, degree of respiratory involvement, presence of pulmonary embolism or deep vein thrombosis, need of ICU admission, oxygen requirement, requirement of mechanical ventilation or extracorporeal membrane oxygenation (ECMO), complications, and duration of the hospitalization and outcome. Finally, according to IEI type, information on the seroconversion was collected.

Serology was evaluated in 30 patients, and seroconversion was documented in 26 patients (2 patients with WAS post-GT, 2 patients with UAD not requiring IgRT, 5 patients with CVID under IgRT, 1 patient with immune dysregulation, 1 patient with Myd88 deficiency, 1 patient with hyperIgM syndrome, 4 patients with 22q11.2DS, 1 patient with ataxia telangiectasia, 1 patient with SCID post-HSCT, 1 patient with SCID (NBAS deficiency), 1 patient with neutropenia, 2 patients with combined immunodeficiency, 1 patient with Good’s syndrome under IgRT, 1 patient with NK deficiency, 1 patient with EO-IBD, and 1 patient with SIgAD). In 4 patients, there was no seroconversion, and these included patients with Good’s syndrome, cDGS, agammaglobulinemia, and CVID. All the patients without seroconversion, except for the patient with Good’s syndrome, were under IgRT. Seroconversion was observed in 5/6 patients with CVID (83.3%).

Information on the duration of viral shedding was available for 60 patients. The mean duration of viral shedding was 28.71 ± 19.42 days (median 23 days; range 6–81 days). Viral shedding was longer in hospitalized compared to non-hospitalized patients (41.46 ± 23.00 vs 25.19 ± 16.95, p 0.0065) (Fig. 7a) and in patients with profound humoral IEI compared to patients with other IEI (31.84 ± 20.93 vs 22.09 ± 10.08 days, p 0.0146) (Fig. 7b).

In the period March 2020 to April 2021, a total of 114 patients infected by SARS-CoV-2, including both children and adults with IEI, were identified in the enrolled centers. Most of the pediatric patients suffered from 22q11.2DS, while most of the adults suffered from humoral immunodeficiencies and in particular CVID. This distribution parallels the distribution of IEI cases in the two groups.

Differently from some of the previously published IEI cohorts, most of the patients (80%) in the present study were managed at home and did not require hospital admission [2, 13, 15]. The hospitalization rate observed in our study was significantly lower compared to some of the previous studies [2, 9, 10, 13]. No difference in the inpatient mortality rate and the ICU admission rate was observed in our cohort compared to previous studies with higher hospitalization rate, implying that the criteria for the hospital admission were similar in the different cohorts. This variability could reflect the knowledge gained during the pandemic and different treatment modalities that are being used currently versus earlier in the pandemic. In our cohort, almost 30% were completely asymptomatic and 50% developed mild to moderate symptoms, and the rate of asymptomatic patients was higher compared to some of the previous studies [2, 9, 10, 13], while no difference was observed with Israelian [14], Turkish [16], Spanish [17], and Iranian cohorts [18]. The variability of the prevalence of asymptomatic cases possibly reflects the different access to testing in different phases of the pandemics and across different countries. The higher rate of asymptomatic patients in our cohort is, indeed, likely due to the greater screening of individuals without symptoms [19]. In fact, after COVID-19 outbreak in Lombardy region, SARS-CoV-2 screening became mandatory to access most Italian hospitals as inpatients or outpatients, to avoid hospital outbreaks. SARS-CoV-2 screening was especially required when aerosol generating procedures were performed. Since most IEI patients require very frequent hospital admissions, it is more likely for asymptomatic IEI patients to be identified. An increasing rate of asymptomatic patients during different phases of the pandemics was also observed in the general population, since at the beginning of the pandemics, the access to the testing was mainly addressed to the most symptomatic patients.

In our cohort, asymptomatic patients included patients with different IEI. Interestingly, among the asymptomatic patients, we also identified a patient with Myd88 deficiency. The patient is of Romani ethnicity and is affected with a classical early onset form caused by a homozygous c.192_194delGGA (p.Glu66del) mutation, associated with a clinical history of severe infections. This was quite surprising in that innate immune response is critical for recognizing and controlling infections through the release of cytokines and chemochines. In SARS-CoV-2 infection, hyperactive cytokine release and cytokine storm are associated with the development of severe forms. A recent study shows that TLR2 and Myd88 are implicated in SARS-CoV-2-induced inflammatory response and are associated with COVID-19 disease severity [20]. This would suggest that Myd88 deficiency may dampen the massive cytokine release associated with the severe forms. Nonetheless, in a recent study on a pediatric cohort of IEI patients, two young brothers with Myd88 deficiency developed pneumonia requiring hospital admission in both cases and oxygen administration in one case [17].

Mean age of the asymptomatic patients was not significantly lower compared to that of patients with mild-moderate symptoms. The most common symptom reported in patients not requiring hospital admission was fever followed by general symptoms and respiratory symptoms. Gastrointestinal symptoms were more rarely observed, and, differently from what observed in the general population, there was no difference among adults and children in the incidence of gastrointestinal symptoms [21, 22]. In our pediatric cohort, we did not identify any case of multisystem inflammatory syndrome in children (MIS-C).

As expected, hospitalization rate was higher in adults than in children, and in the whole cohort, mean age of hospitalized patients was significantly higher than non-hospitalized. In the pediatric cohort, only 2 children required hospital admission: a 1-year-old patient with cDGS and a 2.75-year-old child with NBAS deficiency [23]. Both patients developed COVID-19 symptoms lasting more than 4 weeks, and both recovered without sequelae. Most of the adult patients suffered from humoral immunodeficiencies, and in particular, half of them had CVID. The highest hospitalization rate was observed for Good’s syndrome followed by agammaglobulinemia and CVID. The clinical course of the infection in hospitalized patients was comparable between our study and the international study reported by Meyts et al. [2]. In particular, among the patients requiring hospital admission, the percentage of patients requiring non-invasive ventilation or ICU admission and the inpatient mortality rate were comparable between the two studies. When we compared the baseline laboratory parameters between hospitalized and non-hospitalized patients, we found that B cell count was significantly lower in the former group. However, it should be noted that hospitalized were older than non-hospitalized patients and that 82% of the hospitalized patients suffered from immunodeficiency affecting the humoral compartment, likely explaining the lower B cell count and IgRT as risk factors. As previously reported by Shield et al., in IEI, chronic lung disease represented a risk factor for hospitalization [13]. The overall mortality for COVID in our cohort was comparable to that observed in the Italian general population (3.5 vs 2.5%; p 0.7246), and the mean age of deceased people was much lower in our cohort (48 ± 12.52 vs 80 years). We also compared the age stratified mortality, and we found that the mortality in the 50–60 age range considerable exceeds the mortality from 50 to 60 age group of the Italian population (14.3 vs 0.6%; p < 0.0001) [24] (https://​www.​epicentro.​iss.​it/​coronavirus/​bollettino/​Bollettino-sorveglianza-integrata-COVID-19_​15-dicembre-2021.​pdf). This might be due to the fact that compared to the general population IEI patients have more comorbidities, usually appearing at a younger age. Almost half of the patients in our cohort had preexisting lung disease. No-one under the age of 30 in the IEI cohort died. This is in keeping with other studies that suggest age is the biggest risk factor.

In the general population, the average time to recovery from COVID-19 is highly variable and depends on age and pre-existing comorbidities in addition to illness severity [7, 8]. Individuals with mild infection are expected to recover relatively quickly (e.g., within 2 weeks), whereas many individuals with severe disease have a longer time to recovery (e.g., 2 to 3 months) [7, 8, 11, 12, 25, 26]. In our cohort, we observed the persistence of symptoms beyond 3 weeks in 16.85% of the cases, and, as also observed in the general population, the occurrence of prolonged infection was significantly higher in hospitalized vs non-hospitalized patients (52.63 vs 5.5%, p < 0.0001). Similarly, a longer viral shedding was observed in hospitalized compared to non-hospitalized patients. Among the patients with prolonged infection, all but one patient with SIgAD were under IgRT, and patients with profound humoral immunodeficiency also showed a longer viral shedding compared to other IEI. This confirms what reported in previous studies from the literature showing that persistent infection or relapsing–remitting infection is associated with profound humoral immunodeficiency and the failure to make de novo humoral responses to SARS-CoV-2 [27]. In one patient with XLA from our cohort, readmission was required for the development of worsening dyspnea after a complete resolution of acute COVID-19 phase and negativization of the PRC on the nasopharyngeal swab. A few days after readmission, the nasopharyngeal swab became positive again, and the resolution of the symptoms was obtained with remdesivir and casirivimab/imdevimab [28]. No conclusive data are available on the management of patients with persisting COVID-19 symptoms. Treatment with corticosteroids has been shown to be beneficial in a subset of patients with post-COVID inflammatory lung disease [29]. Longer or multiple courses of remdesivir have been shown to be effective in persistent SARS-CoV-2 pneumonitis in humoral immune defects [30‐32]. The effectiveness of passive antibody administration, including convalescent plasma and monoclonal antibodies, in COVID-19 is still debated, and information on the use of these drugs in immunodeficient patients is still limited. In a recent randomized trial, Simonovich et al. showed no difference in clinical outcome and mortality between convalescent plasma and placebo in the treatment of adult patients with severe SARS-CoV-2 pneumonia [33]. The European Medicines Agency recommend the use of monoclonal antibodies in patients with confirmed COVID-19 who do not require supplemental oxygen and are at high risk of progressing to severe COVID-19. In fact, in a large double-blind study, monoclonal antibody cocktail has been shown to be effective in reducing the viral load especially in patients whose immune response had not yet been initiated [34]. Mira et al. showed rapid recovery of a SARS-CoV-2-infected XLA patient after infusion of COVID-19 convalescent plasma [35]. In our study, information on the treatment was available for 7 out of 10 hospitalized patients with prolonged infection. Two patients were treated with convalescent plasma, 2 were treated with casirivimab/imdevimab, 1 was treated with anakinra, and the remaining 2 were treated with steroids. In 3 cases, these treatments were used in combination with antiviral drugs.

The rapid diffusion of the pandemic prompted studies on the development of new vaccination strategies. Patients with disorders of the immune system were vaccinated with priority. However, information on the efficacy of the vaccine in this population is still limited. We evaluated seroconversion after natural infection in 30 cases. Due to the difference in the techniques used for the analysis in the different labs, we were not able to compare the titer among the patients. Seroconversion was observed in most cases (86.6%). As recently shown in papers looking at the seroconversion after SARS-CoV-2 vaccination in IEI [36, 37], we also observed seroconversion in 5 out of 6 patients with CVID under IgRT and in 1 out of 2 patients with Good’s syndrome under IgRT. This would indicate that, in CVID, a residual functionality of B cell is still available.

In conclusion, in this study, we reported on the clinical course and outcome of SARS-CoV-2 infection in a large cohort of IEI patients, identified over a long period of observation. The identification of the patients through different phases of the pandemics has shown that, similarly to the general population, the clinical course in these patients is often very mild or asymptomatic. In fact, because of the redundancy of the immune system, only a few immunodeficiencies are associated with an increased risk of severe COVID. These include autoimmune polyendocrine syndrome type 1 and the immunodeficiencies involving the type I IFN pathway [38‐44]. The severity and mortality of SARS-CoV-2 infection in this cohort seem to be related to the higher prevalence of comorbidities that usually appear early in life. We also confirmed that patients with humoral defects are able to produce antibodies after the natural infection. However, in some cases, patients with humoral immunodeficiencies may develop persistent infection or relapsing–remitting infection.