SARS-COV-2 infection outcomes in patients with congenital generalized lipodystrophy

Background A new strain of human coronavirus (HCoV) spread rapidly around the world. Diabetes and obesity are associated with a worse prognosis in these patients. Congenital Generalized Lipodystrophy (CGL) patients generally have poorly controlled diabetes and require extremely high doses of insulin. There is no documentation in the literature of cases of COVID in CGL patients. Thus, we aimed to evaluate the prevalence of SARS-CoV-2 infection in CGL patients, and the association of their clinical and metabolic characteristics and outcomes. Methods This is a cross-sectional study carried out between July and October 2020. Clinical data collected were respiratory or other flu-like symptoms, need of hospitalization in the last three months, CGL comorbidities, and medications in use. Cholesterol, triglycerides, glycohemoglobin A1c levels, anti-SARS-CoV-2 antibodies and nasopharyngeal swab for RT-qPCR were also obtained in all CGL patients. Mann-Whitney U test was used to analyze the characteristics of the participants, verifying the non-adherence of the data to the Gaussian distribution. In investigating the association between categorical variables, we used Pearson's chi-square test and Fisher's exact test. A significance level of 5% was adopted. Results Twenty-two CGL patients were assessed. Eight subjects (36.4%) had reactive anti-SARS-CoV-2 antibodies. Only one of these, also presented detectable RT-qPCR. Five individuals (62.5%) were women, median age of 13.5 years (1 to 37). Symptoms like fever, malaise, nausea, diarrhea and chest pain were present, and all asymptomatic patients were children. All subjects had inadequate metabolic control, with no difference between groups. Among positive individuals there was no difference between those with AGPAT2 (75%) and BSCL2 gene mutations (25%) (p > 0.05). No patient needed hospitalization or died. Conclusions We described a high prevalence of SARS-CoV-2 infection in CGL patients with a good outcome in all of them. These findings suggest that at least young CGL patients infected by SARS-COV-2 are not at higher risk of poor outcome, despite known severe metabolic comorbidities.


Diabetology & Metabolic Syndrome
March 2020, the authorities of the state of Ceará, located in northeast region of Brazil, established a series of restrictions to reduce the spread of the infection. These measures culminated in an almost complete blockade of the state [4]. In June 2020, Brazil was the second most affected country around the world [5].
Congenital generalized lipodystrophy (CGL) is a rare disease caused by autosomal recessive inheritance [22]. There are approximately 500 patients described worldwide, with more than 100 cases described in Brazil [23][24][25][26][27]. CGL patients have total or almost total loss of body fat. This abnormality causes ectopic fat accumulation in organs, such as the liver and muscles. Patients evolve with hypertriglyceridemia, severe insulin resistance and poorly controlled diabetes, that usually require extremely high doses of insulin [28].
CGL patients have a greater predisposition to severe infectious diseases which is one of the main causes of mortality among this population [29]. There is no documentation in the literature of cases of SARS-CoV-2 in CGL patients.
Our study aims to evaluate the prevalence of SARS-CoV-2 infection in CGL patients, and the association of their clinical and metabolic characteristics and outcomes.

Study design and participants
This is a cross-sectional study carried out between July and October 2020. Patients with CGL diagnosis enrolled at the Endocrinology Service at the University Hospital Walter Cantídio were invited by telephone to attend for clinical reassessment and participation in the study. This hospital, which is head office of the Brazilian Group for the Study of Inherited and Acquired Lipodystrophies (BRAZLIPO), is a reference on the care lipodystrophy patients. The main inclusion criteria was the CGL clinical diagnosis (total or near total absence of body fat associated with muscle hypertrophy present at birth or soon thereafter, phlebomegaly, acromegaloid facies, increased liver volume, insulin resistance and hypertriglyceridemia) [30]. Currently, 24 CGL patients are followed up. Those who couldn't attend to regular follow up were excluded. The present study was approved by the ethics committee of University Hospital Walter Cantídio (protocol number: 4.255.817). Written informed consent was obtained from all adult participants and written parental consent was obtained for all participants aged under 18.

Data collection
The patients attended an interview and answered a structured questionnaire with clinical and sociodemographic data, smoking habit, presence of DM, dyslipidemia or other CGL comorbidities, and medications in use. They also were asked about the presence of respiratory or other flu-like symptoms and/or need of hospitalization in the last three months.
Blood glucose, cholesterol, and triglyceride levels were obtained after overnight fast and determined according to standard methods using automated equipment. Glycohemoglobin A1c (A1c) values were determined by ion exchange high-performance liquid chromatography (HPLC).
Blood samples for detection of anti-SARS-CoV-2 antibodies, including IgG (Elecsys ® Anti-SARS-CoV-2-Roche Diagnostics) were collected in all patients. Elecsys ® Anti-SARS-CoV-2-Roche Diagnostics is immunoassay for the in-vitro qualitative detection of COVID antibodies in human serum and plasma and has 99.81% specificity [31].
The patients also were referred to a private room by a trained health professional with personal protective equipment to collect the nasopharyngeal swab. Presence of viral RNA was determined in all samples by specific real-time polymerase chain reaction (RT-PCR) assay through the Berlin Protocol-Molecular Kit SARS-CoV2 Biomanguinhos (E / RP). Detectable result indicated the presence of the two RNA regions of the SARS-CoV-2 virus [32].

Statistical analysis
In the numerical variables, the data were presented in median (minimum-maximum). In categorical variables, data were exposed in terms of frequency and prevalence rate. Mann-Whitney U test was used to analyze the characteristics of the participants, verifying the non-adherence of the data to the Gaussian distribution. In investigating the association between categorical variables, we used Pearson's chi-square test and Fisher's exact test. A significance level of 5% was adopted. Statistical analyzes were performed using the statistical program JAMOVI and Microsoft Excel 2016.

Sample description
Of the 24 patients under follow-up, 22 CGL patients from 17 different families were included. Thirteen subjects (59%) were women. Median age was 20 years, ranging from 1 to 42 years. Almost all families lived in the state of Ceará. Eight patients (36%) came from the capital (Fortaleza) or metropolitan region and fourteen individuals (64%) from other cities of Ceará. Only one patient was from state of Rio Grande do Norte. No adult referred alcohol ingestion and only one had smoking habit.
Regarding comorbidities, 21 (95%) patients presented hypertriglyceridemia. Two of these (9.5%) were in fibrate use. Two patients (9%) had hypertension. Eighteen patients (81.8%) had DM with a disease duration of 10 years (0.1-21 years). All diabetics used metformin, except one patient because she was on dialysis. Twelve subjects (75%) were on insulin therapy, with a daily average dose of 1.63 IU/kg (0.4-7.25). Three non-diabetics were using metformin due to insulin resistance.
Clinical and genotypic characteristics of the 22 CGL patients are described in the Table 1.
In this group all adults presented symptoms like fever, malaise, nausea, diarrhea or chest pain in the last 3 months before the visit, and all asymptomatic patients were children. We also observed symptoms among SARS-COV-2-patients without difference between groups (Table 2).
At the time of our evaluation, most patients in both SARS-COV-2+ and SARS-COV-2-had inadequate metabolic control, with no difference between groups ( Table 2).
When comparing only CGL patients under 18 years there was no difference in genotype, clinical or metabolic characteristics between SARS-COV-2+ and SARS-COV-2− subjects (p > 0.05).

Discussion
This is the first study to evaluate the prevalence of SARS-CoV-2 infection in CGL patients and the relationship with its clinical and metabolic profile during the SARS-CoV-2 pandemic.
In the present study, we describe a high prevalence of SARS-CoV-2 infection in CGL patients. This was observed in a predominantly young group, ranging from 1 to 37 years (median 13.5 years). A half of the subjects lived in the capital and metropolitan region of Ceará. Most patients had hypertriglyceridemia and DM, and 72.7% harbor AGPAT2 gene mutations (Subtype 1 CGL).
At the time of our study, 2,65,680 COVID-19 cases had been confirmed in Ceará (Brazil) and the SARS-CoV-2 seroprevalence in Fortaleza (Ceará) was 13.1% [34]. The peak of COVID-19 cases in Ceará occurred between May and June 2020 and the research was carried on later when the incidence was lower. In this context, a higher detection of reactive serologic tests in CGL patients (36.4%) and a lower detection on RT-PCR test was plausible. CDC recommended serologic tests for SARS-CoV-2 infection as an important tool for epidemiologic studies. Unlike RT-PCR or antigen methods that detect acutely infected persons, antibody tests is useful to determine a previous SARS-CoV-2 infection even in a asymptomatic person [35].
Serious forms of COVID-19 have been associated with advanced age and several comorbidities, such as DM, hypertension, obesity, and cardiovascular disease. Such conditions tend to be interrelated in a network of causality that impair their proper understanding [36,37]. CGL is associated with severe metabolic complications, poorly controlled DM that required extremely high doses of insulin [28]. Thus, more serious COVID-19 manifestations would be expected in our series. Several mechanisms have been suggested to explain the increased severity of COVID-19 in DM patients, such as impaired glucose control, diabetes-related immune dysfunction by decreasing viral clearance, reduced neutrophil chemotaxis and presence of monocytes that express higher levels of pro-inflammatory cytokines. Another rationale is the presence of concomitant comorbidities with higher expression of Angiotensin Converting Enzyme 2 (ACE2). The viral binding of SARS-CoV-2 with ACE2 receptors may cause its downregulation, resulting in dysfunction of the renin angiotensin aldosterone system (RAAS), with uncontrolled activity of angiotensin II, which can contribute to acute lung injury [38][39][40]. Despite contradictory evidence, the presence of DM seems to be associated with increased tissue expression of ACE2 [41]. In our series most CGL patients had DM (81%) but none of them presented a poor outcome, even SARS-CoV-2+ patients.
One hypothesis for the asymptomatic disease or mild presentation of the infection found in our group would be the reduction of body fat in these patients, a primary reservoir of SARS-CoV-2 viruses. Obesity has been related to a worse outcome and prognosis of COVID-19.  Adipose tissue demonstrates high expression of ACE2 that SARS-CoV-2 need to enter host cells and which makes adipose tissue a prime reservoir for SARS-CoV-2 viruses, thus increasing the integral viral load [42]. CGL patients have reduced leptin levels, which act in the modulation of the inflammatory response, proliferation and function of T lymphocytes, phagocytosis by macrophages and neutrophil chemotaxis. Thus, leptin deficient patients are more susceptible to infectious diseases [43]. Therefore, it would be expected that these individuals had a higher prevalence of more severe forms of COVID-19 and a high risk of death. However, the predominantly young age of our population may balance these findings. Systematic review of COVID-19 in children showed milder cases and a better prognosis than in adults [44]. One hypothesis would be that the immune system in children can be protected against SARS-CoV-2 because ACE2 enzyme is less mature at a younger age [44,45].
Our patients have the two most common CGL forms, Subtype 1 (mutations in AGPAT2 gene) and Subtype 2 (mutations in BSCL2 gene). AGPAT2 gene is involved in lipogenesis, while BSCL2 is involved in the maturation of preadipocytes and adipocytes. Subtype 2 is considered a more severe form, however there was no difference in prevalence or outcome of SARS-CoV-2 infection in our patients between these subtypes [46][47][48].
COVID-19 infected patients with poor outcome have been associated with lower levels of total cholesterol, HDL and LDL performed during the infectious period [49,50]. In addition to ACE2 receptor, lipoprotein receptors could be another route of entry of SARS-CoV-2 into the host cell. The HDL cholesterol has been inversely correlated with the severity of the disease and was suggested as a potential prognostic blood biomarker [49][50][51][52]. HDL can facilitate a possible route of SARS-CoV-2 entry into the host cell via the SR-B1 receptor. Pretreatment of HEK-293T cells with a potent SR-B1 antagonist ITX5061 strongly inhibited the entry of SARS-CoV-2 into host cells [53]. In our study, levels of total cholesterol, HDL and LDL were not associated with the outcome since almost all patients were not assessed in the acute phase of COVID-19.
This study had some limitations. The retrospective character of the evaluation was based primarily on the    serological diagnosis although the predictive value of the assay used would be up to 95%. We also did not assess the characteristics of exposure, once the objective of the study was not to determine the risk of sickness, but the prevalence of SARS-COV-2 infection in CGL patients, their clinical and metabolic characteristics and their association with outcomes. We expected an increased risk of COVID-19 contamination once the lockdown was not carried out all over the period of the study, and we have a young and economically active population. However, a possible explanation is that most of these patients were jobless.
Also, the small number of patients evaluated may cause some bias. Nonetheless, this is one of the largest series of this rare disease in Brazil and the first study in this field.

Conclusion
In the present study, we described for the first time a high prevalence of SARS-CoV-2 infection in CGL patients with a good outcome in all of them. These findings suggest that at least young CGL patients infected by SARS-COV-2 are not at higher risk of poor outcome, despite of known severe metabolic comorbidities.