Clinical Characteristics of Systemic Lupus Erythematosus in Caucasians and Latin American Hispanics: Data from a Single Tertiary Center

Background Different studies report that systemic lupus erythematosus (SLE) tends to have a more aggressive course in Hispanic patients. In this study, we analysed epidemiologic, clinical, and laboratory characteristics in a cohort of Hispanic and Caucasian lupus patients in the context of Italian health service, which provides free access to care to all citizens, thus mitigating the impact of socioeconomic factors that negatively influence the course of the disease in ethnic minorities. Methods This single-center retrospective study was conducted at the San Martino Hospital “Lupus Clinic” in Genoa, Italy. Patients ≥18 years with a confirmed diagnosis of SLE and definite ethnicity (Hispanic or Caucasian) were recruited. Results A total of 126 patients (90 Caucasians and 36 Hispanics) were enrolled. We compared epidemiologic characteristics, clinical features, autoantibodies profile, and treatment options without evidencing any statistically significant difference between the two groups, except for disease duration, which was higher in the Caucasian group (20.4 years versus 14.2 years in the Hispanic group, P=0.002) and SLICC damage index, which was greater in Caucasian patients (2.11 versus 1.88 in Hispanics, P=0.037), but this difference was no longer significant after correction for disease duration (P=0.096). Conclusions In our cohort, Hispanic ethnicity is not associated with worse disease features and outcomes. Therefore, we speculated that socioeconomic factors, in particular, free access to healthcare, might be more relevant in influencing the course of the disease than genetic background.


Introduction
Systemic lupus erythematosus (SLE) is a chronic autoimmune disease resulting from a complex interaction of genetic, epigenetic, and environmental favouring factors.Te natural history of SLE is typically relapsing-remitting with pleomorphic clinical manifestations potentially involving every organ or tissue [1].
Several studies have shown that some sociodemographic predictors such as ethnicity, gender, age, income, education, and access to healthcare are important variables associated with the epidemiology and the outcome of SLE [2,3].In reference to ethnicity, incidence and prevalence rates are consistently higher among those of African, Hispanic, or Asian descent across studies from diferent countries.Notably, in these same ethnic groups, SLE seems to be more severe with higher disease activity and more damage accrual than in Caucasians [4][5][6].
Immigration can be considered a relatively new phenomenon in Italy, a country historically characterized by emigration.In the last decades, Genoa, similar to other Italian cities, has experienced a signifcant increase in immigration, especially from Latin America.According to the Italian National Institute of Statistics (ISTAT), foreign citizens (about 75000) account for 9.1% of the total population of the metropolitan area of Genoa [7].Te most represented foreign community in Genoa are Ecuadorians (around 20% of foreigners), and also other Latin American populations are present in the territory.Globally, Latin American/Hispanic is the most represented non-White ethnicity in our territory [7].
Several studies have pointed out that Hispanics seem to present a higher incidence of SLE, a more severe course of the disease, and poorer outcomes as compared to Caucasians.Te LUMINA study was based on a prospective cohort specifcally established to analyse minority SLE patients in the United States [8][9][10].In this cohort, Texas Hispanics showed greater disease activity, accelerated damage accrual, increased frequency and severity of renal disease, and diminished survival than Caucasians and Puerto Rican Hispanics.Interestingly, Hispanics from Puerto Rico not only have a lesser proportion of ancestral Native American genes but also beneft from better socioeconomic conditions than Texas Hispanics [9,[11][12][13].In the GLADEL study, Afro-Latin Americans and Mestizos had more severe SLE and develop renal disease at a higher proportion than Whites [14,15].Analysing disease features and outcomes in US SLE patients of Hispanic origin and their Mestizo counterparts in Latin America from the LUMINA and GLADEL cohorts, it emerged that US Hispanics seemed to have a poorer prognosis, despite a comparable genetic background, thus suggesting that socioeconomic factors may account for these discrepancies [16].Similarly, Mexican patients living in Mexico had lower levels of disease activity and damage with respect to Texan Hispanics from the LUMINA cohort, of whom 95% are of Mexican descent [17].Calvo-Alén et al. reported that Hispanics with a strong Amerindian background (US Hispano-Americans) have a more serious disease than that observed in Hispanics from Northern Spain, suggesting that the genetic pool and also socioeconomic diferences between these two Hispanic subgroups probably explain for these fndings [10].Data from the RELESSER registry confrm this observation indicating that Latin-American Hispanic patients have an increased frequency of lupus nephritis and display higher disease severity than European Caucasians Hispanics [18,19].Te multiethnic study profle showed that renal involvement is more frequent among non-Caucasian patients, and, within them, Hispanic patients seem to be at a higher risk of renal damage [20,21].Te CDC, through fve diferent US cohorts, reports that Hispanics have a higher incidence and prevalence of lupus and are at an increased risk of severe manifestations and early development of lupus nephritis, thrombocytopenia, and antiphospholipid syndrome [22].
Ethnicity also impacts drug response.In the short-term induction therapy for lupus nephritis, mycophenolate mofetil (MMF) and intravenous cyclophosphamide (CYC) have a similar overall efcacy; however, a signifcantly higher response to MMF with respect to CYC was observed in patients of Mestizo descent and individuals of Hispanic origin [23].In the EXPLORER trial on rituximab in nonrenal lupus, Hispanic patients achieved signifcantly higher major clinical responses compared with other groups [24].Unfortunately, limited access to these drugs in several Latin American countries, due primarily to cost issues, or in developed countries, owing to disparities in healthcare access often experienced by racial and ethnic minorities, hinders these patients from accessing the most efective medical care.Tis aspect is clearly exemplifed by the recent GLADEL-Pan-American League of Associations of Rheumatology (PANLAR) recommendations, where not only the efcacy but also the costs and availability of a given medication strongly infuence its usage in clinical practice of specifc geographical context [25].
Overall, the abovementioned investigations indicate that Hispanics tend to have an increased prevalence, disease severity, risk of complications, and a worse outcome compared to Caucasian patients.Despite the consistent literature on the subject, the explanation of these disparities has not been univocally established, and diferences in biologic/genetic background (e.g., higher proportion of ancestral Native American genes) and socioeconomic factors (e.g., unfavourable economic status, education, or access to healthcare) might contribute in a nonmutually exclusive manner.
Te purpose of this study was to analyse epidemiological characteristics, clinical and laboratory features, and treatment options in an Italian cohort of SLE patients of Hispanic and Caucasian ethnicity.Tis study was conducted in the context of the Italian health system which provides universal coverage to all citizens, allowing to mitigate, at least in part, the bias due to socioeconomic disparities in healthcare access that typically impact racial/ethnic minorities.[26], and defnite ethnicity (see below) were recruited at San Martino Hospital Lupus Clinic, a dedicated SLE clinic located in Genoa, Italy, between October 2022 and March 2023.Caucasian patients were defned as those born in Italy with a Caucasian phenotype and without known non-Caucasian ancestors.Hispanic patients were defned as those immigrated from Latin American countries with an American-Indian phenotype with Latin American parents and without known African or European ancestors [18].All Hispanic patients had universal health coverage, as is the case with all Caucasian subjects.

2
Autoimmune Diseases Data were extrapolated from medical records and transferred to a fully deidentifed database for statistical analysis.Collected variables included sociodemographic and clinical characteristics, as well as laboratory characteristics.
Clinical manifestations have been defned according to the EULAR/ACR 2019 defnition system and organized in various clinical domains (hematologic, neuropsychiatric, mucocutaneous, serosal, musculoskeletal, and renal) [26].Te EULAR/ACR 2019 score has been assessed at disease onset and at the time of enrolment in the study [26].Te 2023 EULAR/ACR Classifcation Criteria have been adopted in order to defne the presence of antiphospholipid syndrome (APS) [27].
Disease activity has been calculated by the Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K) [28].Low Disease Activity State (LLDAS) was defned as SLEDAI-2K ≤ 4 without activity in major organ systems or new disease activity, Physician Global Assessment (PGA) ≤1, prednisone dose ≤7.5 mg/day, and/or immunosuppressivebiologic-antimalarial drugs on maintenance dose [29].Accumulation of damage has been ascertained through the Systemic Lupus International Collaborative Clinics/American College of Rheumatology Damage Index (SDI), a score system that records irreversible damage that has occurred since the onset of lupus in 12 diferent systems [30].
Current treatment (at last visit) and previous therapeutic regimens were obtained from medical records.

Statistical Analysis.
Overall comparisons of the clinical and immunologic categorical variables among the two ethnic groups were performed using cross tabulations, and their signifcance was assessed by means of the chi-square/ Fisher's exact test.Regarding continuous variables, the comparison among the two groups was assessed using parametric (Student's t-test) or nonparametric (Mann-Whitney U test) test according to the distribution of the analysed variables and the sample size.For some relevant variables, a multivariate analysis was performed using a logistic regression model in order to estimate the adjusted odds ratio of each variable.Spearman's rho coefcient was used to assess correlations.All statistical analyses were performed with Jamovi [31].

Demographic Factors.
Overall, 126 SLE patients were enrolled; 90 of them were Caucasians (71%) and 36 were Hispanics (29%).Among Hispanics, most of them were from Ecuador (64%) and Peru (19%).Te remaining percentage of Hispanic patients (17%) were from other geographical areas of Latin America.All Hispanics enrolled in this study were frst-generation immigrants.As expected, the majority of patients in both ethnic groups were female, and the mean age at the enrolment was 51.6 and 47.8 years in Caucasians and Hispanics, respectively.Among Hispanic patients, 89% received their SLE diagnosis in Italy, while the remaining percentage were diagnosed in their countries of origin.Te comparison of the two ethnic groups showed no signifcant diferences in the univariate analysis for gender and age.Te demographic variables are detailed in Table 1.
At onset, overt disease (fulflling SLE classifcation criteria) was present in 38.9% and 52.7% of Caucasians and Hispanics, respectively (P � 0.154), whereas the remaining patients of both groups presented initially as undiferentiated connective tissue disease and later developed additional clinical/laboratory manifestations of complete SLE.Te mean ACR/EULAR 2019 score assessed at the time of enrolment (last visit) was similar in the two ethnicities, independently from disease duration.
Te percentage of patients with renal involvement was not signifcantly diferent in Caucasians and Hispanics (62.2% vs. 50.0%,respectively).A kidney biopsy was performed in 54 of the 74 patients who developed renal manifestations (relevant proteinuria and/or increase in the serum creatinine), and histologic LN was confrmed in all samples.Te percentage of eligible patients who underwent renal biopsy was similar in the two ethnic groups (71.4% Caucasians vs. 77.8%Hispanics, P � 0.764).Te most common forms of LN observed in our cohort were focal and difuse proliferative glomerulonephritis (classes III and IV) Autoimmune Diseases with no signifcant diferences between the two ethnic groups (77.5% Caucasians vs. 64.3%Hispanics, P � 0.479).
As concern immunological markers of disease, no signifcant diferences in the prevalence of anti-dsDNA and anti-Sm antibodies were found across the two ethnic groups (Table 2).Complement reduction (C3 and/or C4 decreased below the lower laboratory limit) was found in the majority of patients in both populations (Table 2) and associated with a higher prevalence of hematologic manifestations (OR, 5.51, 95% CI, 2.16-14.1,P < 0.001).Caucasians and Hispanics tested positive for antiphospholipid antibodies in at least two determinations in about one-third of patients of each group.Of those, 42.9% Caucasians and 46.2% Hispanics developed clinical manifestations (thrombotic and/or obstetric events), thus fulflling the criteria for APS with no diference among the two ethnicities (Table 2).
As concern disease activity, LLDAS was assessed at the last clinical evaluation, with no signifcant diferences between Caucasians and Hispanics (respectively, 88.9% vs. 80.6%, P � 0.252).Te SDI scores, calculated at the time of enrolment, were signifcantly higher in Caucasian than in Hispanic patients (2.11 ± 1.91 and 1.23 ± 1.31, respectively; P � 0.037) in the univariate analysis.As expected, we found a correlation between disease duration and SDI score (rho � 0.227, P � 0.023).Accordingly, the diference in the SDI score between Caucasians and Hispanics lost statistical signifcance after adjustment for disease duration (P � 0.096 in the multivariate analysis).4 Autoimmune Diseases

Terapy.
During the course of the disease, all patients received diferent treatment associations including corticosteroids (CCs) and/or hydroxychloroquine (HCQ) and/or disease-modifying antirheumatic agents (conventional synthetic and/or biologic DMARDs).A comparison of the two ethnic groups showed no signifcant diferences in treatment strategies employed (Table 3).Regarding treatments evaluated during the last visit, most of the patients were taking low dose of CCs, defned as an equivalent dose of prednisone ≤7.5 mg/day, with no diference in the two ethnicities.Similarly, the percentage of Caucasian and Hispanic patients on antimalarials and/or DMARDs was comparable in the two groups (Table 3).

Discussion
In this study, we analysed demographics, clinical, and laboratory features of a cohort of patients afected by SLE, comparing data from Caucasians and Hispanics.Regarding demographic factors, the two groups were homogeneous in terms of gender and age.As expected, the majority of Hispanic patients were from Ecuador, the most represented foreign community in Genoa [7].In the current literature, Hispanic patients originate from many diferent countries (e.g., multicentric studies) or from a single specifc country (nation-based single-center studies) or country of origin is simply not reported.Since it is now clear that not all "Hispanics" are the same, our data, derived mainly from Ecuadorian patients, might not be directly compared with other studies [32].Importantly, Ecuador's population shows high degrees of Native American genetic ancestry [33], a genetic background evoked as an independent factor in SLE onset and severity [2,[34][35][36]; thus, it might be considered highly representative of the Hispanic patient population.
With regard to clinical characteristics, disease duration was signifcantly longer in Caucasian than in Hispanic patients (21 vs. 14 years, respectively).Tis observation may be explained by the fact that immigration is a recent phenomenon in our region and, consequently, our cohort of Hispanic patients includes subjects with a more recent diagnosis of SLE compared to the Caucasian cohort.
Te cumulative incidence of clinical manifestations related to SLE, including articular, hematologic, renal, and neuropsychiatric disorders, and serositis as well as mucocutaneous lupus, was evaluated in Hispanic and Caucasian patients without evidencing signifcant diferences in the two ethnicities, even after adjustment for disease duration.
Serositis and cytopenias (in particular, thrombocytopenia and leukopenia) are reported in the literature as more frequent in Hispanics than in Caucasians, although to a lesser extent than renal damage [6,12,20].In our study, both populations exhibit a similar prevalence of serosal and hematologic complications, with the latter encompassing single manifestations such as leukopenia, thrombocytopenia, or autoimmune hemolytic anaemia.
In relation to immunologic laboratory features, such as anti-dsDNA antibodies, anti-Sm antibodies, and complement reduction, no diferences emerged comparing the two ethnicities.In agreement with what was reported in the literature, complement reduction was correlated with a higher prevalence of hematologic manifestations [37,38].
Antiphospholipid syndrome (APS) is characterized by vascular thrombosis, gestational morbidity, and/or nonthrombotic manifestations in the presence of antiphospholipid antibodies [39].According to the literature, antiphospholipid antibodies can be detected in up to 40% of SLE patients, but only one-third of them will develop clinical manifestations with no clear diferences among various races/ethnicities [40].Consistently, in our cohort, we fnd a comparable prevalence of aPL positivity and APS, with no diferences between Caucasian and Hispanic SLE patients.
Several authors have reported that Hispanics with SLE experience higher disease activity and disease-related damage than Caucasians [9,11,12,18,19,41].Regarding disease activity, we assessed LLDAS at the last clinical evaluation, fnding no signifcant diferences between the two ethnic groups.Overall, we found that a large proportion of patients in both groups achieved LLDAS at the time of this study.Tis observation may be explained, at least partially, by the long disease duration characterizing our cohort (globally 19.5 years) compared to patients' populations analysed in other studies.Indeed, it is well known that SLE tends to be more active in the frst years after disease onset, while patients with long disease duration show a higher prevalence of low disease activity and remission [42][43][44].Damage accrual related to SLE, assessed by the SDI score, was higher in Caucasians, but this diference disappeared after adjustment for disease duration.Tis observation is not surprising since the damage is correlated to disease duration which, in our cohort, was signifcantly higher in Caucasians [45].
Both Hispanics and Caucasians received comparable therapeutic options during the course of the disease.As widely recommended, the majority of patients were treated with hydroxychloroquine, a drug characterized by multiple positive efects in SLE with an optimal safety profle [46].In our cohort, a substantial proportion of patients either discontinued or were on low-dose steroid therapy, to achieve and maintain this result, and the use of steroid-sparing agents, such as immunomodulating-immunosuppressive drugs, is essential [46].In many countries, the primary constraint in prescribing certain immunosuppressants is related to their cost.In our cohort, due to the policy of the Italian healthcare system, the use of immunosuppressive drugs relies primarily on medical judgement, independent of patients' economic conditions, thus granting the optimal treatment strategy for all patients.
As mentioned, several studies have reported that SLE tends to present a more severe course in Hispanics compared to Caucasians.Unfortunately, the frequent association between non-Caucasian ethnicities and unfavourable socioeconomic status makes biologic and environmental Autoimmune Diseases contributions difcult to disentangle.Te existing data in the literature predominantly stem from studies conducted in countries where migration is a long-standing phenomenon.Consequently, these fndings may not readily translate to countries with a more recent history of immigration, such as Italy.In addition, the majority of these studies are from the American continent where the socioeconomic status of minorities, including access to healthcare, may signifcantly difer compared to European countries, especially Italy, where the healthcare system provides universal coverage and a largely free access to care to all citizens.To the best of our knowledge, our study is the frst investigation conducted in Italy on this subject, and the second one in Europe alongside the Spanish registry RELESSER [18,19].
Tere are limitations in this study, which necessarily inform the interpretation of our results.Firstly, the power of our study is limited by the retrospective design, monocentric nature, and the relatively small sample size.Te low number of patients might have led to a type II statistical error, and data derived from a single center may limit their generalizability and external validity.Furthermore, the crosssectional and retrospective design inherently limits our ability to establish causal relationships and introduces potential bias due to the heterogeneity of the studied populations, particularly regarding variations in disease duration.Tis variability may indeed infuence the expression of time-related clinical manifestations and consequently afect the observed results.Nevertheless, the monocentric structure of our study might potentially provide more homogeneous and uniform data as compared to multicentric cohorts.As noted above, all Hispanic SLE patients enrolled in this study were frst-generation immigrants.Tis may represent a limitation of this study related to a bias known as "healthy migrant efect," a theory postulating that healthy people are more likely to migrate and consequently frst-generation immigrants are healthier than the average person in both the home and host countries [47,48].Te fact that the majority of Hispanic patients have been diagnosed in Italy may be attributed to this phenomenon.Te healthy migrant efect could further account for the observation that our cohort of Hispanic patients have a milder-than-expected disease course, similar to that observed in Caucasians, implying that patients with earlyonset SLE and/or more severe disease might have been unable to migrate.Interestingly, data from GLADEL have shown that older age at diagnosis is associated with a less severe course of disease [3,49].An additional limitation of our study is the omission of potentially relevant variables such as comorbidities or socioeconomical aspects (i.e., yearly income, years of education, home ownership, and home density).However, according to the literature, these elements tend to be unfavourable in Hispanic immigrants and are evoked as an explanation for ethnicity-related diferences among SLE patients, diferences that were not observed in our study [32,50].
Hispanic ethnicity is considered an independent risk factor for severe disease and adverse outcomes in SLE, and many authors have emphasized the important, sometimes prominent, role of unfavourable socioeconomic factors to explain these observations [13,16,[51][52][53].In many studies, particularly those conducted in developed countries, Hispanic patients represent an ethnic minority that often sufers from poor socioeconomic conditions, which have a negative impact on access to healthcare.In contrast to these reports, our cohort of patients beneft from free access to a specialized care center that provides comprehensive care and all available treatment options for SLE, independently from their socioeconomic situations.
In conclusion, we observed similar disease features in Hispanics and Caucasians, suggesting that socioeconomic variables, specifcally healthcare access, might be more infuential on disease course than biologic and genetic background linked to ethnicity.Validation and replication of our fndings in larger studies conducted in similar public health settings are crucial for improving our understanding of the role of ethnicity in SLE.

Data Availability
Te datasets generated and/or analysed during the current study are not publicly available due to individual data privacy but may be available from the corresponding author on reasonable request, in compliance with ethical and legal standards.

Table 2 :
Clinical and laboratory characteristics.Systemic Lupus Erythematosus International Collaborating Clinics.Antiphospholipid positivity was defned as the presence of any of the following: anti-cardiolipin (IgM and/or IgG) and/or anti-β2-glycoprotein (IgM and/or IgG) antibodies and/or presence of lupus anticoagulant (LAC).
Categorical parameters are given as n (%).Continuous variables are given as mean ± SD.P values < 0.05 in bold.ACR/EULAR, American College of Rheumatology-European Alliance of Associations for Rheumatology; dsDNA, double-stranded DNA; LLDAS, lupus low disease activity state; SLICC,

Table 3 :
Previous and current treatments.