Introduction

Systemic lupus erythematosus (SLE) is a multifaceted autoimmune disease driven by dysregulation of the immune system, particularly through the abnormal activation of T and B lymphocytes [1]. That leads to the production of autoantibodies and immune complexes that trigger inflammation, resulting in tissue and organ damage. The disease predominantly affects women of childbearing age [2]. Despite advancements in treatment, mortality rates in SLE remain notably high, particularly among older patients, with infections and cardiovascular complications being the primary causes of death [3, 4]. Thus, SLE diagnosed in patients older than 50 (late-onset SLE – Lo-SLE) presents distinct clinical challenges, including diagnostic delays, a higher multimorbidity burden, and the need for tailored therapeutic strategies to minimize long-term complications associated with ageing [5].

The pathogenesis of Lo-SLE, including sex hormone levels, differs from that of early-onset SLE (Eo-SLE, diagnosed before age 50) [4, 6, 7]. Late-onset SLE represents about 2–20% of all SLE cases, with a lower female-to-male ratio than Eo-SLE [4, 7]. Patients with Lo-SLE typically experience a more insidious disease progression, encounter the classic symptoms less frequently, and exhibit lower disease activity than Eo-SLE [8]. Autoantibody profiles also differ regarding the age of SLE onset [4]. For example, Lo-SLE exhibits lower frequencies of anti-double-stranded DNA (dsDNA), anti-nucleosome, anti-Smith (Sm), and anti-ribonucleoprotein (RNP) antibodies compared to Eo-SLE [7]. Inversely, rheumatoid factor is found more frequently in Lo-SLE [4]. Moreover, older patients often had medical comorbidities, and their initial SLE signs and symptoms could be misinterpreted. Consequently, the disease onset might not be diagnosed promptly [4, 9]. The course of Lo-SLE presents with less organ involvement, but certain manifestations such as arthritis, pleuritis, pericarditis, myositis, pulmonary fibrosis, and associated Sjögren’s disease (SjD) have been reported more frequently [4, 7]. Additionally, Lo-SLE patients may have a higher incidence of thromboembolism with positive laboratory test results for antiphospholipid syndrome (APS) [7]. However, findings on these manifestations remain inconsistent across studies [4, 9]. In some of them, organ damage appears to be more prevalent in the Lo-SLE group, which could be related to iatrogenic factors such as osteoporosis and age-related health issues. Therefore, the question arises about the perception of Lo-SLE as a milder SLE form [7].

In the Polish population, there is still a lack of comprehensive studies exploring clinical and laboratory differences between Lo-SLE and Eo-SLE patients [9]. Therefore, we decided to analyze the clinical and laboratory characteristics of Lo-SLE in a large cohort of SLE patients treated at our hospital and compare them to those of Eo-SLE. By identifying distinct patterns and potential risk factors, we seek to contribute to improved personalized management for SLE patients across different age groups.

Material and methods

Results

Discussion

The present study reveals clinical and laboratory characteristics of a large cohort of SLE patients from a single center, focusing on those diagnosed before the age of 50 compared to those diagnosed at 50 years or older. Our findings revealed that the age of onset significantly impacts the clinical presentation and progression of the disease. These distinctions offer valuable insights into overall prognosis, including the likelihood of organ-related complications and mortality. Our study describes one of the largest Polish cohorts of SLE patients in the literature, comparing various aspects of the disease based on the age of SLE onset.

In general, most SLE patients had early-onset disease, consistent with the current literature on this topic [5]. Late-onset SLE is used in the literature when the disease is diagnosed after the age of 50–65 years, although most studies use 50 years as the threshold [5, 6]. Interestingly, in our cohort of Eo-SLE and Lo-SLE patients, we noted a high prevalence of the female sex in both groups but with a slightly lower female-to-male ratio in the Lo-SLE group. According to the literature, female-to-male ratios in Eo-SLE and Lo-SLE differ, with lower values in Lo-SLE patients [5]. Interestingly, a Spanish multicenter registry cohort (RELESSER group) reported an almost two-fold lower female-to-male ratio in the Lo-SLE group [6]. Likewise, a study on a Polish cohort noted that the prevalence of SLE in males was higher in Lo-SLE compared to younger patients [9]. On the other hand, Aljohaani et al. [8] found similar percentages, similar to our study, of females in late-onset and younger-onset SLE groups (84.9% vs. 86.4%), potentially due to differences in inclusion criteria for the younger-onset group. These distinctions in sex ratios in SLE highlight the potential role of different underlying factors, such as genetic and hormonal influences, in the age-related onset of the disease.

In both groups, hematological and joint symptoms were the most common, which is consistent with previous studies on large SLE cohorts [6, 13]. However, the Eo-SLE cohort exhibited a higher prevalence of a broader range of clinical manifestations, including general symptoms (fever, lymphadenopathy), hematological (leukopenia), skin manifestations (alopecia, urticaria, lupus malar rash), and kidney and central nervous system involvement. This observation aligns with a report published by Riveros Frutos et al. [6], who found that hematological, renal, neuropsychiatric, and skin changes were the most frequent in the Eo-SLE group. Also in line with those results are data presented by Bindroo et al. [14], with alopecia, RP, and lymphadenopathy reported more frequently in Eo-SLE than Lo-SLE. Next, a meta-analysis by Medlin et al. [15] demonstrated, in contrast to our results, that serositis and pleuritis can also be found more frequently in the Eo-SLE group. On the other hand, interstitial lung disease was more frequently observed in Lo-SLE in our data and also in other studies [14, 16]. These findings suggest the possibility of overlap syndromes, such as SLE/SjD, which may contribute to lung involvement in these patients [17]. The same association is further supported by the higher prevalence of anti-SSB antibodies, characteristic of SjD, in Lo-SLE, which was published in the literature [17] and also documented in our data. Furthermore, Catoggio et al. [16] found sicca symptoms more frequent in older SLE patients, which also suggests suspicion of SjD. Unfortunately, data on sicca symptoms or overlapping autoimmune diseases such as SjD were not available for our cohort.

In our data, neuropsychiatric SLE was also less frequent in Lo-SLE than in Eo-SLE patients. This observation is in line with previously published data by Pamuk et al. [18], where symptoms of neuropsychiatric SLE were 1.41-fold more common in younger patients. In contrast, Jeleniewicz et al. [9] did not find differences in the frequency of neuropsychiatric SLE symptoms regarding patient age, likely attributable to the limited sample size of the studied groups.

We also found that Lo-SLE was characterized by a higher prevalence of medical comorbidities and malignancies, which may be linked to the age-related nature of these conditions and has been reported previously by other authors [6, 19, 20]. Moreover, similarly to our findings, specific comorbidities, such as arterial hypertension or dyslipidemia, may also influence the outcome of SLE [6]. Interestingly and consistent with previous data, arterial thromboembolic episodes were observed more frequently in Lo-SLE, specifically myocardial infarction [21]. Additionally, in our stratified analysis, thrombotic events were significantly more frequent in aPLA-positive patients with Eo-SLE, supporting the interpretation of these events as clinical manifestations of APS. In contrast, in Lo-SLE, thromboses occurred at similar rates regardless of aPLA status (positive vs. negative), suggesting they may reflect comorbidities or age-related vascular risk rather than APS.

In our data, the Lo-SLE group was characterized by a lower frequency of positive results for lupus anticoagulant and anti-cardiolipin antibodies in the IgG class. Moreover, SLE-characteristic ANA types were more common in Eo-SLE. This observation largely agrees with data presented in the review by Piga et al. [7], who highlighted a lower prevalence of anti-dsDNA, anti-nucleosome, anti-Sm, anti-RNP antibodies, and lupus anticoagulant positivity in Lo-SLE. Furthermore, this serological profile could influence or reflect the disease course, including treatment mode [7]. We observed a similar high prevalence of anti-dsDNA antibodies in Eo-SLE and Lo-SLE groups as measured by the Crithidia luciliae assay or immunoblotting. However, when these antibodies were assessed by the Crithidia luciliae assay characterized by higher sensitivity and specificity [22], their prevalence in patients from the Eo-SLE group increased. In contrast to our findings, Padovan et al. [23] reported a higher prevalence of anti-dsDNA antibodies in Lo-SLE. However, similar to our results, anti-SSB, and in some cases, anti-SSA antibodies, were more frequently present in Lo-SLE [7]. Another interesting finding of our study concerns the association between anti-RNP antibodies and clinical signs of Eo-SLE but not Lo-SLE. In the former group only, the presence of anti-RNP antibodies was related to the higher frequency of constitutional symptoms, serositis, hematological abnormalities, and RP. This observation aligns with a report by Riveros Frutos et al. [6], in which anti-RNP antibodies were significantly more common in patients with Eo-SLE, particularly in those with systemic symptoms such as fever, hematologic manifestations, serositis, and RP. According to them, anti-RNP antibodies were less common in Lo-SLE patients and were not associated with any specific clinical features [6].

Regarding therapy mode, both subgroups were treated the most frequently with GCs, chloroquine, or hydroxychloroquine, which aligns with EULAR recommendations from 2019 [24]. Furthermore, considering that most Eo-SLE patients presented with more signs of the disease, including kidney involvement and neuropsychiatric manifestations, they required frequent immunosuppressive drugs [6]. Adverse effects of that aggressive treatment, such as the increased risk of infections and organ toxicity, necessitate careful use in older patients [25]. Likewise, drug interactions underscore the importance of individualized pharmacotherapy in optimizing safety and efficacy in Lo-SLE [26].

Finally, we observed a higher mortality rate in Lo-SLE compared to Eo-SLE, which mirrors other data on SLE [6, 27]. In the study by Riveros Frutos et al. [6], mortality was higher in the Lo-SLE group when compared to Eo-SLE patients (14.3% vs. 4.7%, p < 0.001). A study by Viveiros et al. [27] documented that very late-onset SLE patients (≥ 60 years at the time of diagnosis) had the lowest survival rate compared to younger SLE groups, including those with SLE diagnosis between 50 and 59 years.

Study limitations

The retrospective design of this study introduces potential biases in data collection and patient selection, which could affect the reliability of the findings. Additionally, the single-center nature of the study may restrict the applicability of the results to a broader population. Essential lifestyle factors, such as smoking, alcohol consumption, physical activity, and dietary habits, were not accounted for in the database. Furthermore, key laboratory and clinical data, including inflammatory markers, lupus disease activity indices, and sicca symptoms, were unavailable for this cohort. The higher clinical incidence of thrombosis in Lo-SLE, despite lower frequencies of antiphospholipid antibodies, suggests a potential non-immune-mediated etiology or low test sensitivity. Therefore, these results should be interpreted cautiously. Moreover, due to the retrospective nature of the study, the use of disease activity indices, such as SLEDAI, was not feasible. This limitation stemmed from the absence of specific data required for score calculation at potential control time points and the variability in patient follow-up duration. A further limitation of the study is the lack of analysis regarding the impact of specific treatments on disease outcomes. Finally, some observed associations may be coincidental and not indicate causal relationships.

Conclusions

Late-onset SLE patients exhibited distinct clinical characteristics compared to Eo-SLE, with fewer disease-related manifestations, suggesting that these individuals may require less intensive immunosuppressive therapy. Notably, interstitial lung disease was more prevalent among older patients, highlighting a need for closer monitoring of pulmonary complications in this group. Furthermore, Lo-SLE was associated with a higher burden of medical comorbidities, malignancies, and arterial thromboembolic events, emphasizing the importance of a comprehensive approach to managing these patients. The ANA profile in Lo-SLE also differed, showing a reduced variety and frequency of autoantibodies, apart from anti-SSB antibodies, which may indicate a greater likelihood of overlap with SjD. Our findings underscore the need for further research to validate this potential association and optimize care for Lo-SLE patients.