Introduction
Systemic lupus erythematosus (SLE) is a chronic autoimmune disease with diverse clinical manifestations that can affect multiple organ systems [1]. Its diagnosis and management are related to the immunological profile, particularly the presence of specific antinuclear antibodies (ANA) [2].
Although anti-double-stranded DNA (dsDNA) and anti-Sm antibodies are well-established markers specific to SLE and are included in the classification criteria, other antibodies, such as anti-nucleosomes and anti-ribosomal P (anti-P), are still being evaluated for their diagnostic and prognostic significance [3, 4].
Anti-ribosomal antibodies, which are highly specific for SLE, do not have the same diagnostic value as anti-dsDNA or anti-Sm [4]. However, some studies suggest that anti-P may be associated with certain SLE manifestations, such as neuropsychiatric SLE (NPSLE), lupus nephritis (LN), and hepatic involvement [5].
There is a lack of comprehensive data on the prevalence and clinical implications of anti-P antibodies in patients with SLE from African and Arab populations, with only a few studies from Tunisia (2 studies) and Egypt (1 study) providing limited insights [6–8]. Notably, these studies presented controversies that should be clarified in further investigations, and did not explore all specific SLE manifestations in relation to anti-P antibodies. To our knowledge, this study is the first to specifically investigate the clinical implications of isolated anti-P antibodies (anti-P without anti-dsDNA or anti-Sm antibodies) in SLE patients. Thus, this study aimed to compare the clinical manifestations and laboratory findings of SLE patients with and without anti-P antibodies, and to explore the associations between isolated anti-P antibodies and specific SLE features.
Material and methods
Patients
We reviewed the medical records of 75 (7 men, 68 women) SLE patients who were recruited from the departments of internal medicine (n = 40) and rheumatology (n = 35) at the University Hospital of Monastir, Tunisia, between January 2008 and December 2022. All patients fulfilled the updated 1997 American College of Rheumatology (ACR) criteria [9] or Systemic Lupus Erythematosus International Collaborating Clinics (SLICC) criteria for the classification of SLE [10]. The clinical manifestations and blood samples were collected at the time of diagnosis. Neuropsychiatric SLE was defined using the 1999 ACR recommendations for NPSLE syndrome [11]. The diagnosis of LN, which was made at the time of SLE diagnosis, was histologically confirmed by renal puncture biopsy. Systemic lupus erythematosus disease activity was evaluated at the time of diagnosis according to the SLE Disease Score Index 2000 (SLEDAI) [12]. We excluded patients who were likely to have SLE but exhibited some, but not all, ACR 1997 or SLICC 2012 criteria for SLE at the time of evaluation [13]. These patients could be included in the study only if they presented again (throughout the study period) with features fulfilling the classification criteria. We also excluded patients who were positive for anti-dsDNA antibodies by ELISA but negative for Crithidia luciliae immunofluorescence test (CLIF). This is because of the high frequency of false-positive anti-dsDNA results owing to the high sensitivity of ELISA. This is in line with the latest guidelines that recommend the use of CLIF for the identification of anti-dsDNA antibodies [14].
Antinuclear antibody testing
Screening of ANA was performed by an indirect immunofluorescence (IIF) assay using HEp-2 cells (Euroimmun, Lübeck, Germany: ANA cutoff ≥ 1 : 180). Anti-dsDNA antibodies were quantified by ELISA (Biosystems S.A., Barcelona, Spain) and confirmed by IIF assay using CLIF (cutoff > 1 : 20). Anti-nucleosomes, anti-histones, anti-Sm, anti-Sm/RNP, anti-SSA/Ro (Ro60/ Ro52), anti-SSB/La were analyzed by the Euroline ANA-profile (Euroimmun, Lübeck, Germany).
Anti-ribosomal P evaluation
Line blot assay, a semi-quantitative platform capable of detecting multiple autoantibody specificities, is a reliable and validated diagnostic tool [15–17]. The detection of anti-P using this sensitive and specific technique avoids the low sensitivity of IIF for screening these antibodies [18, 19].
Anti-P antibodies were assessed by a specific line blot assay (Euroimmun, Lübeck, Germany) using a purified ribosomal P protein from a native source as an antigen. In this immunoenzymatic semi-quantitative assay, the strips were incubated with 1/100 diluted patient sera. After washing, the strips were incubated with alkaline phosphatase (ALP)-labeled anti-human IgG, and subsequently with 5-bromo-4-chloro-3′-indolylphosphate p-toluidine salt (BCIP)/nitro blue tetrazolium chloride (NBT) substrate for color development. The results were evaluated and interpreted using the EUROLINE Scan software. Line blots are automatically identified, quantified, and registered, and a complete results report is obtained.
Statistical analysis
The comparison between qualitative parameters was performed using the χ2 test or Fisher’s exact test when appropriate. The distribution of quantitative variables was determined using the Kolmogorov-Smirnov test. Normally distributed variables are summarized as mean ± standard deviation, and their means were compared using Student’s t-test. Non-normally distributed variables were summarized as median plus interquartile range, and they were compared using the Mann-Whitney U non-parametric test. All statistical evaluations were performed using SPSS Statistics for Windows (version 21; IBM Corp., USA). A p-value (asymptotic significance [2-sided]) < 0.05 was used to define statistical significance.
Bioethical standards
The study was conducted according to the guidelines of the Declaration of Helsinki. The Ethics Committee of the University Hospital of Monastir, Tunisia confirmed that in this retrospective data analysis no ethical approval is required. Written informed consent was obtained from all patients for anonymous publication of the details of their medical cases.
Results
The mean age at the time of diagnosis of SLE was 39.2 ±16.2 years. Women constituted 90.7% (n = 68) of all the patients, with a sex ratio of men/women = 0.1. In our study, 30 patients (40%) had anti-P. The general features of study subjects are shown in Table I.
Table I
Clinical, biological and therapeutic features and disease activity of the 75 studied systemic lupus erythematosus patients
| Features | Data summary |
|---|---|
| Age [years], mean ±SD | 39.2 ±16.2 |
| Sex ratio (men/women) | 0.1 (7/68) |
| Clinical manifestations [n (%)] | |
| Familial SLE | 5 (6.7) |
| Associated AIDs | 20 (26.7) |
| Cutaneous features | 49 (65.3) |
| Photosensitivity | 33 (44) |
| Pleural effusion | 11 (14.7) |
| Pericarditis | 7 (9.3) |
| Lupus nephritis | 18 (24) |
| Arthritis | 34 (45.3) |
| NPSLE | 25 (33.3) |
| Autoantibodies [n (%)] | |
| ANA | 73 (97.3) |
| Anti-ds DNA | 41 (54.7) |
| Anti-Sm | 19 (25.3) |
| Anti-P | 30 (40) |
| SLEDAI, mean ±SD | 10 ±6.2 |
| Treatment [n (%)] | |
| GCs therapy | 62 (82.7) |
| High doses (> 1 mg/kg) | 17 (22.6) |
| Bolus | 32 (42.6) |
| Antimalarics* [mg/kg] | 65 (86.7) |
| Immunosuppressive | 20 (26.7) |
| Methotrexate [mg/week] | 14 (18.6) |
| Cyclophosphamide [mg/m2] | 14 (18.6) |
| Azathioprine [mg/kg] | 8 (10.6) |
| Mycophenolate mofetil [g/day] | 7 (9.3) |
| Anti-TNF-α** [mg/kg] | 2 (2.6) |
* Ten SLE patients who had not received antimalarics drugs presented with contraindications to this medication, such as visual problems approved by an ophthalmic examination (retinopathy), severe cardiac problems, or severe hepatic biological abnormalities.
AIDs – autoimmune diseases, ANA – anti-nuclear antibodies, anti-P –anti-ribosomal P, ds – double stranded, GCs – glucocorticosteroids, NPSLE – neuropsychiatric SLE, SD – standard deviation, SLE – systemic lupus erythematosus, SLEDAI – Systemic Lupus Erythematosus Disease Activity Index, TNF – tumor necrosis factor.
The mean age of patients with anti-P (anti-P+) was lower than that of patients without anti-P (anti-P–) (33 ±15.3 years vs. 43.3 ±15.6 years, p = 0.006). Anti-P+, compared to anti-P–, presented with a higher prevalence of skin features (26/49 [53.1%] vs. 4/26 [15.4%], p = 0.003), malar rash (18/28 [64.3%] vs. 12/47 [25.5%], p ≤ 0.001), central neurological features (10/15 [66.7%] vs. 20/60 [33.3%], p = 0.01), anemia (26/57 [48.1%] vs. 4/21 [19.0%], p = 0.03), and anti-SSA antibodies (19/33 [57.6%] vs. 11/42 [26.2%], p = 0.006). Interestingly, anti-P+ had a lower presence of SLE/rheumatoid arthritis overlap syndrome (SLE/RA OS) (1/11 [9.1%] vs. 29/64 [45.3%], p = 0.04) (Table II).
Table II
Clinical, biological and therapeutic features and disease activity of SLE patients with or without anti-ribosomal P antibodies
| Features | Patients with anti-P (n = 30) | Patients without anti-P (n = 45) | p | OR, 95% CI |
|---|---|---|---|---|
| Age [years], mean ±SD | 33.03 ±15.3 | 43.3 ±15.6 | 0.006 | |
| Cutaneous features [n (%)] | 26 (86.6) | 23 (51.1) | 0.003 | 6.2, 1.8–20.7 |
| Malar rash | 18 (60) | 10 (22.2) | ≤ 0.001 | 5.2, 1.9–14.4 |
| Discoid lupus* | 5 (16.6) | 3(6.6) | 0.169 | |
| Ulcerations* | 1 (3.3) | 6 (13.3) | 0.232 | |
| Photosensitivity | 15 (50) | 18 (40) | 0.393 | |
| Alopecia | 5 (16.6) | 9 (20) | 0.717 | |
| Serositisa [n (%)] | 5 (16.6) | 7 (15.5) | 0.898 | |
| Pleural effusion | 5 (16.6) | 6 (13.3) | 0.689 | |
| Pericarditis* | 4 (13.3) | 3 (6.6) | 0.427 | |
| Arthritis [n (%)] | 14 (46.6) | 20 (44.4) | 0.85 | |
| Lupus nephritis [n (%)] | 10 (33.3) | 8 (17.7) | 0.122 | |
| Neurological features [n (%)] | 11 (36.6) | 8 (17.7) | 0.06 | |
| Central | 10 (33.3) | 5 (11.1) | 0.018 | 4.0, 1.2–13.2 |
| Headache | 6 (20) | 4 (8.8) | ||
| Seizures | 3 (10) | 0 | ||
| Acute confusional state | 1 (3.3) | 3 (6.6) | ||
| Myelopathy | 0 | 1 (2.2) | ||
| Peripheral | 2 (6.6) | 3 (6.6) | 1 | |
| Psychiatric features* [n (%)] | 4 (13.3) | 5 (11.1) | 1 | |
| Depression | 1 (3.3) | 5 (11.1) | ||
| Delirium | 3 (10) | 0 (0) | ||
| NPSLE features [n (%)] | 12 (40) | 13 (28.8) | 0.317 | |
| Central | 11 (36.6) | 9 (20) | 0.090 | |
| Rhupus syndrome* [n (%)] | 1 (3.3) | 10 (22.2) | 0.042 | 0.1, 0.015–0.9 |
| Hematological disordersb [n (%)] | 27 (90) | 33 (73.3) | 0.139 | |
| Anemia | 26 (86.6) | 28 (62.2) | 0.034 | 3.9, 1.1–13.2 |
| Leucopenia | 13 (43.3) | 11 (24.4) | 0.086 | |
| Lymphopenia | 15 (50) | 14 (31.1) | 0.100 | |
| Thrombopenia* | 3 (10) | 8 (17.7) | 0.509 | |
| ANA [n (%)] | 29 (96.6) | 44 (97.7) | 1 | |
| Anti-ds-DNA | 20 (66.6) | 21 (46.6) | 0.088 | |
| Anti-Sm | 10 (33.3) | 9 (20) | 0.193 | |
| Anti-Sm/RNP | 11 (36.6) | 13 (28.8) | 0.479 | |
| Anti-SSA/Ro | 19 (63.3) | 14 (31.1) | 0.006 | 3.8, 1.4–10.1 |
| Anti-SSB/La | 9 (30) | 10 (22.2) | 0.448 | |
| Anti-histones | 11 (36.6) | 11 (24.4) | 0.255 | |
| Disease activity statusc | ||||
| Weak | 5 (16.6) | 12 (26.6) | ||
| Mild | 12 (40%) | 13 (28.8) | 0.322 | |
| Strong | 12 (40%) | 17 (37.7) | ||
Table III shows that anti-P+ had higher levels of erythrocyte sedimentation rate (ESR: 86.4 ±47.5 mm/ 1st hour vs. 61.2 ±38.5 mm/1st hour, p = 0.021) and IgM-anti-cardiolipin antibodies (IgM-aCL: 18.2 ±23.3 MPL/ml vs. 0.7 ± 2.9 MPL/ml, p = 0.004).
Table III
Serological parameters and activity index of SLE patients with or without anti-ribosomal P antibodies
| Variables | Patients with anti-P (n = 30) | Patients without anti-P (n = 45) | p |
|---|---|---|---|
| ESR [mm/h] | 86.4 ±47.5 | 61.2 ±38.5 | 0.021 |
| CRP [mg/l]* | 10.5 (2.5–44.3) | 17.9 (3.2–46.9) | 0.702 |
| Creatinine [µmol/l] | 89.9 ±54.7 | 64.9 ±18.7 | 0,069 |
| Proteinuria [g/24 h]* | 0.17 (0.05–1.1) | 0.24 (0.12–0.62) | 0.884 |
| C3 [g/l] | 0.9 ±0.3 | 1.2 ±0.3 | 0.05 |
| C4 [g/l]* | 0.16 (0.1–0.2) | 0.24 (0.16–0.3) | 0.229 |
| IgM-aCL [MPL/ml] | 18.2 ±23.3 | 0.7 ±2.9 | 0.004 |
| IgG-aCL [GPL/ml] | 7.2 ±12.3 | 3.8 ±13.7 | 0.415 |
| IgM-aβ2GPI [U/ml] | 13.7 ±45.3 | 0.6 ±2.2 | 0.266 |
| IgG-aβ2GPI [U/ml] | 2.1 ±3.7 | 0.4 ±1.4 | 0.111 |
| SLEDAI | 11.7 ±7.1 | 8.8 ±5.3 | 0.051 |
Comparison of clinical features between patients with isolated anti-ribosomal P antibodies and patients with other lupus markers (Table IV) shows that patients with isolated anti-P had the highest frequency of NPSLE (6/8 [75.0%] vs. 6/22 [27.3%], p = 0.034), and the lowest frequency of LN (0/8 [0%] vs. 9/22 [40.9%], p = 0.029] and arthritis (1/8 [12.5%] vs. 12/22 [54.5%], p = 0.039). In addition, we compared anti-P+ with anti-SSA/Ro and/or anti-SSB/La antibodies and anti-P+ lacking anti-SSA/Ro and anti-SSB/La antibodies. No significant differences were observed between the 2 groups.
Table IV
Comparison of clinical and biological features and disease activity between patients with isolated anti-ribosomal P antibodies and patients with other lupus markers
| Features | Isolated anti-P* (n = 8) | Anti-P with anti-dsDNA, anti-Sm, or both (n = 22) | p | OR, 95% CI |
|---|---|---|---|---|
| Age** [years], mean ±SD | 40.6 ±15.2 | 29 ±13.8 | 0.059 | |
| ESR** [mm/h] | 70.5 ±51.4 | 88.6 ±46.2 | 0.425 | |
| SLEDAI** | 9.1 ±4.4 | 12.4 ±7.9 | 0.282 | |
| Arthritis*** [n (%)] | 1 (12.5) | 12 (54.5) | 0.039 | 0.1, 0.01–0.95 |
| Lupus nephritis*** | 0 | 9 (40.9) | 0.029 | |
| Neurological features*** | 6 (75) | 5 (22.7) | 0.029 | 9.6, 1.45–63.50 |
| Central | 5 (62.5) | 5 (22.7) | 0.083 | |
| Peripheral | 2 (25) | 0 | 0.069 | |
| Psychiatric features*** | 0 | 4 (18.2) | 0.552 | |
| NPSLE*** | 6 (75) | 6 (27.3) | 0.034 | 8, 1.25–51.13 |
| Central | 5 (62.5) | 6 (27.3) | 0.197 | |
| Rhupus syndrome*** | 1 (12.5) | 0 | 0.276 | |
| Anemia*** | 6 (75) | 19 (86.4) | 0.3 |
Discussion
Anti-P antibodies are directed against 3 ribosomal phosphoproteins: P0, P1, and P2 [4]. In this study, we found that anti-P antibodies were significantly associated with central nervous system (CNS) and cutaneous manifestations, hematological abnormalities, and high ESR in patients with SLE. In addition, isolated anti-P antibodies (without anti-dsDNA and anti-Sm antibodies) were associated with a higher frequency of NPSLE features and a lower frequency of renal and articular involvement.
In our study, the prevalence of anti-P antibodies was 40%. However, 2 previous Tunisian studies reported frequencies of 22% and 23.5% [6, 7]. Mahler et al. [20] noted differences in anti-P prevalence among populations. This discrepancy between Tunisian studies can be attributed to 2 major factors. First, our line blot method detected all 3 ribosomal P proteins, whereas Yalaoui et al. [6] used an immunodot with the C-terminal 22 amino acid peptide as an antigen, which is described as less sensitive [3]. Additionally, genetic variability among the Tunisian regions may have influenced these results [21–23]. We also found that the mean age was lower in anti-P+ patients. However, Haddouk et al. [7] and Yalaoui et al. [6] found no association between age and anti-P antibodies. Our study showed a significant association between anti-P antibodies and CNS involvement. However, Haddouk et al. [7] did not find an association between anti-P antibodies and CNS involvement despite the higher neurological involvement in SLE patients with positive anti-P antibodies. Yalaoui et al. [6] also found equal CNS feature distribution in patients with and without anti-P antibodies. These discrepancies may stem from differences in the detection methods and population origins, as explained above. In addition, unlike other studies [24], we used the 1999 ACR classification for NPSLE features. The mechanisms proposed to explain this association between anti-P antibodies and CNS involvement include antigen-directed humoral immunity in the brain, inhibition of protein synthesis, neuronal apoptosis, and inflammatory responses [25–32]. In addition, we found a negative association between anti-P antibodies and SLE/RA OS. To the best of our knowledge, only one study has investigated this relationship; it found that anti-P antibodies were more frequent in SLE patients with RA than in those with SLE alone [33]. Interestingly, our study showed an association between anti-P and IgM-aCL antibodies. Although Yalaoui et al. [6] did not find a statistically significant relationship between these antibodies, Haddouk et al. [7] reported an association between anti-P and IgG-aCL. In fact, in their systematic review and meta-analysis, Shi et al. [34] found an association between anti-P and aCL, without describing a specific mechanism to explain it. Regarding isolated anti-P antibodies, we found that 33.3%, 66.7%, and 26.7% of SLE patients with anti-P antibodies lacked anti-dsDNA, anti-Sm, or both antibodies, respectively. Although we found an association between isolated anti-P antibodies and NPSLE features, a previous meta-analysis [34] and review [35] reported controversial conclusions about the association between anti-P antibodies in general and these features. On the other hand, the association between isolated anti-P and low frequencies of LN and arthritis could be explained by a probable cross-reaction between anti-dsDNA (predictive of LN) and ribosomal P antigens [36, 37] and low expression of P-antigen in fibroblasts, which are the principal cells implicated in arthritis [38, 39].
While our study is limited by its historical design and relatively small sample size, it represents the first comprehensive exploration of specific clinical SLE manifestations in patients with anti-P antibodies from African and Arabic populations. To our knowledge, this is the first study to explore the relationship between isolated anti-P antibodies and SLE features. Using a sensitive and specific detection method (line blot assay) and focusing on patients with SLE at the time of diagnosis have enhanced the study results. The specific features found in SLE patients with isolated anti-P antibodies support the relevance of clustering SLE patients based on specific antinuclear antibodies. Cluster analysis can help identify genetic and environmental factors that contribute to disease development. It can also be used to predict disease progression and tailor treatment plans [40]. In fact, the clinical presentation and management of NPSLE differ if it is an inflammation-induced manifestation due to specific autoantibodies, such as anti-P or thrombotic NPSLE induced by anti-phospholipid antibodies [41].
Conclusions
Our study showed that anti-P antibodies are associated with CNS, cutaneous, and hematological manifestations of SLE. Moreover, isolated anti-P antibodies in patients with SLE are likely associated with a reduced risk of renal and articular involvement. Further studies with larger cohorts are needed to support our findings and explore their pathogenesis.
