Tab Application Banner
  • Users Online: 317
  • Home
  • Print this page
  • Email this page
Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 

 
ORIGINAL ARTICLE
Ahead of print publication  

Autoantibodies in pediatric systemic lupus erythematosus: Cluster analysis and its clinical implications in Indian children


1 Department of Pediatrics, Christian Medical College, Vellore, Tamil Nadu, India
2 Department of Biostatistics, Christian Medical College, Vellore, Tamil Nadu, India
3 Department of Microbiology, Christian Medical College, Vellore, Tamil Nadu, India

Date of Submission20-May-2021
Date of Acceptance12-Nov-2021
Date of Web Publication20-May-2022

Correspondence Address:
Sathish Kumar,
Department of Pediatrics, Christian Medical College, Vellore, Tamil Nadu
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/injr.injr_129_21

  Abstract 


Introduction: In children with Systemic Lupus Erythematosus (SLE), autoantibodies are considered as biomarkers for specific organ involvement or tissue damage. Some autoantibodies are used for diagnosis, disease activity, and few for disease characterisation. By using cluster analysis, we identified antibody clustering in children with SLE with specific subsets of clinical manifestations at the time of diagnosis.
Materials and Methods: All pediatric SLE (pSLE) who fulfilled 4/11 ACR criteria were included in this study. Their autoantibodies profiles were noted. We recruited 212 patients with newly diagnosed pSLE and cluster analysis was done. We identified 3 clusters which were used for analysis.
Results: Cluster 1 had ANA and anti-dsDNA antibodies a low prevalence of all other antibodies. Children in cluster 2 had autoantibodies such as ANA, anti-dsDNA, anti-RNP, and anti-Sm. Cluster 3 was had autoantibodies such as dsDNA, ANA, anti-cardiolipin and anti-SSA antibodies. On analysis, there was statistically significant difference among the 3 clusters for hair loss (P = 0.006), oral ulcers (P = 0.024), arthritis (P = 0.025), neurological symptoms (P = 0.037), renal manifestations (P = 0.003), AIHA (P = 0.012).
Conclusion: In pSLE, autoantibodies clusters with distinct clinical phenotypes. Hence, all autoantibodies should be done at time of diagnosis. It will help in predicting the clinical course of pSLE and also to identify patients at risk of developing major organ involvement later.

Keywords: Autoantibodies, cluster analysis, systemic lupus erythematosus



How to cite this URL:
Vyasam S, Punnen A, Jeyaseelan V, Prakash JJ, Kumar S. Autoantibodies in pediatric systemic lupus erythematosus: Cluster analysis and its clinical implications in Indian children. Indian J Rheumatol [Epub ahead of print] [cited 2022 Jun 26]. Available from: https://www.indianjrheumatol.com/preprintarticle.asp?id=345614




  Introduction Top


Systemic lupus erythematosus (SLE) is a multifactorial autoimmune disease characterized by a varied clinical manifestations.[1] It is known to have an unpredictable course with waxing and waning of symptoms. Due to cumulative damage due to disease, there will be significant negative impact in the quality of life and multi-organ involvement in long term. SLE has a heterogeneous clinical presentation and its course and outcome can vary greatly between patients. Heterogeneity in the clinical picture may be seen in the same individual over the clinical course. SLE is known to involve connective tissues, blood cells, brain, and kidney. Compared to adult SLE, pediatric SLE (pSLE) is known to have more frequent presentation of malar rash, severe renal disease at the onset and higher rates of other organ involvement.[2]

In pSLE, autoantibodies can be used to diagnose or predict disease activity. Some of autoantibodies are associated with specific disease manifestations. For example, anti-dsDNA is associated with lupus nephritis, anti-SSA (Ro) and anti-SSB (La) antibodies is associated with SICCA symptoms, anti-U1RNP antibodies associated with Raynaud's phenomenon and anti-phospholipid antibodies associated with thromboembolic events as seen with adult onset SLE.[3]

Studies on adult SLE have shown that these autoantibodies can exist in pairs or in clusters. There is a scarcity of studies which had looked in to clinical profile and its correlation with autoantibody clustering[4],[5] and those results have been inconclusive. Anti-Sm, anti-ribonucleoprotein (RNP), anti-Ro and anti-La antibodies presence in the population does not seem to differ between pSLE and aSLE.[6] In children with SLE, prevalence of Anti-Sm antibodies, anti-RNP antibodies, anti-Ro antibodies, anti-La antibodies are up to 51%, 37%, 33%, and 15%, respectively, during the course of their disease.[7],[8],[9],[10],[11] However, antibody clustering differs between adult SLE and pSLE.[12]

Cluster analysis of autoantibodies in SLE is also helpful to determine associations between autoantibody clusters and clinical features of pSLE. It and also to predict development of major organ manifestations as mentioned previous studies.[13],[14],[15]

Hence, in this study, we used cluster analysis to identify how autoantibody clustering and their associated with specific subsets of clinical manifestations.


  Materials and Methods Top


This was a retrospective study of newly diagnosed pSLE. All children who were diagnosed with SLE of age group between 1 and 16 years who fulfilled revised ACR classification criteria of SLE at diagnosis were included. All clinical features and immunological tests were defined as per 1997 ACR criteria for SLE.[16] ANA was done by IFA method (Euroimmun, Lubeck, Germany) in a serum dilution of 1: 100. Anti-dsDNA, Anti-Sm, anti-SSA, anti-SSB, anti-RNP, and anti-cardiolipin antibodies were determined by enzyme-linked immunoassay technology (Euroimmun, Lubeck, Germany) which was found to be reliable and accurate with a high level of agreement (>90%) with conventional techniques.[17] Clinical manifestations and autoantibodies data were collected from electronic medical record charts. Our institutional review board and Ethics Committee of CMC, Vellore (IRB No: 9452) reviewed and accepted this project.

Statistical analysis

The objective of cluster analysis is to divide a set of observations into mutually exclusive groups, in order to best represent distinct sets of observations within the sample. Agglomerative hierarchical analysis initially utilized the cluster procedure to generate and profile the clusters. The output was provided which gave the clustering history with the values of the pseudo F and t that were plotted together as a function of cluster number, which allowed estimation of meaningful number of clusters. A non-hierarchical K-means clustering procedure was then used to fine-tune the cluster membership and to ultimately identify groups of patients with similar autoantibody patterns. This method assigns each observation to one disjoint cluster based on the shortest Euclidean distance from the cluster center. The K-means method requires specification of the number of clusters and the cluster analysis was performed 2 times, specifying 3 and 4 clusters (determined by hierarchical analysis and supported by clinical relevance). The results from these analyses were compared with respect to differences in antibody profiles in each cluster and their clinical implications. Chi-square and Fisher's exact test were used for comparisons of the frequencies of clinical and laboratory SLE characteristics among the 3 cluster. All statistical analyses will be done using SAS 9.2 (SAS Institute, Cary, NC, USA). P < 0.05 was considered statistically significant.


  Results Top


We included 212 newly diagnosed children, who fulfilled ACR criteria for SLE. All clinical parameters and autoantibodies were entered in Epidata and were analyzed. Clinical features of pSLE at presentation are mentioned in [Table 1].
Table 1: The presenting symptoms of patients with pediatric systemic lupus erythematosus

Click here to view


Majority of children had arthritis (73%) followed by renal involvement (52%), hair loss (50%), oral ulcers (40%), and malar rash (39%).

Cluster analyses of autoantibodies with various clinical manifestations were done for all pSLE. Initially, we divided them into five major clusters. Akaike information criterion (AIC) is a fined technique based on in-sample fit to estimate the likelihood of a model to predict/estimate the future values. According to AIC, we have identified that only 3 clusters where group analyses can be done. There was no significant difference between 3 cluster groups and 4 cluster groups according to AIC and in cluster quality. In 3 cluster group analysis, the samples were also equally distributed without significant difference; hence we choose 3 cluster group analyses for our study.

As depicted in [Table 2] among the 3 clusters, there was a significant difference in the frequency of anti-dsDNA, anti-RNP, anti-Sm, anti-SSA, anti-cardiolipin, and lupus anticoagulant. If any autoantibody present in at least 50% of pSLE were considered a strong characteristic of the cluster. Because of the relatively low prevalence of lupus anticoagulant and anti SS-B (35.3% of patients in cluster 3), they were not considered an important characteristic of this or any cluster.
Table 2: Cluster analysis: Base line characteristics and autoantibodies frequencies

Click here to view


Children in cluster 1 were characterized by the presence of anti-dsDNA antibodies and ANA and a low prevalence of all other antibodies. The frequency of anti-dsDNA antibodies in this cluster (67.7%) was, however, lower than in cluster 2 (83.3%; P = 0.003) and in cluster 3 (89.20%; P < 0.001). Patients in cluster 2 had multiple autoantibodies: Anti-dsDNA, ANA, anti-RNP, anti-Sm. Children with cluster 3 was characterized by the presence of anti-dsDNA, ANA, anti-Cardiolipin, and anti-SSA antibodies.

Clinical or laboratory features of pSLE among the clusters in shown in [Table 3]. There was a statistically significant difference among the 3 clusters for hair loss (P = 0.006), oral ulcers (P = 0.024), arthritis (P = 0.025), neurological symptoms (P = 0.037), renal involvement (P = 0.003), and AIHA (P = 0.012). Other clinical or laboratory features were not significant.
Table 3: Association of clinical signs and symptoms within clusters

Click here to view


Cluster 1 was characterized by frequent presence of a renal manifestations (64%) than cluster 2 (35%) or cluster 3 (50%) with P = 0.003. This cluster was also characterized by increased neurological manifestations (18%) which was statistically significant compared to cluster 2 (3%) and cluster 3 (16%) with P = 0.03. Autoimmune hemolytic anemia is also more common in cluster 1 (37.6%) compared to cluster 2 (28.9%) and cluster 3 (33.6%) with P = 0.012.

Cluster 2 was characterized by arthritis which is common in cluster 2 (85.2%) compared to cluster 1 (85%) and cluster 3 (74%) with P = 0.025. Cluster 2 had the lowest incidence of renal involvement (35%) when compared to cluster 1 (64%) and cluster 3 (50%) with P = 0.003. This significant result may be due to anti-RNP antibody which may be protective against renal manifestations. Children in this cluster also had lowest incidence of neurological manifestations (3%) compared to cluster 1 (18%) and cluster 3 (11%) with P = 0.03. This cluster children also tend to have increased frequency of Raynaud's phenomenon (11%) compared to other clusters.

Children in cluster 3 tend to have more frequently have hair loss (39%) compared to cluster 1 (35%) and cluster 2 (32%). These cluster patients also tend to have increased incidence of serositis (18%) and photosensitivity (29%) compared to other two clusters.


  Discussion Top


There are few studies on clinical pattern using cluster analysis were reported literature,[15] however this is the first study in India to report autoantibody clustering and we have used a different statistical approach. To identify the association between individual variables, we performed separate association analysis. Clinicians are mostly interested in knowing relation between individual autoantibodies and different disease manifestations. In SLE, autoantibodies plays a major role in its pathogenesis, but it differs among various ethnic groups and in most studies clustering of autoantibodies were observed.[3],[12] Role of autoantibodies were not fully evaluated even though there were a few studies conducted in pSLE.[16],[17],[18],[19],[20] Antibody clustering and clinical associations were not evaluated in any of these studies and clinical characteristics of these clusters vary.

Three different clusters of patient were identified in our study population using autoantibody profile [Table 2]. However, the observation of these three clusters and results of cluster analysis were similar to various reported SLE subsets studies. ANA and anti-ds DNA were grouped under Cluster 1. Anti dsDNA, anti RNP, and anti SM were grouped under Cluster 2. Anti-dsDNA, anti-SSA, and anti-cardiolipin were grouped under Cluster 3. Lupus anticoagulant and anti SS-B was equally distributed in-between the 3 clusters. Hence, the association of these parameters cannot be measured with clinical presentation of SLE. Petri,[12] described that, the patients with aCL, LAC, and anti-dsDNA antibodies formed a unique cluster in aSLE.

There were many studies which have identified subsets of SLE with distinct patterns of organ involvement by cluster analysis. Bokemeyer and Thiele[21] in his study, identified subgroups with severe and milder manifestations using cluster analysis Similar clinical subsets of SLE patients were identified by Jacobsen et al.[22] and Stenszky et al.[23] Both identified clinical different subsets of SLE, where one group had mucocutaneous manifestations, while the other had severe renal disease with heavy proteinuria and renal failure.

The correlation between renal disease and anti dsDNA in SLE is well described in literature. It was shown in various studies a co-existence of anti-dsDNA antibody with frequent and severe lupus nephritis whereas antibody clusters of Sm/RNP was associated with lower occurrence of proteinuria and thrombocytopenia. The association between anti-dsDNA antibodies with serositis was reported by Colburn et al.[21] but, Sultan et al.,[24] did not find any association between anti-dsDNA antibodies and hemolytic anemia or thrombocytopenia.

The following are the observed autoantibody clusters and its relation with various clinical manifestations in our study population. Children in cluster 1 had the highest prevalence of anti-dsDNA and renal and neurological manifestations. Autoimmune hemolytic anemia and hair loss was also high in these patients. It was interesting to note that cases from this cluster had a low prevalence of other autoantibodies. In this cluster, no patients were positive for anti-Sm, anti-SSB, anti-Cardiolipin or anti-RNP. Cluster 2 consisted of 25.5% of all patients. Anti-Sm and anti-RNP were highly predominant in this group with oral ulcers, arthritis and Raynaud phenomenon. The clustering of anti-Sm with anti-RNP has been reported previously.[25] Renal and neurological manifestations compared to other cluster groups were less common. Cluster 3 consisted of 30.7% of patients. This cluster had the highest predominance of anti-cardiolipin, anti-SSA antibodies with hair loss, photosensitivity and serositis.

Cluster 1 showed predominance of Anti ds DNA antibodies which showed a strong association with renal disease. Different studies have already proven the pathogenic role for anti-dsDNA in renal manifestation.[26] Anti-Sm,[27] anti-RNP[28] have been described to have very minimal role in the development of renal disease. Anti-RNP was initially reported in association with MCTD and very rarely had shown clustering with Anti ds DNA autoantibody or renal disease.[29] It is also interesting to observe that patients with renal disease are at lower risk of developing other manifestations.[30]

From our study findings, we can confirm that clustering of autoantibodies has relationship with certain clinical subsets. However, prospective studies are warranted to confirm autoantibody clustering and their effect on disease manifestations.

One of the limitations of our study was that as pSLE were recruited from a tertiary hospital, which may lead to selection bias of children with more severe manifestations. This may explain the larger number of patients in cluster 1 (n = 93) than in cluster 2 (n = 54) or cluster 3 (n = 65). Another limitation of study is lack of long-term follow-up and mortality data.


  Conclusion Top


Our study confirms that in pSLE, autoantibody clusters exists with specific clinical features. The natural course of the disease and its outcome can be influenced by existing antibody clusters. Knowledge about the associations related to autoantibody clustering, clinical features and its associations in pSLE, we can predict the course of illness and identify children at risk of organ damage.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Reveille JD. Predictive value of autoantibodies for activity of systemic lupus erythematosus. Lupus 2004;13:290-7.  Back to cited text no. 1
    
2.
Hanly JG. ACR classification criteria for systemic lupus erythematosus: Limitations and revisions to neuropsychiatric variables. Lupus 2004;13:861-4.  Back to cited text no. 2
    
3.
Hoffman IE, Peene I, Meheus L, Huizinga TW, Cebecauer L, Isenberg D, et al. Specific antinuclear antibodies are associated with clinical features in systemic lupus erythematosus. Ann Rheum Dis 2004;63:1155-8.  Back to cited text no. 3
    
4.
Jurencák R, Fritzler M, Tyrrell P, Hiraki L, Benseler S, Silverman E. Autoantibodies in pediatric systemic lupus erythematosus: Ethnic grouping, cluster analysis, and clinical correlations. J Rheumatol 2009;36:416-21.  Back to cited text no. 4
    
5.
Gilliam BE, Ombrello AK, Burlingame RW, Pepmueller PH, Moore TL. Measurement of autoantibodies in pediatric-onset systemic lupus erythematosus and their relationship with disease-associated manifestations. Semin Arthritis Rheum 2012;41:840-8.  Back to cited text no. 5
    
6.
Hoffman IE, Lauwerys BR, De Keyser F, Huizinga TW, Isenberg D, Cebecauer L, et al. Juvenile-onset systemic lupus erythematosus: Different clinical and serological pattern than adult-onset systemic lupus erythematosus. Ann Rheum Dis 2009;68:412-5.  Back to cited text no. 6
    
7.
Font J, Cervera R, Espinosa G, Pallarés L, Ramos-Casals M, Jiménez S, et al. Systemic lupus erythematosus (SLE) in childhood: Analysis of clinical and immunological findings in 34 patients and comparison with SLE characteristics in adults. Ann Rheum Dis 1998;57:456-9.  Back to cited text no. 7
    
8.
Rood MJ, ten Cate R, van Suijlekom-Smit LW, den Ouden EJ, Ouwerkerk FE, Breedveld FC, et al. Childhood-onset systemic lupus erythematosus: Clinical presentation and prognosis in 31 patients. Scand J Rheumatol 1999;28:222-6.  Back to cited text no. 8
    
9.
Hiraki LT, Benseler SM, Tyrrell PN, Hebert D, Harvey E, Silverman ED. Clinical and laboratory characteristics and long-term outcome of pediatric systemic lupus erythematosus: A longitudinal study. J Pediatr 2008;152:550-6.  Back to cited text no. 9
    
10.
Carreño L, López-Longo FJ, Monteagudo I, Rodríguez-Mahou M, Bascones M, González CM, et al. Immunological and clinical differences between juvenile and adult onset of systemic lupus erythematosus. Lupus 1999;8:287-92.  Back to cited text no. 10
    
11.
Ramírez Gómez LA, Uribe Uribe O, Osio Uribe O, Grisales Romero H, Cardiel MH, Wojdyla D, et al. Childhood systemic lupus erythematosus in Latin America. The GLADEL experience in 230 children. Lupus 2008;17:596-604.  Back to cited text no. 11
    
12.
To CH, Petri M. Is antibody clustering predictive of clinical subsets and damage in systemic lupus erythematosus? Arthritis Rheum 2005;52:4003-10.  Back to cited text no. 12
    
13.
Artim-Esen B, Çene E, Şahinkaya Y, Ertan S, Pehlivan Ö, Kamali S, et al. Cluster analysis of autoantibodies in 852 patients with systemic lupus erythematosus from a single center. J Rheumatol 2014;41:1304-10.  Back to cited text no. 13
    
14.
Li PH, Wong WH, Lee TL, Lau CS, Chan TM, Leung AM, et al. Relationship between autoantibody clustering and clinical subsets in SLE: Cluster and association analyses in Hong Kong Chinese. Rheumatology (Oxford) 2013;52:337-45.  Back to cited text no. 14
    
15.
To CH, Mok CC, Tang SS, Ying SK, Wong RW, Lau CS. Prognostically distinct clinical patterns of systemic lupus erythematosus identified by cluster analysis. Lupus 2009;18:1267-75.  Back to cited text no. 15
    
16.
Lehman TJ, Hanson V, Singsen BH, Kornreich HK, Bernstein B, King K. The role of antibodies directed against double-stranded DNA in the manifestations of systemic lupus erythematosus in childhood. J Pediatr 1980;96:657-61.  Back to cited text no. 16
    
17.
Shergy WJ, Kredich DW, Pisetsky DS. The relationship of anticardiolipin antibodies to disease manifestations in pediatric systemic lupus erythematosus. J Rheumatol 1988;15:1389-94.  Back to cited text no. 17
    
18.
Oshiro AC, Derbes SJ, Stopa AR, Gedalia A. Anti-Ro/SS-A and anti-La/SS-B antibodies associated with cardiac involvement in childhood systemic lupus erythematosus. Ann Rheum Dis 1997;56:272-4.  Back to cited text no. 18
    
19.
Reichlin M, Broyles TF, Hubscher O, James J, Lehman TA, Palermo R, et al. Prevalence of autoantibodies to ribosomal P proteins in juvenile-onset systemic lupus erythematosus compared with the adult disease. Arthritis Rheum 1999;42:69-75.  Back to cited text no. 19
    
20.
Bokemeyer B, Thiele KG. Cluster analysis of 109 patients with systemic lupus erythematosus. Klin Wochenschr 1985;63:79-83.  Back to cited text no. 20
    
21.
Colburn KK, Green LM, Wong AK. Circulating antibodies to guanosine in systemic lupus erythematosus: Correlation with nephritis and polyserositis by acute and longitudinal analyses. Lupus 2001;10:410-7.  Back to cited text no. 21
    
22.
Jacobsen S, Petersen J, Ullman S, Junker P, Voss A, Rasmussen JM, et al. A multicentre study of 513 Danish patients with systemic lupus erythematosus. II. Disease mortality and clinical factors of prognostic value. Clin Rheumatol 1998;17:478-84.  Back to cited text no. 22
    
23.
Stenszky V, Kozma L, Szegedi G, Sonkoly I, Bear JC, Farid NR. Heterogeneity of systemic lupus erythematosus elucidated by cluster analysis. The influence of HLA. J Immunogenet 1986;13:327-40.  Back to cited text no. 23
    
24.
Sultan SM, Begum S, Isenberg DA. Prevalence, patterns of disease and outcome in patients with systemic lupus erythematosus who develop severe haematological problems. Rheumatology (Oxford) 2003;42:230-4.  Back to cited text no. 24
    
25.
Feng X, Zou Y, Pan W, Wang X, Wu M, Zhang M, et al. Prognostic indicators of hospitalized patients with systemic lupus erythematosus: A large retrospective multicenter study in China. J Rheumatol 2011;38:1289-95.  Back to cited text no. 25
    
26.
Hahn BH. Antibodies to DNA. N Engl J Med 1998;338:1359-68.  Back to cited text no. 26
    
27.
Winn DM, Wolfe JF, Lindberg DA, Fristoe FH, Kingsland L, Sharp GC. Identification of a clinical subset of systemic lupus erythematosus by antibodies to the SM antigen. Arthritis Rheum 1979;22:1334-7.  Back to cited text no. 27
    
28.
Migliorini P, Baldini C, Rocchi V, Bombardieri S. Anti-Sm and anti-RNP antibodies. Autoimmunity 2005;38:47-54.  Back to cited text no. 28
    
29.
Mahler M, Kessenbrock K, Szmyrka M, Takasaki Y, Garcia-De La Torre I, Shoenfeld Y, et al. International multicenter evaluation of autoantibodies to ribosomal P proteins. Clin Vaccine Immunol 2006;13:77-83.  Back to cited text no. 29
    
30.
Allen E, Farewell VT, Isenberg DA, Gordon C. A statistical analysis of the interrelationships between disease activity in different systems in systemic lupus erythematosus. Rheumatology (Oxford) 2006;45:308-13.  Back to cited text no. 30
    



 
 
    Tables

  [Table 1], [Table 2], [Table 3]



 

 
Top
 
 
  Search
 
     Search Pubmed for
 
    -  Vyasam S
    -  Punnen A
    -  Jeyaseelan V
    -  Prakash JJ
    -  Kumar S
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussion
Conclusion
References
Article Tables

 Article Access Statistics
    Viewed83    
    PDF Downloaded4    

Recommend this journal