|Year : 2022 | Volume
| Issue : 2 | Page : 142-148
Human leukocyte antigen-B alleles in spondyloarthritides: A single-center, prospective, cross-sectional study
Jeet Hemantkumar Patel1, Ved Chaturvedi2, Lalit Duggal2, Neeraj Jain2, Gurbir Bhandari3, Monika Jain4
1 Arthritis Rheumatology Clinic, Vadodara, Gujarat, India
2 Department of Rheumatology and Clinical Immunology at Sir Ganga Ram Hospital, New Delhi, India
3 Department of Rheumatology, Adesh Institute of Medical Science, Bhatinda, Punjab, India
4 Division of Histocompatibility and Immunogenetics, Institute of Medical Genetics and Genomics, Sir Ganga Ram Hospital, New Delhi, India
|Date of Submission||22-Oct-2020|
|Date of Acceptance||07-Feb-2021|
|Date of Web Publication||19-Mar-2022|
Dr. Jeet Hemantkumar Patel
Arthritis Rheumatology Clinic, 202, Shyam Arena Complex, Sampatrao Colony, Alkapuri, Vadodara - 390 007, Gujarat
Source of Support: None, Conflict of Interest: None
Objective: The objective of this study is to study distribution of human leukocyte antigen (HLA)-B alleles in Spondyloarthritis (SpA) and to find its clinical relevance in the patient management.
Materials and Methods: A prospective, cross-sectional, pilot study was carried out on 100 participants at the single center of Northern India (Sir Ganga Ram Hospital, New Delhi). Assessment in ankylosing spondylitis (ASAS) classification criteria for axial/peripheral spondyloarthritis (SPA) (conventionally AS-ankylosing spondylitis), classification criteria for reactive arthritis (ReA), and CASPAR (classification criteria for psoriatic arthritis) criteria for psoriatic arthritis were used to sub-classify patients. Inflammatory bowel disease (IBD) patients who classified for ASAS criteria for SPA were classified as IBD-related SPA. Demographic data, disease characteristics, and relevant investigations were noted. HLA-B genotyping was carried out by the polymerase chain reaction method. HLA-B genotyping of 100 healthy kidney transplant donors were taken as controls. Fisher's exact test was used. P < 0.05 was considered as statistically significant.
Results: Out of 100 patients, 58 were male and 42 were female. SPA (AS), ReA, PSA, and IBD-related SPA were 65, 12, 16, and 07, respectively. HLA-B 27+ve participants were 57. HLA-B40 and HLA-B52 were present in 25 and 19 participants, respectively. HLA-B27 allele was associated with SPA, PSA, and IBD-related SPA (P < 0.01, 0.034, and 0.037, respectively). In HLA-B27-ve cases, the frequency of HLA-B40 was increased (P value – 0.033, odds ratio [OR] with 95% confidence interval-2.44, [1.09, 5.48]). Similarly, the frequency of HLA-B15 and HLA-B13 were increased in PSA (P value – 0.034, 0.020 with OR – 4.09, 6.33, respectively). The presence of HLA-B27 allele favored axial ± peripheral disease while its absence favored peripheral disease (P value– 0.019). HLA-B27+ve participants had more active disease compared to HLA-B27-ve and HLA-B40+ve cases (P < 0.01 and 0.013, respectively).
Conclusion: The utility of HLA-B alleles other than HLA-B27 is limited in SpA patients' management. To find their use in diagnosis, prognosis and treatment need further studies.xs
Keywords: Genes, human leukocyte antigen, human leukocyte antigen-B alleles, spondyloarthritides
|How to cite this article:|
Patel JH, Chaturvedi V, Duggal L, Jain N, Bhandari G, Jain M. Human leukocyte antigen-B alleles in spondyloarthritides: A single-center, prospective, cross-sectional study. Indian J Rheumatol 2022;17:142-8
|How to cite this URL:|
Patel JH, Chaturvedi V, Duggal L, Jain N, Bhandari G, Jain M. Human leukocyte antigen-B alleles in spondyloarthritides: A single-center, prospective, cross-sectional study. Indian J Rheumatol [serial online] 2022 [cited 2022 Jul 4];17:142-8. Available from: https://www.indianjrheumatol.com/text.asp?2022/17/2/142/339940
| Introduction|| |
Spondyloarthritides (SpA) are a family of heterogeneous inflammatory arthritis with different etiological factors and a strong genetic association with human leukocyte antigen (HLA) B27. They are a group of diseases with axial as well as peripheral arthritis and extra-articular features such as uveitis, psoriasis, and gastrointestinal disturbances. They are mainly classified into spondyloarthritis (SPA) (radiographic and nonradiographic axial SPA (axSPA) and peripheral SPA [pSPA]) colloquially called ankylosing spondylitis (AS), reactive arthritis (ReA), psoriatic arthritis (PSA), and inflammatory bowel disease (IBD)-related SPA. The prevalence of SpA varies from 0.1% to 1.4% across different geographic regions. The frequency of HLA-B27 in Indian SpA patients varies from 30% to 94%, whereas it is present in up to 10% of the general population of India.
The pathogenesis of SpA is multifactorial and involves a number of interlinking pathways. Genetics plays a major role in the pathogenesis of SpA, HLA-B27 being the most affecting gene. Clinical presentation and disease course are influenced by ethnicity, age of onset, sex, and genes., It is estimate that around 2-6% of HLA-B27 positive subjects develop AS and around 10-20% of patients with AS do not carry the HLA-B27 allele., Various studies of different ethnic groups had showed association of different HLA-B alleles other than B27 with SpA. The association of HLA-B40, HLA-B7 CREG antigens (HLA-B22, HLA-B40, HLA-B42, etc.,), HLA-B35 CREG antigens (HLA-B18, HLA-B35, HLA-B51, and HLA-B62), and HLA-B38, HLAB-39, with SpA was reported in the past. In India, the role of HLA-B07 was found in HLA-B27-ve AS patients.
While single Indian study showed the association of HLA-B07 in HLA-B27-ve AS patients, it also showed significantly the low frequency of HLA-B40 in them. Hence, we want to study distribution of HLA-B alleles in SpA and to find their clinical relevance in various subtypes.
| Materials and Methods|| |
The minimum sample size required for the study with the reported proportions of HLA positivity of SpA (0.30) with a precision of 5% and at a significant level of 5% was found to be 323. The formula used was as under:
Since the proposed study was time bound, it was carried out as a pilot study with the minimum number of 100 participants.
Inclusion criteria and clinical assessment
After Ethics committee approval, (EC/06/16/1014, EC approval date: 30th May, 2016) 100 patients of SpA were included from June 2016 to December 2017 at a single center of Northern India (Sir Ganga Ram Hospital, New Delhi). The study was carried out on the diverse population of Indian subcontinent, and even patients from neighbor countries such as Nepal and Afghanistan. Patients who classified for four categories of SpA were enrolled in the study: (1) AS-Axial and/or axial + peripheral spondyloarthritis (axSPA, pSPA) (the short form AS will be used instead of SPA to avoid confusion between SpA and SPA), (2) ReA, (3) PSA, and (4) IBD related SPA. Axial and/or axial + peripheral spondyloarthritis were classified by ASAS classification criteria for axSPA and ASAS classification criteria for pSPA, respectively., ReA and PSA were classified on the basis of preliminary classification criteria for ReA and CASPAR criteria, respectively. Symptomatic colonoscopy-proven IBD patients who fulfilled ASAS classification criteria for axial and/or peripheral diseases were subclassified for IBD-related SPA. Patients <16 years, patients with other inflammatory arthritis (e.g., rheumatoid arthritis etc.,), and mimics of SpA were excluded. Hundred (n = 100) age- and sex-matched healthy kidney donors from same center were used as controls. Demographic data and disease characteristics were noted along with laboratory investigations such as erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). Disease activity was recorded by using scientific calculators for AS disease activity score (ASDAS– ESR, ASDAS– CRP) and disease activity for PSA (DAPSA). For enthesitis activity, Maastricht AS Enthesitis Score (MASES) was used. Patients' follow-up visit data were not recorded.
Imaging and human leukocyte antigen genotyping
Imaging findings of patients were noted. X-ray pelvis (AP) was read by rheumatologists on the basis of modified New York criteria for radiographic grading of sacroiliitis for all the patients. Magnetic resonance imaging sacroiliac joint with fat suppressed sequences was read by one assigned radiologist for every individual patient, blinded to study. HLA B genotyping was carried out by the polymerase chain reaction method using HLA B sequence-specific primer tray kit, and the results were recorded using the standard manufacturing interpretation software.
Continuous variables were expressed as mean ± standard deviation (SD). Categorical variables were analyzed by the frequency distribution, percentages, and contingency tables and assessed by SPSS v. 19.0 (IBM Corp. Released 2010. IBM SPSS Statistics for Windows, Version 19.0. Armonk, NY: IBM Corp.). For comparisons, Fisher's exact test was used. The association between disease characteristics and HLA alleles was determined by the two-tailed P < 0.05, Odds ratio (OR), and confidence interval (CI) of 95%.
| Results|| |
Patients' clinical characteristics
Male: female ratio was 1.3:1. The number of male was more among various subtypes. Female patients were 10 years older than male counter parts (age of female participants – 45.79 ± 11.67 years, age of male participants – 35.90 ± 14.95 years). Similarly, IBD-related SPA had 8–9 years older age compared to other subtypes (e.g., age of AS patients– 39.45 ± 14.37 years, age of IBD related SPA– 48.57 ± 14.01 years). Delay in the diagnosis was 2–3 years across all subgroups except ReA in which there was maximum delay of 2 weeks only. Sacroiliac joints (SI) joint, lumbo-sacral spine, knee, and ankle joints were the most affected joints (66%, 56%, 73%, and 60%, respectively). On the contrary, small joints of feet, thoracic spine, and hip joints were the least affected joints (13%, 17%, and 18%, respectively). The rest of patients' clinical features are given in [Table 1].
Human leukocyte antigen-B genotyping of cases and controls
Frequencies of different HLA-B alleles of cases and controls are depicted in [Figure 1]. In controls, HLA-B35 was the most common allele (N = 35/100), whereas HLA-B27 positivity was seen in 10% of participants. In cases, HLA-B27 was the most common allele (57%) followed by HLA-B40 (23%) and HLA-B52 (18%). Participants with HLA-B27 allele were counted in HLA-B27+ve group irrespective of presence of second allele (e.g., HLA-B40, HLA-B15, etc.,). The association of HLA-B27 with SpA was statistically significant (P value-0.0001, OR–11.93, CI 5.56–25.61), whereas frequencies of HLA-B35 and HLA-B18 were significantly decreased in overall cases (P value–0.0014, 0.0029).
|Figure 1: Bar diagram of various human leukocyte antigen-B alleles in 100 cases of Sponyloarthritides and 100 controls|
Click here to view
Human leukocyte antigen-B alleles in various subclasses of spondyloarthritides
Various subgroups and their important allelic associations are given in [Table 2]. HLA-B27 allele was associated AS, PSA, and IBD-related SPA, although its association with ReA was not found while comparing allelic distribution of subgroup cases with controls. In regards to clinical characteristics, this study showed that HLA-B27+ve cases had axial disease while HLA-B27-ve cases had higher chances for peripheral disease (P value– 0.019). HLA-B27-ve patients had also more chances of having ReA and PSA (P value– 0.027, 0.029 respectively) while HLA-B27+ve individuals had higher chances of AS (P value– 0.0003). Difference between major allelic subgroups is given in [Table 3]. There was no major difference between HLA-B27+ve and HLA-B40+ve participants in terms of extra-articular features such as enthesitis, dactylitis, uveitis, colitis, or psoriasis (P value – 0.63, 0.46, 0.12, 0.35, and 0.68, respectively).
|Table 3: Association of human-leukocyte antigen B alleles with disease characteristics|
Click here to view
Human leukocyte antigen-B27 and human leukocyte antigen-B40 alleles and disease activity
While comparing with HLA-B27-ve patients, HLA-B27+ve patients had significantly higher mean ± SD for EMS, ESR, CRP, ASDAS-ESR, ASDAS-CRP (P value– 0.003, 0.019, 0.018, <0.01, <0.01 respectively). HLA-B27+ve patients also had higher mean ± SD for peripheral pain, ASDAS-ESR and ASDAS-CRP when compared to HLA-B40+ve patients (P value– 0.036, 0.013, and 0.018, respectively). While there was no major difference in relation to HLA-B alleles for back pain, Visual Analog Scale (VAS), patient global assessment for disease VAS, DAPSA, and MASES.
| Discussion|| |
Study of HLA-B alleles in SpA including all subtypes revealed that HLA-B27 was the major allele of clinical use. It was associated with various disease characteristics and disease activity. On the other hand, we found few non-HLA-B27 alleles that had limited use in the context of SpA management. The frequency of HLA-B40 was higher in HLA-B27-ve cases that supports previous studies., However, it was opposite to the Indian study that showed significantly less frequency of HLA-B40 in HLA-B27 negative AS compared to controls. Possible reason might be that our study included all subtypes while other Indian study had included only AS patient and inclusion criteria for AS were also different from our study. Other studies had also revealed a role of HLA-B18 and HLA-B35 (HLA-B35 CREG antigens) and HLA-B38 (split product of HLA-B16) in AS, albeit in our study, their frequencies were less in AS.
[Table 4] shows various genetic studies in different ethnic populations and their results. Our study, another Indian study and other world studies corroborates the fact that SpA is a heterogeneous group of disorders and carry complex polygenic nature. India has varied climatic places, different descendants, complex marriage system, different food pattern, and multiple ecological and nonecological factors that affect health and hygiene. Therefore, allelic distribution and their disease association are nonuniform in the context of SpA. Even in different world studies, the association of non-HLA-B27 alleles was not uniform globally.
HLA-B27 was present in 57% of cases in this study which was similar to HLA-B27 prevalence in Indian SpA patients (68%). HLA-B27 was present in 70.8%, 37.5%, 31.3%, and 25% of cases of AS, IBD related SPA, PSA, and ReA, respectively. These findings of HLA-B27 prevalence in subtypes of SPA were similar to that of the world literature. HLA-B15 and HLA-B13 were found to be associated with PSA in the present study which were contrary to the association of HLA-B16 and HLA-B17 with PSA in some of the studies.,
HLA-B27 was associated with many clinical characteristics including articular and extra-articular manifestations (e.g., enthesitis and dactylitis) in our study that is supported by world literature. Although HLA-B27 has been associated with more prevalence of sacroiliac joint involvement and ankylosis, we did not find this association in our study. This result was opposite to world studies;, however, possible reasons could be inclusion of all SpA subtypes and smaller sample size.
HLA-B40 allele was not influencing for any particular disease manifestations in our study that corroborates with world studies.,, In some studies, HLA-B15 was associated with peripheral arthritis, enthesitis, more severe disease, periodontitis, and Behcet's disease. On the contrary to these findings, HLA-B15 was associated with uveitis and PSA in the present study.
Our study supports the fact that HLA-B27 is associated with early disease onset with more severe phenotype. In our study, HLA-B27 + ve patients had longer duration of illness, higher markers of inflammation (CRP and ESR), higher disease activity indices (ASDAS-CRP and ASDAS-ESR) and more knee joint involvement compared to HLA-B27-ve cases. This could be the possible reason for higher DMARDS use in HLA-B27+ve subjects compared to HLA-B27-ve participants. No other HLA-B alleles including HLA-B15, had affected disease severity and disease activity in the present study. This is contrary to one Latin American study, in which HLA-B15 was associated with higher BASDAI (Bath AS Disease Activity Index), BASFI (Bath AS Functional Index), and more peripheral disease.
Small sample size is a limitation of the study. Limited number of some subtype patients was also one of the drawbacks like ReA had 12 patients and IBD related SPA had only 7 participants. This study was carried out at tertiary care hospital and high disease activity was the main reason to visit the hospital. Hence, average disease activity was higher in our study. Early cases might be missed as they reached to our center on the later stage of the disease. Further data are required to specify the role of different HLA-B alleles in SpA management in Indian patients, especially with participants from all the parts of India.
| Conclusion|| |
Inference from the previous studies and the present study is that non-HLA-B27 alleles have doubtful utility in terms of diagnosis, prediction of clinical manifestations or disease severity, and treatment part. HLA-B40 is one such allele with increased frequency in HLA-B27 negative cases; however, clinical usefulness is doubtful. Further study with adequate number of participants is required to validate the results of the present study.
We would like to thank sincerely to our statistician, Ms Parul Takkar working in Research Department at Sir Ganga Ram Hospital, New Delhi. We also thank to staff members of HLA genetic lab for their support in HLA genotyping.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Pilar SR. Spondyloarthritis: Pathogenesis, clinical manifestations, diagnosis and management. Eur Med J 2016;1:96-102.
Ehrenfeld M. Geoepidemiology: The environment and spondyloarthropathies. Autoimmune Rev 2010;9:A325-9.
Parasannanavar DJ, Rajadhyaksha A, Ghosh K. Application of a simple in-house PCR-SSP technique for HLA-B* 27 typing in spondyloarthritis patients. Arthritis 2013;2013:504109.
Khan MA. An overview of clinical spectrum and heterogeneity of spondyloarthropathies. Rheum Dis Clin North Am 1992;18:1-0.
Lau CS, Burgos-Vargas R, Louthrenoo W, Mok MY, Wordsworth P, Zeng QY. Features of spondyloarthritis around the world. Rheum Dis Clin North Am 1998;24:753-70.
van der Linden SM, Valkenburg HA, de Jongh BM, Cats A. The risk of developing ankylosing spondylitis in HLA-B27 positive individuals. A comparison of relatives of spondylitis patients with the general population. Arthritis Rheum 1984;27:241-9.
Braun J, Bollow M, Remlinger G, Eggens U, Rudwaleit M, Distler A, et al
. Prevalence of spondylarthropathies in HLA-B27 positive and negative blood donors. Arthritis Rheum 1998;41:58-67.
Robinson WP, van der Linden SM, Khan MA, Rentsch HU, Cats A, Russell A, et al
. HLA-Bw60 increases susceptibility to ankylosing spondylitis in HLA-B27+patients. Arthritis Rheum 1989;32:1135-41.
Khan MA. B7-CREG and ankylosing spondylitis. Br J Rheumatol 1983;22:129-33.
Wagener P, Zeidler H, Eckert G, Deicher H. Increased frequency of HLA-Bw35 CREG antigens in HLA-B27 negative ankylosing spondylitis. Br J Rheumatol 1983;22:134-5.
Khan MA, Kushner I, Braun WE. B27-negative HLA-BW16 in ankylosing spondylitis. Lancet 1978;1:1370-1.
Yamaguchi A, Tsuchiya N, Mitsui H, Shiota M, Ogawa A, Tokunaga K, et al
. Association of HLA-B39 with HLA-B27-negative ankylosing spondylitis and pauciarticular juvenile rheumatoid arthritis in Japanese patients. Evidence for a role of the peptide-anchoring B pocket. Arthritis Rheum 1995;38:1672-7.
Devraj JP, Shankarkumar U, Ghosh K. Increased frequency of HLA-B7 among B27-negative seronegative spondarthritis patients from Mumbai, Western India. Br J Biomed Sci 2009;66:25-7.
Rudwaleit M, Landewé R, van der Heijde D, Listing J, Brandt J, Braun J, et al
. The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part I): Classification of paper patients by expert opinion including uncertainty appraisal. Ann Rheum Dis 2009;68:770-6.
Rudwaleit M, van der Heijde D, Landewé R, Akkoc N, Brandt J, Chou CT, et al
. The Assessment of SpondyloArthritis International Society classification criteria for peripheral spondyloarthritis and for spondyloarthritis in general. Ann Rheum Dis 2011;70:25-31.
Braun J, Kingsley G, van der Heijde D, Sieper J. On the difficulties of establishing a consensus on the definition of and diagnostic investigations for reactive arthritis. Results and discussion of a questionnaire prepared for the 4th
International Workshop on Reactive Arthritis, Berlin, Germany, July 3-6, 1999. J Rheumatol 2000;27:2185-92.
Chandran V, Schentag CT, Gladman DD. Sensitivity specificity of the CASPAR criteria for psoriatic arthritis in a family medicine clinic setting. J Rheumatol 2008;35:2069-70.
Machado P, Landewé R, Lie E, Kvien TK, Braun J, Baker D, et al
. Ankylosing Spondylitis Disease Activity Score (ASDAS): Defining cut-off values for disease activity states and improvement scores. Ann Rheum Dis 2011;70:47-53.
Helliwell PS, FitzGerald O, Fransen J, Gladman DD, Kreuger GG, Callis-Duffin K, et al
. The development of candidate composite disease activity and responder indices for psoriatic arthritis (GRACE project). Ann Rheum Dis 2013;72:986-91.
Heuft-Dorenbosch L, Spoorenberg A, van Tubergen A, Landewé R, van ver Tempel H, Mielants H, et al
. Assessment of enthesitis in ankylosing spondylitis. Ann Rheum Dis 2003;62:127-32.
van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361-8.
Londono J, Santos AM, Peña P, Calvo E, Espinosa LR, Reveille JD, et al
. Analysis of HLA-B15 and HLA-B27 in spondyloarthritis with peripheral and axial clinical patterns. BMJ Open 2015;5:e009092.
van Gaalen FA, Verduijn W, Roelen DL, Böhringer S, Huizinga TW, van der Heijde DM, et al
. Epistasis between two HLA antigens defines a subset of individuals at a very high risk for ankylosing spondylitis. Ann Rheum Dis 2013;72:974-8.
Díaz-Peña R, Blanco-Gelaz MA, Njobvu P, López-Vazquez A, Suárez-Alvarez B, López-Larrea C. Influence of HLA-B*5703 and HLA-B*1403 on susceptibility to spondyloarthropathies in the Zambian population. J Rheumatol 2008;35:2236-40.
Imre K, Koó E, Seszták M, Bosák V, Bitterova O. [Genetics in patients with psoriatic arthritis]. Orv Hetil 2006;147:2415-9.
Wei JC, Tsai WC, Lin HS, Tsai CY, Chou CT. HLA-B60 and B61 are strongly associated with ankylosing spondylitis in HLA-B27-negative Taiwan Chinese patients. Rheumatology (Oxford) 2004;43:839-42.
Vargas-Alarcón G, Londoño JD, Hernández-Pacheco G, Pacheco-Tena C, Castillo E, Cardiel MH, et al
. Effect of HLA-B and HLA-DR genes on susceptibility to and severity of spondyloarthropathies in Mexican patients. Ann Rheum Dis 2002;61:714-7.
Gladman DD, Brubacher B, Buskila D, Langevitz P, Farewell VT. Differences in the expression of spondyloarthropathy: A comparison between ankylosing spondylitis and psoriatic arthritis. Clin Invest Med 1993;16:1-7.
Purrmann J, Zeidler H, Bertrams J, Juli E, Cleveland S, Berges W, et al
. HLA antigens in ankylosing spondylitis associated with Crohn's disease. Increased frequency of the HLA phenotype B27, B44. J Rheumatol 1988;15:1658-61.
Mielants H, Veys EM, Joos R, Noens L, Cuvelier C, De Vos M, et al
. HLA antigens in seronegative spondyloarthropathies. Reactive arthritis and arthritis in ankylosing spondylitis: Relation to gut inflammation. J Rheumatol 1987;14:466-71.
Sheehan NJ. HLA-B27: What's new? Rheumatology (Oxford) 2010;49:621-31.
Marzo-Ortega H, McGonagle D, O'Connor P, Hensor EM, Bennett AN, Green MJ, et al
. Baseline and 1-year magnetic resonance imaging of the sacroiliac joint and lumbar spine in very early inflammatory back pain. Relationship between symptoms, HLA-B27 and disease extent and persistence. Ann Rheum Dis 2009;68:1721-7.
Stein JM, Machulla HK, Smeets R, Lampert F, Reichert S. Human leukocyte antigen polymorphism in chronic and aggressive periodontitis among Caucasians: A meta-analysis. J Clin Periodontol 2008;35:183-92.
Choukri F, Chakib A, Himmich H, Hüe S, Caillat-Zucman S. HLA-B*51 and B*15 alleles confer predisposition to Behçet's disease in Moroccan patients. Hum Immunol 2001;62:180-5.
Feldtkeller E, Khan MA, van der Heijde D, van der Linden S, Braun J. Age at disease onset and diagnosis delay in HLA-B27 negative vs. positive patients with ankylosing spondylitis. Rheumatol Int 2003;23:61-6.
[Table 1], [Table 2], [Table 3], [Table 4]