Systemic lupus erythematosus with high disease activity identification based on machine learning

OBJECTIVE

Clinical evaluation of systemic lupus erythematosus (SLE) disease activity is limited and inconsistent, and high disease activity significantly, seriously impacts on SLE patients. This study aims to generate a machine learning model to identify SLE patients with high disease activity.

METHOD

A total of 1014 SLE patients with low disease activity and 453 SLE patients with high disease activity were included. A total of 94 clinical, laboratory data and 17 meteorological indicators were collected. After data preprocessing, we use mutual information and multisurf to evaluate and select the importance of features. The selected features are used for machine learning modeling. Performance of the model is evaluated and verified by a series of binary classification indicators.

RESULTS

We screened out hematuria, proteinuria, pyuria, low complement, precipitation, sunlight and other features for model construction by integrated feature selection. After hyperparameter optimization, the LGB has the best performance (ROC: AUC = 0.930; PRC: AUC = 0.911, APS = 0.913; balance accuracy: 0.856), and the worst is the naive bayes (ROC: AUC = 0.849; PRC: AUC = 0.719, APS = 0.714; balance accuracy: 0.705). Finally, the selection of features has good consistency in the composite feature importance bar plot.

CONCLUSION

We identify SLE patients with high disease activity by a simple machine learning pipeline, especially the LGB model based on the characteristics of proteinuria, hematuria, pyuria and other feathers screened out by collective feature selection.

SHP2: its association and roles in systemic lupus erythematosus

OBJECTIVE

Systemic lupus erythematosus (SLE) is an autoimmune inflammatory disease. Src homology 2 domain containing protein tyrosine phosphatase (SHP2) is a member of the protein tyrosine phosphatases (PTPs) family. To date, relationship between SHP2 and SLE pathogenesis is not elucidated.

METHOD

We measured plasma levels of SHP2 in 328 SLE patients, 78 RA patients, 80 SS patients and 79 healthy controls by ELISA, and discussed association of SHP2 in SLE patients, potential of plasma SHP2 as a SLE biomarker. Moreover, histological and serological changes were evaluated by flow cytometry, HE/Masson examination, immunofluorescence test in pristane-induced lupus mice after SHP2 inhibitor injection to reveal role of SHP2 in lupus development.

RESULTS

Results indicated that SHP2 plasma levels were upregulated in SLE patients and correlated with some clinical, laboratory characteristics such as proteinuria, pyuria, and may be a potential biomarker for SLE. After SHP2 inhibitor treatment, hepatosplenomegaly and histological severity of the kidney in lupus mice were improved. SHP2 inhibitor reversed DCs, Th1, and Th17 cells differentiation and downregulated inflammatory cytokines (IL-4, IL-6, IL-10, IL-17A, IFN-γ and TNF-α) and autoantibodies (ANA, anti-dsDNA) production in pristane-lupus mice.

CONCLUSION

In summary, SHP2 correlated with SLE pathogenesis and promoted the development of lupus.

Gene polymorphisms of SIRT1 in patients with rheumatoid arthritis

AIM

Previous studies have shown that silent information regulator 1 (SIRT1) expression is elevated in rheumatoid arthritis (RA) patients. However, whether gene polymorphisms in SIRT1 gene associated with RA in a Chinese Han population remains to be discussed.

METHOD

In this case-control study, 529 RA patients and 700 healthy controls were selected, and association of 11 SIRT1 gene polymorphisms (rs12415800, rs3740051, rs932658, rs3740053, rs7895833, rs10509291, rs33957861, rs7069102, rs2273773, rs3818292, rs1467568) with RA susceptibility was evaluated.

RESULTS

Frequency of GA+GG genotype of rs3740051 in RA patients was significantly lower than that in healthy controls (P = .037). Frequencies of GC and GC+GG genotypes of rs7069102 were significantly lower than those in healthy controls (P = .036, P = .047). Frequencies of GA and GA+GG genotypes of rs1467568 were lower in RA patients as compared to those in healthy controls (P = .011, P = .013). For rs2273773, RA patients who carried the T allele had higher number of tender joints than patients who carried the C allele (P = .033). Other polymorphisms did not associate with RA risk.

CONCLUSION

The findings suggest that rs3740051, rs7069102 and rs1467568 variants in SIRT1 gene are related to RA susceptibility in a Chinese Han population.

Gene polymorphisms of LGALS2, LGALS3 and LGALS9 in patients with rheumatoid arthritis

Rheumatoid arthritis (RA) is a complicated rheumatic autoimmune disease. Lectin, galactoside-binding soluble, 2 (LGALS2), LGALS3 and LGALS9, three members of the galectin family, play potential roles in autoimmune diseases, including RA. However, association of genetic polymorphisms of LGALS2, LGALS3 and LGALS9 with RA risk in a Southern Chinese Han population has not been elucidated. A case-control study was conducted herein, including 500 RA patients and 650 healthy individuals of Southern Chinese Han origin. Twelve single nucleotide polymorphisms (SNPs), including rs7291467 for the LGALS2 gene, rs4644, rs4652, rs1009977, rs2274273 and rs17128183 for the LGALS3 gene, and rs4795835, rs3763959, rs4239242, rs3751093, rs732222 and rs4794976 for the LGALS9 gene, were genotyped. Polymorphisms were genotyped using the KASP method. Frequencies of rs1009977 genotype TG and rs3751093 genotype GA of LGALS3 gene were significantly different between RA patients and healthy controls (P = 0.049, P = 0.033). Allele T and genotypes TT and TT + TG of rs4794976 for LGALS9 gene were significantly correlated with RA risk (P = 0.017, P = 0.012, P = 0.041). Subgroup analysis revealed that rs1009977, rs2274273 and rs17128183 polymorphisms of LGALS3 gene and rs4795835 polymorphism of LGALS9 gene were correlated with several RA clinical manifestations (all P < 0.05). In addition, haplotype GCGTT showed an increased risk for RA (OR = 1.216, 95% CI: 1.028-1.438, P = 0.023), whereas haplotype GCGTG showed a reduced risk for RA susceptibility (OR = 0.779, 95% CI: 0.625-0.971, P = 0.026). In conclusion, LGALS3 and LGALS9 gene polymorphisms may associate with RA predisposition in a Southern Chinese Han population.

Association between IL-37 and Systemic Lupus Erythematosus Risk

Interleukin-37 (IL-37) is an anti-inflammatory cytokine. In our former study, we found increased plasma IL-37 levels in systemic lupus erythematosus (SLE) patients. However, relationship between IL-37 levels and clinical laboratory characteristics of SLE patients has not been elucidated. In addition, association of IL37 gene polymorphism with SLE risk needs to be discussed. A group of 580 individuals (220 SLE patients and 360 healthy controls) in a Southern Chinese Han population were recruited. Plasma IL-37 levels were evaluated using enzyme-linked immunosorbent assay (ELISA). Four single-nucleotide polymorphisms (rs3811047, rs2723186, rs2723176 and rs4364030) of IL37 gene were genotyped. Relationship of IL-37 expression, IL37 gene polymorphisms and clinical characteristics was discussed. We found that plasma levels of IL-37 were negatively associated with SLE disease activity index (SLEDAI) (r_s = -0.352, P = .001), and were higher in less active patients compared with active patients (P = .003). Decreased levels of IL-37 were found in SLE patients with discoid rash when compared to patients who did not have this symptom (P < .001). Plasma IL-37 levels were significantly lower in patients with hypocomplementemia comparing to those without this feature (P = .009). Levels of IL-37 in SLE with positive proteinuria were lower than patients with negative proteinuria (P = .046). Furthermore, allele distribution of rs2723186, rs4364030 between SLE cases and healthy individuals was significantly different (P = .001, P = .010, respectively). Genotype of rs4364030 was different between SLE cases and controls (P = .015). Haplotype analysis revealed that the frequency of haplotype CG (rs2723176 (C) +rs2723186 (G)) was higher in SLE, as compared with healthy individuals (P = .002). In conclusion, the plasma levels of IL-37 were related to SLE severity, and IL37 gene polymorphisms (rs2723186, rs2723176 and rs4364030) may associate with SLE susceptibility.

Association of IL-35 expression and gene polymorphisms in rheumatoid arthritis

Interleukin (IL)-35 is the newest member of the IL-12 family. It is expressed in many immune cells and has been recognized as a novel inflammatory cytokine that may have bifunctional properties. Recent findings have indicated that the expression of IL-35 is abnormal in rheumatoid arthritis (RA) patients. However, the results were inconsistent. In this study, 400 RA patients were recruited to evaluate serum levels of IL-35 in a Chinese Han population by enzyme-linked immunosorbent assay. The association of IL-35 gene polymorphisms and RA genetic susceptibility was investigated in 400 RA patients and 612 healthy controls. The results showed that serum levels of IL-35 were elevated in 100 RA patients compared to 51 healthy controls, relating to disease activity and synovial fluid IL-35 expression in the training cohort. Another independent 300 RA patients and 369 other rheumatic disease patients (98 lupus, 95 osteoarthritis, 95 gout, 42 Sjogren's syndrome and 39 ankylosing spondylitis patients) confirmed that serum levels of IL-35 were elevated in RA patients, and serum IL-35 has good diagnostic ability for differentiating RA from the other rheumatic diseases. The genotyping of 10 IL-35 polymorphisms, including rs2227314, rs2243115, rs2243123, rs2243131, rs568408, rs583911, rs428253, rs4740, rs9807813 and rs4905, revealed that rs2227314, rs2243131, rs9807813, and rs583911 were correlated with RA risk. Different genotypes (rs2227314, rs583911, and rs9807813) exhibited different expression of IL-35. These findings demonstrate that serum levels of IL-35 are increased in RA patients and that IL-35 polymorphisms are correlated with RA risk.

Association of MBL2 gene polymorphisms and systemic lupus erythematosus susceptibility: A meta-analysis

OBJECTIVES

Mannose binding lectin (MBL) gene single nucleotide polymorphisms have been associated with systemic lupus erythematosus (SLE) risk with inconsistent results. This study aimed to explore whether MBL2 A\B, A\C, A\D, A\O, L\H and Y\X polymorphisms affected SLE susceptibility.

METHODS

A meta-analysis was performed on 20 studies, containing allelic contrast, additive, dominant and recessive models. Odds ratio (OR) was calculated to reflect the effect of association.

RESULTS

A total of 64 pooled comparisons were conducted, including 7194 SLE patients and 7401 healthy controls. The meta-analysis inducted a significant association between allele B and SLE (OR = 0.766, 95% CI = 0.681-0.862, P < .001). The genotype BB in the additive model and AB + BB in the recessive model both reduced the risk of SLE (OR = 0.611, 95% CI = 0.422-0.882, P = .009; OR = 0.806, 95% CI = 0.688-0.944, P = .008). Regarding A\O polymorphisms, results revealed statistical differences in allelic contrast, additive model and recessive models (OR = 0.826, 95% CI = 0.732-0.931, P = .002; OR = 0.737, 95% CI = 0.557-0.977, P = .034 and OR = 0.793, 95% CI = 0.683-0.921, P = .002, respectively). As for L\H, meta-analysis revealed that allele H and genotype HH both decreased SLE susceptibility in allelic contrast and dominant models (OR = 1.463, 95% CI = 1.097-2.007, P = .018; OR = 1.383, 95% CI = 1.124-1.701, P = .002). Stratification by ethnicity indicated that allele H related to SLE in European populations (OR = 0.736, 95% CI = 0.617-0.879, P = .001), and the recessive model correlated with SLE in Asians (OR = 0.808, 95% CI = 0.667-0.979, P = .03).

CONCLUSION

The present study suggests that A\B and A\O polymorphisms were associated with SLE susceptibility, and the allele H may be a protective factor in SLE.

[Application of CT angiography for detecting perforator vessels in the anterolateral thigh flap surgery]

Objective:To investigate the effectiveness of CT angiography for detecting perforator vessels in the anterolateral thigh flap surgery.Method:Eight patients with oral and maxillofacial soft tissue defects were repaired with the anterolateral thigh flap. Preoperative CT angiography was conducted to detect perforator vessels. The perforator vessel with large diameter, abundant blood flow, and long pedicle was selected to design the flap.Result:The position of perforator vessels during operation were completely consistent with those detected by CT angiography. The anterolateral thigh flap for 8 cases tally survived.One case had the vascular crisis，which was survived after surgical exploration. The flap ranged from 10 cm×5 cm to 6 cm×4 cm in size. Donor recipient site was not severe complications.Conclusion: CT angiography can be used to locate the perforator vessels preoperatively and accurately，which is beneficial for surgeons to design the flap and reduce surgical complications. The anterolateral thigh flap is an effective method of repairing oral and maxillofacial soft tissue defects.