Systemic lupus erythematosus phenotypes formed from machine learning with a specific focus on cognitive impairment

Outcome clusters and their stability over 1 year in patients with SLE: self-reported and performance-based cognitive function, disease activity, mood and health-related quality of life

OBJECTIVE

To determine if self-reported fatigue, anxiety, depression, cognitive difficulties, health-related quality of life, disease activity scores and neuropsychological battery (NB) cluster into distinct groups in patients with SLE based on symptom intensity and if they change at 1-year follow-up.

METHODS

This is a retrospective analysis of consecutive consenting patients, followed at a single centre. Patients completed a comprehensive NB, the Beck Anxiety Inventory, Beck Depression Inventory, Fatigue Severity Scale, Short-Form Health Survey Physical Component Summary and Mental Component Summary scores and the Perceived Deficits Questionnaire. Disease activity was assessed by Systemic Lupus Erythematosus Disease Activity Index 2000. Ward's method was used for clustering and principal component analysis was used to visualise the number of clusters. Stability at 1 year was assessed with kappa statistic.

RESULTS

Among 142 patients, three clusters were found: cluster 1 had mild symptom intensity, cluster 2 had moderate symptom intensity and cluster 3 had severe symptom intensity. At 1-year follow-up, 49% of patients remained in their baseline cluster. The mild cluster had the highest stability (77% of patients stayed in the same cluster), followed by the severe cluster (51%), and moderate cluster had the lowest stability (3%). A minority of patients from mild cluster moved to severe cluster (19%). In severe cluster, a larger number moved to moderate cluster (40%) and fewer to mild cluster (9%).

CONCLUSION

Three distinct clusters of symptom intensity were documented in patients with SLE in association with cognitive function. There was a lower tendency for patients in the mild and severe clusters to move but not moderate cluster over the course of a year. This may demonstrate an opportunity for intervention to have moderate cluster patients move to mild cluster instead of moving to severe cluster. Further studies are necessary to assess factors that affect movement into moderate cluster.

Systemic lupus in the era of machine learning medicine

Artificial intelligence and machine learning applications are emerging as transformative technologies in medicine. With greater access to a diverse range of big datasets, researchers are turning to these powerful techniques for data analysis. Machine learning can reveal patterns and interactions between variables in large and complex datasets more accurately and efficiently than traditional statistical methods. Machine learning approaches open new possibilities for studying SLE, a multifactorial, highly heterogeneous and complex disease. Here, we discuss how machine learning methods are rapidly being integrated into the field of SLE research. Recent reports have focused on building prediction models and/or identifying novel biomarkers using both supervised and unsupervised techniques for understanding disease pathogenesis, early diagnosis and prognosis of disease. In this review, we will provide an overview of machine learning techniques to discuss current gaps, challenges and opportunities for SLE studies. External validation of most prediction models is still needed before clinical adoption. Utilisation of deep learning models, access to alternative sources of health data and increased awareness of the ethics, governance and regulations surrounding the use of artificial intelligence in medicine will help propel this exciting field forward.

Serum S100A8/A9 and MMP-9 levels are elevated in systemic lupus erythematosus patients with cognitive impairment

Objective

Cognitive impairment (CI) is one of the most common manifestations of Neuropsychiatric Systemic Lupus Erythematosus (NPSLE). Despite its frequency, we have a limited understanding of the underlying immune mechanisms, resulting in a lack of pathways to target. This study aims to bridge this gap by investigating differences in serum analyte levels in SLE patients based on their cognitive performance, independently from the attribution to SLE, and exploring the potential for various serum analytes to differentiate between SLE patients with and without CI.

Methods

Two hundred ninety individuals aged 18-65 years who met the 2019-EULAR/ACR classification criteria for SLE were included. Cognitive function was measured utilizing the adapted ACR-Neuropsychological Battery (ACR-NB). CI was defined as a z-score of ≤-1.5 in two or more domains. The serum levels of nine analytes were measured using ELISA. The data were randomly partitioned into a training (70%) and a test (30%) sets. Differences in the analyte levels between patients with and without CI were determined; and their ability to discriminate CI from non-CI was evaluated.

Results

Of 290 patients, 40% (n=116) had CI. Serum levels of S100A8/A9 and MMP-9, were significantly higher in patients with CI (p=0.006 and p=0.036, respectively). For most domains of the ACR-NB, patients with CI had higher S100A8/A9 serum levels than those without. Similarly, S100A8/A9 had a negative relationship with multiple CI tests and the highest AUC (0.74, 95%CI: 0.66-0.88) to differentiate between patients with and without CI.

Conclusion

In this large cohort of well-characterized SLE patients, serum S100A8/A9 and MMP-9 were elevated in patients with CI. S100A8/A9 had the greatest discriminatory ability in differentiating between patients with and without CI.

The Effects of Systemic Lupus-Related Cognitive Impairments on Activities of Daily Living and Life Role Participation: A Qualitative Framework Study

OBJECTIVE

Cognitive impairment (CI) in systemic lupus erythematosus (SLE) negatively impacts health-related quality of life leading to activity limitations. This qualitative study aimed to (1) explore the effect of SLE-related CI on activities of daily living and life role participation and (2) describe factors influencing activity restriction and life role participation.

METHODS

Semistructured, in-depth interviews of lived experience of CI in SLE were conducted with 24 participants with SLE. Sociodemographic and clinical data, and objective and subjective cognitive function, were collected to characterize participants. A qualitative thematic content analysis was undertaken guided by a framework analytical approach.

RESULTS

Participants reported problems in multiple cognitive domains, with multiple perceived causes. CI was felt to impact work, social, domestic, and family life, health, and independence. Five overarching themes were represented in the data: (1) characterization of SLE-reported CI, (2) perceived cause of CI, (3) perceived impact of CI on activities of daily living and life role participation, (4) adaptations for managing CI, and (5) influence of CI adaptations on activities of daily living and life role participation.

CONCLUSION

This study provides a better understanding of the patient experience of CI in SLE, how it impacts their lives, and what coping strategies they employ. It highlights the long-term challenges those with CI in SLE undergo and provides evidence for the urgent need to implement multidisciplinary treatment options. When managing CI, it may be beneficial to evaluate and understand available psychosocial support resources to help identify and reinforce relevant adaptations to improve health-related quality of life.

SVM-Based Model Combining Patients' Reported Outcomes and Lymphocyte Phenotypes of Depression in Systemic Lupus Erythematosus

BACKGROUND

The incidence of depression in patients with systemic lupus erythematosus (SLE) is high and leads to a lower quality of life than that in undepressed SLE patients and healthy individuals. The causes of SLE depression are still unclear.

METHODS

A total of 94 SLE patients were involved in this study. A series of questionnaires (Hospital Depression Scale, Social Support Rate Scale and so on) were applied. Flow cytometry was used to test the different stages and types of T cells and B cells in peripheral blood mononuclear cells. Univariate and multivariate analyses were conducted to explore the key contributors to depression in SLE. Support Vector Machine (SVM) learning was applied to form the prediction model.

RESULTS

Depressed SLE patients showed lower objective support, severer fatigue, worse sleep quality and higher percentages of ASC%PBMC, ASC%CD19+, MAIT, TEM%Th, TEMRA%Th, CD45RA+CD27-Th, TEMRA%CD8 than non-depressed patients. A learning-based SVM model combining objective and patient-reported variables showed that fatigue, objective support, ASC%CD19+, TEM%Th and TEMRA%CD8 were the main contributing factors to depression in SLE. With the SVM model, the weight of TEM%Th was 0.17, which is the highest among objective variables, and the weight of fatigue was 0.137, which was the highest among variables of patients' reported outcomes.

CONCLUSIONS

Both patient-reported factors and immunological factors could be involved in the occurrence and development of depression in SLE. Scientists can explore the mechanism of depression in SLE or other psychological diseases from the above perspective.

Application of Machine Learning Models in Systemic Lupus Erythematosus

Systemic Lupus Erythematosus (SLE) is a systemic autoimmune disease and is extremely heterogeneous in terms of immunological features and clinical manifestations. This complexity could result in a delay in the diagnosis and treatment introduction, with impacts on long-term outcomes. In this view, the application of innovative tools, such as machine learning models (MLMs), could be useful. Thus, the purpose of the present review is to provide the reader with information about the possible application of artificial intelligence in SLE patients from a medical perspective. To summarize, several studies have applied MLMs in large cohorts in different disease-related fields. In particular, the majority of studies focused on diagnosis and pathogenesis, disease-related manifestations, in particular Lupus Nephritis, outcomes and treatment. Nonetheless, some studies focused on peculiar features, such as pregnancy and quality of life. The review of published data demonstrated the proposal of several models with good performance, suggesting the possible application of MLMs in the SLE scenario.

Cognitive Performance in Patients with Systemic Lupus Erythematosus Using the Ped-ANAM

Computerized batteries have been widely used to investigate cognitive impairment (CI) in patients with SLE. The aim of this study was to evaluate the cognitive performance of patients with SLE in relation to healthy controls using the Pediatric Automated Neuropsychological Assessment Metrics (Ped-ANAM) battery. In addition, we aimed to examine differences in Ped-ANAM scores according to age of disease onset, presence of disease activity, and disease damage. We included 201 consecutive adult-onset (aSLE) and childhood-onset SLE (cSLE) patients who were being followed at the hospital's rheumatology outpatient clinic and 177 healthy controls. We applied the percentage of correct answers on the Ped-ANAM subtests and the Performance Validity Index (PVI) metric and correlated them with the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) and Systemic Lupus Erythematosus Damage Index (SDI). Then, we established their relationships with neuropsychiatric systemic lupus erythematosus (NPSLE). We observed CI in a total of 38 (18.9%) SLE patients and 8 (4.5%) healthy controls (p < 0.001). CI was observed in eight (19.5%) cSLE patients and 32 (20%) aSLE patients (p = 0.8175). Individual analysis of the aSLE subtests showed a significant difference in all subtests compared to healthy controls; the greatest differences were in matching to sample (p < 0.001) and memory search ( p < 0.001). In the cSLE group, we observed a difference in the code substitution subtests (p = 0.0065) compared to the healthy controls. In the evaluation of clinical outcomes, disease activity was significantly correlated with CI in cSLE (r = 0.33; p = 0.042) and aSLE (r = 0.40; p = 0.001). We also observed an association between disease activity and neuropsychiatric manifestations (p = 0.0012) in aSLE. In conclusion, we determined that cognitive dysfunction, mainly in memory and attention, was more prevalent in patients with SLE. In both the cSLE and aSLE groups, disease activity was associated with worse cognitive function. This is the first study to use the Ped-ANAM in Brazil. Longitudinal studies are necessary to determine how the Ped-ANAM will perform over time.