Comprehensive disease control in systemic lupus erythematosus

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.

Fragility fractures in lupus patients: Associated factors and comparison of four fracture risk assessment tools

OBJECTIVE

Osteoporosis (OP) and fragility fractures (FF) are common comorbidities in patients with systemic lupus erythematosus (SLE). This study aimed to (1) assess the prevalence of these conditions in a cohort of SLE patients (2) evaluate the risk factors associated with FF, and (3) compare the accuracy of four different FF risk assessment algorithms to determine which performs better in this specific rheumatologic population.

MATERIALS AND METHODS

We conducted a cross-sectional study with SLE women who underwent bone mineral density assessment by dual-energy X-ray absorptiometry (DEXA) within 3 months of their last visit. Conventional radiology methods were used to evaluate the presence of FF. The 10-year risk of osteoporotic fractures was estimated using four tools: DeFRA, FRAX (adjusted for GC dosage), GARVAN, and QFracture. The comparison of these computational tools was analyzed by the area under the receiver operating characteristic (ROC) curves.

RESULTS

We analyzed 86 SLE patients with a median age of 56 years (IQR 12.1) and a median age at diagnosis of 34 years (IQR 17.2). The median T-score values at the femoral neck and lumbar spine were -1.6 (IQR 0.9) and -1.7 (IQR 1.1), respectively. Of the patients, 33 (38.4%) had OP, with 13 patients (15.1%) experiencing FF. Univariate analysis showed that the presence of FF was associated with thrombocytopenia (p = .01), hemolytic anemia (p = .0001), and the intake of cyclosporine A (p = .002), cyclophosphamide (p = .006), and rituximab (p = .001). The median 10-year risk of major FF for the four calculation tools were as follows: DeFRA 9.85 (IQR 8.6); FRAX GC 8.8 (IQR11.7); GARVAN 12 (IQR 8.2); QFracture 4.1 (IQR 5.8). We observed a significant correlation among all instruments evaluated (p < .0001); in particular, the best correlation was recorded between the FRAX GC and the DeFRA (r = 0.85). DeFRA was the best tool for this population with an AUC of 0.94 (p < .0001, CI 0.88-1).

CONCLUSIONS

OP is a common comorbidity in SLE patients, even in younger patients. FF appears to be more frequent in patients with hematologic involvement. The comparison of the four algorithms shows that DeFRA is the most accurate tool and should be applied to SLE patients.

Clinical utility of mesenchymal stem/stromal cells in regenerative medicine and cellular therapy

Mesenchymal stem/stromal cells (MSCs) have been carefully examined to have tremendous potential in regenerative medicine. With their immunomodulatory and regenerative properties, MSCs have numerous applications within the clinical sector. MSCs have the properties of multilineage differentiation, paracrine signaling, and can be isolated from various tissues, which makes them a key candidate for applications in numerous organ systems. To accentuate the importance of MSC therapy for a range of clinical indications, this review highlights MSC-specific studies on the musculoskeletal, nervous, cardiovascular, and immune systems where most trials are reported. Furthermore, an updated list of the different types of MSCs used in clinical trials, as well as the key characteristics of each type of MSCs are included. Many of the studies mentioned revolve around the properties of MSC, such as exosome usage and MSC co-cultures with other cell types. It is worth noting that MSC clinical usage is not limited to these four systems, and MSCs continue to be tested to repair, regenerate, or modulate other diseased or injured organ systems. This review provides an updated compilation of MSCs in clinical trials that paves the way for improvement in the field of MSC therapy.

Organ damage in Systemic Lupus Erythematosus patients: A multifactorial phenomenon

The prevention of chronic damage, especially in early disease phases, remains an unmet need in the management of Systemic Lupus Erythematous (SLE) patients, despite the application of a so-called treat-to-target strategy. The high proportion of SLE patients developing chronic damage suggests a multifactorial aetiology. Thus, besides disease activity, other factors may contribute to the development of damage. The revision of data published so far underlines that, next to disease activity, it is possible to identify other factors playing a relevant role in damage development and progression. In summary, the presence of antiphospholipid antibodies and drugs used to treat SLE patients, in particular glucocorticoids, is strongly associated with SLE-related damage. Furthermore, recent data suggests the possible role of genetic background in determining the development of specific organ damage, in particular renal and neurological. Nonetheless, demographic factors, such as age, sex and disease duration could exert a role along with the presence of comorbidities. The contribution of different factors in determining damage development suggests the need for new outcomes to assess a comprehensive disease control including not only the assessment of disease activity, but also the evaluation of chronic damage development.

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.

Current state and completeness of reporting clinical prediction models using machine learning in systemic lupus erythematosus: A systematic review

OBJECTIVE

We carried out a systematic review (SR) of adherence in diagnostic and prognostic applications of ML in SLE using the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) Statement.

METHODS

A SR employing five databases was conducted from its inception until December 2021. We identified articles that evaluated the utilization of ML for prognostic and/or diagnostic purposes. This SR was reported based on the PRISMA guidelines. The TRIPOD statement assessed adherence to reporting standards. Assessment for risk of bias was done using PROBAST tool.

RESULTS

We included 45 studies: 29 (64.4%) diagnostic and 16 (35.5%) prognostic prediction- model studies. Overall, articles adhered by between 17% and 67% (median 43%, IQR 37-49%) to TRIPOD items. Only few articles reported the model's predictive performance (2.3%, 95% CI 0.06-12.0), testing of interaction terms (2.3%, 95% CI 0.06-12.0), flow of participants (50%, 95% CI; 34.6-65.4), blinding of predictors (2.3%, 95% CI 0.06-12.0), handling of missing data (36.4%, 95% CI 22.4-52.2), and appropriate title (20.5%, 95% CI 9.8-35.3). Some items were almost completely reported: the source of data (88.6%, 95% CI 75.4-96.2), eligibility criteria (86.4%, 95% CI 76.2-96.5), and interpretation of findings (88.6%, 95% CI 75.4-96.2). In addition, most of model studies had high risk of bias.

CONCLUSIONS

The reporting adherence of ML-based model developed for SLE, is currently inadequate. Several items deemed crucial for transparent reporting were not fully reported in studies on ML-based prediction models.

REVIEW REGISTRATION

PROSPERO ID# CRD42021284881. (Amended to limit the scope).

The Impacts of the Clinical and Genetic Factors on Chronic Damage in Caucasian Systemic Lupus Erythematosus Patients

Objective: The purpose of this study was to determine the distribution of organ damage in a cohort of systemic lupus erythematosus (SLE) patients and to evaluate the roles of clinical and genetic factors in determining the development of chronic damage. Methods: Organ damage was assessed by the SLICC Damage Index (SDI). We analyzed a panel of 17 single-nucleotide polymorphism (SNPs) of genes already associated with SLE, and we performed a phenotype−genotype correlation analysis by evaluating specific domains of the SDI. Results: Among 175 Caucasian SLE patients, 105 (60%) exhibited damage (SDI ≥1), with a median value of 1.0 (IQR 3.0). The musculoskeletal (26.2%), neuropsychiatric (24.6%) and ocular domains (20.6%) were involved most frequently. The presence of damage was associated with higher age, longer disease duration, neuropsychiatric (NP) manifestations, anti-phospholipid syndrome and the positivity of anti-dsDNA. Concerning therapies, cyclophosphamide, mycophenolate mofetil and glucocorticoids were associated with the development of damage. The genotype−phenotype correlation analysis showed an association between renal damage, identified in 6.9% of patients, and rs2205960 of TNFSF4 (p = 0.001; OR 17.0). This SNP was significantly associated with end-stage renal disease (p = 0.018, OR 9.68) and estimated GFR < 50% (p = 0.025, OR 1.06). The rs1463335 of MIR1279 gene was associated with the development of NP damage (p = 0.029; OR 2.783). The multivariate logistic regression analysis confirmed the associations between TNFSF4 rs2205960 SNP and renal damage (p = 0.027, B = 2.47) and between NP damage and rs1463335 of MIR1279 gene (p = 0.014, B = 1.29). Conclusions: Our study could provide new insights into the role of genetic background in the development of renal and NP damage.