Machine learning predicts stem cell transplant response in severe scleroderma

[Autologous peripheral stem cell transplantation in systemic sclerosis: An important step forward, but we must temper our enthusiasm!]

Three prospective randomized studies have demonstrated the efficacy of autologous hematopoietic stem cell (HSC) transplantation in systemic sclerosis (SSc) on survival. These results encourage us to offer this therapy to patients who have a rapidly progressive disease and who have early symptoms but no advanced visceral involvement. HSC autograft can thus be discussed in patients with diffuse cutaneous SSc with a duration of the disease since the first visceral manifestations (cutaneous, cardiac, digestive, pulmonary, or renal) excluding Raynaud's phenomenon of less than 5 years. However, the indications for HSC autograft in SSc validated at European level and in the national diagnostic and care protocol (PNDS) are broader and some of these indications are debatable, in particular in patients with worsening diffuse interstitial lung disease. These indications are discussed in a reasoned way, taking into account the level of evidence and the toxicity of the HSC autograft.

Is cyclophosphamide still the gold standard in early severe rapidly progressive systemic sclerosis?

Cyclophosphamide (CYC) has been a gold standard of treatment for severe progressive Systemic Sclerosis (SSc), especially in patients with concomitant interstitial lung disease (ILD). This approach was based on results of several interventional studies, including randomized control trials, which mainly addressed SSc-ILD as a primary end point and skin involvement as a second one. The use of CYC is time-limited due to significant adverse events. More recently, other immunosuppressive and biological agents showed efficacy but better safety profile in patients with SSc and SSc-ILD. With regards to other end-points, post-hoc analyses, systematic reviews and metalysis showed that CYC had limited influence on patients' quality of life, event-free survival and mortality. Comprehensive patient's stratification according to a molecular, cellular and phenotypic pattern may help in choosing of personalized medicine with more ambitious treatment effect and should be the future direction. According to the above available data and even if scientific evidence may be missing, experts' opinion has changed the attitude to CYC as an anchor drug in the management of severe SSc. Indeed, CYC has been pushed to the second and even third treatment option after mycophenolate mofetil, tocilizumab or rituximab. This position became obvious during debate on this topic at CORA meeting 2023.

[Autologous peripheral stem cell transplantation in systemic sclerosis: An important step forward, but we must temper our enthusiasm!]

Three prospective randomized studies have demonstrated the efficacy of autologous hematopoietic stem cell (HSC) transplantation in systemic sclerosis (SSc) on survival. These results encourage us to offer this therapy to patients who have a rapidly progressive disease and who have early symptoms but no advanced visceral involvement. HSC autograft can thus be discussed in patients with diffuse cutaneous SSc with a duration of the disease since the first visceral manifestations (cutaneous, cardiac, digestive, pulmonary, or renal) excluding Raynaud's phenomenon of less than 5 years. However, the indications for HSC autograft in SSc validated at European level and in the national diagnostic and care protocol (PNDS) are broader and some of these indications are debatable, in particular in patients with worsening diffuse interstitial lung disease. These indications are discussed in a reasoned way, taking into account the level of evidence and the toxicity of the HSC autograft.

The cancer-associated glycan polysialic acid is dysregulated in systemic sclerosis and is associated with fibrosis

OBJECTIVE

Systemic sclerosis (SSc) is a rare but deadly disease characterized by autoimmunity, vasculopathy, and fibrosis. Fibrotic complications associated with SSc correlate with severe morbidity and mortality. Previous studies in SSc have identified fibroblasts as the primary drivers of fibrosis; however, the mechanism(s) promoting this are not well understood. Aberrant glycosylation, particularly polysialylation (polySia), has been described as a prominent feature of aggressive cancers. Inspired by this observation, we aimed to determine if polySia is dysregulated in various forms of SSc.

METHODS

All patients with SSc met the 2013 ACR/EULAR. Patients were sub-classified into limited cutaneous (lSSc, N = 5 or 46 patients for polySia quantification in the dermis or serum; respectively), diffuse cutaneous (dSSc, N = 11 or 18 patients for polySia quantification in the dermis or serum; respectively), or patients with dSSc treated with an autologous stem cell transplantation (post-ASCT, N = 4 patients for quantification in the dermis). Dermal polySia levels were measured via immunofluorescence microscopy in 10 μm dermal sections, quantified in each group (healthy volunteers (HC), lSSc, dSSc, and post-ASCT) and correlated with skin fibrosis (via the modified Rodnan skin score (mRSS)). Similarly, serum polySia was quantified in each group, and correlated with the mRSS.

RESULTS

Dermal polySia levels were highest in patients with dSSc (compared to HC < 0.001), and correlated with the degree of fibrosis in all of the groups (P = 0.008). Serum polySia was higher in all SSc groups (p < 0.001) and correlated with the severity of mRSS (p < 0.0001).

CONCLUSION

Polysia is more abundant in the skin and sera from patients with SSc and correlates with the degree of skin fibrosis. The aberrant expression of polySia highlights its potential use as a biomarker in patients with progressive forms of SSc. Dysregulated polySia levels in SSc further emphasizes the cancer-like phenotype present in SSc, which may promote fibrosis and immune dysregulation.

Deep learning-enabled quantification of simultaneous PET/MRI for cell transplantation monitoring

Current methods of in vivo imaging islet cell transplants for diabetes using magnetic resonance imaging (MRI) are limited by their low sensitivity. Simultaneous positron emission tomography (PET)/MRI has greater sensitivity and ability to visualize cell metabolism. However, this dual-modality tool currently faces two major challenges for monitoring cells. Primarily, the dynamic conditions of PET such as signal decay and spatiotemporal change in radioactivity prevent accurate quantification of the transplanted cell number. In addition, selection bias from different radiologists renders human error in segmentation. This calls for the development of artificial intelligence algorithms for the automated analysis of PET/MRI of cell transplantations. Here, we combined K-means++ for segmentation with a convolutional neural network to predict radioactivity in cell-transplanted mouse models. This study provides a tool combining machine learning with a deep learning algorithm for monitoring islet cell transplantation through PET/MRI. It also unlocks a dynamic approach to automated segmentation and quantification of radioactivity in PET/MRI.

Three Distinct Transcriptional Profiles of Monocytes Associate with Disease Activity in Scleroderma Patients

OBJECTIVE

Patients with diffuse cutaneous systemic sclerosis (dcSSc) display a complex clinical phenotype. Transcriptional profiling of whole blood or tissue from patients are affected by changes in cellular composition that drive gene expression and an inability to detect minority cell populations. We undertook this study to focus on the 2 main subtypes of circulating monocytes, classical monocytes (CMs) and nonclassical monocytes (NCMs) as a biomarker of SSc disease severity.

METHODS

SSc patients were recruited from the Prospective Registry for Early Systemic Sclerosis. Clinical data were collected, as well as peripheral blood for isolation of CMs and NCMs. Age-, sex-, and race-matched healthy volunteers were recruited as controls. Bulk macrophages were isolated from the skin in a separate cohort. All samples were assayed by RNA sequencing (RNA-seq).

RESULTS

We used an unbiased approach to cluster patients into 3 groups (groups A-C) based on the transcriptional signatures of CMs relative to controls. Each group maintained their characteristic transcriptional signature in NCMs. Genes up-regulated in group C demonstrated the highest expression compared to the other groups in SSc skin macrophages, relative to controls. Patients from groups B and C exhibited worse lung function than group A, although there was no difference in SSc skin disease at baseline, relative to controls. We validated our approach by applying our group classifications to published bulk monocyte RNA-seq data from SSc patients, and we found that patients without skin disease were most likely to be classified as group A.

CONCLUSION

We are the first to show that transcriptional signatures of CMs and NCMs can be used to unbiasedly stratify SSc patients and correlate with disease activity outcome measures.

Differentially expressed genes in systemic sclerosis: Towards predictive medicine with new molecular tools for clinicians

Systemic sclerosis (SSc) is a rare and chronic autoimmune disease characterized by a pathogenic triad of immune dysregulation, vasculopathy, and progressive fibrosis. Clinical tools commonly used to assess patients, such as the modified Rodnan skin score, difference between limited or diffuse forms of skin involvement, presence of lung, heart or kidney involvement, or of various autoantibodies, are important prognostic factors, but still fail to reflect the large heterogeneity of the disease. SSc treatment options are diverse, ranging from conventional drugs to autologous hematopoietic stem cell transplantation, and predicting response is challenging. Genome-wide technologies, such as high throughput microarray analyses and RNA sequencing, allow accurate, unbiased, and broad assessment of alterations in expression levels of multiple genes. In recent years, many studies have shown robust changes in the gene expression profiles of SSc patients compared to healthy controls, mainly in skin tissues and peripheral blood cells. The objective analysis of molecular patterns in SSc is a powerful tool that can further classify SSc patients with similar clinical phenotypes and help predict response to therapy. In this review, we describe the journey from the first discovery of differentially expressed genes to the identification of enriched pathways and intrinsic subsets identified in SSc, using machine learning algorithms. Finally, we discuss the use of these new tools to predict the efficacy of various treatments, including stem cell transplantation. We suggest that the use of RNA gene expression-based classifications according to molecular subsets may bring us one step closer to precision medicine in Systemic Sclerosis.

Clinical and Molecular Findings After Autologous Stem Cell Transplantation or Cyclophosphamide for Scleroderma: Handling Missing Longitudinal Data

OBJECTIVE

Among individuals with systemic sclerosis (SSc) randomized to cyclophosphamide (CYC) (n = 34) or hematopoietic stem cell transplantation (HSCT) (n = 33), we examined longitudinal trends of clinical, pulmonary function, and quality of life measures while accounting for the influence of early failures on treatment comparisons.

METHODS

Assuming that data were missing at random, mixed-effects regression models were used to estimate longitudinal trends for clinical measures when comparing treatment groups. Results were compared to observed means and to longitudinal trends estimated from shared parameter models, assuming that data were missing not at random. Longitudinal trends for SSc intrinsic molecular subsets defined by baseline gene expression signatures (normal-like, inflammatory, and fibroproliferative signatures) were also studied.

RESULTS

Available observed means for pulmonary function tests appeared to improve over time in both arms. However, after accounting for participant loss, forced vital capacity in HSCT recipients increased by 0.77 percentage points/year but worsened by -3.70/year for CYC (P = 0.004). Similar results were found for diffusing capacity for carbon monoxide and quality of life indicators. Results for both analytic models were consistent. HSCT recipients in the inflammatory (n = 20) and fibroproliferative (n = 20) subsets had superior long-term trends compared to CYC for pulmonary and quality of life measures. HSCT was also superior for modified Rodnan skin thickness scores in the fibroproliferative subset. For the normal-like subset (n = 22), superiority of HSCT was less apparent.

CONCLUSION

Longitudinal trends estimated from 2 statistical models affirm the efficacy of HSCT over CYC in severe SSc. Failure to account for early loss of participants may distort estimated clinical trends over the long term.

A genomic meta-analysis of clinical variables and their association with intrinsic molecular subsets in systemic sclerosis

OBJECTIVES

Four intrinsic molecular subsets (inflammatory, fibroproliferative, limited, normal-like) have previously been identified in SSc and are characterized by unique gene expression signatures and pathways. The intrinsic subsets have been linked to improvement with specific therapies. Here, we investigated associations between baseline demographics and intrinsic molecular subsets in a meta-analysis of published datasets.

METHODS

Publicly available gene expression data from skin biopsies of 311 SSc patients measured by DNA microarray were classified into the intrinsic molecular subsets. RNA-sequencing data from 84 participants from the ASSET trial were used as a validation cohort. Baseline clinical demographics and intrinsic molecular subsets were tested for statistically significant associations.

RESULTS

Males were more likely to be classified in the fibroproliferative subset (P = 0.0046). SSc patients who identified as African American/Black were 2.5 times more likely to be classified as fibroproliferative compared with White/Caucasian patients (P = 0.0378). ASSET participants sera positive for anti-RNA pol I and RNA pol III autoantibodies were enriched in the inflammatory subset (P = 5.8 × 10-5, P = 9.3 × 10-5, respectively), while anti-Scl-70 was enriched in the fibroproliferative subset. Mean modified Rodnan Skin Score (mRSS) was statistically higher in the inflammatory and fibroproliferative subsets compared with normal-like (P = 0.0027). The average disease duration for inflammatory subset was less than fibroproliferative and normal-like intrinsic subsets (P = 8.8 × 10-4).

CONCLUSIONS

We identified multiple statistically significant differences in baseline demographics between the intrinsic subsets that may represent underlying features of disease pathogenesis (e.g. chronological stages of fibrosis) and have implications for treatments that are more likely to work in certain SSc populations.

Hematopoietic Cell Transplantation for Systemic Sclerosis-A Review

Systemic sclerosis (SSc) is an autoimmune, multi-organ, connective tissue disease associated with significant morbidity and mortality. Conventional immunosuppressive therapies demonstrate limited efficacy. Autologous hematopoietic stem cell transplantation (HCT) is more efficacious but carries associated risks, including treatment-related mortality. Here, we review HCT as a treatment for SSc, its efficacy and toxicity in comparison to conventional therapies, and the proposed mechanisms of action. Furthermore, we discuss the importance of and recent developments in patient selection. Finally, we highlight the knowledge gaps and future work required to further improve patient outcomes.

Precision medicine: the precision gap in rheumatic disease

For many oncological conditions, the application of timely and patient-tailored targeted therapies, or precision medicine, is a major therapeutic development that has provided considerable clinical benefit. However, despite the application of increasingly sophisticated technologies, alongside advanced bioinformatic and machine-learning algorithms, this success is yet to be replicated for the rheumatic diseases. In rheumatoid arthritis, for example, despite an array of targeted biologic and conventional therapeutics, treatment choice remains largely based on trial and error. The concept of the 'precision gap' for rheumatic disease can help us to identify factors that underpin the slow progress towards the discovery and adoption of precision-medicine approaches for rheumatic disease. In a rheumatic disease such as rheumatoid arthritis, it is possible to identify four themes that have slowed progress, solutions to which should help to close the precision gap. These themes relate to our fundamental understanding of disease pathogenesis, how we determine treatment response, confounders of treatment outcomes and trial design.

The Use and Utility of Machine Learning in Achieving Precision Medicine in Systemic Sclerosis: A Narrative Review

Background: Systemic sclerosis (SSc) is a rare connective tissue disease that can affect different organs and has extremely heterogenous presentations. This complexity makes it difficult to perform an early diagnosis and a subsequent subclassification of the disease. This hinders a personalized approach in clinical practice. In this context, machine learning (ML), a branch of artificial intelligence (AI), is able to recognize relationships in data and predict outcomes. Methods: Here, we performed a narrative review concerning the application of ML in SSc to define the state of art and evaluate its role in a precision medicine context. Results: Currently, ML has been used to stratify SSc patients and identify those at high risk of severe complications. Additionally, ML may be useful in the early detection of organ involvement. Furthermore, ML might have a role in target therapy approach and in predicting drug response. Conclusion: Available evidence about the utility of ML in SSc is sparse but promising. Future improvements in this field could result in a big step toward precision medicine. Further research is needed to define ML application in clinical practice.

A systematic review of the predicted outcomes related to hematopoietic stem cell transplantation: focus on applied machine learning methods' performance

INTRODUCTION

: Hematopoietic stem cell transplantation (HSCT) is a critical therapeutic procedure in blood diseases, and the investigation of HSCT data can provide valuable information. Machine learning (ML) techniques are novel and useful data analysis tools that have been applied in many studies to predict HSCT survival and estimate the risk of transplantation.

AREAS COVERED

: A systematic review was performed with a search of PubMed, Science Direct, Embase, Scopus, and the European Society for Blood and Marrow Transplantation, the Center for International Blood and Marrow Transplant Research, and the American Society for Transplantation and Cellular Therapy publications for articles published by September 2020.

EXPERT OPINION

: After investigating the results, 24 papers that met eligibility criteria were included in this study. The applied ML algorithms with the highest performance were Random Survival Forests (AUC=0.72) for survival-related, Random Survival Forests and Logistic Regression (AUC=0.77) for mortality-related, Deep Learning (AUC=0.8) for relapse, L2-Regularized Logistic Regression (AUC=0.66) for Acute-Graft Versus Host Disease, Random Survival Forests (AUC=0.88) for sepsis, Elastic-Net Regression (AUC=0.89) for cognitive impairment, and Bayesian Network (AUC=0.997) for oral mucositis outcome. This review reveals the potential of ML techniques to predict HSCT outcomes and apply them to developing clinical decision support systems.

Life after Autologous Hematopoietic Stem Cell Transplantation for Systemic Sclerosis

Stem cell transplantation has been investigated as treatment for severe and progressive systemic sclerosis (SSc) for the past 25 years. To date, more than 1000 SSc patients have been transplanted worldwide. Overall and event-free survival have increased over the years, reflecting stricter patient selection criteria and better clinical management strategies. This review addresses long-term outcomes of transplanted SSc patients, considering phase I/II and randomized clinical trials, as well as observational studies and those assessing specific aspects of the disease. Clinical outcomes are discussed comparatively between studies, highlighting advances, drawbacks and controversies in the field. Areas for future development are also discussed.

Current Concepts on the Pathogenesis of Systemic Sclerosis

From the clinical standpoint, systemic sclerosis (SSc) is characterized by skin and internal organ fibrosis, diffuse fibroproliferative vascular modifications, and autoimmunity. Clinical presentation and course are highly heterogenous and life expectancy variably affected mostly dependent on lung and heart involvement. SSc touches more women than men with differences in disease severity and environmental exposure. Pathogenetic events originate from altered homeostasis favored by genetic predisposition, environmental cues and a variety of endogenous and exogenous triggers. Epigenetic modifications modulate SSc pathogenesis which strikingly associate profound immune-inflammatory dysregulation, abnormal endothelial cell behavior, and cell trans-differentiation into myofibroblasts. SSc myofibroblasts show enhanced survival and enhanced extracellular matrix deposition presenting altered structure and altered physicochemical properties. Additional cell types of likely pathogenic importance are pericytes, platelets, and keratinocytes in conjunction with their relationship with vessel wall cells and fibroblasts. In SSc, the profibrotic milieu is favored by cell signaling initiated in the one hand by transforming growth factor-beta and related cytokines and in the other hand by innate and adaptive type 2 immune responses. Radical oxygen species and invariant receptors sensing danger participate to altered cell behavior. Conventional and SSc-specific T cell subsets modulate both fibroblasts as well as endothelial cell dysfunction. Beside autoantibodies directed against ubiquitous antigens important for enhanced clinical classification, antigen-specific agonistic autoantibodies may have a pathogenic role. Recent studies based on single-cell RNAseq and multi-omics approaches are revealing unforeseen heterogeneity in SSc cell differentiation and functional states. Advances in system biology applied to the wealth of data generated by unbiased screening are allowing to subgroup patients based on distinct pathogenic mechanisms. Deciphering heterogeneity in pathogenic mechanisms will pave the way to highly needed personalized therapeutic approaches.

Transcriptomic Evaluation of Juvenile Localized Scleroderma Skin With Histologic and Clinical Correlation

OBJECTIVE

Juvenile localized scleroderma (LS) is an autoimmune disease of the skin whose pathogenesis is not well understood due to the rarity of the disease. This study was undertaken to determine the skin transcriptome in skin biopsy tissue from children with juvenile LS compared to pediatric healthy controls, with identification of significant molecular targets using RNA sequencing (RNA-Seq). In this study, differentially expressed genes (DEGs) were assessed for correlations with histopathologic and clinical features in children with juvenile LS, and were used to group the children into distinct genetic clusters based on immunophenotype.

METHODS

RNA-Seq was performed on sections of paraffin-embedded skin tissue obtained from 28 children with juvenile LS and 10 pediatric healthy controls. RNA-Seq was carried out using an Illumina HTS TruSeq RNA Access library prep kit, with data aligned using STAR and data analysis using a DESeq2 platform. A standardized histologic scoring system was used to score skin sections for the severity of inflammation and levels of collagen deposition. Histologic scoring was completed by 2 pathologists who were blinded with regard to the status of each sample. Spearman's rank correlation coefficients were used to assess significant correlations between DEG expression profiles and skin histologic findings in patients with juvenile LS.

RESULTS

We identified 589 significant DEGs in children with juvenile LS as compared to healthy controls. Hierarchical clustering was used to demonstrate 3 distinct juvenile LS immunophenotype clusters. The histologic scores of skin inflammation (based on numbers and categories of inflammatory cell infiltrates) were significantly correlated with the expression levels of HLA-DPB1, HLA-DQA2, HLA-DRA, and STAT1 genes (r_s > 0.5, P < 0.01). Collagen thickness correlated with the expression levels of collagen organization genes as well as with genes found to be correlated with the severity of inflammation, including genes for major histocompatibility complex (MHC) class I, MHC class II, and interferon-γ signaling.

CONCLUSION

Among children with juvenile LS, 3 distinct genetic signatures, or clusters, were identified. In one cluster, inflammation-related pathways were up-regulated, corresponding to the histologic skin inflammation score. In the second cluster, fibrosis-related pathways were up-regulated. In the third cluster, gene expression in the skin corresponded to the patterns seen in healthy controls. Up-regulation of HLA class II genes was observed within the first cluster (characterized by predominant inflammation), a feature that has also been observed in the peripheral blood of patients with morphea and in the skin of patients with systemic sclerosis.

Hematopoietic stem cell transplantation in systemic sclerosis: Yes!! BUT

Systemic sclerosis is a chronic multisystem, autoimmune disease with high mortality, and to date, has no effective approved therapy. For patients with the most severe forms of systemic sclerosis, autologous hematopoietic stem cell therapy has been proven to be effective, as demonstrated by three randomized controlled clinical trials and at least two registry studies. In the controlled studies, autologous hematopoietic stem cell therapy was shown to improve mortality, skin involvement, quality-of-life, and function compared with patients treated with intravenous cyclophosphamide. There is significant transplant-related mortality associated with autologous hematopoietic stem cell therapy, as well as infectious and noninfectious serious adverse events, including cardiac and renal compromise. Although the results of autologous hematopoietic stem cell therapy clinical trials to date are highly encouraging, a number of important unanswered questions remain. These include the following: What is the optimal transplant regimen; can the spectrum of systemic sclerosis patients who respond to autologous hematopoietic stem cell therapy be expanded; how can disease relapse following autologous hematopoietic stem cell therapy be predicted and managed; can treatment-related toxicity and mortality be mitigated; can response or toxicity be predicted; and finally, is there a place for allogeneic stem cell transplants in systemic sclerosis? Systemic sclerosis is an autoimmune disease with multiple immunological, vascular, and fibrotic abnormalities. Immunosuppressive therapy is frequently used, but to date its efficacy has been generally modest. Stem cell transplantation is the most potent immunosuppressive therapy currently available, and as discussed below, may have an important role in the management of systemic sclerosis.

Genetic Signatures From RNA Sequencing of Pediatric Localized Scleroderma Skin

The purpose of this study was to explore the skin transcriptional profile in pediatric localized scleroderma (LS) to provide a better understanding of the altered immune and fibrotic pathways promoting disease. LS is a progressive disease of the skin and underlying tissue that causes significant functional disability and disfigurement, especially in developing children. RNA sequencing (RNAseq) technology allows for improved understanding of relevant cellular expression through transcriptome analysis of phases during LS disease progression (more active/inflammatory vs. inactive/fibrotic) and also permits the use of RNA extracted from existing paraffin-embedded skin tissue, which is important in pediatrics. A strong correlation was observed between the comparison of genes expressed between fresh (RNAlater) and paraffinized skin in healthy and LS subjects, supporting the use of paraffinized tissue. LS gene signatures compared to healthy controls showed a distinct expression of an inflammatory response gene signature (IRGS) composed of IFNγ-, IFNα-, and TNFα-associated genes. GSEA^© enrichment analysis showed that the IRGS, including interferon-inducible chemokines such as CXCL9, CXCL10, CXCL11, and IFNγ itself, was more highly expressed in LS patients with more inflammatory lesions. The use of paraffinized skin for sequencing was proven to be an effective substitute for fresh skin by comparing gene expression profiles. The prevalence of the IFNγ signature in the lesion biopsies of active LS patients indicates that these genes reflect clinical activity parameters and may be the promoters of early, inflammatory disease.