What Comes before Scleroderma?

Abstract

While the classification criteria for systemic sclerosis (SSc) have been carefully delineated, the definition of what comes before meeting classification criteria is not so well understood. In some ways, it is similar to “pre-rheumatoid arthritis” where a reasonable definition has been developed and the downstream early treatment of “pre-rheumatoid arthritis” is being tested. However, for SSc, there may well be a very early SSc phase before any, but constitutional symptoms occur. This preclinical phase is very poorly understood or described. The very early diagnosis of systemic sclerosis (VEDOSS) has been defined, but there remain multiple questions surrounding VEDOSS, including when and how to treat patients with this diagnosis. Despite progress, there are no fully validated biomarkers or genetic predictors for disease evolution. Moreover, although VEDOSS patients with Raynaud’s phenomenon (RP), autoantibodies and SSc capillaroscopic pattern could be easily followed up, and no targeted cohort study to achieve these ends has been developed. Such a cohort study is very much needed, but it would require documenting all appropriate clinical, genetic, and autoimmune measures, followed for at least 5 and perhaps more years, using a randomized menu of treatments.

1. Introduction

1.1 Personalized or precision medicine

The use of the term “precision medicine” has recently become increasingly popular as has “personalized medicine”. They conceptualize the combination of individualized clinical, laboratory, imaging, and genetic data, to fit a specific treatment to a specific patient [1]. This is particularly useful in the management of complex and heterogeneous diseases such as rheumatoid arthritis (RA) or systemic sclerosis (SSc) in which advances in the identification of epigenomic, transcriptomic, and proteomic factors that change during the progression of these diseases or in response to treatment have increased their application for optimal patient management.

1.2 Outline

This chapter first gives the background regarding pre-connective tissue disease. Next, it uses rheumatoid arthritis as an example of the successful use of this concept. Thereafter, the chapter examines pre-scleroderma and very early diagnosis of systemic sclerosis (VEDOSS) and their potential in the future treatment of scleroderma.

2. Background

2.1 Early diagnosis of SSc

Despite knowing that classification criteria are not diagnostic criteria, such classification criteria are regularly used for diagnosis. Several attempts have been made over the years to overcome the diagnostic delay due to the clinical use of the American College of Rheumatology (ACR) 1980 classification criteria, as diagnostic criteria. These criteria require, as a major criterion, skin involvement for identifying SSc patients and starting to treat them. It is of note that skin thickening has been identified as the hallmark of SSc and has unintentionally led to relatively late diagnosis. This was because sufficient skin involvement to meet these classification criteria (1980) often occurred only after several years of disease and after visceral involvement has occurred, dooming clinical trials and effective treatment.

2.2 Typical elements of very early SSc

The recent identification of some typical elements of the very early stages of SSc, named as “red flags,” with various combinations of Raynaud’s phenomenon (RP), positive anti-nuclear antibodies (ANA), puffy fingers, abnormal nailfold capillaroscopy, and SSc specific antibodies [2], has shifted attention toward patients with SSc characteristics prior to classic skin thickening.

2.3 Problems inherent to the present SSc diagnosis

Unfortunately, these measures, while sensitive, are not highly specific [3] (sensitivity: 0.75; specificity: 0.72) [4]. Given this lack of specificity, it would be necessary to generate a better prediction model about the likelihood that patients will develop SSc. In 2013, the SSc ACR/EULAR (American College of Rheumatology/European Alliance for Rheumatology) classification criteria [5] were published, demonstrating increased sensitivity and specificity, (sensitivity: 0.91; specificity: 0.92) compared to ACR 1980 SSc classification criteria, not absolutely requiring skin fibrosis. These classification criteria resulted in earlier diagnosis of SSc, creating a “window of opportunity” for earlier treatment before significant visceral involvement or damage has occurred.

3. Introduction to pre-RA

It may be asked why should a section on rheumatoid arthritis be in a chapter on SSc? The issue of pre-RA is the one which has been well researched and much is known about its implications. Therefore, examining and understanding “pre-RA” is a template for considering “pre-SSc” and may lead to a better understanding of whether “pre-SSc” exists and what might be done to treat if it does.

The concept of “Pre-RA” is well known. This is a disease stage in which classification criteria are not yet met but the probability of developing RA is high. While the specifics of this “pre-RA” diagnosis have changed over time, the basic concept is to treat prior to significant symptomatic disease and prevent progression to classification/diagnostic criteria, thus preserving the quality of life and preventing long-term anatomic deterioration [6].

3.1 Characterizing “pre-RA”

The term “pre-RA” applies to events that occur prior to the clinical onset of RA as defined by ACR/EULA, classical [7] or during an asymptomatic phase [8, 9]. It is marked, however, by abnormal immune function and responses, even in the absence of clinical signs of autoimmune tissue injury [10].

Several factors in addition to clinical factors contribute to the different fates of patients presenting with apparently similar characteristics. There are genetic predispositions, such as the human leukocyte antigen (HLA) DR genotypes, environmental factors, such as smoking, and serological characteristics (cyclic citrullinated peptide (CCP) and rheumatoid factor) [11].

Rheumatoid arthritis has a preclinical period during which genetic and environmental factors interact, possibly in a sequential manner, to initiate and propagate the autoimmune process, leading to tissue inflammation and injury. During this period, disease-related autoantibodies, such as rheumatoid factor (RF) and anti-citrullinated peptide antibodies (ACPA), can develop even in the absence of clinical signs and symptoms of tissue injury. At a later stage, some but not all patients may experience minimal symptoms or signs that are considered nonspecific or unclassifiable for any rheumatic disease [11].

The EULAR study group recommends that, in prospective studies, individuals at risk of developing RA should be described as having the following five stages [12, 13, 14].

1. Genetic risk factors for RA

2. Environmental risk factors for RA

3. Autoimmunity associated with RA

4. Symptoms without clinical evidence of arthritis

5. Unclassified arthritis

3.1.1 Genetic risk factors

The shared epitope refers to a conserved amino acid sequence within the antigen-binding groove of the HLA-DRB1 molecule. In Caucasians, this sequence is present in several HLA-DRB1 alleles, including *0401, *0404, *0405, *0408, *0410, *0413, *0416, *0419, and *0421. These alleles share a common amino acid motif at positions 70 to 74 (Q/R-K/R-R-A-A)*. The increased risk of RA associated with HLA has an odds ratio of approximately 6, in Caucasians. Also, a family history of RA increases the risk of disease by 3 to 10 times [15, 16].

3.1.2 Environmental risk factors

Several environmental risk factors have been identified that contribute to the susceptibility of developing rheumatoid arthritis (RA). These factors include the following.

3.1.2.1 Smoking

Smokers have a higher risk of developing RA, and smoking also tends to worsen the disease progression and response to treatment. For example, smokers carrying two HLA-SE copies face a 40-fold increased risk of RA, highlighting the interplay of genetic and environmental factors in its development. Moreover, the risk remains elevated for up to 20 years after quitting smoking [17].

3.1.2.2 Infections

Chronic periodontal disease (periodontitis) has been linked to a higher risk of developing RA. Martinez-Martinez et al. found periodontal bacterial DNA in all synovial fluid samples and 83.5% of serum samples from RA patients. Despite this, synovial fluid samples from RA patients did not yield bacterial growth, and bacterial DNA was not identified in leukocytes [18].

Additionally, some bacterial and viral infections, such as Porphyromonas gingivalis (P. gingivalis), Proteus mirabilis (P. mirabilis), Epstein-Barr virus (EBV), and mycoplasma, have been implicated as potential triggers of RA in susceptible individuals [19]. A study indicates that individuals with elevated serum IgG antibody levels against early antigen of EBV, even if the antibody is found in the preclinical phase, are likely to develop RA. This implies, increased EBV reactivation cycles during this phase [20].

3.1.2.3 Obesity

Adipose tissue produces pro-inflammatory molecules that contribute to chronic inflammation and adiponectin (released by fat cells) levels in RA. Hyperlipidemia linked to obesity plays a role in RA development, particularly in women. Obesity, leads to increase in RA susceptibility through metabolic and endocrine pathways [21]. This involves elevated secretion of proinflammatory cytokines and adipokines by adipocytes, along with disturbances in sex hormone metabolism, resulting in heightened estrogen levels due to increased aromatase activity in adipose tissue [22]. Although some studies diverge, most affirm a link between adiponectin and inflammation markers like CRP. Recent research also associates adiponectin with C-reactive protein (CRP) levels and confirms a positive correlation with the DAS28 disease activity score [21, 23].

3.1.2.4 Hormonal factors

In investigating RA development, numerous studies have explored hormonal factors. A higher occurrence of RA in women compared to men (2 to 3:1) peaks at menopause. Although not well understood, it is believed that hormones may influence the immune response and contribute to the development of RA [24].

3.1.2.5 Dietary factors

While the relationship between diet and RA is complex, some dietary factors have been studied in relation to RA risk. For example, omega-3 fatty acids found in certain fish and plant sources have an anti-inflammatory effect. Kremer et al. carried out controlled studies, showing that greater than or equal to 3.0 g omega-3 acids (found in abundance in fish) had a mild anti-inflammatory effect [25]. On the other hand, diets high in red meat and processed foods have been associated with an increased risk of RA [26].

These environmental risk factors do not directly cause RA but rather contribute to the overall risk of this disease and may interact with genetic factors to trigger the development of the disease.

3.2 Serological precursors of pre-RA

Serological precursors of pre-rheumatoid arthritis can help in identifying individuals who may progress to RA. Here are some important serological precursors associated with pre-RA.

3.2.1 Rheumatoid factor (RF)

Rheumatoid factor (RF) can be detected in pre-RA individuals [27]. For example, among 79 patients with RA who were blood donors and had pre-RA samples, rheumatoid factor has the positive medium of 4.5 years (range: 0.1–13.8) before diagnosis [28].

3.2.2 Anti-cyclic citrullinated peptide antibodies

Anti-cyclic citrullinated peptide antibodies (anti-CCP target proteins that have undergone citrullination): Anti-CCP antibodies are highly specific (85–99% specific when found) for RA and are often present in pre-RA individuals [29, 30]. Studies have shown that individuals who are positive for anti-CCP antibodies have a 20–70% chance of developing RA within 2 to 5 years [29, 31, 32, 33].

Its presence is sensitive, but nonspecific and tends to occur more proximate to the appearance of clinical symptoms, usually appearing, within 1 year of clinical disease. However, 40% of RA patients have normal levels. In a large observational study of over 9000 patients, 26% had discordant ESR and CRP, making them unreliable for predicting joint damage progression [34, 35].

3.2.3 Other autoantibodies

Besides RF and anti-CCP antibodies, can be present in the blood of pre-RA individuals. These include anti-carbamylated protein antibodies (anti-CarP) and anti-mutated citrullinated vimentin antibodies (anti-MCV) [36]. Their presence may indicate an increased risk of developing RA. A recent study found a 42% sensitivity and 96% specificity of anti-CarP in anti-CCP and RF-negative RA patients [37].

It should be remembered that the presence of these serological precursors does not guarantee that an individual will develop RA. However, their detection can help identify individuals who may benefit from closer monitoring and early intervention to prevent or minimize the progression of the disease.

3.3 Diagnosing pre-RA

Compared to rheumatoid factor (RF), anti-CCP antibodies seem to provide a better prediction model for future RA development. When both RF and anti-CCP are found together, the hazard ratio of developing RA is 2.87 (1.22–6.76) [30, 31, 32, 38]. The wide range of the confidence interval reflects the variability of disease progression among different individuals.

3.3.1 Additional factors

To improve the accuracy of prediction, additional factors can be considered. Certain genetic factors, such as the presence of specific human leukocyte antigen (HLA) alleles such as those noted above, can further increase the risk of developing RA. The presence of the HLA-DRB1 shared epitope is highly correlated with anti-CCP-positive RA development. HLA-DRB1 SE alleles are much more common in patients with anti-CCP (82–89.6%) compared to those without (53–70%). Additionally, the presence of other symptoms related to arthritis, even in the absence of clearly inflamed joints upon physical examination, can be considered when assessing the likelihood of progression to RA [39].

3.4 Using pre-RA to consider therapy

The PRAIRI study, which investigated the effects of rituximab versus placebo in anti CCP- and RF-positive individuals who did not have RA, but might have had arthralgias, demonstrated a delay in the onset of arthritis by approximately 12 months for those who received rituximab [40]. This suggests that targeted interventions have the potential to prevent or delay the development of full-blown RA [41]. Currently, there are several studies, in the pre-RA period to prevent or delay the onset of RA. These studies investigated drugs like methotrexate and abatacept as potential interventions [40, 41, 42].

These tests and predictive models provide the probability of a patient progressing to RA but are not infallible. Clinical history, examination, understanding of social situations, etc. (i.e., clinical judgment) are also necessary before embarking on treatment “and no treatment is without at least potential adverse effects” including hepatotoxicity, birth defects, heightened infection risk, lymphoma, and increased skin cancer risk. Regular monitoring is crucial during RA treatment.

Summary: The extensive research done in pre-RA is worth reviewing as it provides the template when considering pre-SSc. Thus, when considering pre-SSc, some factors to consider might include genetics, the environment (e.g., smoking and hormonal factors), antibodies, and autoantibodies which could make diagnosing pre-SSc and its therapy practical.

4. Characterizing SSc

The concept of personalized medicine in SSc is to identify precise predictors of disease that can be targeted to either prevent or significantly delay the onset of clinical SSc. The above considerations in pre-RA can sometimes be applied below.

4.1 Clinical characterization

SSc is an unpredictable autoimmune disease with a high rate of morbidity and mortality as well as high socioeconomic costs because it has a chronic and debilitating course [43, 44]. This disease can be devastating and have a profound impact on quality of life and life expectancy., with a standardized mortality ratio of 3.5 [45, 46]. The great heterogeneity of Refs. [47, 48, 49] both clinical manifestations at the onset and during disease evolution requires looking for potential predictors starting from its preclinical and asymptomatic phases (very early SSc). This, in turn, requires the development of effective personalized/precision medicine strategies, one of which is the discovery, and use of SSc biomarkers, that can be identified early before the onset of skin fibrosis and organ damage.

4.2 Response to treatment

Despite presenting with similar initial clinical symptoms, SSc patients often exhibit varying responses to treatment with a standardized mortality ratio of 3.5 [45, 46]. Patients primarily experiencing an inflammatory phase may benefit from immunomodulatory treatments, while those in the fibrotic phase may require anti-fibrotic interventions. In cases where patients exhibit a combination of fibrosis and inflammation, combination therapy may be necessary. Moreover, the majority of patients need vasoactive therapies, beginning during the very early phases of the disease because they have Raynaud’s phenomenon (RP) and vascular impairment.

This multiplicity of pathways requires the administration of appropriate therapies targeting the underlying disease mechanistic pathways in a timely manner. Ultimately, targeting such pathways may halt/delay SSc disease progression. Having in mind such a preventive rationale, may pave the way for more innovative medications/approaches in certain SSc phenotypes. By integrating molecular characteristics with clinical phenotypes, innovative and more detailed stratification systems could significantly improve therapeutic approaches in personalized medicine, ultimately leading to better patient outcomes.

The great heterogeneity of SSc manifestations [47] at the onset and during disease evolution requires one to look for potential predictors of SSc starting from its preclinical and asymptomatic phases (very early SSc). This, in turn, requires the discovery and validation of SSc biomarkers that can be identified early, before the onset of clinical manifestations.

4.3 Autoantibodies

SSc-specific autoantibodies, including anti-centromere (ACA), anti-topoisomerase (ATA (ScL-70)), and anti-RNA polymerase III antibody (anti-RNAP III), are markers of disease progression and internal organ involvement [50]. These autoantibodies, which are commonly observed in SSc patients, typically occur independent of each other (Table 1). Unfortunately, the use of autoantibodies only contributed a small amount to effective risk stratification [51]. A cluster analysis, utilizing a large database and considering clinical and serologic variables from 120 EUSTAR centers (comprising 6927 patients), revealed that the dichotomous classification of SSc patients as lcSSc or dcSSc was insufficient. A significant proportion of lcSSc patients (39%) and dcSSc patients (19%) clustered in a discordant manner. To overcome this limitation, the study incorporated data in the presence of organ damage to predict the risk of further organ damage or death. As a result, the study identified six different clusters characterized by more homogeneous clinical phenotypes [52].

Clinical stratification and antibody profile of each cluster [52]

Cluster 1 (1186 pts): female, older onset, GI involvement, lcSSc, ACA (less ILD); Cluster 2 (720 pts): ILD, PH, lcSSc, ACA, ATA; Cluster 3 (1243 pts): younger onset, lowest mRSS, less aggressive, lcSSc, ACA > ATA; Cluster 4 (1673): older onset, DU, cardiac, lung, MSK, GI involvement, lcSSc, ATA > ACA; Cluster 5 (1249): male, younger onset, multi-organ involvements (cardiac, lung, GI, joint), dcSSc, ATA > ACA; Cluster 6 (856 pts): male, youngest onset, most aggressive, multi-organ involvement (cardiac, lung, renal, GI, MSK), dcSSc, ATA

Scleroderma-antibodies distinctive clusters [50]

ACA: lcSSc, PAH; ATA: dcSSc, ILD; anti-RNAP III: lcSSc, SRC; anti-PM-Scl: PM/DM overlap, arthritis overlap, ILD; anti-Th/To: lcSSc, ILD, PAH; anti-Ku: muscle and joint involvement; anti-U1RNP: overlap syndromes; anti-U3RNP: dcSSc, muscle involvement, PAH; anti-U11/U12RNP: ILD

Monocyte subsetted clusters [53]

Cluster 1 (high CD16+ monocyte, low memory B-cell subsets): more in lcSSc; Cluster 2 (high classical monocytes): dcSSc, high mRSS; Cluster 3 (high memory B cells): usually showed less skin involvement; Cluster 4 (low classical monocytes): usually showed less skin involvement

Table 1.

Proposed classification systems in systemic sclerosis based on published studies.

lc, limited cutaneous; dc, diffuse cutaneous; ACA, anti-centromere antibody; ATA, anti-topoisomerase antibody; ANA, anti-nuclear antibody; SSc, systemic sclerosis; Ab, antibody; ILD, interstitial lung disease; PAH, pulmonary arterial hypertension; PH, pulmonary hypertension; RP, Raynaud’s phenomenon; SRC, scleroderma renal crisis; PM, polymyositis; DM, Dermatomyositis; NFC, nailfold capillaroscopy; MSK, musculoskeletal; GI, gastrointestinal; mRSS, modified Rodnan skin score; DU, digital ulcer.

4.4 Pre SSc

4.4.1 Molecular stratification (monocyte sub setting)

The monocyte subset was incorporated by van der Kroef et al. [53]. In SSc characterization, they reported that prior to the onset of skin fibrosis and other organ manifestations, patients with RP, SSc-specific autoantibody positivity, and/or specific nailfold video capillaroscopy (NVC) patterns exhibited distinct frequencies of immune cell subsets. Through cluster analysis, it was demonstrated that circulating immune cell populations could differentiate SSc subsets into four distinct clusters: cluster 1 (characterized by high CD16+ monocytes and low memory B cells), cluster 2 (showing increased classical monocytes), cluster 3 (exhibiting increased memory B cells), and cluster 4 (displaying lower classical monocytes) [53]. These clusters were associated with different clinical features, such as limited cutaneous involvement in cluster 1, no skin involvement in clusters 3 and 4, and an enrichment of patients with ILD and diffuse cutaneous involvement in cluster 2. Mankinde et al. took a slightly different approach. Using an early SSc data registry and unbiased cluster analysis of monocytes, they found a stable transcriptional signature among three monocyte subsets. Although there were no differences in skin involvement among groups A, B, and C, at baseline, groups B and C had worse lung involvement [54].

4.4.2 SSc gene signature phenotyping

Gene signature phenotyping identified three main intrinsic gene subsets: fibroproliferative, inflammatory, and normal-like. Serial skin biopsies demonstrated that these intrinsic gene subsets were inherent and stable features of the disease in a given patient, indicating distinct pathogenic processes among SSc patients [55, 56]. A recent study by Skaug et al. [57] further reported that immune cell and fibroblast signatures in early dcSSc changed over time, showing a trend toward normalization as these signatures declined during follow-up. This finding has implications for patient stratification in future clinical trials focusing on early-stage disease. The intrinsic gene subsets were consistently observed across different skin biopsy sites, regardless of clinical involvement (thickened or normal skin) [58, 59]. Furthermore, these intrinsic gene subsets were found to be conserved across tissues such as the esophagus and skin, highlighting shared pathogenic processes in SSc across different tissues. However, the issue of microenvironment may influence gene expression, as functional genomic network analysis conducted by Taroni et al. identified a distinct lung-specific innate immune process, indicating that certain gene pairs are more likely to interact in specific tissues compared to others [58, 59].

4.5 Very early diagnosis of systemic sclerosis (VEDOSS)

Previously, in SSc, various terms have overlapped, that is, “early SSc,” “VEDOSS,” pre-scleroderma, and undifferentiated connective tissue disease (UCTD) at risk for systemic sclerosis (UCTD-risk-SSc) [44, 48, 49, 60, 61]. These terms may identify different clinical scenarios that have something in common and are positioned in the very early phase of SSc. In fact, the term pre-SSc today identifies a moment in time where the disease may be represented by the presence of vague symptomatology (e.g., fatigue) with/without Raynaud’s phenomenon (RP) and specific antibodies only. Clearly, having only RP does not mean that SSc already exists, but the presence of specific antibodies, with or without RP, should raise the suspicion that the patient will develop VEDOSS or definite SSc. Its usefulness for diagnostic purposes is limited due to its low specificity as RP may also be found in other connective tissue diseases such as mixed connective tissue disease (MCTD), undifferentiated connective tissue disease (UCTD), systemic lupus erythematosus (SLE), dermatomyositis/polymyositis, Sjögren’s syndrome (SS), vasculitis, and RA [62, 63, 64, 65]. The latency between RP and the onset of the first non-RP-symptom could help define this time period as the Pre-SSc clinical phase although its specificity is low and its duration may be 5 years or more. Other parameters should be considered in the assessment of endothelial dysfunction, like capillaroscopic abnormalities, serum biomarkers (such as intercellular adhesion molecule-1, vascular cell adhesion molecule, anti-endothelial antibodies-1, or E-selectin or puffy fingers (PF)) [66]. In fact, puffy fingers have been proposed as an early sign of pre-scleroderma or VEDOSS and represent a continuum between these two entities and SSc [47]. Lescot suggested that PF may signify a vascular phenomenon such as is found in lcSSc or an early inflammatory phenomenon leading to dcSSc.

4.6 From ACR 1980 to VEDOSS criteria

In this gray area, in which the defined colors of the disease are still absent, and it is not yet possible to draw the patient’s future, a very early diagnosis in a pre-clinical disease phase, and timely treatments, is the cornerstone for preventing disease evolution. If we review the road traveled by rheumatologists to achieve early diagnosis in SSc, we can see its beginnings in 1991 with the introduction of the term “early SSc” by Steen and Medsger [67]. They identified disease stages preceding the development of irreversible vascular damage and atrophic lesions in patients with definite diffuse (<3 years from the first non-RP symptom) or limited SSc (<5 years), then also focusing on a temporal concept (number of years since the diagnosis).

In 1996, the term “pre-scleroderma” was proposed for the first time by Fine LG et al., and it identified patients with RP, digital ischemic abnormalities, and nailfold capillaroscopy findings (NFC) abnormalities or disease-specific circulating autoantibodies (i.e., anti-topoisomerase antibody, anticentromere antibody, anti RNA polymerase 3, antifibrillarin, anti-Th/To, or anti-PM-Scl) [60]. A few years later, a similar concept was named limited SSc (lSSc) by LeRoy and Medsger to describe patients with RP and either SSc-specific autoantibodies or SSc-type NFC pattern [61]. No other symptom/sign/laboratory findings were mentioned as necessary for the earliest diagnosis of SSc. Some patients with lSSc and without skin involvement, but with SSc NFC abnormalities, disease-specific antibodies, and/or internal organ involvement were called “SSc sine scleroderma” even if none of the registries published thus far have considered it [43]. Recently, Valentini et al. proposed the term “Undifferentiated Connective Tissue Disease at risk for Systemic Sclerosis” (UCTD-risk-SSc), to define a condition characterized by RP and either SSc-specific autoantibodies or a capillaroscopic scleroderma pattern or both, but without satisfying classification criteria for SSc, and also without features consistent with SSc sine scleroderma [68].

Two of the most important stages of this road toward defining an early diagnosis were the identification of the aforementioned “red flags,” through a Delphi consensus study conducted by the European League Against Rheumatism (EULAR) Scleroderma Trial and Research (EUSTAR) among SSc experts. The consensus study proposed criteria for very early diagnosis of SSc (VEDOSS) [49] in 2011. This was followed by the publication of new ACR/EULAR classification criteria in 2013, in which the score of 9 was considered valid for classifying patients as SSc. This new definition allowed SSc-specific antibodies (ACA, ATA, and anti-RNA polymerase 3), capillaroscopy alterations (early signs of SSc), and more advanced features such as skin involvement, pulmonary involvement, and digital lesions.

The VEDOSS criteria defined ANA positivity, RP and PF as signs which should raise suspicion for SSc; if these were positive, it was suggested to examine SSc-specific antibodies (anticentromere, anti-topoisomerase (Scl-70), or anti-polymerase 3) and capillaroscopy. These criteria were validated in 2021 by a multicenter study that investigated the VEDOSS criteria to detect progression to classification criteria for SSc. This study evaluated 553 RP patients meeting the 5 VEDOSS criteria but not SSc criteria. At 5 years, 254 of the 553 (52.4%) were classified as EULAR/ACR 2013 SSc (Table 2). The study also reported that the absence of ANA decreased the risk of progression to SSc, (5-year progression 10–8%), while ANA positivity together with the presence of PF showed the highest rate of progression to 2013 ACR/EULARSSc criteria (79%). Particularly, ANA-positive patients with RP, SSc-specific autoantibodies, and PF had the highest progression rate over time (94.1%), followed by patients with ANA positive, specific autoantibodies, and NVC abnormalities (82.2%).

	Censored before 5 years (n = 299)	Five-year follow-up completers (n = 254)	P value	Total (n = 553)
Male	29 (9.7%)	17 (6.7%)	0.22	46 (8.3%)
Female	270 (90.3%)	237 (93.3%)	237 (93.3%)	507 (91.7%)
Age	43.63 (14.4)	48.7 (15.2)	<0.001	45.9 (15.0)
Duration of Raynaud’s phenomenon (years)	4.0 (1.7–10.0)	4.0 (1.5–10.3)	0.98	4.0 (1.7–10.0)
ANA positive	187/293 (63.8%)	214/251 (85.3%)	<0.001	401/544 (73.7%)
Systemic sclerosis-specific autoantibody positive	77/280 (27.5%)	131/247 (53.0%)	<0.001	208/527 (39.5%)
Anticentromere antibody positive	64/275 (23.3%)	100/244 (41.0%)	<0.001	164/519 (31.6%)
Anti-scleroderma-70 positive	11/280 (3.9%)	28/245 (11.4%)	<0.001	39/525 (7.4%)
Anti-RNA polymerase III positive	2/109 (1.8%)	4/71 (5.6%)	0.21	6/180 (3.3%)
Puffy fingers	48/298 (16.1%)	48/242 (19.8%)	0.31	96/540 (17.8%)
Nailfold capillaroscopy abnormalities	105/286 (36.7%)	77/219 (35.2%)	0.78	182/505 (36.0%)

Table 2.

Characteristics of patients in the VEDOSS project [48].

In clinical practice, when defining VEDOSS, several issues need to be addressed

1. “How many and which individual aspects of VEDOSS predict specific visceral involvement in SSc?

2. When should immunosuppressive treatment be started in VEDOSS?

3. Can starting vasoactive treatment very early prevent/delay the onset of pulmonary arterial hypertension or other vascular complications such as digital ulcers?

4. Can we define what is the most appropriate follow-up timing in VEDOSS patients to avoid stressing patients without organ involvement?

5. Can other tests be useful for diagnosing VEDOSS or to define progression (e.g., lung ultrasound)?

6. Can hydroxychloroquine be useful to prevent progression?

These questions, and no doubt others, await research, assuming the acceptance of a uniform very early scleroderma definition (e.g., VEDOSS and SSc sine scleroderma).

5. Intrinsic gene subsets influence treatment response

The use of molecular phenotyping in SSc patients holds promise for guiding therapeutic approaches by tailoring treatments based on the individual’s unique intrinsic gene subsets.

For example, it has been observed that SSc patients in the fibroproliferative gene subset tend to respond positively to tyrosine kinase inhibitors (TKIs), such as imatinib and nilotinib, which target tyrosine kinases involved in fibrotic pathways and reduced the expression of genes associated with the fibroproliferative subset in dcSSc patients. Similarly, higher baseline expression of genes related to TGFbR and PDGFR signaling in SSc patients correlated with positive responses to nilotinib. More recent trials analyzing the response to dasatinib also revealed that patients who showed improvement were predominantly from the fibroproliferative or normal-like subsets, while those who did not respond well were mainly from the inflammatory subsets [69, 70, 71].

Janus kinase (JAK), which is involved in cytokine signaling, has been implicated in the pathogenesis of SSc. Pre-clinical studies suggested its role in transmitting pro-inflammatory or profibrotic signals to target cells. Gene expression profiling analysis has confirmed elevated IL6/JAK/STAT gene signatures in skin biopsies of dcSSc patients belonging to the inflammatory subset, compared to healthy individuals.

The preliminary efficacy of tofacitinib, primarily an inhibitor of JAK1/3, was seen in the treatment of dcSSc patients with refractory skin involvement. A pilot study conducted at a single center evaluated the use of tofacitinib in a case series of 10 patients. The results showed a significant improvement in modified Rodnan skin score (mRSS) within the first month, indicating its potential as an effective immunosuppressant for progressive skin thickness in dcSSc [72]. Ongoing Phase I/II randomized controlled trials (NCT03274076) by Khanna et al. have shown initial results indicating the safety of tofacitinib and a trend toward mRSS improvement. Further studies are necessary to confirm the efficacy of tofacitinib and evaluate its response in relation to inflammatory and fibrotic gene signatures [73].

TGFb signaling targeted treatment (Fresolimumab) demonstrated efficacy in patients with high baseline levels of the TGFb-regulated gene thrombospondin-1 (THBS1). In those patients, THBS1 levels declined along with improved skin scores [74]. Taroni et al. conducted a functional genomic meta-analysis, using publicly available gene expression data from clinical trials of various therapeutics including MMF and fresolimumab. The analysis revealed that improvers on fresolimumab had high baseline levels of TGFb-related genes, while non-improvers had elevated levels of immune-related genes at baseline. Conversely, MMF improvers had high baseline levels of immune-related genes that decreased after treatment [59]. This study emphasizes the importance of genome-wide gene expression data collected during clinical trials.

Patients who responded to immunosuppressive medications were more likely to be associated with inflammatory gene subsets. For instance, responders to mycophenolate mofetil (MMF), which targets lymphocyte proliferation, belonged to the inflammatory gene subset (four out of seven improvers), while non-improvers were associated with the fibroproliferative gene subset (two subjects) [75]. Similarly, responders to abatacept, which inhibits CD28 T-cell activation, had positive responses among patients in the inflammatory gene subset. Improvers who responded to abatacept (four out of five), were primarily associated with the inflammatory gene subset and had higher baseline levels of CD28 signaling. On the other hand, the non-improver (one patient) belonged to the normal-like gene subset and exhibited lower baseline levels of CD28 signaling [76].

In summary, these findings demonstrate the relevance of intrinsic gene subsets and their relationship with specific treatment responses. They emphasize the potential for using genomic data to inform treatment decisions and personalize therapeutic approaches in SSc patients.

6. Classification according to abnormal nailfold capillaroscopy findings (NFC) patterns

Abnormal NFC can be classified as early, active, or late, providing valuable insights into the overall progression of the disease. The severity of NFC patterns has been identified as a predictive factor for future severe organ involvement, although it requires years to manifest. This slowly changing risk tends to increase as the NFC pattern progresses from early to late, even after accounting for disease duration, subset, and vasoactive medications [49, 77, 78, 79].

NFC and laboratory tools (gene signatures and autoantibodies) into a classification system have the potential to develop a more comprehensive and predictive approach. This integration can inform treatment decisions and ultimately lead to improved personalized/precision patient outcomes [80].

7. Conclusion

While the classification criteria for SSc have been carefully delineated, the definition of what comes before meeting classification criteria is not so well understood. For systemic sclerosis, there may well be a very early SSc phase before any but constitutional symptoms occur, associated with varying degrees of genetic and serological signals. The very early diagnosis of systemic sclerosis (VEDOSS) has been defined and is useful. However, there remain multiple questions surrounding VEDOSS, including when and how to treat patients with this diagnosis.

The identification of validated biomarkers and genetic predictors for disease susceptibility and progression would allow risk stratification of patients and subsequent tailored clinical and therapeutic management and an efficient use of resources. However, despite progress, there are as yet no fully validated biomarkers or genetic predictors for disease evolution.

Moreover, although VEDOSS patients with RP, autoantibodies, and SSc capillaroscopic pattern could be followed up, we have not yet developed a cohort study, documenting all appropriate clinical, genetic, and autoimmune measures, followed for at least 5 and perhaps more years, using a randomized menu of treatments (Table 3).

Proportion fulfilling 2013 ACR-EULAR criteria		ANA	Ssc-Ab	SSc pattern on NFC	Puffy fingers
In the presence of	ANA	58.9%	70.2%	75.0%	79.0%
	SSc-Ab	70.2%	70.2%	82.2%	94.1%
	SSc pattern on NVC	75.0%	82.2%	70.1%	69.2%
	Puffy fingers	79.0%	94.1%	69.2%	70.8%
In the absence of	ANA	10.8%	31.0%	40.4%	47.5%
	SSc-Ab	31.0%	31.0%	41.9%	49.6%
	SSc pattern on NFC	40.4%	41.9%	41.5%	50.9%
	Puffy fingers	47.5%	49.6%	50.9%	47.9%

Table 3.

Frequency of progression as calculated in the presence or absence of the VEDOSS criteria alone or in combination [47, 48, 49].

Acknowledgments

I wish to acknowledge the diligent and ongoing work of Claudia Real, my executive assistant, without whom this chapter would never have been completed.

Disclosures

Bellando-Randone S.—none declared; Suliman Y.—none declared; Anwar S.—none declared; Furst D.E.—none declared.