Localized scleroderma (LS) is the most common type of scleroderma in children. LS differs from systemic sclerosis (SSc) by typically being confined to the skin and subcutaneous tissue, with only rare involvement of the internal organs 12. There are neither sensitive nor specific laboratory tests for measuring disease activity, and monitoring is done almost exclusively by clinical assessment, which is often challenging. Identifying a laboratory marker of disease activity will aid in the management of affected children.

The pathogenesis of the skin fibrosis is far from clear; like other autoimmune diseases it is thought to involve an environmental trigger in an immuno-susceptible host leading to inflammation and damage, with increased production and deposition of collagen 3. The Transforming Growth Factor β family of cytokines (TGFβ) plays a major role in modulation of the immune response, and is a critical counter-inflammatory/regulatory cytokine. TGFβ1 is also a central mediator in fibrosis and angiogenesis, playing an important role in the fibrotic process in scleroderma 3. Several studies have found elevated serum TGFβ1 concentrations in both generalized and localized forms of sclerosis 456. Skin TGFβ1 levels are more difficult to quantitate and in one study, no differences were found in affected skin from those with SSC 6 or LS 7. Our study objectives were to determine the peripheral blood levels of TGFβ1 in children with LS and to determine the relationship of serum levels to disease activity.

Fifty-five patients in various sub-type groups (table 1) were studied. The mean age (± SD) at disease onset was 9.2 ± 3.6 years, (range of 0.7 to 19.2 years, median 8.7 years). The mean disease duration at the time of TGFβ1 sampling was 4.9 ± 3.7 years.

The mean serum TGFβ1 concentrations were significantly higher in patients than controls: 51393 ± 33953 pg/ml, (range 2949 – 141965 pg/ml), vs. 9825 ± 5287 pg/ml (range 4725 – 16761 pg/ml) (p < 0.0001). There was no significant difference between the TGFβ1 measurements between the 4 different subtypes as described above (Fig 1). (Mean TGFβ1 for LIN face = 39991 pg/ml, GM = 49053 pg/ml, M = 55,939 pg/ml, LIN limb = 58,481 pg/ml, F (3,51) = 0.714, p = 0.55.)

There was no difference in TGFβ1 serum concentrations between the 21 patients in the active phase of the disease (51374 pg/ml), to the 23 patients with inactive disease (50985 pg/ml), or the 11 indeterminate phase patients (52284 pg/ml); F (2,52) = 0.0053, p = 0.99 (Fig. 2).

In a multiple regression model, there was no significant relationship between TGFβ1 serum concentrations and patient demographics (age, disease duration) or basic laboratory markers (ESR, eosinophil count, IgG concentrations or ANA),

Our study clearly demonstrates that serum concentratons of TGFβ1 are very elevated in pediatric LS patients at all times during evolution of their disease. The measurements were elevated in all the disease subtypes, and there was no correlation between the disease activity and the cytokine levels. This is in accord with previous studies that reported increased TGFβ1 serum concentrations in adults with SSc and LS 456.

TGFβ1 mediates fibrosis through downstream modulators. It is derived from many cellular sources including members of the immune system and those in other organ systems including activated endothelial cells. A critical downstream effect is stimulation of fibroblasts and excessive extra-cellular matrix deposition 3. TGFβ1 is secreted as a pro-peptide (latent form) and enzymatically cleaved into the active form 8, and is a potent chemoattractant for human skin fibroblasts 9. TGFβ1 also up-regulates synthesis of matrix metalloproteinase inhibitors, which inhibit collagenase activity 10.

In our study, there was no correlation between the serum concentrations of TGFβ1 and LS disease activity or disease subtype. There has been much work regarding an appropriate classification system for localized scleroderma. Interestingly, our data suggest that regardless of classification system used, this is a cohesive group of patients with elevation of circulating TGFβ1 as a common feature. Interestingly, high levels were also found in all phases of the disease included inactive states. Many different hypothesis can be invoked to explain this finding including modulation at the receptor level 11, as increased expression of TGFβ1 receptors are reported in skin fibroblasts of SSc and LS patients 121314 and continued production of TGFβ1 by endothelial cells as a repair/wound healing response in inactive lesions. Alternatively, immune sources of TGFβ1 may be responsible for production of TGFβ1 as a regulatory/counter-inflammatory cytokine keeping the disease in the quiescent phase. Or it may simply by that peripheral blood levels of TGFβ1, a cytokine known to affect changes locally in only limited areas of affected skin, do not represent biochemical activity in affected skin; a scenario common to cytokines and enzymes which are tightly regulated at the tissue level. Another practical explanation of the lack of correlation between disease activity and TGFβ blood levels may be related to our limited clinical ability to determine disease activity accurately. It may be overly simplistic to assume a direct relationship between TGFβ1 and disease, as with other processes in the complex milieu of the human body, multiple mechanisms are at play to regulate the balance between extracellular matrix production and degradation.

Potential limitations of our study include population selection, specifically our control population. Our control group consisted of children with atopic dermatitis. LS is an inflammatory skin disease as is atopic dermatitis, thus making children with atopic dermatitis our choice for a specificity control for non- specific inflammation of the skin. Circulating levels of TGFβ1 in this population may be different from levels in healthy children. In fact, some reports suggest an association between low TGFβ1 responder genotype 15 and low TGFβ1 mRNA production by circulating leukocytes in some children with atopic dermatitis 16.

Increased TGFβ1 serum concentrations appear to be common to all subtypes of children with LS suggesting a potentially useful role for TGFβ1 as a biomarker of the disease. Lack of correlation of circulating TGFβ1 concentration with disease activity preclude their use clinically to guide therapy in affected children.

The author(s) declare that they have no competing interests.

YU – participated in its design, and coordination of the study, and writing the manuscript.

BMF – participated in the design of the study and performed the statistical analysis, and helped to draft the manuscript.

BRK – participated in the design of the study, data collection, helped to draft the manuscript.

RML – participated in the design of the study, data collection, helped to draft the manuscript

RSMY -Senior author- participated in the design of the study, data collection and TGFβ1 analysis, and writing the manuscript.

All authors participated in writing, read and approved the final manuscript