Background/Purpose:  Localised scleroderma can occur in
overlap with systemic sclerosis (SSc); nodular and keloidal morphea are rare subtypes
but can be disfiguring and challenging to treat. Published literature suggests
that keloid scarring and SSc skin share TGFbeta regulated gene expression alterations
and that nodular and keloidal morphea represent a spectrum of keloid formation
in SSc.  This study aims to determine the pathological differences between these
conditions and addresses the hypothesis that bone marrow derived fibroblast
progenitors (fibrocytes) link them.

Methods: We examined a clinical database of 2200 patients
with SSc defined by ACR/EULAR classification criteria to identify patients with
a clinical or pathological diagnosis of localised scleroderma. The demographics,
clinical features and laboratory parameters of those with nodular or keloidal
morphea were examined in detail. We identified structural and biochemical
differences by examining paired skin biopsies from these lesions and standard
forearm biopsy site in SSc (n=6).  Skin biopsies from patients with both early
and established diffuse cutaneous SSc were used as controls.

Results: 49 patients (2.2%) had both localised scleroderma
and SSc. Data from this group were broadly comparable to the SSc cohort,
excepting a higher prevalence in females (female:male ratio 11:1). Three
subtypes of localised scleroderma were prevalent: plaque morphea (67%); linear
morphea (8.0%) and nodular or keloidal morphea (25%). Plaque and linear morphea
patients had predominantly  limited cutaneous SSc (88% and 75% respectively)
while the majority of nodular/keloidal morphea patients had diffuse cutaneous
SSc (92%, p<0.001). There were no clear serological associations. Keloid
nodules were localised on the upper chest rather than areas of high SSc skin
disease activity. Examination of histology in 5 of 12 patients with nodular/keloidal
morphea identified acute inflammatory infiltrates, mucin deposition and altered
distribution of collagen compared to SSc skin biopsies. Expression analysis of
TGFbeta regulated genes identified multiple significant differences in
expression within morphea lesions and at other sites in the same patients
(figure 1). A significant increase in CCl2 expression may be associated with
fibrocyte recruitment.

Conclusion: This is the largest analysis of localised scleroderma
in SSc. There are associations of subtypes of localised scleroderma with
subsets of SSc which are not related to serology. Paired examination of
histological samples and gene expression of nodular or keloidal SSc skin identifies
increased TGFbeta downstream activity compared with other biopsy sites
suggesting that this mediator may be a key driver of localised scleroderma in
SSc.