Background
Unlike UVB radiation that can penetrate at the most into the papillary dermis, longer wavelengths in the UVA region have the capacity to reach the subcutis as well. Accordingly, as well as due to its lesser antiproliferative activity, UVB irradiation has not been established in the treatment of sclerotic disorders except for occasional cases of graft-versus-host disease (GvHD) [
1,
2]. Hence, this review examines different modalities of UVA phototherapy in the treatment of connective tissue diseases and related disorders.
The term irradiance (e.g., in mW/cm
2), which is the most commonly used term in photobiology, relates to the subject (e.g., patient) struck by the irradiation. In photobiology, the time integral of the irradiance is commonly expressed as fluence (e.g., in J/cm
2), or even more loosely as dose [
3,
4]. Even though mostly combined with 8-methoxypsoralene or other photochemotherapeutic agents, broad-band UVA irradiation (315–400 nm), containing both UVA1 (340–400 nm) and UVA2 (315–340 nm), was used as monotherapy e.g. in the treatment of atopic dermatitis [
5]. However, this phototherapeutic option was being replaced more frequently by the usage of irradiation devices which allow a more effective treatment by the administration of selected spectra. By eliminating shorter wavelengths in the UVA2 region adverse effects such as erythema are minimized and therapeutically effective higher UV doses can be administered. Thus, today broad-band UVA may play a subordinated role in modern phototherapy anymore, except for the combined application with psoralenes, even though it is still used for a large number of patients due to its wide availability and although it has not yet been directly compared with UVA1 for many sclerosing disorders. Conventional UVA1 treatment emitts wavelengths mainly between 340 and 400 nm, but may also produce scattered radiation >530 nm including infrared irradiation (780–3000 nm). Three different pattern of UVA1 dosage have been described: high-dose UVA1 phototherapy applying UVA1 doses ranging from 90–130 J/cm
2 single (975–1840 J/cm
2 cumulative) irradiation, medium-dose UVA1 phototherapy admitting doses between 20–90 J/cm
2 single (300–975 J/cm
2 cumulative) UVA1 and last but not least low-dose UVA1 phototherapy administering single UVA1 doses of ≤ 20 J/cm
2 or a cumulative doage ≤ 300 J/cm
2, respectively [
6‐
9]. Conventional UVA1 phototherapy may be accompanied by extensive heat load predominantly generated by infrared irradiation (780–3000 nm) and/or insufficient cooling systems of the phototherapy devices. For this reason, within the last years, lavish new UVA1 equipment was developed containing a special filtering and cooling system, in which a considerable amount of wavelengths >530 nm are eliminated and consequently the heat load due to heat-producing infrared radiation is strongly diminished [
10]. Since more than two decades, the combination of oral 5- or 8-methoxypsoralen followed by broad-band UVA exposure is an effective treatment option in a widespread number of indications such as psoriasis and cutaneous T cell lymphoma [
11‐
13]. Psoralens specifically belong to the best characterized agents of photosensitizing chemicals. Following its administration low-dose UVA irradiation is applied including an individual progression depending on the formation of the so-called PUVA erythema. In general, the overall mean cumulative dose has been found to comprise ≤ 400 J/cm
2 UVA [
14]. In contrast to systemic PUVA therapy topical PUVA, for example applied as PUVA bath, PUVA shower or PUVA cream, is mainly characterized by absent/reduced systemic side effects and the restriction to a selective limited lesional area [
15]. As a consequence, systemic PUVA has mostly been replaced by topical PUVA treatment representing an efficient well-tolerable alternative to oral methoxypsoralen administration. Generally, overall mean cumulative dosage has been found to comprise ≤ 200 J/cm
2 UVA [
14].
Different types of UVA phototherapy were introduced as an innovative and promising therapeutic option in the treatment of inflammatory diseases such as atopic dermatitis and more recently in therapy of lymphoproliferative disorders such as cutaneous T cell lymphoma and related skin affections. No more than seven years ago, subsequent to promising clinical results of extracorporeal photochemotherapy and psoralene plus UVA (PUVA) in systemic sclerosis (SSc) and morphea [
16‐
19], first investigations have verified the therapeutic value of UVA1 irradiation for the treatment of localized scleroderma (LS) [
20‐
22].
However, encouraged by the clinical success and the diversity of immunomodulatory effects achieved by the use of UVA phototherapeutic regimens in a great number of different indications, additional studies focused on investigating the efficacy of UVA phototherapy in the treatment of a widespread range of sclerotic skin diseases [
23]. Furthermore, series of UVA exposure may be used in the treatment of lupus erythematosus (LE), even if known as a photosensitive condition.
Today, different forms of UVA phototherapy are widely used and have subsequently developed into a treatment modality of importance within the field of dermatology and rheumatology at least as an adjunctive treatment and, beside the 'initial indications', may also or even especially be indicated as a successful alternative in the treatment of skin manifestations of connective tissue diseases and related disorders.
Methods
For this systematic review we concentrated on the therapeutic use of UVA application in humans. The computerized bibliographic database PubMed (includes all citations from Medline and additional life science journals) without time limits (January 1966 to April 2004) was screened for original papers, case reports, letters, reviews and book articles on UVA/UVA1/PUVA. As main key words we used "UVA", "UV-A", "UVA phototherapy", "UVA1", "UV-A1" "UVA1 phototherapy", "PUVA", "PUVA phototherapy", "PUVA photochemotherapy", "ultraviolet A", "ultraviolet A1", "sclerosis", "scleroderma", "sclerosus", "sclerotic", "sclerodermoid", "morphea", and "lupus erythematosus". Other sources included monographs, textbooks, and the reference lists from all the articles retrieved. All abstracts were read and selected by two authors (F.B., T.G.) Inclusion or exclusion of articles were based on consensus. Relevant data including study design, number of patients, duration of treatment, clinical outcome, cumulative UVA doses, and adverse effects were retrieved from the articles, summarized and briefly discussed as follows.
Discussion
Beside a high number of different side indications, systemic and especially topical PUVA treatment have been shown to provoke a remarkable clearance of psoriatic plaques and infiltrated lesions of cutaneous T cell lymphoma. On the other hand, UVA1 phototherapy achieved practical value in the treatment of inflammatory and malignant T cell related skin diseases. Typical indications regularly include exacerbated atopic dermatitis, cutaneous T cell lymphoma, parapsoriasis or mucinosis follicularis due to the induction of T cell apoptosis and dermal immunoregulation. By reason of notable collateral induction of dermal collagenase activity, UVA irradiation was subsequently introduced as a treatment alternative in LS and other sclerotic collagenoses in both dermatology and rheumatology.
As far as the extensive accumulation of collagen is concerned several investigations demonstrated the long-wave UVA-induced stimulation of the synthesis of specific mRNA-levels of various matrix-metalloproteinases in cultured human fibroblasts, probably due to their lower antioxidant capacity and involvement of the protein kinase C pathway [
21,
102‐
110]. Simultaneously, an elevation of interstitial collagenase m-RNA and protein expression can be determined immunohistochemically and by the use of nucleic-acid in-situ hybridization in dermal fibroblasts [
21,
111]. Furthermore, several studies provided evidence that at least UVA1 irradiation induces the formation of several cytokines and soluble factors e.g. interleukin-1 and/or interleukin-6 stimulating the synthesis of collagenase, while some immunomodulatory cytokines remain unaltered [
112,
113].
On the other hand, UVA1 irradiation has been shown to initiate apoptotic cell death in dermal T lymphocytes [
114,
115]. A shift of the balance between protooncogenes (e.g., bcl-2) and tumor suppressor genes (e.g., p53) towards the induction of apoptosis seems to be one of the major effects of UVA1 irradiation [
10]. Beside the involvement of singlet oxygen as an early intermediate in collagenase induction, oxidative stress has also been proven to induce lipid peroxidation in cytoplasmatic membranes and to be responsible for DNA damage [
116,
117]. Therefore, one causal factor of membrane alterations might be the (P)UVA-induced generation of reactive oxygen species, such as singlet oxygens or superoxide anions and hydroxyl radicals leading to lipid peroxidation, structural and functional modifications of membranes characterized by altered fluidity, increased permeability and inactivation of cellular enzymes and transport proteins [
116‐
119]. Furthermore, singlet oxygen is able to open mitochondrial megachannels, releasing apoptosis initiating factor (AIF) and cytochrom c heading towards indirect DNA damage in T cells [
116,
120,
121]. Besides, UVA1-induced apoptosis is triggered by receptor mechanisms, e.g. by the alternative activation of the FAS/FAS-ligand (APO-1, CD95) system in peripheral T cells [
114,
116,
122].
If and to what extent additional modulations of impaired endothelial cells might also contribute to the posttherapeutic clinical and histological improvement still has to be investigated [
123]. Nevertheless, the efficacy of different regimens of UVA phototherapy might probably, at least in parts, be due to the mechanisms of action as mentioned above.
Inasmuch as LE and other autoimmune disorders are concerned, a transparent mechanism of action remains obscure. Immunohistologically, LE is at least characterized by an inflammatory T cell derived infiltration mainly of the T helper subtype combined with an impairment of T and B lymphocyte regulation, dysregulated dendritic cell abnormalities and defective clearance of immune complexes and autoantigens [
124‐
126]. Analogous to the sclerosing disorders as mentioned above, induction of apoptosis in T lymphocytes may also represent the committed step of UVA exposure in LE. Simultaneously, UVA has been shown to directly affect presence, function and morphology of dermal and epidermal Langerhans cells, which may result in a suppression of cell-mediated immunity and a disruption of autoreactive T lymphocyte, B cell and Langerhans cell stimulating processes [
127‐
131]. Controversially, UV radiation is often associated with exacerbating skin eruptions and photosensitivity is actually a diagnostic criterion of LE. Nevertheless, in how far improvement of skin condition after repeated irradiation might be due to adaptive decreased oxidative stress upon subsequent UVA exposures has to be evaluated [
131].
Despite an extensive therapeutic administration of UVA irradiation, relatively little data are available concerning possible acute and long-term side effects. Usually, patients exhibit a dose-dependent tanning of the skin, which has been described to appear following a single minimal pigmenting dose of 50 J/cm
2 UVA1 [
132,
133], whereas UVA1-induced erythema can usually only be observed after ≥ 90 J/cm
2 single dose UVA1 [
134]. After UVA1 exposure, individuals of all skin types appear to develop more or less immediate pigment darkening that is due to a reversible photochemical reaction (oxidation of melanin and its precursors and metabolites). In contrast, even after a comparably lower dosage, broad-band UVA irradiation is able to provoke erythema and delayed tanning which is due to an enzymatically controlled production of melanin polymers. After topical or systemic PUVA treatment patients exhibit the characteristic methoxsalen dose-dependent PUVA erythema ranging from three to six days following irradiation [
135,
136]. Additionally, systemic PUVA is often associated with nausea or vomitus. However, by considering exclusion criteria such as an autoimmune disease associated abnormally increased photosensibility, solar and heat induced urticaria or a history of polymorphous light eruption and, especially in the case of PUVA therapy, by the consequent posttherapeutic use of potent sun protection, additional clinically relevant acute side effects may usually not be expected.
As far as long-term side effects are concerned no definite prediction has yet been taken. Controlled studies dealing with the carcinogenesis induced by broad-band UVA sources are still rare. Nevertheless, the induction of dermal hyperplastic elastic fibers resulting in early skin ageing following a cumulative dose of 4000–8000 J/cm
2 UVA1 seems to be imminent [
6]. Additionally, as already reported above, the induction of collagenases released by dermal fibroblasts is known to be an important cofactor within this process [
9,
137]. On the other hand, carcinogenesis of UVA1 irradiation is still poorly understood. Taking the elimination of the potentially procarcinogenetic wavelength ranging from 315–320 nm into account, one might speculate on the possibly lower risk of UVA irradiation [
138]. Indeed, possible melanogenetic long-term effects of long-wavelength UV irradiation (induction of malignant melanomas) have previously been discussed [
139,
140]. Simultaneously, animal studies suggested the induction of squamous cell carcinomas even though provoked by 220 kJ/m
2 for a period of 265 days [
6]. Until today, no assignment to humans could be achieved. However, at least concentrated PUVA photochemotherapy has been found to be related to potential mutagenesis and the increased occurrence of squamous cell carcinomas as well as malignant melanomas in psoriatic patients [
9,
141‐
143].
Moreover, recent investigations concerning the effects of UVA1 irradiation on human dermal endothelium revealed the initiation of apoptotic cascades even after a comparably low dose of single 80 J/cm
2 UVA1 radiation. Due to the penetration of up to 20% of UVA to the level of dermal vasculature, the induction of the programmed cell death cascades may develop to one of the main side effects of UVA phototherapeutic strategies [
144].
Even though especially in case of LE sunlight exposure has been postulated to induce exacerbation in as much as half of the patients, courses of UVA irradiation can also be used in such disease. In this respect, UVA2 and UVB seem to be responsible for the induction of LE eruptions [
145]. Nevertheless, different studies provide strong evidence that apoptosis associated with a shift of the balance between p53 and bcl-2, simultaneously one of the main mechanisms of action concerning UVA phototherapy, may play a role in the pathogenesis and activity of LE and might to correlate with the sequential progress of LE skin lesions [
146,
147].
As the peak of dermatological therapeutic usage of UVA irradiation still seems to rise, scientific research engagement is needed in order to rate its potential long-lasting negative impact. Therefore, until the evaluation of firm data UVA phototherapy should most likely be restricted in the number of cycles per year, treatment should be supervised by an experienced dermatologist and a UV pass book should be issued.
Conclusion
Today, by considering this widespread range of clinical and experimental studies, one might clearly conclude that different regimens of UVA phototherapy have simultaneously been developed to effective, often well-tolerated and beneficial therapeutic strategies in the treatment of a variety of sclerotic skin diseases such as SSc, LS, chronic GvHD, extragenital LSA or sclerodermoid rarities and other disorders affecting the connective tissue. In this respect, the controlled application of UVA irradiation seems to exhibit a comparably tolerable risk/benefit ratio as a minimum in case of these precise indications. Furthermore, UVA phototherapy might also be considered as an optional treatment in both the cutaneous and systemic forms of LE, although the mechanism of action remains difficult to understand.
Nevertheless, therapeutic application of UVA phototherapy, especially of long-wave UVA1 phototherapy, is still 'under construction' as to its limited availability apart from selective centers of excellence and mostly uncontrolled pilot investigations or case reports especially as far as PUVA or conventional UVA1 phototherapy in sclerodermic skin affections or autoimmune disorders are concerned. From our point of view there is no doubt that UVA phototherapy could by far be much more frequently used in the treatment of connective tissue lesions of the skin and, due to its ability to affect dermal vascular structure, even of extracutaneous manifestations. However, additional research efforts are required to determine an exemplary clear-cut catalogue of indications responding to UVA irradiation. Therefore, oncoming controlled randomized studies evaluating the efficacy of UVA phototherapy in connective tissue diseases should not only focus on the assessment of further innovative indications, but also on the comparison between phototherapeutic agents and conventional immunosupressive/-modulating regimens as for example systemic glucocorticoids, azathioprine, methotrexate or cyclophosphamide as well as on the confirmation of former uncontrolled reports, not only because of the fact that the natural history of fibrotic disorders includes a period of inflammation/rapid induration followed by a prolonged period of regression even in untreated patients, but also in order to expand its usage to a widely available treatment option.
Authors' contributions
FB conceived of this investigation including its methods and manuscript structure, performed the comprehensive literature search including data extraction and interpretation, and finished the paper. TG participated in the literature search. A.K. conceived of the study. P.A. participated in its design and coordination.
All authors read and approved the final manuscript.