Corresponding author: Dr L.E. Rhodes University of Manchester Clinical Sciences Building Hope Hospital Manchester, M6 8HD, UK Tel: 44 161 7871150 Fax: 44 161 7871304 e-mail: email@example.com
Polymorphic light eruption (PLE) is a highly prevalent photosensitivity disorder, estimated to affect 11–21% people in temperate countries. Typically, PLE appears as a recurrent pruritic eruption comprising papules and/or vesicles and/or plaques, which occurs on photo-exposed skin sites following sun exposure, and which heals without scarring. Commoner in females, the aetiology is uncertain, although there is evidence of an immune basis. We perform a review of the prophylaxis and treatment of this condition. While sun protection, corticosteroids and desensitization phototherapy are the mainstays of management, a range of anti-inflammatory and immunomodulatory agents are reported.
Polymorphic light eruption (PLE) appears to have been first described in 1817 by Robert Willan, who used the term ‘eczema solare’ (1). Later, Rasch coined the term ‘polymorphous light eruption’ (2). It is the most common photosensitivity disorder in white Caucasians, estimated to affect 11–21% of the population in the Northern Hemisphere (3–6). Its incidence in other populations is largely unknown, although it has been reported to occur in 5% of white Australians (3). The onset of the disease is typically in the second to third decade of life. More females are affected than males, in the ratio of approximately 3 : 1 (3–5, 7, 8), although higher ratios of 6/7 : 1 have also been reported (9, 10).
The aetiology, clinical features and investigation of PLE have been extensively reviewed (10) and will only be briefly considered here. While the exact mechanism of the disorder remains unknown, studies have suggested a delayed-type hypersensitivity immunological reaction (11), with some evidence of failure of ultraviolet (UV)-induced local immunosuppression (12). Recent genetic studies suggest a strong inherited component (13, 14). The skin eruption is most commonly papulo-vesicular (Fig. 1), but may also take other forms, including plaques, bullae and purpura. It usually occurs in a cyclical fashion, beginning in spring and resolving by autumn. Many sufferers improve by late summer with ‘hardening’, i.e. increased tolerance, of the skin. Some patients with a higher threshold for rash provocation may only experience symptoms on sunny holidays, while those most severely affected may also be symptomatic in the winter months. Most skin lesions develop after a latent period following sun exposure, typically several hours, but a minority have more rapid onset of symptoms. The lesions then resolve spontaneously after a few days to a week or two, without scarring. The diagnosis of PLE is usually made clinically, but may be supported in atypical cases by histological and phototest findings. A positive photo-provocation test by means of exposure to UVA and/or UVB emitting lamps can be very helpful (Fig. 2). The action spectrum for provocation of the disease is more often in the UVA than UVB waveband, with many patients sensitive to both (9, 16, 17). Although it is reported to be positive in 50–74% cases (9, 15–17), the provocation yield can be improved by repeated testing (18). Photopatch testing may reveal co-existing sunscreen allergies (19), which will influence management.
Treatment of PLE
For the majority of patients with PLE, the rash is mild and self-limiting and quickly settles within a few days of sun avoidance. The proportion of people with PLE who seek medical attention is estimated to be less than 26% (3–5), although this still represents a large number of people, considering that the disorder may affect up to 21% of the population. Treatments may be prophylactic or suppressive, and their application depends on the severity of the disease, availability of equipment and patient choice.
FIRST LINE TREATMENTS FOR PLE
For all sufferers, preventive management is advised during sunny weather, by avoidance of intense UV exposure between 11.00 and 15.00, use of protective clothing and application of sunscreen. The tightness of the weave of clothing fabrics determines the amount of photoprotection – thus while a typical crepe blouse gives a sun protection factor (SPF) of only 5, a jersey T-shirt has an SPF of 32 and a silk blouse has an SPF of 280 (20). From a practical viewpoint, holding the fabric up against bright lighting will give the patient some indication of the tightness of the weave and hence its suitability for protection. It should be noted that the transmission of UV radiation for all fabrics is increased when a fabric is wet (21).
Older generation sunscreens that are protective primarily against UVB may not provide adequate protection against provocation of PLE (22). The potentially misleading term, SPF, of a sunscreen refers to protection against the erythemogenic effect of UVB, but does not address UVA protection; sunscreens with high SPF values do not confer equal protection against provocation of PLE (23). Unfortunately, there is no uniform international system for assessment of UVA protection by sunscreens, although in some instances, this is indicated by a star system (one star – least protection; four stars – maximum protection). New generation broad-spectrum sunscreens have a high SPF, together with longer wavelength UVA protection, and have been reported to confer total or partial protection in up to 90% of PLE sufferers (24, 25). In a study of 45 clinic patients (24), the regular application of a broad-spectrum sunscreen (Uvistat Ultrablock 30®) was reported to prevent PLE in 90% patients either totally or partly over the summer months. Another study showed that 25 patients who had SPF 60 sunscreen (containing Mexoryl SX®, Parsol 1789 and titanium oxide) applied at the manufacturers' test thickness, i.e. 2 mg/cm2, to 5×5 cm areas on the chest, did not develop PLE with repeated provocation over 5 days (25).
Patient education in appropriate sunscreen application technique is important. Previous studies have demonstrated that photosensitive patients only apply approximately a quarter of the thickness of cream used under the manufacturers' test conditions and that certain exposed body sites (ears, temples, posterior and lateral neck) are often missed (26). Education can result in sustained improvement in patient sunscreen application technique (27). However, UV absorbers in chemical sunscreens are now the most common cause of positive photopatch tests (19), and potential photoallergies to sunscreens should be considered in cases of worsening or atypical PLE (28).
While no trial has made a detailed examination of the efficacy and use of topical steroid in PLE, it is widely used in this disorder. It is reported that potent topical steroid may be helpful in relieving itch, and that it is helpful in some, but not all, patients (29). Used alone, it may be adequate for people who have mild episodes of PLE. It may also be used in combination with phototherapy or photochemotherapy, to reduce the incidence and severity of provoked rash during treatment (30, see UVB section).
SECOND LINE TREATMENTS FOR PLE
Systemic corticosteroids may be needed to settle attacks of PLE. In a randomized, double-blind, placebo-controlled trial (31), 21 PLE patients were given a supply of both prednisolone (25 mg) and placebo tablets prior to a sunny vacation. Half the patients used the corticosteroid first if they experienced PLE symptoms and half used the placebo first, and they were instructed to switch to the alternative therapy if symptoms persisted after 48 h. Of the 10 patients requiring medication, eight who took prednisolone first or transferred to it from placebo found that the itch and rash settled more quickly compared with the two patients who took placebo first and remained on it (2.8 vs. 5.4 days for itch, 4.2 vs. 7.8 days for rash). One patient experienced adverse effects of transient gastrointestinal upset and depressed mood. The authors suggested the use of 25 mg prednisolone daily for 4–5 days at the onset of an attack. However, the potential long-term side effects of repeated courses of prednisolone must be considered, particularly since some patients may take several sunny holidays per year. There may therefore be a place for the cautious use of prednisolone for those who suffer from occasional and sufficiently symptomatic attacks of PLE on vacation, and in the absence of any contraindication.
In milder cases of PLE, a self-conditioning programme by graduated exposure to sunlight in springtime may be appropriate (32). In more severe PLE, medically supervised conditioning/desensitisation treatment remains the mainstay of the treatment (Fig. 3). A course of psoralen and UVA therapy (PUVA), narrowband (NBUVB) or broadband UVB (BBUVB) phototherapy, usually administered in early spring, can be effective as prophylactic treatment (Table 1a,b). The mechanisms by which phototherapy induces photoprotection are not fully understood. In addition to photoprotection by induction of melaninization and epidermal thickening, a range of UV-induced immunomodulatory and anti-inflammatory effects are reported (33). UVB appears to produce its effects largely via epidermal keratinocytes and Langerhan cells, while UVA penetrates more deeply into the dermis to have a greater direct influence on fibroblasts, endothelial cells and T-lymphocytes. Both types of radiation modulate adhesion molecule expression and induce soluble mediators, including α-melanocyte-stimulating hormone, IL-10 (which suppresses the production of interferon-γ) and prostaglandin E2. Prostaglandin E2 influences the expression of costimulatory molecules on the antigen-presenting cell surface and prevents the activation of T cells. Moreover, UVR induces apoptosis in skin-infiltrating T cells (33).
Table 1a. Summary of UVB phototherapy trials in PLE
With prednisolone cover for first 2 wks of phototherapy
Ortel (1986) (43)
3×/wk for 3–4 wks
Jansen (1982) (36)
3×/wk for 3–4 wks, then 1–2×/wk
13–42 (mean 23)
3×/wk for 3–4 wks, then 1–2×/wk
13–24 (mean 20)
Parrish (1979) (35)
Prospective parallel (vs. β-carotene)
2–3×/wk for 4–12 wks
Patient selection for PUVA determined by geographical factors
Gschnait (1978) (34)
4×/wk until tanned
All five patients spent holiday in intense solar radiation – two had slight recurrence
Psoralen and UVA therapy (PUVA)
The benefit of PUVA in PLE was demonstrated in several earlier uncontrolled studies (34–36). Gschnait et al. (34) first hypothesised that the induction of melanin pigmentation offered photoprotection. In a series of five PLE patients, 8-MOP PUVA was given 4× per week for 6–10 sessions in May and June; and allowed all patients to spend their vacations in intense sun. Later, Parrish et al. (35) demonstrated that PUVA was more effective than β-carotene (see below) in 29 patients, while Jansen et al. (36) reported that 12/13 of patients receiving systemic PUVA and 10/13 those receiving bath PUVA had clinical benefit. However, when oral trimethoxypsoralen (TMP)-UVA was compared with UVA alone in a randomized, double-blind controlled trial of 22 patients (6), no significant difference in clinical benefit was found.
In an uncontrolled, prospective trial of BBUVB, Addo and Sharma (37) compared two groups of 10 patients, receiving either three or five times weekly treatment. While 2/10 of the group receiving five treatments a week developed severe adverse reactions requiring temporary cessation of therapy, the three times weekly group had minimal reactions only. Another uncontrolled prospective trial of eight patients found a high incidence of provocation (38%) and erythema (100%) episodes with five times weekly treatment (38). Both studies observed a high remission rate, while the three times weekly regime may be better tolerated than the five times weekly protocol.
NBUVB vs. BBUVB
NBUVB therapy might be anticipated to cause less unwanted disease provocation than BBUVB, due to the lack of emitted UVA radiation. However, there are no reported comparison studies of NBUVB with BBUVB in PLE.
PUVA vs. BBUVB
In a randomized, double-blind controlled trial (39), PLE patients received 8-MOP plus UVA (n=13), UVB plus placebo tablets (n=13) or low dose UVA plus placebo tablets (n=12). The groups were broadly comparable at baseline. Three times weekly treatments were administered for 6 weeks. Using self-assessment of treatment efficacy, 92% patients considered PUVA successful compared with 62% with UVB. These findings were supported by Addo and Sharma (37), who reported complete remission in 89% of patients treated with PUVA compared to 69% treated with UVB.
PUVA vs. NBUVB
Thrice-weekly NBUVB for 5 weeks has been shown to be as effective as oral 8-MOP PUVA in reducing the frequency and severity of PLE symptoms (40, 41). In a randomized double-blind comparative trial, 13 patients were randomized to receive NBUVB (and placebo tablets) and 12 were randomized to receive PUVA during March and April (40). All patients had a history of frequent PLE episodes from UK sun exposure, and had not benefited from the use of broad-spectrum sunscreens. The patients then recorded a weekly diary of PLE occurrence, severity and restriction of outdoor activity from May to September. When correlated with individual solar exposure (monitored by polysulphone film badges), the authors concluded that there was no significant difference in the efficacy of NBUVB and PUVA. Compared to oral PUVA, NBUVB has several advantages; absence of risk of gastrointestinal upset with oral 8-MOP or need for protective spectacles post-exposure, ability to use in pregnancy and in childhood, and controversially, possibly reduced photocarcinogenic potential.
UVA1 and UVA/UVB
The use of UVA1 (340–400 nm) and UVA/UVB (300–400 nm), as found in sunbeds, was examined in 31 PLE patients (42). After 10 sessions of phototherapy, 48% of subjects experienced PLE during 10 weeks of follow up, although it tended to be less severe than in previous years. Histidine and urocanic acid skin levels were increased during and immediately after treatment, suggesting biochemical sun protection, but this rise was not sustained at 10 weeks.
General considerations in photo(chemo)therapy for PLE
Courses of phototherapy/photochemotherapy are generally given over 5–6 weeks, although the regimes vary between centres. Starting doses may depend on minimal erythemal dose (MED) or minimum phototoxic dose (MPD) testing, and are frequently 50–70% of these measured erythemal thresholds. On the other hand, fixed dose regimens aim to use a starting dose too low to cause erythema or to provoke the disease in the majority of patients. Incremental increases are frequently prescribed at alternate UV doses. Following each treatment course, regular sun exposure throughout summer to maintain the level of hardening is advised, as the benefit may otherwise be lost within 4–6 weeks.
Management of a provocation episode
The risk of provoking a PLE episode, particularly with the first few photo(chemo)therapy exposures (41, 43), is estimated to be around 50% (41). Bilsland et al. (40) reported provocation rates of 62% and 50% for NBUVB and PUVA, respectively. This group's strategy was to manage moderate provocation with repetition of the previous dose and severe provocations by omission of a treatment, both followed thereafter by lower dose increments. Topical (30) or systemic (35) steroids may be required. Indeed, some clinicians have advised administration of prednisolone at the outset of phototherapy in order to improve tolerability (44). Alternatively, a potent topical steroid may be applied prophylactically to treated sites immediately after exposure in order to reduce the risk of provocation. A small double-blind study showed that routine application of betamethasone dipropionate or clobetasol propionate immediately after TL-01 UVB exposure prevented provocation in 5/7 patients (30).
Carcinogenic potential of photo(chemo)therapy
Psoralen and UVA therapy
The carcinogenic potential of PUVA is highlighted by a prospective study of a cohort of 1380 USA psoriasis patients who were first treated in 1975/1976. The risk of cutaneous squamous cell carcinoma (SCC) was increased 11-fold with high dose exposure of >260 treatments compared with patients who had received 160 or fewer treatments (P<0.01), 10 years after their first PUVA treatment (45). Indeed, this cancer risk persists after discontinuation of therapy (46), while the risk of basal cell carcinoma is only substantially increased in patients exposed to very high doses of PUVA (≥337 treatments) (46). The risk of melanoma is increased 15 years after the first PUVA treatment, is dose-dependent (≥250 treatments) (47) and appears to increase with the passage of time (48). The PUVA-related increased risk of skin cancer is corroborated by an 11-year retrospective survey of 4799 Swedish patients (49) with various skin diagnoses. The risk of developing cutaneous SCC increased dramatically above 1200 J/cm2 or 200 PUVA treatments (approximately 30-fold in men). Both cohorts give evidence of increased internal malignancy (49, 50), but of different systems.
A meta-analysis of photocarcinogenicity in BBUVB-treated psoriasis patients (51) gave an estimate of excess incidence of nonmelanoma skin cancer of two per 100 patients treated per year. This risk is higher in PUVA-treated patients (52, 53). Murine studies conducted in the 1990s indicated that the photocarcinogenic potential of NBUVB was up to two times greater than BBUVB (54–56), and an expert panel concluded that NBUVB was probably two to three times more carcinogenic per MED than BBUVB (57). Clinically, however, the relative risks of NBUVB and BBUVB will depend on dose regimens applied. Predictions cannot be made for PLE, since we do not know the comparative efficacy, including numbers and doses of treatment needed, of BB and NBUVB in this disorder.
When to prescribe photo(chemo)therapy in PLE
In summary, NBUVB and PUVA appear equally effective in PLE (40), while BBUVB may be marginally less effective (37, 39). However, the risk of photocarcinogenesis and also the time, cost and inconvenience involved in the treatment, should be weighed against the benefits. For many PLE sufferers, phototherapy may be reserved for years in which there are special holidays/occasions, with a minority of the more severely affected patients receiving annual treatment.
THIRD LINE TREATMENTS FOR PLE
The clinical effectiveness of azathioprine in PLE was demonstrated in two patients unresponsive to other treatments, lending support to an underlying immunological basis of the disorder (58). After 3 months, phototesting revealed increased minimal erythemal responses. Azathioprine may be considered appropriate for patients who are exquisitely sun-sensitive, in whom sunscreens are ineffective and who cannot tolerate phototherapy, but is contraindicated in female patients trying to conceive. Other side effects are myelosuppression and gastrointestinal upset, and a monitoring of full blood count, renal and liver function are mandatory. The clinical benefit of cyclosporin in PLE was reported in a patient treated for co-existing psoriasis (59).
OTHER TREATMENTS FOR PLE
Omega-3 fatty acids
E coli filtrate
Hydroxychloroquine has membrane-stabilizing properties, leading to a range of anti-inflammatory effects, including inhibition of proteolytic enzymes in connective tissue (60). Hydroxychloroquine was reported to have a mild benefit vs. placebo in PLE in a double-blind controlled trial of 28 matched patients (61). The dose of hydroxychloroquine used was 400 mg daily for the first month and 200 mg thereafter for a total of 12 weeks. The severity of rash reduced by two points on the visual analogue scale (P<0.01), but PLE was not abolished in most patients. An earlier study by Corbett et al. (62) of 120 PLE patients taking chloroquine 400 mg daily for 12 weeks found only a modest reduction in skin irritation. In both studies, full autoantibody screening had not been available and it is possible that some cases of subacute lupus erythematosus, which is responsive to hydroxychloroquine, might have been misdiagnosed as PLE. Side effects of the drug include gastrointestinal upset and oculotoxicity. Corneal deposits are asymptomatic, dose-related and not a contraindication to continuing treatment, whereas retinopathy is irreversible and the most serious complication (63). A maximum daily dosage of 6.5 mg/kg lean body weight (usually 400 mg daily), is recommended, with baseline and annual assessment for visual symptomology (testing of near visual acuity using a standard reading chart), and referral to an ophthalmologist if there is visual impairment at baseline or if changes in visual acuity develop on treatment (64).
β-carotene is a scavenger of singlet oxygen and free radicals, and is also capable of absorbing 400 nm radiation. While there is some evidence that it may increase the MED to UV (65, 66), the reported clinical benefit in PLE is generally unimpressive. While it was suggested that β-carotene improved sun tolerance in PLE (67), in most studies the benefit was minimal, or when used in combination with topical sunscreen (35, 68, 69). Furthermore, in controlled trials (62, 70), there was no demonstrable benefit over placebo. Hence, it is not a standard treatment in PLE.
This was reported to be moderately effective in an uncontrolled trial of 42 patients, who also used topical sunscreens of SPF 6–17 (71). When taken at a dose of 2–3 g daily before sun exposure, 60% patients reported complete abolition of symptoms despite continued sun exposure, while 26% reported no benefit. Apart from mild fatigue in some patients, the treatment was well tolerated. The authors proposed an underlying error in tryptophan metabolism in PLE. However, nicotinamide did not prevent provocation of PLE lesions with UVA and UVB in 14 patients (72).
The use of thalidomide in PLE has been quoted in reviews (73–75), which have not clearly distinguished actinic prurigo from PLE. In a case series of 25 patients (76) with ‘PLE’, 88% good to excellent lesion healing was reported with thalidomide. However, from the clinical details presented, it is clear that most, if not all, the patients suffered from actinic prurigo or ‘hereditary PLE’ as it was previously known in people of American Indian descent (77). A range of immunosuppressive and anti-inflammatory properties of thalidomide have been seen in vivo and in vitro (78). Its teratogenic effects are well known, and it is not recommended for use in fertile women of child-bearing age. The other major adverse effect is peripheral neuropathy, estimated to complicate 21–50% cases; rarely, it may also cause exfoliative dermatitis and vasculitis.
ω-3 polyunsaturated fatty acids
Preliminary studies of the effect of oral ω-3 polyunsaturated fatty acids (PUFAs) in PLE showed that this dietary supplement is a potential treatment for this disorder (79). ω-3 PUFAs are found in oily fish such as herrings and sardines, and are generally lacking in the Westernized diet. In uncontrolled studies, 10 g mixed ω-3 PUFAs (MaxEPA®, Seven Seas, Hull, UK), taken as five capsules twice daily day for 3 months, significantly increased the MED to UVB, and significantly increased the threshold for UVA-provocation of rash. Many patients additionally reported clinical benefit and have continued to take the supplements annually during the spring and summer months. More recently, significant protection against UVB-induced erythema and cellular damage by a purified ω-3 PUFA, eicosapentaenoic acid, has been confirmed in a double-blind randomized study in healthy volunteers (80). ω-3 PUFAs may be acting in PLE to modulate inflammatory and immune responses via their demonstrated ability to influence UV-generated skin PGE2 levels in PLE (79), by acting as a buffer to reactive oxygen species (81) or by effects on cell signalling events (82).
A randomized, double-blind controlled study involving 30 PLE patients showed that three topical antioxidant mixtures (comprising α-glycosylrutin, tocopherol acetate and ferulic acid), significantly reduced the development and severity of PLE induced experimentally (83). This suggests that the underlying pathogenetic mechanisms of PLE may involve UV-induced reactive oxygen species formation. A randomized controlled study involving 27 PLE patients showed that the topical application of α-glucosylrutin before sun exposure reduced the severity of PLE symptoms, although this was in combination with SPF 15 sunscreen (84). However, in a small double-blind controlled trial involving nine patients, prior systemic administration of vitamin C and vitamin E for 8 days did not reduce the severity of photo-provoked PLE (85).
E. coli filtrate
Colibiogen® is a commercially available E. Coli filtrate used for the treatment of gastrointestinal and allergic conditions. A small preliminary study involving nine patients suggested that its intramuscular, but not oral, administration significantly increased the provocation threshold of PLE to UVA (86). More work is needed to establish its efficacy in clinical situations. The mechanisms underlying its effects are unknown, but it appears to be a biologic response modifier.
There are wide-ranging treatments reported for this very common and sometimes disabling photosensitivity condition, indicating the challenges of its treatment. Prophylactic measures, such as sun avoidance and application of sunscreen are important, although of limited benefit to many individuals. Topical steroids are used during attacks of the disease, but systemic steroids may be warranted and are effective. For the more severely affected patient, desensitization phototherapy is the mainstay of management. However, this is time-consuming and conveys long-term risks. Hydroxychloroquine and β-carotene have minimal, if any, benefit. Immunosuppressive agents may be considered, and appear effective, in the most afflicted patients. It is difficult to draw firm conclusions about the efficacy of many potential treatments in PLE due to the study design and difficulty in assessing outcome measures. Several potential treatments require further double-blind, randomized controlled trials to assess clinical efficacy. Further research into the underlying mechanisms of the condition may permit a more targeted treatment approach.
Dr T.C. Ling was supported by the European Union Framework V Progamme, project no. QLK4–CT01–0015.