Inclusion criteria were consecutive subjects 18 years or older diagnosed of PLE between 2014 and 2015 which reported benefit with natural hardening and who accepted the 2 days per week regime during one month period.

The diagnosis of PLE had been previously established by clinical history, phototesting and laboratory tests (including antinuclear antibodies (ANAs) and porphyrins to rule out lupus erythematous (LE), erythropoietic protoporphyria (EPP) and other sun induced skin diseases). None of the subjects were taking photosensitizing drugs at the time of diagnosis. All subjects presented frequent episodes of PLE and had not benefited from the use of broad-spectrum sunscreens, oral beta-carotene or antihistaminic drugs. Sex, age, clinical presentation, time since the onset of the disease, serum antinuclear antibodies and vitamin D levels, and the result of phototest were registered in our database. The result of the phototest was classified between normal or pathologic, defined as a decreased Minimal Erythema Dose (MED) to UVB or an abnormal UVA reaction.18

This is a retrospective, open planned and non-randomized study to assess the efficacy of UVB phototherapy in prevention of polymorphic light eruption.

Phototesting was performed using a template on uninvolved skin on the patient's back, and subsequent exposure to different doses of UVA and UVB (Waldmann® UVA 700L and UV 801 BL). Each patient was evaluated 24h later for the development of erythema. The minimal erythema dose (MED) was defined as the lowest dose of UVB, or UVA, that produced perceptible erythema, covering the entire irradiated area.18 Photoprovocation was not performed.

Desensitization treatment courses were offered to each subject once a year in spring in 2014 and in 2015. All subjects were whole body irradiated with Waldmann® UV 7001K light stand calibrated annually. The standard treatment protocol for narrowband UVB (TL-01) is shown in Table 1. The starting dose was 0.15J/cm2. This was followed by a 20% incremental dosage, twice a week, during four weeks. The total number of sessions was eight and the final dose achieved was 0.53J/cm2. Each course of treatment was carefully documented including the total number of treatment sessions administered and the frequency of PLE flares. If PLE was provoked, oral corticosteroids were administered without discontinuation of the treatment.

Subjects were told to keep exposing themselves to sunlight through over the summer after the desensitization protocol. As the photo-protective effects of desensitization treatments are temporary with a complete loss of the hardening phenomenon once subjects are no longer exposed to sunlight or the intensity of sunlight decreases (e.g. winter); we considered each treatment course as an independent one, and did not perform a separate analysis of first and second course. Treatment courses in different years were considered not cumulative and were analyzed separately.

Subjects were followed up both after the end of the desensitization treatment and at the beginning of autumn. Each subject's response to treatment during the subsequent months was tabulated and analyzed. Subjects were asked whether they have had flares during the summer, considering complete responders those who reported from 0 to 1 flare and partial responders those who reported 2 flares. Subjects who had more than 2 flares were considered non-responders. Taking extra treatment during the summer such as oral beta-carotene or antihistaminic drugs was also registered.

Statistical analyses were performed with SPSS (version 20.0; SPSS Inc., Chicago, IL). Descriptive analysis of the sample was performed, including percentages for categorical variables, and mean, minimum, maximum and standard deviation values for continuous variables. Comparisons of continuous variable means were performed using Student's exact t-test when variables followed a normal distribution. Comparisons of discrete variable means were performed using the Mann–Whitney non-parametric test. Comparisons between category variables were performed with tests and Fisher corrections were performed when required.

Between 2014 and 2015, 16 subjects with the diagnosis of polymorphic light eruption (PLE) were seen at our Photobiology Unit fulfilled the inclusion criteria for desensitization and agreed to complete the treatment protocol. Desensitization treatment with narrow band UVB (NBUVB) was offered to 15 subjects. The remaining one was not finally included because of personal history of melanoma.

Regarding sex distribution, there were 4 men (26.7%) and 11 women (73.3%). The mean age was 42.4 years (range 16–62). All of them had decreased serum levels of 25-hydroxyvitamin-D3 (<25ng/ml) in blood tests.

Sixty-six per cent of subjects were found to be ANA positive, but only 3 of them had levels >1:160. However, none of them met the diagnostic criteria for lupus erythematous (LE) of the American College of Rheumatology.19

The mean duration of PLE at presentation was 6.13 years (range, 1–15 years). There were no differences between the duration of the disease and the age and sex of the subjects (p>0.05).

Phototesting revealed that the minimal erythema dose for UVB was decreased in 64.8% of subjects tested while the remaining 35.7% fell within normal limits.18 The result of phototesting did not show differences (p>0.05) between age and sex of subjects.

Nine subjects were treated both in 2014 and in 2015 and 6 started phototherapy in 2015. Therefore, a total of 24 desensitization treatment courses were performed between 2014 and 2015. All treatment courses were administered in spring: 8.3% started in March, 50% in April, 33.3% in May and 8.3% in June. All subjects completed their desensitization courses.

Polymorphous light eruption (PLE) was provoked in 8 treatment courses (33.3%). All of the induced PLE were mild, and only 3 of them (12.5%) required the administration of 20mg of oral prednisone in decreasing doses during 10 days. The induced flares resolved in all subjects, and the interruption of the desensitization treatment course was not required in any case.

The evaluation of response was performed both after finishing the desensitization treatment and at the beginning of autumn. The effect of hardening was sustained during the follow up after 21 desensitization treatment courses, as subjects reported a complete remission of symptoms during all the sunny Spanish summer. The remaining 3 desensitization treatment courses were ineffective and subjects continued having disease outbreaks. The nine subjects, who received 2 desensitization treatment courses, were complete responders both in 2014 and 2015.

There was no association between the maintained response to the desensitization treatment and demographic characteristics of the subjects. No association was found either between response and serum levels of 25-hydroxyvitamin-D3, positive antinuclear antibodies (ANAs) in laboratory tests or the result of previous subjects’ phototest (Table 2). Female sex and low vitamin D levels were associated with a better response although this could not be considered statistically significant due to the high percentage of female subjects and the suboptimal vitamin D levels in all the cases.

We found a statistically significant association (p<0.05) between the years of progression of the polymorphous light eruption (PLE) at presentation and the complete remission of symptoms after the desensitization treatment (Table 3). The mean time from when the PLE began was 5.9 years (std. deviation 4.318) in the 87.5% of subjects that presented a sustained response during the summer, while the mean time was 8 years (std. deviation 6.245) in those who did not respond.

With respect to the year's period of treatment, there were no differences (p>0.05) between the sustained response to the desensitization and the month in which photo hardening was performed.

Nearly half of the cases (45.8%) spent the summer without need for any other treatment. In 11 treatment courses (45.8%) subjects attempted to strengthen the response with oral beta-carotene, and only in 2 cases (8.4%) antihistaminic drugs were taken. However, there were no differences (p>0.05) between the maintained response to photohardening and the use of extra treatments after the desensitization course.

The authors declare that no experiments were performed on humans or animals for this study.

The authors declare that they have followed the protocols of their work centre on the publication of patient data.

The authors have obtained the written informed consent of the patients or subjects mentioned in the article. The corresponding author is in possession of this document.