Frequency of skin cancer among psoriasis, vitiligo, and mycosis fungoides patients treated with narrowband ultraviolet B phototherapy

Narrowband ultraviolet B (NB‐UVB) phototherapy is a popular and relatively contemporary treatment option. However, only a few studies to date have explored the potential risk of skin cancer following NB‐UVB treatment.


| INTRODUC TI ON
[3][4] Despite its high therapeutic efficacy, the long-term effects of phototherapy may potentially include photo-carcinogenesis, as 90% of nonmelanotic skin cancer (NMSC) stem from UV radiation. 5,6rcinogenesis has been previously described with psoralen plus ultraviolet light A (PUVA) therapy. 7Similarly, a potential relationship between NMSC and NB-UVB has been hypothesized but not yet proven.10][11][12][13][14][15][16][17][18][19][20] This study investigated the potential long-term risk of skin cancer in patients treated with NB-UVB for psoriasis, vitiligo, and mycosis fungoides.This study was approved by the local institutional review board, which waived the requirement for informed consent given the retrospective nature of the study and utilization of de-identified patient data.

| RE SULTS
A total of 767 patients treated with NB-UVB were included in the study, including 509 with psoriasis, 122 with vitiligo, and 136 with mycosis fungoides.The mean age and skin cancer ratio according to diagnosis are shown in Table 1.
The mean follow-up time was 12.88 ± 5.58 years.The mean number of treatments was 99.09 ± 70.26 for patients who developed skin tumors versus 94.79 ± 104.56 for those who did not.However, this difference was not statistically significant (p = .81).
During the follow-up, 34 patients (4.43%) developed cancerous skin lesions.The mean age at diagnosis of skin cancer was 64.68 ± 13.12 years.The mean time from the commencement of treatment to tumor diagnosis was 8.1 ± 2.9 years.A significant positive relationship was found between the advancing age and the development of skin cancer in a multivariate analysis (p < 0.05).A lighter skin type (mostly types 2 and 3) was identified as a significant risk factor (p < 0.05).
The study cohort included 509 patients with psoriasis with a mean age of 47.86 ± 16.57 years (Table 1), of whom 20 (3.93%) were diagnosed with skin cancer at a mean age of 63.95 ± 11.14 years.
The study cohort also included 122 patients with vitiligo (mean age, 41.63 ± 15.49 years; Table 1), among whom three (2.46%) were diagnosed with skin cancer at a mean age of 49.67 ± 9.61 years.
The highest occurrence of skin cancer was observed among patients with mycosis fungoides; 11 (8.09%) of the total 136 patients (mean age, 41.63 ± 15.49 years; Table 1) were diagnosed with skin cancer at a mean age of 70.09 ± 14.58 years.Conversely, patients with vitiligo had the lowest occurrence of skin cancer (2.46%).
However, this difference was not statistically significant.
Individual data for patients with psoriasis, mycosis fungoides, and vitiligo, who developed skin cancer, are presented in Tables 2-4.
These tables overview the patients' demographics, skin type, number of treatments, skin cancer type, follow-up time, and the time elapsed until the cancer diagnosis.

TA B L E 1
Comparison between patient demographics and skin cancer ratios in psoriasis, vitiligo, and mycosis fungoides cohorts.In this cohort study, none of the three groups (psoriasis, vitiligo, and mycosis fungoides) showed an association between the number of NB-UVB treatments and development of skin cancer.7][18][19][20][21][22] The highest skin cancer ratio was observed in the mycosis fungoides group (8.09%), followed by the psoriasis (3.93%) and vitiligo (2.46%) groups.However, the intergroup differences were not statistically significant.Furthermore, a significant correlation was observed between older age and skin cancer occurrence.
In recent decades, there has been an increase in skin cancers, with an incidence of 3%-8% in Caucasian populations in Europe, the United States, Canada, and Australia. 21The incidence of nonmelanotic skin cancer in the general population of Israel is unknown.
The ratio of skin cancer in our study (4.34%) aligns with general population findings from studies conducted in populations with skin types and latitudes comparable to those in Israel. 22,23This supports our finding that there is no relationship between NB-UVB treatments and skin cancer.The relatively higher occurrence in the mycosis fungoides group in our study may be attributed to the higher mean age of this group compared to the other groups.It is unlikely that NB-UVB was the cause of the higher cancer occurrence in this group, as the mean number of treatments for patients diagnosed with cancer was lower than in those who were not.
In our study, skin cancer developed within a mean period of 8 years from the onset of treatment.This timeline could suggest a possible correlation between NB-UVB treatments and the development of skin cancer, given that carcinogenesis typically progresses slowly over time. 24Nevertheless, patients who developed skin cancer had a significantly higher mean age, a significant risk factor for skin cancer in this study population.
Few studies have explored the relationship between skin cancer and NB-UVB treatments.Many of these studies reached similar conclusions, supporting the results of our study, with no significant risk of skin cancer in patients treated with NB-UVB.In a study from Scotland 17 involving 3867 patients (55% of whom had psoriasis) treated with NB-UVB and followed for up to 22 years, no significant relationship with skin cancer was found.Another study from Germany 12 included 195 patients with psoriasis who were followed up for up to 10 years; of them, 126 were treated with NB-UVB.This study observed no skin cancers among the treated patients other than a single in situ melanoma identified in the first year of treatment.[27][28] In contrast, a recent study by Chia et al., 29 in Singapore, including 3730 patients (psoriasis, 58%; vitiligo, 42%), showed an increase in skin cancer incidence, mostly among patients with psoriasis treated with NB-UVB, relative to the local general population.A significant correlation with skin malignancy was also found for the cumulative therapy dose and previous systemic treatments.
However, in line with our results, there was no correlation with the number of treatments.A possible explanation offered is that substantially higher phototherapy doses are required for darker Asian skin phototypes, along with an ascertainment bias due to increased surveillance.In another study published in 2016, 30 six of 50 patients with psoriasis (12%) treated with NB-UVB were diagnosed with skin cancer.This was particularly evident in patients who received more than 200 treatments; however, the study population was relatively small.An additional study from Scotland 15 including 1908 patients (48% with psoriasis) treated with NB-UVB demonstrated a small but significant increase in the incidence of basal cell carcinoma (BCC) compared with the control group.This was attributed to possible diagnostic bias, as the patients were extensively followed up, and the possible tendency of patients with a good response to UVB to intentionally increase their exposure to sunlight and, consequently, their cancer risk.Furthermore, disregarding the three patients with BCC who underwent excision within the first months of the initial NB-UVB treatment rendered the increase non-significant.
One of the main limitations of our study is that the cumulative radiation dose over time was not tracked.Therefore, the number of treatments per patient was the only indicator of the exposure levels.
Another limitation is that the analysis in this study did not include age adjustment, which would have required a larger study population.Furthermore, this study did not include a control group that was not treated with NB-UVB; consequently, isolating the effect of treatment on carcinogenesis was challenging.
In conclusion, we found no significant difference in the skin cancer rate among patients with psoriasis, vitiligo, and mycosis fungoides treated with NB-UVB, nor did we observe any association with the number of treatments during at least 10 years of follow-up.
The occurrence of skin cancer in the study population was similar to that expected in the general population.Thus, it appears that NB-UVB is a relatively safe treatment option for these patients.Age is a significant risk factor for carcinogenesis; therefore, older patients treated with NB-UVB should be followed up carefully.As we could not definitively exclude NB-UVB as a potential trigger for carcinogenesis, future studies should employ a prospective design, including radiation dose tracking, control groups, age adjustment, and a larger study population.

AUTH O R CO NTR I B UTI O N S
Our cohort study included patients exposed solely to NB-UVB phototherapy for psoriasis, vitiligo, and mycosis fungoides in the phototherapy units of two university hospitals in Israel in 2000-2005.Patients were followed up for a minimum duration of 10 years, until the beginning of 2016.Data were collected from the personal electronic medical records of the aforementioned institutes and/or from the patients' health maintenance organizations.The collected data included age, sex, skin type, dermatological diagnosis (psoriasis, vitiligo, or mycosis fungoides), number of NB-UVB treatments per body area, use of concomitant medication, and follow-up duration.For patients diagnosed with skin cancer, the time interval between the initiation of treatment and the development of cancer and its subtype were recorded.Statistical analyses were conducted using R in SAS, version 9.4 (SAS Institute).Descriptive statistics are presented as mean ± standard deviation or proportion, as appropriate, for quantitative variables.Fisher's scoring algorithm was used to analyze the relationship between the number of NB-UVB treatments and skin cancer by patient group.As a relatively small number of skin cancers was anticipated, we decided not to analyze the results for each cancer type separately.All reported p-values are two-sided.Values of p < .05were considered statistically significant.
Mimouni designed, drafted, interpreted, and revised this work.Julia Shulman designed, acquired data, analyzed, and interpreted this work.Awni Abu Unes acquired data, analyzed, and interpreted this work.Lev Pavlovsky and Felix Pavlotsky designed, supervised, and critically revised this work.All authors approved the submitted version of this work and agreed to be personally accountable for their own contributions.