Carcinogenic risks of psoralen UV-A therapy and narrowband UV-B therapy in chronic plaque psoriasis: a systematic literature review.

Oral 8-methoxypsoralen-UV-A (PUVA) and narrowband UV-B (NB-UVB or UVB TL-01) are effective and widely used treatments for chronic plaque psoriasis. Although the role of PUVA therapy in skin carcinogenesis in humans with psoriasis has been clearly demonstrated, there is still controversy regarding the risk of skin cancer with NB-UVB. Furthermore, there is no clear evidence about the maximum cumulative number of sessions not to be exceeded in a lifetime. To assess the respective cutaneous carcinogenic risks of PUVA or NB-UVB in psoriasis; to estimate the respective dose-relationship between skin cancers and PUVA or NB-UVB; to estimate a maximum number of sessions for PUVA or NB-UVB not to be exceeded in a lifetime. A systematic literature search was carried out in Medline, Embase and Cochrane Library databases from1980 to December 2010 in English and French, with the keywords 'Psoriasis' AND 'UVB therapy' AND 'UVA therapy' AND 'cancer' AND 'skin' OR 'neoplasm' OR 'cutaneous carcinoma' OR 'melanoma'. Of 243 identified references, 49 published studies were included. Most of them (45/49) concerned PUVA therapy, with 41 assessing the risk of non-melanoma skin cancers (NMSC) following PUVA. All publications referring to the US prospective PUVA follow-up study revealed an increased risk of NMSC with the following characteristics: risk most pronounced for squamous cell carcinomas developing even with low exposures and increasing linearly with the number of sessions, tumors occurring also on non-exposed skin including invasive penile tumors, risk persisting after cessation of treatment. An increased risk of basal cell carcinomas was observed in patients receiving more than hundred PUVA sessions. The four prospective European studies selected in our review and most of the pre-1990 European and US retrospective studies failed to find a link between exposure to PUVA and skin cancer. Only the most recent cohorts, including three large long-term retrospective European studies comparing records with their respective national cancer registries reported on an independent increased risk of NMSC with PUVA, The risk was lower as compared to the US prospective PUVA follow-up study. Six studies assessed the risk of melanoma following PUVA therapy: two of the three US publications coming from the same PUVA prospective follow-up study revealed an increased risk with more than doubled incidence of both invasive and in situ melanoma among patients exposed to at least 200 PUVA treatments compared with patients exposed to lower doses, whereas the three retrospectives European studies, comparing the incidence of melanoma in PUVA users with national cancer registers, did not find any increased risk of melanoma. No increased risk of skin cancer was evidenced in the four studies specifically assessing the potential carcinogenic risk of NB-UVB. There is an increased risk of skin cancer following PUVA, shown by both US and European studies. The greater risk measured by the US studies may be at least partly explained by high UVA dose exposure and the lighter phototypes of the treated patients. The lack of prospective studies in psoriasis patients treated with NB-UVB constitutes a barrier to the robust assessment of carcinogenic risk of this phototherapy technique.