Non-melanoma skin tumors occurring on a Japanese psoriatic patient treated with longterm photochemotherapy

Effect of ethnicity on the risk of developing nonmelanoma skin cancer following long-term PUVA therapy

BACKGROUND

Research demonstrating an increased incidence of skin cancer with psoralen plus ultraviolet A (PUVA) therapy has reflected the Caucasian experience. Our objective was to review the literature on skin cancer risk associated with long-term PUVA therapy in non-Caucasians.

METHODS

Our analysis included 4,294 long-term PUVA patients in Japan, Korea, Thailand, Egypt, and Tunisia with a follow-up period of at least 5 years.

RESULTS

The relative risk of PUVA patients developing nonmelanoma skin cancer relative to general dermatology outpatients was 0.86 [CI 0.36-1.35].

CONCLUSIONS

There does not appear to be an increased risk of nonmelanoma skin cancer with long-term PUVA therapy in Asian and Arabian-African populations. Therefore, in phototherapy risk assessment, it is important to consider the patient's skin phototype and the potential protection that more pigmented skin may confer.

Micronucleus evaluation in mitogen-stimulated lymphocytes of PUVA treated patients

PUVA describes the treatment of patients with psoralens plus an exposure to a source of UV light of 320-400 nm (UVA). Contradictory results have been reported on the chromosomal damage of PUVA when assayed by sister chromatid exchange (SCE) method. Micronucleus (MN) test is used to detect both clastogenic (breaking) and aneugenic (abnormal segregation) effect of physical/chemical agents on the chromosomes. No data have been found on the MN formation in the cells of PUVA treated patients. Frequency of micronuclei in 72 hours cultivated/mitogen-stimulated lymphocytes of patients have been evaluated at zero time and after 20, 40, 60 sessions of PUVA treatment. While the beginning MN frequency was approximately 0.22% (n=23), it raised to approximately 0.32 (n=23), approximately 0.42 (n=14) and approximately 0.53% (n=10) corresponding respectively to 20, 40 and 60 sessions. These sessions correspond reciprocally to 54+/-23, 172+/-48, 300+/-61 joules/cm2 of UVA and 13, 26, 39 mg/kg of 8-metoxypsoralen (8-MOP). While large interindividual variances were apparent, highly significant differences have been observed between initial MN frequency and after that of the 20, 40 and 60 sessions, (p = 0.000, p = 0.004, p = 0.005, reciprocally, Wilcoxon two-related samples test). The coefficient of correlation between MN frequency and UVA doses starting from zero to 60 sessions of treatment has been found as r = 0.61. This indicates a significant relationship between UVA doses and MN frequencies. However, MN inducibility and synergistic property of 8-MOP with UVA should be taken into account. Gradual MN increase during different sessions of PUVA treatment shows that--once appeared--a part of MN at least persist in the cells of patients from a few days to a few weeks. Smoking as a confounding factor seems to increase MN frequency (p = 0.053, Mann-Whitney U-test) in the beginning population, taken as the control population. This is the first report on the kinetics of MN formation during different sessions of PUVA treatment. Based on our results, we concluded that PUVA treatment causes a detectable chromosome damaging effect on the relatively profound cells/tissues of its human users. Therapists should be careful with its use, especially on the patients who may be more susceptible to carcinogenesis (e.g. immunosuppressed and/or elderly subjects).