Psoriasis, its treatment, and cancer in a cohort of Finnish patients.

This study was designed to estimate the relative cancer risk of patients with moderate to severe psoriasis, with reference to different treatments. A cohort of 5687 hospitalized patients with psoriasis obtained from the Finnish Hospital Discharge Register in 1973-84 was linked with the records of the Finnish Cancer Registry. Standardized incidence ratios for cancer were calculated by dividing the observed number of cases by the expected cases, which were based on the national sex-specific and age-specific cancer incidence rates. By the end of 1995, 533 cancer cases were observed in the cohort. The overall cancer incidence was increased (standardized incidence ratio 1.3, 95% confidence interval 1.2-1.4). The estimated relative risks were highest for Hodgkin's disease (standardized incidence ratio 3.3, 95% confidence interval 1.4-6.4), squamous cell skin carcinoma (standardized incidence ratio 3.2, 95% confidence interval 2.3-4.4), non-Hodgkin's lymphoma (standardized incidence ratio 2.2, 95% confidence interval 1.4-3.4), and laryngeal cancer (standardized incidence ratio 2.9, 95% confidence interval 1.5-5.0). The role of prior oral antipsoriatic medications or phototherapy on the development of these cancers was assessed in a nested case-control study, for which 67 cases and 199 sex and age matched controls were selected from the psoriasis cohort. The relative risks were estimated using conditional logistic regression analysis. Oral 8-methoxy-psoralen plus ultraviolet-A radiation therapy and the use of retinoids were associated with an increased risk of squamous cell skin carcinoma (relative risk adjusted for the other treatment variables 6.5, 95% confidence interval 1.4-31, and 7.4, 95% confidence interval 1.4-40, respectively), whereas none of the treatments could be linked with the occurrence of non-Hodgkin's lymphoma.

Trioxsalen bath PUVA did not increase the risk of squamous cell skin carcinoma and cutaneous malignant melanoma in a joint analysis of 944 Swedish and Finnish patients with psoriasis.

It has been suggested that trioxsalen bath and ultraviolet (UV) A (PUVA) is associated with a very low or no risk of non-melanoma skin cancer, but the numbers of patients in individual studies have been limited. In order to attain statistically relevant information about the cancer risk associated with trioxsalen bath PUVA, two follow-up studies were combined and the joined cancer incidence was analysed among 944 Swedish and Finnish patients with psoriasis. The mean follow-up time for skin cancer was 14.7 years. Standardized incidence ratios (SIR) were calculated as a ratio of observed and expected numbers of cases. The expected numbers of cases were based on the national cancer incidence rates in the respective countries. There was no excess of squamous cell skin carcinoma [SIR 1.1, 95% confidence interval (CI) 0.2-3.2] or malignant melanoma (SIR 0.9, 95% CI 0.1-3.2) in the combined cohort. Basal cell skin carcinoma was not studied. The incidence of all non-cutaneous cancers was not increased (SIR 1.1, 95% CI 0.8-1.4). A threefold excess risk of squamous cell skin carcinoma after trioxsalen bath PUVA could therefore be excluded, which is a markedly lower risk than that associated with oral 8-methoxypsoralen PUVA. The result needs to be confirmed in a future follow-up, however, as the number of patients with high PUVA exposures was low.

Basal cell skin carcinoma and other nonmelanoma skin cancers in Finland from 1956 through 1995.

OBJECTIVE: To study trends of nonmelanoma skin cancer in Finland. DESIGN: Descriptive analysis of incidence and mortality rates for basal cell skin carcinoma (BCC) and other non-melanoma skin cancers (NMSCs) from 1966 and 1956, respectively, through 1995 in relation to sex, age, anatomical distribution, place of residence, and occupation. SETTING: Data were obtained from the nationwide Finnish Cancer Registry, to which reporting of skin cancer is compulsory. PATIENTS: Inhabitants of Finland (5.1 million in 1998). MAIN OUTCOME MEASURES: Age- and sex-specific incidence and mortality rates and overall rates adjusted for age to the world standard population; occupation-specific standardized incidence ratios, with the total Finnish population as reference. RESULTS: The age-adjusted incidence rate in 1991 through 1995 for BCC was 49 per 100,000 person-years in men and 45 in women. For NMSC it was 8.7 in men and 5.3 in women. Both cancer types showed an increasing trend in incidence rates. The proportion of tumors in the face, scalp, and neck was 59% for BCC and 67% for NMSC. The incidence rate of NMSC increased from north to south, while there was no great urban-rural or occupational variation in the occurrence of NMSC. The incidence rate for BCC was higher in urban than in rural regions. Farmers, forestry workers, and fishermen showed low incidence of BCC, whereas occupations with a high level of education or compulsory health checkups and medical care occupations appeared to have an increased incidence of BCC. The mortality rate for BCC in 1991 through 1995 was 0.08 per 100,000 person-years in men and 0.05 in women, and for NMSC, it was 0.38 in men and 0.23 in women. The mortality trend was decreasing for both cancer types. CONCLUSIONS: The incidence of NMSC is fairly low in Finland, accounting for 3.5% of all new cancer cases. Conversely, BCC is the most common cancer type. The incidence trend is increasing for both skin cancer types, but mortality remains low.

Expression of p53 protein before and after PUVA treatment in psoriasis.

We investigated the effect of the potentially carcinogenic psoralen plus UVA radiation (PUVA) therapy on the expression of p53 in skin of psoriatic patients. p53 antibodies DO7 and Pab240, antibodies against PCNA and Ki67 and the avidin-biotin immunoperoxidase complex method were used in the immunohistochemical staining of biopsy samples from non-lesional and lesional skin of 23 patients who received either trioxsalen bath PUVA or oral 8-methoxypsoralen PUVA. Biopsies were taken before and after a PUVA course. A modest expression of p53 was seen in psoriatic lesions in 17/21 patients before any treatment, probably as a physiological reaction to the hyperproliferation. Both p53 and the proliferation markers Ki67 and PCNA followed the same pattern, being more frequent in psoriatic lesions than in non-lesional skin. Exposure to PUVA induced an increase in p53 expression in non-lesional skin in 14/19 patients, putatively as a response to DNA damage caused by PUVA. In psoriatic lesions about half of the patients showed increased and half decreased expression of p53. The latter finding might be explained by decreased proliferation activity of the healing epidermis. In conclusion, p53 nuclear positivity in non-lesional skin after PUVA treatment is likely to be induced by DNA damage caused by PUVA, while in psoriatic lesions it could be a result of the combined effect of decreasing epidermal proliferation and DNA-damage.

Cancer incidence among Finnish psoriasis patients treated with 8-methoxypsoralen bath PUVA.

BACKGROUND: Long-term oral 8-methoxypsoralen (8-MOP) and UVA (PUVA) therapy increases the risk of nonmelanoma skin cancer and possibly also of cutaneous malignant melanoma. Topical application of 8-MOP PUVA induces malignant tumors in rodent skin, but little is known about its carcinogenicity in human skin. OBJECTIVE: Our purpose was to investigate the carcinogenicity of 8-MOP bath PUVA in humans. METHODS: This was a cohort study of 158 patients with psoriasis, for whom 8-MOP bath PUVA had been initiated during 1979 to 1992. The average number of 8-MOP bath PUVA treatments was 36 (range, 6 to 204) and the mean cumulative UVA dose was 92 J/cm2 (range, 3 to 884 J/cm2) by the end of 1995. The patients were not treated with any other forms of PUVA. Cancer incidence subsequent to 8-MOP bath PUVA up to the end of 1995 was determined by linking the cohort with the records of the Finnish Cancer Registry. The standardized incidence ratios (SIR) were calculated for skin cancer and some common internal cancers, using the expected numbers of cases based on the regional cancer incidence rates. RESULTS: There was one case of basal cell carcinoma, but no cases of other types of skin cancer. A total of 6 noncutaneous cancers were observed (SIR, 1.3; 95% confidence interval, 0.5 to 2.8). CONCLUSION: No association between cutaneous cancer and 8-MOP bath PUVA was found, but the statistical power of this study alone is not adequate to warrant definite conclusions. The results can be used in a meta-analysis as soon as other studies on the carcinogenicity of 8-MOP bath PUVA are published.