Background The use of narrowband ultraviolet (UV) B phototherapy to treat psoriasis and other disorders has increased markedly since the TL-01 lamps were introduced in the 1980s. While broadband UVB phototherapy has generally been considered to be a relatively safe treatment, some concern has been raised about the potential increased skin cancer risk with narrowband UVB.
Objectives The likelihood of a patient who is free of nonmelanoma skin cancer (NMSC) at the start of phototherapy developing a malignancy after a certain follow-up period will be dependent not only on the carcinogenic potential of the treatment but also on the age-conditional probability of natural occurrence. We were interested to explore the potential difficulty of designing studies to separate these two events.
Methods Mathematical models were developed that combined age-conditional probabilities of developing NMSC due to natural causes with the risk of inducing these cancers from narrowband UVB phototherapy in order to estimate the excess number of cancers resulting from this therapeutic intervention in a cohort of patients.
Results Within-department studies will be most unlikely to demonstrate that the number of NMSCs observed in follow-up studies is significantly different from that expected in an untreated population, even for a follow-up period of 20 years.
Conclusions Determination of the carcinogenic potential associated with narrowband UVB will require large multicentre studies typically involving several thousand new patients per year and followed up for 10 years or more.