Incidence of cutaneous melanoma has been increasing. We recently found that intake of citrus fruits, especially grapefruit, was associated with increased risk of developing melanoma and believe that psoralens in citrus fruits may be responsible for the association. An oral 8-methoxy psoralen (8-MOP; methoxsalen) in combination with ultraviolet-A radiation (Psoralen and UVA Light [PUVA] therapy) is used to treat skin disorders and found to be associated with increased incidence of melanoma. Psoralens are a group of naturally occurring linear furocoumarins (furanocoumarins) in fruits and vegetables and are potentially phototoxic and photocarcinogenic; they intercalate DNA and photochemically induce mutations. The levels of furocoumarins present in our diets, while normally well below levels that would cause evident acute phototoxicity, do cause pharmacologically relevant drug interactions, known as the `grapefruit effect'. Furthermore, there are no efficient routes for the metabolic detoxification of psoralens in humans. Consequently, low levels of psoralens, which are not completely excreted, may occur in our skin awaiting suitable photon interactions to create mutational lesions potentially resulting in melanoma or other skin cancers. Citrus fruit consumption has increased substantially in the US. However, there is a lack of epidemiological data on dietary psoralen intake in relation to melanoma risk because food composition database for dietary psoralens are not available. Data are lacking on blood and urine psoralen levels in humans. Therefore, in this proposal, we plan to capitalize on existing dietary data collected in large cohort studies (where the positive association between citrus fruit intake and melanoma risk was found) and propose to construct a food composition database on dietary psoralens through direct assessment of several common psoralens in psoralen-rich foods including citrus fruit items and to evaluate dietary psoralen intake in relation to melanoma risk with at least 1,000 melanoma cases. We also propose to conduct a pilot study to investigate whether we can detect these psoralens in archived blood and urine samples of large cohort studies and to assess laboratory variability, stability due to delays in processing, and within-person variability of the blood and urinary psoralens. We are uniquely positioned to most efficiently address the hypotheses. The results of the investigation will pave the way for future studies that will capitalize on these findings and cn lead to studies of plasma or urine psoralen levels to unveil the roles of naturally occurring psoralens in carcinogenesis. Because dietary guidelines for cancer prevention encourage consumption of fruits and vegetables, clarification of the effect of naturally occurring psoralens on carcinogenesis has public health implications as we have experienced through the findings of `grapefruit effect'.