The mycotoxin ochratoxin A (OTA) is a rodent carcinogen produced by species of the ubiquitous fungal genera Aspergillus and Penicillium. OTA is found in a variety of food items and as a consequence is also found in human plasma (average concentrations found in this study: 0.1±1 ng OTA/ml plasma). To improve the scienti®c basis for cancer risk assessment the toxicokinetic pro®le of OTA was studied in one human volunteer following ingestion of 395 ng 3 H-labeled OTA (3.8 lCi). A two-compartment open model consisting of a central compartment was found to best describe the in vivo data. This two-compartment model consisted of a fast elimination and distribution phase (T 1/2 about 20 h) followed by a slow elimination phase (renal clearance about 0.11 ml/min.) and a calculated plasma half-life of 35.55 days. This half-life was approximately eight times longer than that determined previously in rats. In addition, the intraindividual¯uctuation of OTA plasma levels was investigated in eight individuals over a period of 2 months. The concentrations determined ranged between 0.2 and 0.9 ng OTA/ml plasma. The plasma levels in some individuals remained nearly constant over time, while others varied considerably (e.g. increase of 0.4 ng/ml within 3 days, decrease of 0.3 ng/ml within 5 days) during the observation period. This intraindividual¯uctuation in OTA plasma levels, which may represent di erences in OTA exposure and/or metabolism, as well as the large di erence in plasma half-life in humans compared to rats must be taken into consideration when the results of rat cancer study data are extrapolated to humans for risk assessment purposes.

Abstract--Ochratoxin A (OA) is a nephrotoxic and nephrocarcinogenic mycotoxin which is predominantly produced by the two ubiquitous fungal genera, Aspergillus and Penicillium. OA is found in foodstuffs, predominantly in cereals but also in coffee beans. Inconsistent results have been published regarding the influence of roasting on the OA content in roasted beans and the transfer into the coffee brew. In the present study an HPLC method was used for the detection of OA in green and roasted coffee beans as well as in the coffee brew. For qualitative confirmation and quantification of low OA levels in roasted coffee beans and coffee brew an additional clean-up step by immunoaffinity column was applied before HPLC analysis. In green coffee beans OA was detected in 13 out of 25 commercial samples analysed (detection limit, 0.5/~g OA/kg). Roasting (250°C, 150 sec) of naturally contaminated green beans or beans inoculated with A. ochraceus resulted only in a small reduction in the OA level. OA was also found to be eluted into the brew. Of 40 coffee brews prepared from commercially available samples OA was detected in 18 brews by HPLC and/or additional immunoaffinity column clean-up in the range of 0.4 to 7.8 pg OA/kg equivalent ground coffee. Our preliminary results suggest, therefore, that regular coffee consumption may contribute to exposure of humans to OA.

No abstract