Enzyme immunoassays for the detection of mycotoxins in plant-based milk alternatives: pitfalls and limitations

Plant-based milk alternatives (PBMAs) are a potential source of mycotoxin uptake. To ensure food safety, simple and rapid testing methods of PBMAs for mycotoxins are therefore required. This study investigated the applicability of enzyme immunoassay (EIA) methods for direct testing of PBMAs without sample extraction. Mycotoxin analyses included aflatoxin B-1 (AFB(1)), sterigmatocystin (STC), ochratoxin A (OTA), deoxynivalenol (DON), and T-2/HT-2-toxin (T-2/HT-2). It was found that the PBMA matrix negatively affected the EIA to varying degrees, thus affecting the reliability of the results. A dilution of PBMAs of at least 1:8 was necessary to overcome matrix interference. This resulted in calculated detection limits of 0.4 mu g/L (AFB(1)), 2 mu g/L (STC), 0.08 mu g/L (OTA), 16 mu g/L (DON), and 0.4 mu g/L (T-2/HT-2). After analysis of 54 PBMA products from German retail stores, positive results in at least one test system were obtained for 23 samples. However, most positive results were near the calculated detection limit. Control analyses of selected samples by LC-MS/MS for AFB(1), STC, and OTA qualitatively confirmed the presence of trace amounts of STC in some samples, but quantitative agreement was poor. It was concluded that the high diversity of ingredients used in PBMAs led to a highly variable degree of sample matrix interference even in a 1:8 dilution. Since the use of higher dilutions conflicts with the need to achieve low detection limits, the application of EIA for routine mycotoxin analysis in PBMA for mycotoxins requires further study on the development of a feasible sample preparation method.