Impact of matrix complexity on HPLC MS/MS and immunoassay based rapid test of mycotoxins in corn and poultry feed

Authors: Pokoo-Aikins, Anthony; Mitchell, Trevor; Read, Quentin; Glenn, Anthony

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 5/1/2026
Publication Date: N/A
Citation: N/A

Interpretive Summary: N/A

Technical Abstract: Mycotoxin screening is critical in the poultry feed industry because these toxins can be detrimental to animal production and health when they are ingested. As a result, there are mechanisms put in place to periodically test for these toxins in feed and feed ingredients to make sure that the levels are acceptable for animal consumption. The most common testing methods include High Performance Liquid Chromatography–Mass Spectrometry (HPLC-MS/MS) and Immunoassay-based Rapid Test. HPLC-MS/MS is generally the gold standard which offers a higher sensitivity and detects multiple toxins, but it requires more testing time, trained staff, and greater cost compared to Immunoassay-based Rapid Test which is a fast, lower cost, single toxin per test that is good for testing simple matrixes. This study compared the performance of HPLC-MS/MS with a rapid, antibody-based detection system that uses lateral flow test kits to quantify mycotoxins in corn (n=45), broiler feed (n=46), and layer feed (50) samples. The corn samples were analyzed for fumonisins (FUM), deoxynivalenol (DON), aflatoxins (AFLA), and zearalenone (ZEA), whereas the feed samples were analyzed for AFLA, DON and FUM. The samples were collected from various locations in the southeastern United States. Descriptive statistics used to analyze the data showed that the rapid methodology tended to overestimate FUM concentrations and underestimated ZEA concentration compared to the LC-MS/MS standard. The absolute errors for the rapid method tended to be less than 0.05 ppm for AFLA, less for DON, and less than 3.0 ppm for FUM and less than 1.0 ppm for ZEA. The percentage errors for the rapid method were often considerably higher than HPLC-MS/MS. HPLC-MS/MS was able to detect AFLA in 58% of the corn samples versus 98% with the rapid method, FUM was 98% for both methods. The rapid method detected both AFLA and DON in 98% and 6.5% AFLA and 83% DON in broiler feed samples. Both testing methods were able to detect 100% FUM in both broiler and layer feed samples. These results show that the rapid method consistently detected AFLA in samples that HPLC-MS/MS did not detect. This was true across all sample types, but the effect was more pronounced in feed samples. While, rapid mycotoxin testing methods provide numerous economic and ease-of-use benefits for observing mycotoxin levels in the real world, LC-MS/MS remains the gold standard due to its analytical accuracy.