Assessment of Risk for Interference by Circulating Biotin in Samples Received for High Sensitivity Troponin-T, Thyrotropin, and for Prostate Specific Antigen Testing by Immunoassays.

BACKGROUND

Biotin interference in streptavidin-based immunoassays is known and may lead to erroneous results and thus to diagnostic error. The recent increase in reports of biotin interference in immunoassay-based testing has been attributed to increased intake of biotin supplements by the public and to the high dose biotin therapy in patients with neurological and inherited disorders. Circulating biotin levels greater than 20 ng/mL are reported to exhibit interference in high sensitivity troponin T (hs-TnT), thyroid stimulating hormone (TSH), and in prostate specific antigen (PSA) among other assays when using our Cobas® 6000 immunoassay analyzer (Roche Diagnostics, IN, USA). This study aims to examine the risk for biotin interference among our patient population.

METHODS

Serum and plasma leftover samples from 183 different patients were collected following completion of hs-TnT (53 samples), TSH (45 samples), and PSA (85 samples) testing. Aliquots were stored frozen at -20°C until analysis. Biotin concentrations in these samples were measured using an ELISA (ALPCO, Salem, NH, USA) according to the manufacture's protocol. Samples with biotin levels of 20 ng/mL or greater were considered as high-risk samples (HRS) for biotin interference.

RESULTS

The overall concentrations of biotin in our patients' samples ranged from 0.02 ng/mL to 11.38 ng/mL (median 0.42 ng/mL). The median and (range) biotin concentrations in hs-TnT, TSH, and PSA samples were 0.27 ng/mL (0.02 - 6.86 ng/mL), 0.39 ng/mL (0.08 - 11.38 ng/mL), and 0.47 ng/mL (0.09 - 7.73 ng/mL), respectively. Although there was no significant difference between biotin levels in samples for TSH or PSA measurement (p = 0.85), biotin in samples for PSA and for hs-TnT and in samples for TSH and hs-TnT were significantly different (p = 0.049 and 0.089), respectively. None of the samples had biotin levels greater than or equal to 20 ng/mL.

CONCLUSIONS

Using representative samples with requests for hs-TnT, TSH, and PSA testing, where reliable performance for the selected assays at their lowest measurement range is required for clinical intervention, among our study population the risk was considered minimal as their circulating biotin levels were less than 20 ng/mL. However, educating clinicians and laboratory users regarding the potential of biotin interference is always recommended.