Anomalous serum thyroxin measurements with the Abbott TDx procedure

Abstract

We describe a patient whose serum gave consistently low results for thyroxin as measured with the Abbott TDx but normal results by radioimmunoassay. Further clinical and laboratory studies did not support the low TDx results. Anti-thyroxin antibodies did not explain the discordant results, and we found no evidence of heterophilic antibodies in the patient's serum. We were unable to identify the cause of the problem.

Immunoassay Reagents for Thyroid Testing. 1. Synthesis of Thyroxine Conjugates

“…The Abbott TDx T4 FPIA, used in the Levine (7) report, was developed using an antibody to an immunogen represented by structures 2-4, the result of a nonspecific coupling of L-T4 through both the carboxyl and amino groups to the carrier protein, bovine serum albumin (BSA). The FPIA was optimized using d-T4 tracer 17. In a sense this tracer was heterologous to all the immunogen structures produced, since the D enantiomer was used.…”

“…1043-1802/94/2905-0459S04.50/0 ism (7). It was postulated that the FPIA tracer used in the assay might be binding to endogenous immunoglobulin G in the patient sample.…”

“…The reaction mixture was then vacuum filtered to remove insoluble 1,3-dicyclohexylurea, affording 16 mL of filtrate. Then 4 mL (0.08 mmol) of the filtrate containing the active ester (7) was added to a stirred solution of bovine serum albumin (250 mg, 0.0037 mmol) dissolved in 0.05 M sodium phosphate (10 mL, pH = 8.0) and DMF (10 mL). After being stirred for 3 days the reaction was dialyzed against 0.05 M sodium phosphate (4 L, pH = 8.0) for 24 h and then water (4 L) for 24 h. The dialyzed solution was then lyophilized to afford the desired L-thyroxine immunogen (8) (297 mg).…”

“…The filtrate was combined with 5-(aminomethyl)fluorescein hydrobromide (13) (884 mg, 2.00 mmol) and triethylamine (1.8 mL, 13 mmol), and the reaction was stirred for 16 h, under nitrogen, in the dark. The solvent was removed in vacuo, and the residue was purified by preparative reversed phase C18 HPLC, eluting with water/methanol/acetic acid (25/75/0.4, v/v), to afford 1.51 g (61%) of the desired íerí-butyl ester protected tracer as an orange solid: 1H NMR (300 MHz, DMSO-d6) ó 10.13 (s, 2H), 9.29 (s, 1H), 8.43 (m, 1H), 7.85 (s, 1H), 7.83 (s, 2H), 7.68 (d, 1H, J = 5 Hz), 7.12-7.28 (m, 2H), 7.07 (s, 2H), 6.68 (s, 2H), 6.54 (s, 4H), 4.36-4.61 The ferf-butyl ester protected tracer (15, 1.464 g, 1.19 mmol) was dissolved in methylene chloride/trifluoroacetic acid (30 mL 1/1, v/v) and stirred for 5 h, and the solvent was removed in vacuo. The crude product was purified by preparative reversed phase C18 HPLC, eluting with water/methanol/acetic acid (25/75/0.4, v/v) to yield 1.01 g (72%) of the desired L-thyroxine tracer ( 16) as an orange solid: NMR (300 MHz, DMSO-d6) ó 10.0-10.3 (broad s, 2H), 8.31 (t, 1H, J = 2 Hz), 7.86 (s, 2H), 7.83 (s, 1H), 7.64 (d, 1H, J = 7 Hz), 7.15-7.30 (m, 2H), 7.08 (s, 2H), 6.68 (s, 2H), 6.55 (s, 4H), 4.31-4.59 (m, 2H), 3.28-3.55 (m, 3H), 2.82-2.99 (m, 2H); MS (FAB) (M + H)+ caled for C38H27N2O10I4 1178.7840, found 1178.7834; HPLC [25: 75:0.4, water:methanol:acetic acid; 240 nm] retention time 8.1 min, 99%.…”