A necessary adjustment of protocol for use of DPC coat-a-count total testosterone assay with saliva.

S testosterone (T) assessment has been used in varied medical (1) and psychological testing (2) contexts. As with other analytes, this is most efficiently done using a direct assay format with an immobilized, highly specific antibody. Many studies have chosen DPC’s coated tube assay for total T (Coat-a-Count Total Testosterone, Diagnostic Products Corp., Los Angeles, CA, USA) and their 1986 protocol for saliva to meet these needs. That protocol makes two changes in the serum diagnostic protocol. First, instead of a 50 mL sample of undiluted serum standards or controls, 200 mL is used for raw saliva, or serum standards and controls that are diluted 20-fold in water. Second, the antibody binding period is extended from 3 h at 37°C to 16–24 h at room temperature. Both serum and saliva protocols add 1.0 mL of I-T tracer solution so final assay volumes are 1.05 mL and 1.2 mL, respectively. Stated sensitivity (limit of detection) of the DPC kit is ;4 ng/dL (0.14 nM, 40 pg/mL) or 2 pg/tube for serum with an analytical range of 4–1,600 ng/dL (2–800 pg/assay tube). No alteration of assay sensitivity is noted in the DPC salivary protocol but the analytical range is shifted to 1–80 ng/dL (2–160 pg/assay tube). During validation checks of this assay using DPC’s saliva protocol, the kit reagents were unable to reliably measure human salivary total T. Therefore, we now question the validity of results in studies using this kit with DPC’s saliva protocol without initial validation and modification. Besides this caution, we offer a simple solution to rectify the problem and to generate reliable results from saliva using this assay kit. We used DPC’s coated tube assay for total T and their protocol for saliva in the context of a psychological evaluation of power motivation in college-age young men (Human Use Clearance on 4/2/97 by IRB M1084-01, Harvard University). Published values of salivary T have been reviewed by Read (3). Generally, adult males average 150–500 pM (43–144 pg/ mL) including results from studies using a “gold standard” reference method, gas chromatography/ mass spectrometry (4). The highest T is anticipated among young adult males such as those under study (3,5). Adult females averaged , 50 pM (, 14 pg/mL) in the same studies noted by Read (3) and Gould et al. (4). Given these values, 200 mL of adult male saliva should contain ;8–29 pg of T while female saliva should contain , 3 pg. It is unlikely the DPC kit and protocol will generate reliable results for female samples as they fall at, or below, the assay detection limit. Samples from males would lie at $ 80% B0 in an analytical region of low estimation precision. A preliminary quality control test showed DPC’s current salivary T protocol inadequate. A parallelism check consisting of duplicates of a 6-step serial dilution [1:1.25–1:40, in 0.05 M phosphate buffered 0.15 M saline (PBS), pH 7] of a saliva pool collected in-house from young adult male volunteers was run along with duplicates of an identical serial dilution of the same pool spiked at each dilution with a constant 160 pg pure T/mL. Expected results for the simple dilution series are ;1–140 pg/mL (cf. 3,4), however, a nearly constant 54 6 11 pg/mL (mean 6 SD) was observed regardless of dilution factor; 16 other undiluted saliva samples from other young males measured 44 6 23 pg/mL. The spiked saliva dilutions differed consistently from the simple saliva dilutions (190 6 44 pg/mL observed; 160 pg/mL expected). As this is a direct assay with no sample Correspondence: Kenneth L. Campbell, Department of Biology, University of Massachusetts at Boston, 100 Morrissey Blvd., Boston, MA 02125-3393. E-mail: kenneth. [email protected]. Manuscript received June 24, 1998; revised and accepted September 8, 1998. Diagnostic Products Corporation, 5700 West 96th Street, Los Angeles, CA 90045. Deceased, March 27, 1998. Clinical Biochemistry, Vol. 32, No. 1, 83–85, 1999 Copyright © 1999 The Canadian Society of Clinical Chemists Printed in the USA. All rights reserved 0009-9120/99/$–see front matter