An LC-MS-MS method was validated for the quantitation of a beta(3) agonist (A) in human urine to support Phase I studies. A was designed to accelerate metabolism for weight reduction. During assay development a significant loss of A was apparent from frozen urine quality control samples. The addition of 0.75% bovine serum albumin (BSA) in urine (v/v) was required to maximize the recovery of A from urine. Urine samples were basified and extracted into methyl t-butyl ether-isopropyl alcohol (90:10, v/v). The organic layer was washed, evaporated, reconstituted, and injected onto a 5 cm, C8 HPLC column prior to MS-MS analysis. The standard curve was linear from 5 to 500 ng/ml. Intraday precision for peak area ratios from BSA urine samples at seven separate concentrations over a range of 5-500 ng/ml (n=5) was <4.0% and calculated concentrations were within 91-115% of nominal concentrations. Interday precision for BSA urine quality control (QC) samples at four separate concentrations (n=10 of each) was <5.0% and individual calculated concentrations were within 90-111% of nominal concentrations. This work emphasizes that potential metabolites and quality control standards should be prepared and assayed as early as possible in method development, especially before the sample collection section of the clinical protocol is prepared. The methods described here have wide utility to other compounds containing basic benzene sulfonamides and to beta3 agonist candidates.