BACKGROUND: Recent studies have shown that 3'-deoxy-3'-[(18)F] fluorothymidine ([(18)F]FLT)) uptake depends on endogenous tumour thymidine concentration. The purpose of this study was to investigate tumour thymidine concentrations and whether they correlated with [(18)F]FLT uptake across a broad spectrum of murine cancer models. A modified liquid chromatography-mass spectrometry (LC-MS/MS) method was used to determine endogenous thymidine concentrations in plasma and tissues of tumour-bearing and non-tumour bearing mice and rats. Thymidine concentrations were determined in 22 tumour models, including xenografts, syngeneic and spontaneous tumours, from six research centres, and a subset was compared for [(18)F]FLT uptake, described by the maximum and mean tumour-to-liver uptake ratio (TTL) and SUV. RESULTS: The LC-MS/MS method used to measure thymidine in plasma and tissue was modified to improve sensitivity and reproducibility. Thymidine concentrations determined in the plasma of 7 murine strains and one rat strain were between 0.61 +/- 0.12 muM and 2.04 +/- 0.64 muM, while the concentrations in 22 tumour models ranged from 0.54 +/- 0.17 muM to 20.65 +/- 3.65 muM. TTL at 60 min after [(18)F]FLT injection, determined in 14 of the 22 tumour models, ranged from 1.07 +/- 0.16 to 5.22 +/- 0.83 for the maximum and 0.67 +/- 0.17 to 2.10 +/- 0.18 for the mean uptake. TTL did not correlate with tumour thymidine concentrations. CONCLUSIONS: Endogenous tumour thymidine concentrations alone are not predictive of [(18)F]FLT uptake in murine cancer models.