The tricyclic carboxamide N-[2-(dimethylamino)ethyl]acridine-4-carboxamide (DACA) is a DNA-intercalating agent capable of inhibiting both topoisomerases I and II and is currently in Phase II clinical trial. Many related analogues have been developed, but despite their potent in vitro cytotoxicities, they exhibit poor extravascular distribution. As part of an ongoing drug development program to obtain related "minimal intercalators" with lower DNA association constants, we have compared the biodistribution and metabolite profiles of the prototype compound, DACA, with three analogues to aid rational drug selection. All of these compounds share a common structural feature, N-dimethyl side chain, which was radiolabeled with the positron-emitting radioisotope, carbon-11. This strategy was selected because it allows promising candidates emerging from preclinical studies in animals to be evaluated rapidly in humans using positron emission tomography (PET). The acridine DACA, the phenazine SN 23490, the pyridoquinoline SN 23719, and the dibenzodioxin SN 23935 were found to be cytotoxic in in vitro assays with an IC50 of 1.4-1.8 microM, 0.4-0.6 microM, 1.3-1.6 microM, and 24-36 microM, respectively, in HT29, U87MG, and A375M cell lines. Ex vivo biodistribution studies with carbon-11 radiolabeled compounds in mice bearing human tumor xenografts showed rapid clearance of 11C-radioactivity (parent drug and metabolites) from blood and the major organs. Rapid hepatobiliary clearance and renal excretion were also observed. There was low [<5% of injected dose/gram (%ID/g)] and variable uptake of 11C-radioactivity in three tumor types for all of the compounds. Tumor (U87MG) to blood 11C-radioactivity for [11C]DACA, [11C](9-methoxyphenazine-1-carboxamide (SN 23490), [11C]2-(4-pyridyl)quinoline-8-carboxamide (SN 23719), and [11C]dibenzo[1,4]dioxin-1-carboxamide (SN 23935) at 30 min were 2.9 +/- 1.1, 2.3 +/- 0.6, 2.6 +/- 0.6, and 0.7 +/- 0.2, respectively. For SN 23719, the distribution of 11C-radioactivity in normal tissues and tumors determined ex vivo was in broad agreement with that determined in vivo by whole body PET scanning. [11C]DACA was rapidly and extensively metabolized to several plasma metabolites and a major tumor metabolite. In contrast, [11C]SN 23935, [11C]SN 23490, and [11C]SN 23719 showed less extensive metabolism. In the tumor samples, the parent [11C]DACA and [11C]SN 23935 represented between 0.3 and 1.5%ID/g, whereas [11C]SN 23490 and [11C]SN 23719 represented between 1.5 and 2.8%ID/g. In conclusion, by using a strategy with 11C-labeling, we have determined the tissue distribution and metabolic stability of novel tricyclic carboxamides with the view of selecting analogues with potentially better in vivo activity against solid tumors. SN 23490 and SN 23719 had more favorable distribution and metabolic stability compared with DACA and SN 23935 and may warrant further development. The radiolabeling strategy used allows ex vivo and in vivo evaluation of promising anticancer agents in animals and offers the potential of rapid translation to studies in humans using PET.