Summnu.aThe strategy of combining carbogen breathing and nicotinamide to overcome chronic and acute hypoxia respectively is being evaluated clinically. The effects of both agents individually and in combination on relative perfusion of 400-700 mm3 RIF-I tumours and normal tissues were measured by 'Rb extraction.Carbogen breathing alone for 6 min increased relative tumour perfusion by 50-700,<0 compared with control at flow rates of 50 to 200 ml min-'. but the effect was lost at 300 ml min-'. All flow rates also produced similar increases in relative perfusion of lung, of between 36% and 58%. and smaller increases in skin, of between 200 and 3400. The minimum breathing time at 150mlmin-' to produce a significant increase in relative tumour perfusion was 4.5 min. and the effect was maintained up to 9 min. Nicotinamide alone at 1000mgkg-' 60min before assay did not alter relative tumour perfusion. Comparing the combination of nicotinamide with 6 min carbogen breathing at 150 ml min-with carbogen breathing alone showed no difference in relative tumour perfusion: increases were of 36% and 42% respectively. Nicotinamide-induced alterations in microcirculation associated with reduction of acute hypoxia have therefore not been detected by 'Rb extraction. The perfusion-enhancing effect of carbogen in this tumour is probably an important component of its radiosensitising ability. in addition to its known ability to increase the oxvgen-carrving capacity of the blood. and should be taken into consideration in clinical studies.

SummaryFlavone acetic acid (FAA) showed impressive effects against murine solid tumours but no activity in clinical studies. The mechanism of action in mice may involve damage to tumour vasculature or immunomodulation, and these effects may be species-specific. Alternatively, concentrations of FAA achieved in mouse tumours may be higher than in human tumours. It is important to resolve this issue since it raises important questions about the relevance of in vitro versus in vivo tumour screens and the development of FAA analogues. As part of a Cancer Research Campaign Phase II study of metastatic melanoma in which 8.4 g m 2 FAA was given as a 6 h infusion, six tumour biopsies were obtained from four patients. FAA tumour concentrations were determined by HPLC and compared with subcutaneous murine solid tumours within the same analytical laboratory. Tumour/plasma percentages (range 26-61%; mean ± SD, 43.9 ± 11.4%) were similar to those in mice, as was the area under the curve (AUC) extrapolated to infinity and the AUC above the putative activity threshold of 100 jig ml . We conclude that the exposure of drug-refractory human melanoma tissue to FAA was comparable to that of sensitive mouse tumours. This suggests that reduced penetration of FAA into human tumours is unlikely to explain the lack of antitumour activity observed in clinical studies and that differences in mechanism of action are predominant.Flavone acetic acid (FAA) is a synthetic flavonoid with impressive activity in preclinical testing against murine tumours including some quite refractory to conventional agents (Corbett et al., 1986;Plowman et al., 1986;Bibby et al., 1988). It attracted substantial interest because of the likely involvement of a unique mechanism of action (Cummings & Smyth, 1989;Workman, 1989;Bibby, 1991). The precise mode of anti-tumour cytotoxicity is uncertain but may involve indirect effects (Finlay et al., 1988) on tumour vasculature (Eveloch et al., 1988;Zwi et al., 1989;Bibby et al., 1989;Murray et al., 1989) or immunological mechanisms Urba et al., 1988;.Despite encouraging preclinical results, FAA proved completely ineffective in Phase I and II clinical trials (Kerr et al., 1987;1989;Kaye et al., 1990). The reason for this is unclear. Differences in pharmacokinetics have been observed between mouse, dog and man (Cummings & Smyth, 1989;Kerr et al., 1989;Zaharko et al., 1986, Damia et al., 1988Gouyette et al., 1988;Chabot et al., 1989). However, plasma concentrations similar to those in mice (>100 gLg ml-l) were observed in human plasma at the doses used in the Phase II studies. Therefore differences relating to tumour penetration may be responsible. Alternatively, the vascular or immunomodulatory mechanisms may depend on species-specific receptors. It is important to resolve this issue since it raises questions about the relevance of in vitro versus in vivo tumour screens and the development of FAA analogues.We report the human tumour FAA concentrations achieved in patients with metastatic melanoma treated in the Cancer Resea...