Summary The effects of different doses of angiotensin 11 (0.02 to 0.5 jlg kg-' min-' on mean arterial blood pressure, tissue blood flow and tissue vascular resistance were investigated in BD9 rats. Blood flow was measured using the uptake of 12511 or 4C-labelled iodoantipyrine ('25I-IAP and '4C-IAP). Spatial heterogeneity of blood flow within tumours, before and after angiotensin II infusion, was also measured using '4C-IAP and an autoradiographic procedure. Mean arterial blood pressure rose steeply with angiotensin II dose. Blood flow to skeletal muscle, skin overlying the tumour, contralateral skin, small intestine and kidney tended to decline in a dose-dependent manner. Blood flow to the tumour was also reduced (to 80% of control values) but there was no dose response. Blood flow to the heart was slightly increased and blood flow to the brain was unaffected by angiotensin II. Vascular resistance, in all tissues, was increased by angiotensin II infusion. The increase in tumour tissue was similar to that found in skeletal muscle and small intestine and is likely to be caused by a direct vasoconstricting effect of the drug rather than autoregulation of tumour blood flow in the face of an increase in perfusion pressure. The reduction in overall blood flow at the highest perfusion pressure was due to a preferential effect of angiotensin II at the tumour periphery. These results show that some tumours, at least, can respond directly to the effects of vasoactive agents.

Summary Tumour blood flow was characterised in a 'tissue-isolated' rat tumour model, in which the vascular supply is derived from a single artery and vein. Tumours were perfused in situ and blood flow was calculated from simultaneous measurement of (1) venous outflow from the tumour and (2) uptake into the tumour of radiolabelled iodo-antipyrine (IAP). Comparison of results from the two measurements enabled assessment of the amount of blood 'shunted' through the tumours with minimal exchange between blood and tissue. Kinetics of IAP uptake were also used to determine the apparent volume of distribution (VDapp) for the tracer and the equilibrium tissue-blood partition coefficient (A). A was also measured by in vitro techniques and checks were made for binding and metabolism of IAP using high-pressure liquid chromatography. VDapp and A were used to calculate the perfused fraction (a) of the tumours. Tumour blood flow, as measured by IAP (TBFIAP), was 94.8 ± 4.4% of the blood flow as measured by venous outflow, indicating only a small amount of nonexchanging flow. This level of shunting is lower than some previous estimates in which the percentage tumour entrapment of microspheres was used. The unperfused fraction ranged from 0 to 20% of the tumour volume in the majority of tumours. This could be due to tumour necrosis and/or acutely ischaemic tumour regions. For practical purposes, measurement of the total venous outflow of tumours is a reasonable measure of exchangeable tumour blood flow in this system and allows for on-line measurements. Tracer methods can be used to obtain additional information on the distribution of blood flow within tumours.

Summary The perfusion characteristics of the P22 carcinosarcoma were investigated in tissue-isolated tumour preparations in the ovarian and inguinal fat pads of BD9 rats. Tumours were perfused with a physiological buffer of known viscosity and changes in perfusion pressure were recorded at different perfusion rates in an ex vivo system. At perfusion pressures exceeding [30][31][32][33][34][35][36][37][38][39][40] The potential importance of differentially modifying tumour perfusion as a means of enhancing some forms of cancer therapy has been recently reviewed (Jirtle, 1988;Hirst & Wood, 1989;Jain, 1990). The delivery of oxygen and other radiosensitisers is enhanced when tumour perfusion is increased, as is the delivery of chemotherapeutic agents to the tumour, whilst reducing tumour blood flow has been shown to have value in the response of tumours to hyperthermia. The identification of those factors which may be modified to produce a preferential change in tumour perfusion could have important implications for therapy.Tumour perfusion rate, q, is dependent on the pressure gradient across the tumour vascular bed, AP, and on the resistance to flow, FR, imposed by the geometric resistance of the vasculature, z, and the viscosity of the perfusing fluid, ij:Ex vivo perfusion of tissue-isolated tumours supplied by a single artery and drained by a single vein permits determination of both FR and z, if i is known, of a tumour over a range of perfusion pressures (Sevick & Jain, 1989a,b). In the rat, there are two suitable sites in which tissue-isolated tumours can be grown, the ovarian and inguinal fat pads (Gullino & Grantham, 1961;Grantham et al., 1973).One of the many problems encountered in trying to predict the outcome of various forms of cancer therapy is the variability of the response of tumours of the same type located in different sites to physiological manipulations (Hirst et al., 1991). Using the two isolated tumour models, the opportunity exists to characterise the physiological parameters governing perfusion in each site and this may provide an indication as to the mechanism behind the site dependency of tumour response to treatment. Materials and methods Animals and tumourA transplanted rat carcinosarcoma, designated P22, was used for these experiments. This tumour arose in the treated site Ovarian fat pad Ovarian-isolated tumours were implanted using the technique established by Gullino & Grantham (1961), using parafilm as the enclosing material.Inguinal fat pad A 1 -2 cm incision was made in the skin overlying the inner right thigh of male rats. The fat pad supplied by the epigastric artery was identified and cut free so that no contralateral supply was possible. The epigastric vessels were carefully cleared of any fat and connective tissue between the fat pad and the femoral vessels. The inguinal fat pad was cut so that a piece of fat ; 3-5 mm3 was left attached to the vascular pedicle. Two I mm3 tumour fragments were placed in the fat pad which was subsequently enclosed in a specially designed silic...

Sumnman-A tissue-isolated preparation of the P22 rat carcinosarcoma was used to investigate the tumour vascular response to angiotensin II (ATII). In particular. the relative importance of systemic and local tumour factors was assessed by comparing tumour vascular resistance during systemic administration of ATII and during administration directly into the tumour-supplying artery. The effect of hypervolaemia on tumour vascular resistance was determined as well as the effect of ATII on oxygen metabolism. Tumour vascular resistance was increased by ATII in a dose-dependent manner. The response was biphasic with an initial peak in resistance followed by a lower plateau phase. Systemic administration of ATII was more effective in increasing tumour vascular resistance than direct administration. This suggests that systemic administration is not causing any reopening of previously collapsed tumour blood vessels. Further evidence for this is that hypervolaemia caused no reduction in tumour vascular resistance and that there was no difference in oxygen extraction by tumours between groups treated with systemically and directly administered ATII. A heterogeneous distribution of ATII receptors in the P22 tumour is a more likely explanation for the known heterogeneity of blood flow response to ATII.Kewords: P22 tumour; angiotensin II: vascular resistance; blood flow; oxygen metabolism Intravenous infusion of angiotensin II (ATII) has been found to increase tumour blood flow relative to most normal tissues, and this has led to the concept of 'hypertension chemotherapy'. in which ATII is used to improve the relative delivery of chemotherapeutic agents to tumours (Susuki et al., 1981;Takematsu et al.. 1985;Noguchi et al., 1988;Kobayashi et al.. 1990Kobayashi et al.. . 1991Anderson et al., 1991;Kerr et al., 1992; Mutoh et al., 1992). However, in absolute terms, ATII has been found to increase blood flow in some tumours (Tokuda et al.. 1990; Honr et al., 1991;Tanda et al., 1991) while decreasing it in others (Jirtle et al., 1978;Tozer and Shaffi, 1993). A decrease in absolute tumour blood flow could compromise delivery of drug to tumour microregions. Whether tumour blood flow increases or decreases is dependent upon the balance between drug-induced hypertension ansing from systemic vasoconstriction and local vasoconstriction induced within the tumour itself (since blood flow = perfusion pressure . vascular resistance). Presumably, this balance varies with tumour type. although the underlying factors which determine response remain unclear. It is important to identify these factors in order to predict the response of a particular tumour to ATII.We have previously found that ATII causes a significant increase in the vascular resistance of early generation, subcutaneous transplants of the P22 rat carcinosarcoma Tozer et al., 1994a). We also found that the degree of vasoconstriction induced by ATII was dependent on pretreatment blood flow. That is, in whole tumours, vasoconstriction was greatest where blood flow to control tumours...

Summary The ability of cancer cells to survive at a distance from blood vessels should be dependent on the local supply of nutrients to each vessel. The corded growth of tumour cells around blood vessels within regions of necrosis in the RH carcinoma in the mouse allows the limit to which cells can be supported by individual vessels to be observed. The thickness of individual tumour cords was measured in conventionally stained tumour sections using a scanning technique to determine the distance between the blood vessel wall and the most distant viable cell adjacent to necrosis. Cord radius was found to vary with the oxygen supply conditions. Control animals had a mean radius of 105±2 ;Lm while animals that had breathed 10% oxygen had significantly narrower cords (93 ± 3 tLm after 48 h) and animals breathing 100% oxygen had significantly wider cords (117 ± 31tm after 24 h). Mice (1970). In mice exposed to 10% oxygen for 48 h the tumours (mammary carcinomas) had cords which were narrower than those in controls. The time course of cord shrinkage was not studied in these experiments though it is reasonable to assume that it would not be an instantaneous process and must proceed at a rate determined amongst other factors by the metabolism of the cells and their tolerance of hypoxia.One of the practical implications for radiotherapy of these effects is that a reduction in oxygen supply conditions would be expected to produce only a transient change in the number of radiobiologically hypoxic cells. We have previously speculated (Hirst & Wood, 1987) that the adaptation of radiosensitivity that occurs over time when tumour-bearing animals are made anaemic could be accounted for by the death of cells at the periphery of corded structures, or at least those most distant from the supplying vessels, leading to a reduction in cord radius and the re-establishing of a lower hypoxic fraction similar to that before anaemia was induced. We have studied the effects of breathing oxygen, at both higher and lower than normal tensions, and of anaemia on the radius of cords in a slow growing mouse carcinoma. Our results show that PO2 in the inspired gas has a marked effect on cord radius though acute anaemia does not. Materials and methods Animals