Semi-automated software for the three-dimensional delineation of complex vascular networks.

The understanding of tumour angiogenesis is of great importance in cancer research, as is the tumour response to vascular-targeted drugs. This paper presents software aimed at aiding these investigations and other situations where linear or dendritic structures are to be delineated from three-dimensional (3D) data sets. This software application was written to analyse the data from 3D data sets by allowing the manual and semi-automated tracking and delineation of the vascular tree, including the measurement of vessel diameter. A new algorithm, CHARM, based on a compact Hough transform and the formation of a radial map, has been used to locate vessel centres and measure diameters automatically. The robustness of this algorithm to image smoothing and noise has been investigated.

Increased cellular hypoxia and reduced proliferation of both normal and leukaemic cells during progression of acute myeloid leukaemia in rats.

The microenvironmental changes in the bone marrow, spleen and liver during progression of the transplantable promyelocytic leukaemia in the Brown Norwegian rat (BNML) have been studied. We used flow cytometry to estimate cellular hypoxia and proliferation based on in vivo pulse-labelling with a mixture of 2-nitroimidazole linked to theophylline (NITP) and bromodeoxyuridine (BrdUrd). The leukaemic cells were identified with the RM124 antibody. In rats inoculated with leukaemic cells the fraction of RM124+ cells was significantly increased from day 20 onwards in the spleen and from day 27 in the bone marrow and liver, reaching a level of 65-87% in these organs at day 32. At day 32, the NITP+ fraction of RM124+ cells had increased significantly in the bone marrow and spleen to 88% and 90%, respectively. The corresponding fractions of NITP+ normal cells reached 63% and 65%, respectively. From day 13 to day 32, the DNA-synthesizing (BrdUrd+) fraction of RM124+ cells in the bone marrow decreased significantly from 52% to 25%, and of normal cells from about 20% to 6%. In the bone marrow and spleen at day 27 and 32, the S-phase and G2/M-phase fractions according to DNA content were higher for the NITP+ than for the NITP- cells. This could partly be explained by an impaired cell cycle progression due to hypoxia. Nevertheless, we found indications of leukaemic cells that were simultaneously labelled with NITP and BrdUrd, in the bone marrow and spleen. These latter findings suggest that in contrast to normal cells some of the leukaemic cells can proliferate even during hypoxia, and this subpopulation may consequently renew and expand the leukaemic cell load.

Spatial relationship between hypoxia and the (perfused) vascular network in a human glioma xenograft: a quantitative multi-parameter analysis.

PURPOSE: To quantitatively study the spatial distribution of tumor hypoxia in relation to the perfused vasculature. METHODS AND MATERIALS: Using a human glioma xenograft model, nude mice were administered two different hypoxia markers (NITP or pimonidazole) and the perfusion marker Hoechst 33342. Frozen tumor sections were sequentially scanned for perfusion, hypoxia, and vasculature, respectively, to quantitate perfusion, vasculature, and hypoxia parameters in the same section. RESULTS: All tumors showed incomplete perfusion. Both NITP and pimonidazole stained the same hypoxic tumor areas. No statistically significant differences between the two markers were observed. The density of the perfused vessels was inversely related to the hypoxic fraction. At critical distances from perfused vessels, hypoxia occurred. These data suggest that predominantly diffusion-limited hypoxia was detected, based on the spatial distribution of nearby vessels. Also, the proportion of hypoxia distributed over arbitrary zones of 50 microm around perfused vessels was calculated. The largest proportion of hypoxia was found at distances beyond 100 microm from perfused vessels. CONCLUSION: With the multiple staining and functional microscopic imaging technique described here, the spatial relationship between perfused vessels and hypoxia was quantified in whole tumor cross-sections. The usefulness of this histologically-based method to quantitate morphological and physiological aspects of the tumor microenvironment was evaluated.

Vascular architecture and microenvironmental parameters in human squamous cell carcinoma xenografts: effects of carbogen and nicotinamide.

BACKGROUND AND PURPOSE: A better understanding of the vascular architecture and the microenvironmental parameters (VAMP) will allow the identification of tumours that can be more effectively treated by intensified fractionated radiotherapy or modifiers of blood flow and oxygenation or combinations of these approaches. MATERIALS AND METHODS: Proliferation (BrdUrd), vascular architecture (endothelial marker), perfusion (Hoechst 33342) and oxygenation (NITP) were studied in two human laryngeal squamous cell carcinoma tumour lines grown as xenografts in nude mice. The effects of carbogen and nicotinamide on these parameters were evaluated. RESULTS: Carbogen treatment resulted in a decrease of the number of perfused blood vessels from 66% to 55% in one of the two tumour lines. In this tumour line nicotinamide prevented this reduction of tumour blood flow by carbogen. In both tumour lines the labelling index (LI) decreased after treatment with carbogen for 1 h, from 11-13% to 5-7%. Both tumour lines showed a drastic reduction of hypoxia by carbogen alone or by carbogen plus nicotinamide. CONCLUSIONS: In both laryngeal squamous cell carcinoma xenograft tumour lines carbogen was very effective in reducing diffusion limited hypoxia. Only in one of the two tested tumour lines carbogen also caused a reduction of tumour blood perfusion, which could be compensated for by nicotinamide. In addition, carbogen reduced tumour cell proliferation. The fact that differences in response to nicotinamide and carbogen were observed and that they can be studied in vivo provides a basis for further development of a 'predictive profile' which will guide the clinician to select the optimal treatment for individual patients or groups of patients.

Use of 2-nitroimidazoles as bioreductive markers for tumour hypoxia.

Tumour hypoxia is thought to contribute to some failures of radiotherapy to achieve local control. Polarographic measurements of tumour oxygenation have been shown to predict clinical response to radiotherapy and patient survival. Hypoxia is also involved in many common types of normal tissue morbidity. However, at present there is no widely used method of measuring hypoxia in the clinic, or for individualizing therapy on the basis of tumour or tissue oxygenation. The bioreductive metabolism of 2-nitroimidazoles provides a way of labelling hypoxic cells in vivo and a variety of isotopic labels have been proposed for the non-invasive detection of bound metabolites of these markers. Several 2-nitroimidazoles with immunologically identifiable side-chains have been described and conventional immunostaining procedures can be used to locate their metabolites, bound to hypoxic cells in histological sections. Use of fluorescent immunoreagents allows flow cytometric assessment of hypoxia and multiple colour fluorescent staining allows hypoxia to be correlated with other markers on a cell by cell basis. 2-Nitroimidazole hypoxia markers show considerable promise for clinical use in diagnosing hypoxia and their use could allow rational application of hypoxia-related therapies to those patients most likely to benefit from them.

Induction of lethal mutations in experimental tumours after single and fractionated irradiations in vivo.

PURPOSE: To investigate the prolonged reduction in cellular viability (lethal mutations) of surviving cells following irradiation of tumours in vivo and to test the effects of fractionation on the expression of lethal mutations. MATERIALS AND METHODS: A mouse mammary carcinoma (CaNT) was treated with single dose or fractionated X-ray treatments in vivo and survival quantified with an in vitro excision assay soon after irradiation and at various times up to 35 days after in vitro propagation of the surviving cells. RESULTS: A dose-dependent reduction in the plating efficiency was observed in cells isolated from irradiated tumours up to 35 days and many cell generations after irradiation. Considerable heterogeneity in plating efficiency was observed in clonal cell lines isolated from individual colonies grown from irradiated tumours. Delayed expression of lethal damage was observed after fractionated irradiation, although recovery of cellular fitness was greater than after irradiation with single doses (reported previously) suggesting that this form of damage is affected by inter-fraction repair. At equi-toxic doses, delayed expression of lethal damage was similar after three compared with two fractions of radiation per day (reported previously). CONCLUSIONS: These effects indicate that conventional excision assays of tumour cell viability under-estimate the total lethal damage caused by irradiation and have implications for modelling of the response of tumours to radiotherapy. The effect of fractionation on expression of this type of damage implies the involvement of repair processes. Therefore the repair proficiency may affect the balance between the immediate and delayed reduction of viability in irradiated cells.

Cell cycle distribution of hypoxia and progression of hypoxic tumour cells in vivo.

Hypoxia was assessed in three murine tumour models in vivo by measuring the incorporation of 7-(4'-(2-nitroimidazole-1-yl)-butyl)-theophylline (NITP), an immunologically identifiable hypoxia marker that binds bioreductively to cells under low-oxygen conditions. Proliferating cells were labelled in the same tumours by administering the thymidine analogue bromodeoxyuridine (BrdUrd). The relative hypoxia in each cell cycle phase of cells isolated from tumours was assessed by addition of propidium iodide with analysis by flow cytometry. There was no relationship between tumour volume and hypoxia in either the anaplastic sarcoma SaF or the poorly differentiated carcinoma CaNT and only a slight negative correlation in moderately well-differentiated carcinoma Rh. The G1/G0 phase contained the greatest number of aneuploid hypoxic cells (aneuploid hypoxia ranging from less than 1% up to 40%, 38% and 71% in SaF, CaNT and Rh respectively), although there were significant amounts of hypoxia present in S- and G2/M phases for all three tumours examined. However, the highest proportion of hypoxia occurred in the G2/M phase, in which up to 60% of the cells were hypoxic. Simultaneous measurement of hypoxia, proliferation and DNA content using a novel triple-staining flow cytometry method showed that hypoxic cells could actively participate in the cell cycle. In addition, the cell cycle distribution of NITP and BrdUrd labelling showed that hypoxic cells could progress through the cell cycle, although their rate of progression was slower than that of better oxygenated cells.

Nicotinamide pharmacokinetics in humans: effect of gastric acid inhibition, comparison of rectal vs oral administration and the use of saliva for drug monitoring.

The effect of inhibiting gastric acid secretion on nicotinamide pharmacokinetics was studied in five volunteers with the intent of reducing the large variations observed previously in the time to and magnitude of peak plasma concentrations. Plasma levels were determined using a standard high-performance liquid chromatography (HPLC) method after an oral dose of 3 g of nicotinamide either alone or preceded by pretreatment with omeprazole. Suppression of gastric acid production had no significant effect on the rate of uptake or on the peak levels achieved. To bypass gastric acidity, the rectal route was also assessed using a suppository in four volunteers and one patient undergoing radiotherapy. Absorption was slow and variable and much lower plasma levels were observed than after oral dosing. Thus, no improvement in the pharmacokinetics of nicotinamide was observed using either of these two approaches. Parallel estimations were made using a novel and non-invasive method for monitoring nicotinamide pharmacokinetics in saliva. A large and variable fraction of the total amount of nicotinamide-related material in saliva was found to be nicotinic acid, a metabolite not normally found in human plasma. This conversion was inhibited by the use of a chlorhexidine mouthwash, indicating that the oral flora was responsible for its production. The time to peak levels of nicotinamide or of nicotinamide plus nicotinic acid in saliva correlated well with that in plasma. However, peak concentrations for nicotinamide alone were significantly lower than in plasma, and very variable, whereas for nicotinamide plus nicotinic acid saliva levels were 20-30% higher, but more consistent. Although there are some practical difficulties in quantitatively handling saliva, the method is very useful for monitoring nicotinamide pharmacokinetics and for assessment of compliance with nicotinamide treatment.

Comparison of two techniques for detecting tumour hypoxia: a fluorescent immunochemical method and an in vitro colony assay.

The aim of this study was to compare the percentage of hypoxic cells obtained with two methods: an in vitro colony assay and a new method based on immunodetection of a marker for hypoxic cells (NITP) which could be used in patients. These studies have been carried out using one rodent tumour EMT6 (a mammary carcinoma) and one human tumour HRT18 (a rectal adenocarcinoma). The hypoxic cell fraction was assessed in control mice and in mice receiving two treatments: 250 mg/kg nicotinamide + carbogen, and 250 mg/kg nicotinamide + carbogen + 4 ml/kg perflubron emulsion. The two treatments increased the radiosensitivity of the two cell lines, nicotinamide plus carbogen plus perflubron emulsion having the greatest radiosensitising effect. For untreated and treated tumours, the percentage of hypoxic cells obtained with the in vitro colony assay were comparable to those obtained with immunodetection using NITP. Whatever the treatment, NITP detection was a convenient test to detect the hypoxic cell fraction in the two solid tumours we have studied.

Vascular perfusion and hypoxic areas in RIF-1 tumours after photodynamic therapy.

The influence of photodynamic therapy (PDT) on vascular perfusion and the development of hypoxia was investigated in the murine RIF-1 tumour. Image analysis was used to quantify changes in perfusion and hypoxia at 5 min after interstitial Photofrin-mediated PDT. The fluorescent stain Hoechst 33342 was used as an in vivo marker of functional vascular perfusion and the antibody anti-collagen type IV as a marker of the tumour vasculature. The percentage of total tumour vasculature that was perfused decreased to less than 30% of control values after PDT. For the lower light doses this decrease was more pronounced in the centre of the tumour. The observed reduction in vascular perfusion showed a good linear correlation (r = 0.98) with previously published tumour perfusion data obtained with the 86Rb extraction technique. The image analysis technique provides extra information concerning the localisation of (non)-perfused vessels. To detect hypoxic tumour areas in vivo, an immunohistochemical method was used employing NITP [7-(4'-(2-nitroimidazol-1-yl)-butyl)-theophylline]. A large increase in hypoxic areas was found for PDT-treated tumours. More than half the total tumour area was hypoxic after PDT, compared with < 4% for control tumours. Our studies illustrate the potential of image analysis systems for monitoring the functional consequences of PDT-mediated vascular damage early after treatment. This provides direct confirmation that the perfusion changes lead to tissue hypoxia, which has implications for the combined treatment of PDT with bioreductive drugs.

Nicotinamide pharmacokinetics in normal volunteers and patients undergoing palliative radiotherapy.

The influence of nicotinamide formulation on absorption characteristics and incidence of adverse side-effects has been studied in normal volunteers and in patients undergoing radiotherapy. Escalating single or repeated oral doses of nicotinamide were administered in tablet or liquid form under fasting or non-fasting conditions. Drug absorption was slowed both by the presence of food in the stomach and by the administration of nicotinamide in tablet form compared with when it was dissolved in orange juice. Peak concentrations were generally slightly higher following the liquid preparation, but the incidence of adverse side-effects (chiefly nausea) was increased. A single dose of 9 g (88-97 mg/kg) nicotinamide in tablet form was well tolerated in two fasting normal volunteers, and in patients, doses of up to 133 mg/kg as tablets were tolerated twice/week for three weeks. Daily administration of 80 mg/kg nicotinamide was tolerated when given as tablets, but not in a liquid formulation. Neither the peak concentration nor the area under the concentration/time curve (AUC) of nicotinamide, nor the main metabolites of nicotinamide appeared to correlate with the incidence of toxicity.

Pharmacokinetics and binding of the bioreductive probe for hypoxia, NITP: effect of route of administration.

The novel compound 7(-)[4'-(2-nitroimidazol-l-yl)-butyl]-theophylline (NITP) can be used as an immunologically detectable probe for hypoxic cells. Because of the limited water solubility of NITP, it has been administered dissolved in peanut oil with 10% dimethylsulphoxide (DMSO). A new aqueous formulation has been devised, based on a 50% solution of a modified beta-cyclodextrin (Molecusol HPB), which increases the water solubility of NITP 10-fold. The pharmacokinetics of NITP in plasma and tumours have been compared following oral and intraperitoneal (i.p.) administration of the NITP in Molecusol, i.p. administration of NITP dissolved in peanut oil + 10% DMSO and injection of a near-saturated aqueous solution of the drug intravenously via the tail vein or i.p. or directly into the tumours. Binding of the marker to hypoxic cells within tumours was also measured after the different routes of administration. The Molecusol vehicle was unexpectedly toxic when administered i.p., but there was no toxicity from NITP dissolved in Molecusol when administered orally. Binding of the drug within tumours was seen for both the peanut oil + 10% DMSO and Molecusol formulations and for both oral and intraperitoneal routes. Binding of NITP within tumours has also been observed following direct injection of the drug, with minimal whole-body exposure to NITP. However, the bound metabolites of NITP within tumours were localised to the injection site, suggesting that direct injection is unlikely to be a useful method of administering bioreductive hypoxia markers. The data in this paper demonstrate that bound metabolites of the hypoxia marker NITP can be detected in tumours following oral administration of an aqueous formulation of NITP, and suggest that oral administration could be a satisfactory administration route for clinical studies with NITP.

Simultaneous triple staining for hypoxia, proliferation, and DNA content in murine tumours.

Hypoxia and proliferation rate are two important biological factors influencing the outcome of radiotherapy regimes for solid tumours. Hypoxic cells are more resistant to radiation than aerobic cells and a rapidly dividing tumour may repopulate faster during treatment. Clinical trials are underway to assess the importance of both these parameters. In this article we describe a method to simultaneously measure hypoxia and proliferation using multiparameter flow cytometry. Hypoxic cells were detected using a bioreductively bound marker with an immuno-recognisable side-chain, NITP and proliferation was measured by bromodeoxyuridine (BrdUrd) incorporation. These parameters were related to cell cycle position by measuring total DNA content with 7-aminoactinomycin D. The data were analysed using single laser excitation on a bench top flow cytometer. Simultaneous measurement of the three parameters shows the presence of cells which have incorporated BrdUrd and are also hypoxic by the criterion of NITP binding. The murine SaF tumour has a relatively constant aneuploid labelling index of 24%. However, the level of aneuploid hypoxia was variable ranging from 0.8 to 40.9% with a mean value of 15.6%. Within the BrdUrd labelled population there is a range of hypoxia from 2.8 to 28.5% (mean 15.1%); this represents 0.7 to 6.6% of the total tumour population. There are approximately twice as many oxygenated cells than hypoxic cells actively in the cell cycle. In vivo tumours contain cells with S phase DNA content which do not incorporate BrdUrd. This cell population has equivalent proportions of hypoxic and oxic cells. However, there are up to 12-fold more hypoxic cells in the unlabelled S than the BrdUrd labelled population. These data show that proliferation and hypoxia can be measured simultaneously using flow cytometry and the technique may form the basis of a predictive assay for these two important biological determinants of radiotherapy outcome.

Potential bioreductively activated hypoxia probes and post-irradiation radiosensitizers related to NITP.

NITP (1) is an effective marker of hypoxia in tumours for both microscopy and cell sorting studies and, additionally, the compound shows post-irradiation sensitization, probably by inhibition of repair of radiation damage to DNA. However, NITP does not have the substitution pattern which the immunochemical reagents are raised to recognize and the compound has very low solubility in water. We report the synthesis of an isomer (13) of NITP which has the desirable substitution pattern and is also soluble in very weak aqueous base. The successful synthesis of 13 uses a nitrosation and cyclization of a substituted uracil (16), but earlier approaches from 5 and 12 yielded the pyridoxanthine derivative 6. The preparative use of nitro group displacement reactions from 8-nitrocaffeine is shown to be a useful entry to a range of 8-substituted caffeines and is utilized to obtain two derivatives of NITP which carry aliphatic amine chains, i.e. 34 and 35.

Development of bioreductive markers for tumour hypoxia.

Hypoxic cells in tumours can be identified by exposing them to an immunologically identifiable 2-nitroimidazole (NITP) with a theophylline substituent which becomes bioreductively metabolised and binds to cellular macromolecules in the absence of oxygen. A range of monoclonal and polyclonal antibodies raised against theophylline or caffeine can identify cells containing bound adducts of NITP, in some cases with higher specificity than the standard product used. An alternative approach utilizes the very high specificity of FITC-avidin as a reagent to detect metabolic binding of a 2-nitroimidazole with a biotinylated side-chain (NIB), with the advantage of a single-step staining protocol. Both proliferating and hypoxic cell populations within tumours can be identified by simultaneous staining for incorporation of NITP and BrdUrd and this has shown that some cells incorporate both markers, suggesting that there is some overlap between the proliferating and hypoxic cell compartments.

Contribution of lethal mutations to excision assays for tumour cell survival.

Conventional assays of cell survival determine only the proportion of colony-forming cells, assuming that all such cells are equivalent. However, cells surviving irradiation are reported to have lower plating efficiencies than unirradiated controls, suggesting an additional component of cellular damage that is ignored in conventional survival assays, but which could contribute to therapeutic outcome. Therefore we have examined the contribution of this additional form of damage to excision assays for cell survival in experimental tumours following both single dose and fractionated irradiation (10F/5 days) in vivo. Plating efficiencies were considerably lower for the long-term descendents of irradiated compared with non-irradiated cells. Expression of delayed reproductive death was reduced after fractionated radiation doses, only appearing after a substantial number of 3.4 Gy fractions had accumulated. Thus estimates of survival derived from single clonogenicity assays may underestimate the reduction in cell viability from a particular treatment. This could compromise assays for intrinsic radiosensitivity and mathematical modelling of the efficacy of treatment regimens.

Bioreductive markers for hypoxic cells: 2-nitroimidazoles with biotinylated 1-substituents.

The interference by oxygen with the bioreductive metabolism and binding within cells of 2-nitroimidazoles has been used to identify hypoxic cells. Three novel compounds were synthesized with a 1-substituent containing a biotin moiety. Bound adducts of these compounds could be identified in hypoxic cells in vitro by the biotin binding proteins, avidin or streptavidin, labeled with fluorescein. The metabolism and discrimination of these compounds between well-oxygenated and hypoxic cells was evaluated by flow cytometry. Ester or amide links between the 2-nitroimidazole and the biotin were degraded in the presence of mouse serum, but a compound with a C5 hydrocarbon link was stable, and this compound was suitable for evaluation in an in vivo tumor model.

Growth of fibroblasts as a potential confounding factor in soft agar clonogenic assays for tumour cell radiosensitivity.

Soft agar clonogenic assays are considered to be a standard method for measuring tumour cell radiosensitivity and it has been widely reported that fibroblast contamination does not occur. We report here that human fibroblasts can proliferate to form colonies in a modified form of the Courtenay-Mills soft agar clonogenic assay. It was observed that early passage skin fibroblasts could form colonies in soft agar, although the plating efficiencies were reduced compared with growth on plastic. It was demonstrated that normal lung could proliferate in agar with similar plating efficiencies to fresh tumours and that fibroblastic cells were present in these cultures. Characterisation of primary lung tumour cultures also showed that fibroblastic cells were present in these cultures. Characterisation of primary lung tumour cultures also showed that fibroblastic cells were present which lacked epithelial features and which resembled closely the cells found in cultures of normal lung. This is an important finding for workers using soft agar assays to culture human tumour cells and is of interest in understanding the processes of normal growth control of human fibroblasts.

Pharmacokinetics of varying doses of nicotinamide and tumour radiosensitisation with carbogen and nicotinamide: clinical considerations.

Plasma concentrations, after administration of varying doses of nicotinamide, were measured in CBA male mice using a newly-developed high performance liquid chromatography assay. In all dose groups, peak levels were observed within the first 15 min after an i.p. administration of 0.1, 0.2, 0.3 or 0.5 mg g-1 of nicotinamide. There was a clear dose-dependent increase in plasma concentration with increasing dose, with almost a five-fold lower concentration (1.0 vs 4.9 mumol ml-1) achieved with a dose of 0.1 mg g-1 compared with 0.5 mg g-1, respectively. The half-life of nicotinamide increased from 1.4 h to 2.2 h over the dose range (P < 0.01). Comparisons with previous pharmacokinetic data in humans show that clinically-relevant oral doses of 6 and 9 g in humans give plasma levels slightly higher than those achieved at 1 h with doses of 0.1 to 0.2 mg g-1 in mice. Tumour radiosensitisation with carbogen alone, and with carbogen combined with varying doses of nicotinamide (0.05 to 0.5 mg g-1), was investigated using a 10-fraction in 5 days X-ray schedule. Relative to air-breathing mice, a statistically significant increase in sensitisation was observed with both a local tumour control and with an in vivo/in vitro excision assay (P < or = 0.007). With the local control assay, a trend was observed towards lower enhancement ratios (ERs) with decreasing nicotinamide dose (from 1.85 to 1.55); carbogen alone was almost as effective as when combined with 0.1 mg g-1 of nicotinamide. With the excision assay, ERs for carbogen combined with nicotinamide increased with decreased levels of cell survival. At a surviving fraction of 0.02, enhancement ratios of 1.39-1.48 were obtained for carbogen plus 0.1 to 0.3 mg g-1 of nicotinamide. These were lower than those seen with the two higher doses of 0.4 to 0.5 mg g-1 (ERs = 1.63-1.69).