
	Version PDF

Radiosensibilisation de cellules tumorales hypoxiques par production endogène de monoxyde d’azote

Bulletin du Cancer. Volume 95, Numéro 3, 282-91, mars 2008, Synthèse

DOI : 10.1684/bdc.2008.0592

Résumé

Summary

Auteur(s) : Mark De Ridder, Gretel Van Esch, Benedikt Engels, Valeri Verovski, Guy Storme , Oncologisch centrum UZ Brussel, Department of Radiation Oncology, Laarbeeklaan 101, B-1090 Brussels, Belgium.

Résumé : L’hypoxie est une propriété typique du micromilieu tumoral et une cause principale de radiorésistance. Pendant ces dernières dizaines d’années, plusieurs stratégies ont été développées pour améliorer l’oxygénation de tumeurs solides au cours du traitement radiologique, par exemple la respiration d’oxygène sous conditions hyperbares (3 atmosphères) et l’amélioration de la perfusion tumorale par nicotinamide, qui est utilisée en combinaison avec l’administration de carbogène et un traitement radiologique hyperfractionné pour inhiber la repopulation des cellules tumorales (ARCON). D’autres stratégies sont l’utilisation de radiosensibilisateurs qui miment l’effet d’oxygène au niveau de l’ADN, comme les nitro-imidazoles, et l’utilisation de drogues bioréductives comme la mitomycine C et la tirapazamine. Une méta-analyse des études randomisées qui ont évalué une forme de radiosensibilisation hypoxique montre un net avantage au niveau local et de la survie, surtout pour les tumeurs ORL et du col utérin. Pourtant, aucune des ces stratégies est utilisée en routine clinique, pour raisons pratiques et suite à leur toxicité. Nous avons développé une stratégie alternative, qui exploite les propriétés du micromilieu tumoral pour cibler la radiosensibilisation. La plupart des tumeurs solides expriment la synthase inductible de NO (iNOS) qui produit une quantité importante de monoxide d’azote (NO). Cette enzyme est induite, entre autres, par les lipopolysaccharides bactérielles (LPS) et par le lipide A, une de ses dérivées qui est utilisée dans les vaccins. Nous avons démontré que la production intracellulaire de NO par iNOS radiosensibilise des cellules tumorales hypoxiques à des concentrations extracellulaires non toxiques. De plus, la transcription d’iNOS est soutenue par l’hypoxie chronique et par des cytokines pro-inflammatoires comme l’interféron γ. Pour cela, nous avons proposé l’infiltrat tumoral pro-inflammatoire comme nouvelle cible pour des stratégies de radiosensibilisation et avons identifié deux modes d’action : 1) les cellules de l’infiltrat tumoral pro-inflammatoire (macrophages, cellules T/NK) peuvent être une source de cytokines qui induisent iNOS dans les cellules tumorales avoisinants \; 2) des macrophages tumorales proprement stimulées peuvent devenir une source intratumorale importante de production de NO. Nos recherches actuelles sont concentrées sur l’intégration de stratégies immunitaires dans des protocoles de radiosensibilisation.

Mots-clés : hypoxie, radiosensibilisation, synthase inductible de monoxide d’azote, infiltrat tumoral pro-inflammatoire, lipide A

Illustrations

ARTICLE

Auteur(s) : Mark De Ridder, Gretel Van Esch, Benedikt Engels, Valeri Verovski, Guy Storme

Oncologisch centrum UZ Brussel, Department of Radiation Oncology, Laarbeeklaan 101, B-1090 Brussels, Belgium

Article reçu le 20 Juin 2007, accepté le 24 Août 2007

Oxygen and tumor cell radioresponse

The response of cells to ionizing radiation is strongly dependent upon oxygen, as illustrated for EMT-6 mouse mammary carcinoma cells (figure 1). In this figure, the cell surviving fraction is shown as a function of radiation dose, administered under aerated conditions and under anoxic conditions, which were achieved in a nitrogen based atmosphere. This allows us to determine the oxygen enhancement ratio (OER), which is classically calculated at a surviving fraction of 0.1 (OER = radiation dose in anoxia/radiation dose in air). For most mammalian cells, the OER for γ-irradiation is around 3. The mechanism responsible for the enhancement of radiation damage by oxygen is called “the oxygen fixation hypothesis”. Briefly, when radiation is absorbed in a biological material, free radicals are produced (figure 2). These radicals can be produced directly in the DNA, or indirectly in water molecules and diffuse far enough to damage the DNA. It is the fate of the free radicals produced in the DNA (DNA•) that determines the biological damage. In the presence of oxygen, DNA-OO• is produced and further processed to DNA-OOH. This results in a changed chemical composition of the DNA and thus fixation of the DNA damage. In the absence of oxygen, DNA• can react with H⁺, restoring the DNA in its original form. To exert its effect on radiosensitivity, oxygen must be present during or within milliseconds after radiation [1, 2].

Tumor oxygenation and radiotherapy outcome

The most widely accepted technique for assessing tumor oxygenation is the measurement of tissue oxygen tension (pO₂) by polarographic needle electrodes (Eppendorf electrodes), which are introduced in the tumor and moved forward by an automatic stepping motor [3, 4]. This approach allows to measure the pO₂ along several electrode tracks. Generally, minimally 50 pO₂ readings are performed, from which the “median pO₂” is calculated. Tumor hypoxia may also be expressed as the “hypoxic proportion” (percentage of pO₂ readings < 2.5 or 5 mmHg) or as the “hypoxic subvolume” (percentage of pO₂ values below 2.5 or 5 mmHg multiplied by the total tumor volume). These parameters have been used to study the relation between tumor oxygenation on radiotherapy outcome in cervical carcinomas and head-and-neck tumors essentially (table 1). In summary, these studies show that cervix carcinomas and head-and-neck cancers are poorly oxygenated, with a median pO₂ of about 10 mmHg, and that low pO₂ before radiotherapy is the most significant adverse prognostic factor. Low and highly heterogeneous levels of oxygenation were found in other types of solid tumors as well, and are considered to be the most important signature of tumor microenvironment [5]. The etiology of tumor hypoxia will be discussed in the next paragraphs.
Table 1 The prognostic significance of pre-treatment tumor oxygenation. All reported patients were treated by radiotherapy or chemoradiotherapy with curative intent. Hypoxia had a significant detrimental effect on the mentioned endpoints. S: survival; RFS: relapse free survival; DFS: disease free survival; LC: locoregional control) [73-82]

Publication	N	Parameter	pO₂ (mm Hg)	Follow up (months)	Endpoints
Cervix cancer
Höckel et al., 1993	31	Median pO₂	< 10	19	S, RFS
Fyles et al, 1998	74	Hypoxic fraction	< 5	14	DFS
Knocke et al, 1999	51	Median pO₂	< 10	36	DFS, LC
Rofstad et al, 2000	32	Hypoxic subvolume	< 5		S, DFS, LC
Sundfor et al, 2000	40	Hypoxic subvolume	< 5	33	S, DFS, LC

Head-and-neck cancer
Nordsmark et al, 1996, 2000	70	Hypoxic fraction	< 2.5	24	LC
Brizel et al, 1997	28	Median pO₂	< 10	12	DFS
Stadler et al, 1997	59	Hypoxic subvolume	< 5	8	S
Rudat et al, 2001	134	Hypoxic fraction	< 2.5	2	S

Pathophysiology of tumor hypoxia

The growth and survival of cells in solid tumors is dependent on the adequate supply of oxygen and nutrients, which diffuse from the blood vessels and are consumed by the tumor cells. In order to meet their increasing oxygen and nutrient demand, tumors develop their own blood supply. However, the neovasculature is morphologically and functionally abnormal, and is generally unable to meet the increasing demands, resulting in a diffusion limited chronic hypoxia. As a result we get a so-called “corded structure”, which was first described by Thomlinson and Gray back in 1955, based on their observations of a metabolic oxygen gradient relative to blood vessels [6]. Tumor cells can roughly be divided in two categories with regard to their oxygenation and radiosensitivity. Cells lying near the capillaries, within the diffusion distance of oxygen (± 100 μm) are well oxygenated and radiosensitive. Cells lying at the edge and beyond the diffusion distance are hypoxic and radioresistant. We modelled this metabolic oxygen gradient in our radiosensitizing experiments by irradiating tumor cell in “micropellets”. Essentially, 0,5 x 10⁶ tumor cells in 100 μl of medium were placed in conical plastic tubes and pellets (± 300 μm thick) are produced by centrifugation at 300 g for 5 min, and kept on ice. Metabolic oxygen depletion in pellets is induced by a 3 min incubation at 37 °C prior to radiation. This model provides an oxygen enhancement ratio of at least 2.5, which indicates a mean level of oxygenation below 0.5 % [7].

In the late 1970s, Brown postulated that another type of hypoxia, being transient in nature, existed in solid tumors as well [8]. This was later confirmed and shown to result from temporary cessations in blood flow [9]. The mechanisms responsible for the intermittent closure of tumor blood vessels are not entirely understood yet. They may include plugging of blood vessels by blood cells or by circulating tumor cells, collapse of blood vessels in regions with high interstitial pressure, spasms and spontaneous vasomotion in incorporated host arterioles. This temporary closing of blood vessels results in perfusion limited acute hypoxia.

Strategies to overcome hypoxia induced radioresistance

Horsman and Overgaard performed a meta-analysis of all randomized trials in which some form of hypoxic modification was performed in solid tumors undergoing radiotherapy with curative intent [10]. They identified 91 trials, reporting more than 11 000 patients in total. The trials involved hypoxic cell radiosensitizers (n = 53), hyperbaric oxygen (n = 31), a combination of both (n =1), oxygen or carbogen breathing (n = 5) and blood transfusion (n = 1). Tumor sites were head-and-neck (n = 29), cervix (n = 20), bladder (n = 16), brain (n = 13), lung (n = 10), esophagus (n = 2) and mixed (n = 1). Overall, hypoxic modification significantly improved local control (odds ratio 1.29) and survival (odds ratio 1.19), without significantly affecting the rate of distant metastasis or radiotherapy related complications. The improvement remained significant when evaluating the trials with hypoxic cell radiosensitizers and hyperbaric oxygen separately. Analysis according to site showed only a significant improvement for head-and-neck and cervix cancer.

Improving tumor oxygenation

One of the earliest clinical attempts to eliminate hypoxia induced radioresistance involved patients breathing high oxygen content gas under hyperbaric conditions (3 atmosphere) [11]. The largest clinical trial with hyperbaric oxygen has been conducted by the British Medical Research Council, which randomized 1669 patients between radiotherapy with or without hyperbaric oxygen [12]. Hyperbaric oxygen significantly improved both survival and local control after radiotherapy for head-and-neck tumors and for advanced carcinomas of the cervix. In carcinomas of the bronchus there seemed to be some improvement in survival but this was not statistically significant. In carcinoma of the bladder hyperbaric oxygen showed no benefit.

Another way to improve the oxygen availability is to optimize the hemoglobin concentration. Indeed, it is well established that hemoglobin concentration is an important prognostic factor for the response to radiotherapy, especially in head-and-neck tumors, cervix carcinomas and bladder cancer [13, 14]. Generally, anemic patients have a reduced locoregional tumor control and survival. Based on these observations, the effect of transfusion in anemic patients was evaluated in the Danish head-and-neck cancer studies (Dahanca) 5, 6 and 7 [15, 16]. These trials confirmed the prognostic significance of the hemoglobin concentration. In addition, patients with low hemoglobin level were randomized to receive blood transfusions or not. Surprisingly, correction of anemia did not result in a significantly improved local control. This is supported by data from animal experiments, showing only a clear benefit of blood transfusion to mice when the interval between transfusion and irradiation is shorter than 24 h. This can be explained by the hypothesis that solid tumors posses adaptation mechanisms to anemia, which counteract the beneficial effects of transfusion on radiotherapy outcome [17]. The use of erythropoietin is another approach to correct anemia, and was evaluated in randomized placebo controlled trials in cervix and head-and-neck cancer [18, 19]. These trials demonstrated an impaired local control in the erythropoietin arm. Hence, we consider anemia to be a surrogate marker for poor prognosis rather than a pathophysiological cause of radioresistance. We do not support the general idea that non-symptomatic anemia should be corrected prior to radiotherapy.

Prevention of perfusion-limited acute hypoxia

Another way to improve tumor oxygenation is to prevent perfusion-limited acute hypoxia. Experimental studies showed that nicotinamide, a vitamine-B3 analog, can effectively radiosensitize murine tumors. In vitro studies suggested that this enhancement of radiation damage was the result of an inhibition of repair mechanisms. However, mouse studies demonstrated that the primary mode of action involves a reduction in tumor hypoxia. This was explained by a significant decrease in transient occlusions of blood vessels and thereby of perfusion-limited acute hypoxia. Nicotinamide was evaluated in clinical trials primarily in combination with carbogen breathing to decrease chronic hypoxia, and with accelerated radiotherapy to counteract tumor repopulation. Kaanders et al. recently reported such a phase II trial of ARCON (Accelerated Radiation CarbOgen and Nicotinamide) in 215 patients with advanced head-and-neck cancer. The 3-year locoregional control rates were excellent (e.g. 80 % for T3/4 carcinomas of the larynx), and the toxicity was acceptable [20]. Further evaluation of ARCON is necessary, to define in which clinical setting it might be beneficial.

Hypoxic cell radiosensitizers

In 1969, Adams and Cooke introduced the concept of hypoxic cell radiosensitizers, which are chemicals that mimic oxygen and enhance thereby radiation damage [21]. They demonstrated that the efficiency of radiosensitization is directly related to the electron affinity. These compounds are, in contrast to oxygen, not metabolized by tumor cells and can therefore diffuse further away from the capillaries than oxygen. Because these drugs mimic the effect of oxygen, they are supposed not to increase the toxic effects of radiotherapy to the well oxygenated normal tissues surrounding the tumor. The most widely studied group of hypoxic cell radiosensitizers are the nitroimidazoles. Laboratory experiments showed complete reversal of hypoxia induced tumor cell radioresistance by misonidazole without increasing the response of aerated tissues. These encouraging results led to a boom of clinical trials exploring the clinical radiosensitizing potential of misonidazole in the late 1970s. However, the results of these clinical trials have been generally disappointing. The most important factor underlying the failure of misonidazole to achieve useful advantage is undoubtedly the low plasma concentrations achievable with the permitted dose of this neurotoxic drug [22].

Nevertheless, a significant benefit was seen in some trials, such as the Dahanca 2 trial [23]. In this trial 626 patients with pharynx and larynx carcinoma were randomized to two different split-course radiation regimens and given either misonidazole or placebo during the initial 4 weeks of treatment. Overall, the misonidazole treated group did not have a significantly better control rate than the placebo group. However, a significant benefit was found in patients with pharynx carcinomas. Misonidazole induced significant peripheral neuropathy in 26 % of the treated patients, whereas other drug related side effects were minimal. In the Dahanca 5 trial, a less toxic nitroimidazole compound, nimorazole (Naxogin^®), was used [15]. Briefly, 422 patients with invasive carcinoma of the supraglottic larynx and pharynx were randomized to receive nimorazole or placebo, in association with conventional primary radiotherapy. With a median follow up of 112 months, the nimorazole group showed a significantly better locoregional control rate than the placebo group and a lower cancer-related death rate, without major side-effects. As a consequence, nimorazole has become part of the standard treatment schedule for head-and-neck tumors in Denmark.

More recently, a number of nitroimidazole derivatives, such as pimonidazole, EF3 and EF5, were developed as hypoxia markers. These compounds undergo intracellular reduction, bind mainly to thiol containing proteins and are thereby trapped in hypoxic cells. Specific antibodies allow their detection by immunohistochemistry, immunofluorescence and flow cytometry. In head-and-neck squamous-cell carcinoma, piminidazole staining has been demonstrated to correlate with decreased 2-year local control rates after radiotherapy [24]. Evans et al. demonstrated that EF5 binding correlates with pO₂ measurements by Eppendorf needle electrodes, and may be predictive for recurrence in brain tumors treated by surgery and postoperative radiotherapy [25, 26]. Another possible method to detect nitroimidazole adducts is the use of radioactive tracers. This approach has some intrinsic advantages, including its non-invasiveness, the possibility to evaluate the whole tumor volume and its repeatability. In addition, image fusion techniques and the use of intensity modulated and image guided radiotherapy allow to delineate hypoxic radioresistant sub-target volumes for delivering a partial tumor boost [27, 28]. [¹⁸F]EF3 is such a promising tracer. It can be detected by PET, and is currently being evaluated in a phase I clinical trial in head-and-neck cancer patients at the St-Luc University Hospital in Brussels [29]. Dynamic contrast enhanced MRI offers an alternative for delineating hypoxic subvolumes [30, 31].

Bioreductive drugs

Selective killing of hypoxic cells can also be achieved with bioreductive drugs, which undergo intracellular reduction to form active cytotoxic species under low oxygen tension. Two groups of bioreductive drugs, the quinines (e.g. mitomycin-C) and the organic nitroxides (e.g. tirapazamine), have been introduced in clinic.

Laboratory studies demonstrated that the activation of mitomycin-C by bioreduction under hypoxic conditions, leads to the formation of products that crosslink DNA and produce thereby cell killing. Animal studies indicated that mitomycin-C can be used in combination with radiotherapy to kill the hypoxic fraction of a solid tumor, while radiotherapy alone eradicates the oxygenated fraction. This led to the clinical evaluation of mitomycin-C in combination with radiotherapy in head-and-neck cancer patients [32]. Pilot studies reported an improvement in local tumor control, without an increase in normal tissue toxicity. However, these preliminary findings were not confirmed by larger clinical trials. There are two obvious reasons for this failure. First, mitomycin-C has a relative small differential killing effect between aerobic and hypoxic cells. Second, in most clinical trials mitomycin-C is only administered once or twice during the course of radiotherapy. Attempts to find more efficient quinones led to the development of porfiromycin and EO9, which are currently being evaluated.

The most promising group of bioreductive drugs are the organic nitroxides, of which tirapazamine is the lead compound. This molecule is metabolized by intracellular reductases to form a transient oxidizing radical that can be efficiently scavenged by molecular oxygen in aerated tissues to reform the parent compound. In the absence of oxygen, the oxidizing radicals abstract protons from the DNA to form DNA radicals and finally strand breaks. Tirapazamine shows substantial selective cytotoxicity for anoxic cells; it is approximately 100-fold more cytotoxic to anoxic than to oxygenated cell cultures. Animal studies showed a beneficial effect for combinations of tirapazamine with cisplatin and with radiation. Clinical phase II and III trials of tirapazamine combined with cisplatin in malignant melanoma and non-small cell lung cancer suggested a synergistic effect. Phase I and II trials of tirapazamine combined with radiation in patients with advanced head-and-neck cancer reported high local control rates and acceptable toxicity [33, 34]. The efficacy and toxicity of tirapazamine concurrent with radiotherapy is further being evaluated in several clinical trials.

NO-based radiosensitizing strategies

As early as 1957, Howard-Flanders showed that the authentic NO-gas is an efficient radiosensitizer of hypoxic bacteria, and postulated fixation of radiation induced DNA damage, thus mimicking the effects of oxygen on DNA lesions, as primary mechanism [35, 36]. An alternative mechanism might be interaction of NO with iron-sulphur containing enzymes, resulting in inhibition of mitochondrial respiration and sparing of the natural radiosensitizer oxygen [37].

In the early 1990s, Mitchell et al. evaluated the radiosensitizing potential of the NONOates, which have a X-[N(O)NO]-structure. When X is a secondary amine group, 2 molecules of NO per molecule of NONOate are spontaneously generated when dissolved in aqueous media. The NO level produced by the NONOates can easily be controlled, since the generation of NO is constant, predictable and independent from the pO₂. Mitchell et al. demonstrated that spontaneous release of NO by diethylamine nonoate (DEA/NO) and dipropylamine nonoate (PAPA/NO) radiosensitizes hypoxic Chinese hamster V79 lung fibroblast to a similar extent as oxygen [36]. Griffin et al. reported comparable activity for DEA/NO and spermine nonoate (SPER/NO), with enhancement ratios of 2.8-3.0 in hypoxic SCK mammary carcinoma cultures [38]. However, the generation of NO by the NONOates is not tumor selective and their in vivo application would result in high NO levels in the circulation, causing vasoactive complications (septic-like shock).

To be able to decrease the concentration of NO in the circulation, our laboratory explored the possibility to exploit the hypoxic tumor microenvironment for selective generation of NO by bioreduction. We showed that bioreduction of sodium nitroprusside (SNP) and S-niroso-N-acetylpenicillamine (SNAP) results in generation of NO and radiosensitization of hypoxic mouse mammary carcinoma and human pancreatic cancer cells [7, 39]. In our model of metabolic induced hypoxia, the radiosensitizing effect of NO was close to that of oxygen, while no radiosensitization was observed in aerobic cells. In an attempt to further decrease the extracellular concentration of NO, we decided to explore the possibility to endogenously generate NO inside tumor cells. Our laboratory was the first to demonstrate that the inducible isoform of nitric oxide synthase (iNOS), activated by cytokines (IL1β + IFNγ) in aerobic conditions, is capable of radiosensitizing tumor cells through endogenous production of NO (figure 3) [40]. We found that the iNOS pathway has a serious advantage over NO-donors, since a comparable level of radiosensitization was achieved at a 10 to 30-times lower extracellular NO background, which is favorable for in vivo applications.

Regulation of the iNOS promotor

To choose an optimal, clinically relevant iNOS induction schedule, we studied the transcriptional activation of iNOS in murine EMT-6 mammary carcinoma cells. It is well documented that full activation of the iNOS gene requires cooperation of two promoter regions, located from - 40 to - 300 bp (region I) and – 900 to – 1100 bp (region II) upstream of the TATA box (figure 4). Standard combinations like IFNγ + IL1β, prime iNOS activation via an interferon-responsive element (ISRE) or an IFNγ activated site (GAS) in region II. To maintain a high iNOS transcription rate, a second stimulus toward region I is required and can be provided by IL1β or LPS via the NFκB signaling pathway [41]. This pathway involves kinase IKKα-induced phosphorylation and proteasome-mediated degradation of the inhibitory IκB subunit in the cytoplasm followed by translocation of the p50/p65 NFκB complex to the nucleus and its binding to the iNOS promoter. In macrophages, the second stimulus may be provided by hypoxia via hypoxia-inducible factor-1 (HIF1) that has a binding site in region I [42].

We identified a similar effect of hypoxia on iNOS induction in tumor cells. Indeed, 16 h pre-incubation of EMT6 cultures in 1% oxygen (average of the tumor microenvironment) indirectly modulated the radioresponse through cytokine-inducible iNOS. Low concentrations of cytokines, which were not active in aerobic cells became efficient inducers of iNOS in chronic hypoxia. This in turn drastically improved the radioresponse of EMT6 cells (up to 2.5-times), collected from chronically hypoxic cultures and dropped into micropellets, our model of metabolic hypoxia that was described above. Hence, hypoxia has a dual on the radioresponse of EMT6 tumor cells: it directly induces radioresistance, and may indirectly modulate radiosensitivity through up-regulation of the iNOS/NO pathway [43].

The dual role of NFκB in tumor cell radioresponse

It is time to face the possibility of conflicting roles of NFκB in the radioresponse of tumor cells, keeping in mind that NFκB signaling is triggered by diverse stimuli and involved in the regulation of multiple downstream genes. Extensive literature strongly suggested that dysregulation or constitutive activation of NFκB is linked to tumorigenesis, angiogenesis and metastasis, and that it protects tumor cells from radiation damage [44, 45]. Consistently, NFκB inhibition has been used as an approach to radiosensitize tumor cells, aiming at stimulating apoptosis and/or inhibiting DNA repair. This approach did work, but disruption of NFκB signaling by indomethacin, by the proteasome inhibitors MG132 and PS341 or by transfection with the super-repressor IκBα resulted in a moderate if not marginal radiosensitization by 1.2 to 1.4 times [46-48]. The range of radiosensitization covered by NFκB inhibition was not impressive but, we told, it might be amplified under hypoxic conditions, which were missing in all reports mentioned above. An idea behind was (a) it is difficult to further improve the radioresponse of already radiosensitive aerobic tumor cells, and (b) hypoxic rather than aerobic cells are an obstacle for radiotherapy. Therfore, we treated EMT6 cells with phenylarsine oxide (PAO) or lactacystin (a proteasome inhibitor) to inhibit NFκB, and afterwards analyzed their hypoxic cell radioresponse. Once again, the increase in radioresponse was rather moderate (up to 1.4-fold), but the picture turned upside down when NFκB was pre-activated by bacterial lipopolysaccharide (LPS) to imitate its increased basal activity in tumors. Instead of radiosensitization, PAO and lactacystin drastically impaired (by > 2 times) the radioresponse of hypoxic EMT6 cells through disruption of NFκB signaling towards the iNOS gene [49]. Taken together, our findings suggest that the radiosensitizing effects of NFκB inhibitors may be seriously compromised through iNOS, one of its downstream targets. Secondly, this counteraction may be unmasked only in hypoxia, which is a major cause of tumor radioresistance. Finally, the benefit of radiosensitization obtained through activated NFκB signaling towards the iNOS gene (2 to 2.5-fold) is much more promising than that caused by NFκB inhibition (1.4-fold). This comparison encouraged us to look for a good preclinical candidate to activate the NFκB/iNOS/NO pathway.

Lipid A analogues and hypoxic tumor cell radiosensitization

During the last decade, the mechanism and profile of the NFκB signaling pathway has been clarified in many types of mammalian cells, activated by diverse stimuli such as TNFα, IL1β and LPS. Perhaps one of the best studied and widely used stimulus was and still remains to be LPS, the major component of the outer membrane of gram-negative bacteria. This endotoxin and immunostimulator is known to activate monocytes/macrophages through the TLR4/MyD88-receptor complex, which finally results in the NFκB controlled secretion of a variety of pro-inflammatory mediators including NO, TNFα and IL6 [50]. An attempt to use LPS in cancer immunotherapy was not successful, since strong septic-like toxicity could not be prevented at a dose as low as 4 ng/kg [51]. Because of that, the bioactive component of LPS, lipid A, was isolated and its simplified synthetic analogs MPL (monophosphoryl lipid A), ONO4007 and OM174 were developed and examined in preclinical and clinical trials. These derivatives appeared to be much better tolerated, while preserving the immunomodulating activities of LPS [52]. Although anticipated, the role of NFκB in lipid A signaling towards iNOS has not been studied in macrophages in detail, whilst tumor cells were simply out of the scope of interests. The questions we raised were: (1) Is lipid A as active as LPS in iNOS induction and radiosensitization of EMT6 tumor cells? (2) Does lipid A cause radiosensitization at plasma relevant concentrations? (3) Does lipid A activate iNOS in EMT6 tumor cells through NFκB signaling?

As we deduced from the LPS chemistry, lipid A did closely mimic the activity of LPS to induce iNOS in hypoxia but not in normoxia. What was similar and predictable, the range of radiosensitizing effects up to 2.5-fold observed in hypoxic but not aerobic EMT6 tumor cells [53]. What was different and worthy to mention, lipid A displayed a 30 to 100-times less potency and hence was routinely used at 3-30 μg/ml, compared with 0.1 μg/ml LPS. Such concentrations were, however, plasma achievable in cancer patients, as the phase I trials of ONO4007 has recently shown [54]. What was unclear and worthy to explore, the role the NFκB signaling pathway towards the iNOS gene, considering the considerable shift in active concentrations. Two conclusions emerged from our transfection experiments using an NFκB tandem and an array of plasmids containing either deletions or mutations at critical points in the two regions of the iNOS promoter that bind transcription factors. First, the role of NFκB is crucial in the lipid A-induced transcriptional activation of iNOS, since the activity of the NFκB tandem and the wild type iNOS promoter was significantly increased. Second, lipid A-driven signaling predominantly targets the proximal but not distal NFκB binding site of the iNOS promoter, which fits to the background of LPS [41, 55]. Therefore, the difference between LPS and lipid A observed in our experiments presumably reflected a modulated affinity of lipid A for the TLR4 receptor, as a result of the carbohydrate core removal from LPS. The similar if not identical profile of lipid A/LPS activities was very helpful for us to profit from the extensive background on the very toxic, and sometimes fatal, substance LPS in the domain of immunology. Why we turned to immunology being convinced radiobiologists is explained in the next paragraph.

Role of the proinflammatory tumor infiltrate in radioresponse

The tumor microenvironment has other particular aspects besides hypoxia. Indeed, solid tumors contain a complex network of inflammatory cells (e.g., macrophages, T/NK-cells), which are re-programmed to stimulate (rather than to inhibit) tumorigenesis, through the secretion of several growth and pro-angiogenic factors [56]. Such a mechanism was described for breast, cervix and bladder carcinomas, wherein an increased density of tumor-associated macrophages was correlated with poor prognosis. In the past, an idea to exploit inflammatory cells in tumor immunotherapy was regularly revisited but never realized. Surprisingly, an idea to exploit inflammatory cells in tumor radiotherapy was even never visited, although, the inflammatory mediators TNFα and NO have been repeatedly used as biochemicals to radiosensitize tumor cells [7, 36, 39, 57, 58]. We decided to explore whether the proinflammatory tumor infiltrate may be exploited for tumor cell radiosensitization.

As a first step, the macrophage-like RAW 264.7 cell line was used to model the interaction between pro-inflammatory and tumor cells [59]. RAW 264.7 macrophages were activated with plasma-relevant concentrations of lipid A, and the conditioned medium was screened for cytokine secretion by Elisa and for the oxidative NO metabolite nitrite. The use of RAW 264.7 cells appeared to be a valid experimental model, since the spectrum of released mediators was generally matching that of native macrophages [60]. Indeed, TNFα and IL6 were produced at the highest rates. Both cytokines are known to be co-expressed by monocytes/macrophages during the acute phase of inflammation and immunity. Second, IL12 and IL18 were produced as well, which illuminates paracrine signaling from macrophages to T/NK cells, leading to the secondary production of IFNγ [61, 62]. This cascade was omitted in our model, which was based on a pure macrophage like cell line. Two possible mechanisms of radiosensitization were identified: (a) indirect, through induction of the iNOS/NO pathway in EMT6 cells by macrophage-released mediators, and (b) direct, through induction of the iNOS/NO pathway in RAW 264.7 macrophages and diffusion of NO to bystander tumor cells (figure 5).

Macrophages and hypoxia appear to be independent partners rather than occasional friends that normally co-localize in the same tumor regions. Indeed, hypoxia may up-regulate the iNOS expression in tumor cells only if a basal level of iNOS or a high transcriptional inducibility of iNOS is displayed, because of the genetic pattern of the tumor cells. Let us name such a pattern as iNOS-positive, which seems to be true for at least some types of tumors [63-71]. What about iNOS-negative tumors or tumor cell subpopulations, wherein the iNOS/NO pathway is genetically silenced? In such a case, neither hypoxia nor macrophage-released cytokines look promising anymore. This is the situation wherein macrophages are supposed to make the difference, since they are an alternative source of NO. To our knowledge, an iNOS-negative pattern of macrophages has little if any chance to exist, regarding that macrophages are naturally programmed to deliver a high output of NO upon proper activation. It appears from our experimental data, that this activation may require IFNγ next to lipid A, to benefit from their synergism based on the cooperative signaling of NFκB and JAK/STAT towards the iNOS promoter [41]. To examine this, we modeled the immune cells in the proinflammatory infiltrate by splenocytes, which contain both macrophages and IFNγ producing T/NK-cells. Stimulation of splenocytes by the lipid A analogue OM174 resulted in the production of IL12, IL18 and IFNγ. This caused up to 2.1-fold radiosensitization of EMT6 tumor cells, which was associated with the iNOS-mediated production of NO. Both iNOS activation and radiosensitization were counteracted by neutralizing antibodies against IFNγ or against IL12/18 [72]. Lipid A analogues appear therefore be a potential new class of tumor cell radiosensitizers through activation of the IFNγ/iNOS pathway. This finding indicates a rationale for combining immunostimulatory and radiosensitizing strategies in the future.

References

1 Howard-Flanders P, Moore D. The time interval after pulsed irradiation within which injury to bacteria can be modified by dissolved oxygen. I. A search for an effect of oxygen 0.02 second after pulsed irradiation. Radiat Res 1958 ; 9 : 422-37.

2 Michael BD, Adams GE, Hewitt HB, Jones WB, Watts ME. A posteffect of oxygen in irradiated bacteria : a submillisecond fast mixing study. Radiat Res 1973 ; 54 : 239-51.

3 Vaupel P, Schlenger K, Knoop C, Hockel M. Oxygenation of human tumors : evaluation of tissue oxygen distribution in breast cancers by computerized O2 tension measurements. Cancer Res 1991 ; 51 : 3316-22.

4 Lartigau E, Le Ridant AM, Lambin P, Weeger P, Martin L, Sigal R, et al. Oxygenation of head and neck tumors. Cancer 1993 ; 71 : 2319-25.

5 Brown JM, Wilson WR. Exploiting tumour hypoxia in cancer treatment. Nat Rev Cancer 2004 ; 4 : 437-47.

6 Thomlinson RH, Gray LH. The histological structure of some human lung cancers and the possible implications for radiotherapy. Br J Cancer 1955 ; 9 : 539-49.

7 Verovski VN, Van den Berge DL, Soete GA, Bols BL, Storme GA. Intrinsic radiosensitivity of human pancreatic tumour cells and the radiosensitising potency of the nitric oxide donor sodium nitroprusside. Br J Cancer 1996 ; 74 : 1734-42.

8 Brown JM. Evidence for acutely hypoxic cells in mouse tumours, and a possible mechanism of reoxygenation. Br J Radiol 1979 ; 52 : 650-6.

9 Chaplin DJ, Olive PL, Durand RE. Intermittent blood flow in a murine tumor : radiobiological effects. Cancer Res 1987 ; 47 : 597-601.

10 Horsman MR, Overgaard J. 2002. Overcoming tumour radioresistance resulting from hpoxia. In : Steel GG, ed. Basic clinical radiobiology. New York : Arnold, 2002.

11 McEwen JB. Clinical trials of hyperbaric oxygen and radiotherapy. Br J Radiol 1968 ; 41 : 556.

12 Radiotherapy and hyperbaric oxygen. Report of a Medical Research Council Working Party. Lancet 1978 ; 2 : 881-4.

13 Fein DA, Lee WR, Hanlon AL, Ridge JA, Langer CJ, Curran WJ, et al. Pretreatment hemoglobin level influences local control and survival of T1-T2 squamous cell carcinomas of the glottic larynx. J Clin Oncol 1995 ; 13 : 2077-83.

14 Pollack A, Zagars GK, Dinney CP, Swanson DA, von Eschenbach AC. Preoperative radiotherapy for muscle-invasive bladder carcinoma. Long term follow-up and prognostic factors for 338 patients. Cancer 1994 ; 74 : 2819-27.

15 Overgaard J, Hansen HS, Overgaard M, Bastholt L, Berthelsen A, Specht L, et al. A randomized double-blind phase III study of nimorazole as a hypoxic radiosensitizer of primary radiotherapy in supraglottic larynx and pharynx carcinoma. Results of the Danish Head and Neck Cancer Study (Dahanca) Protocol 5-85. Radiother Oncol 1998 ; 46 : 135-46.

16 Overgaard J, Hansen HS, Specht L, Overgaard M, Grau C, Andersen E, et al. Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck : DAHANCA 6 and 7 randomised controlled trial. Lancet 2003 ; 362 : 933-40.

17 Hirst DG. Anemia : a problem or an opportunity in radiotherapy? Int J Radiat Oncol Biol Phys 1986 ; 12 : 2009-17.

18 Temkin SM, Hellmann M, Serur E, Lee YC, Abulafia O. Erythropoietin administration during primary treatment for locally advanced cervical carcinoma is associated with poor response to radiation. Int J Gynecol Cancer 2006 ; 16 : 1855-61.

19 Henke M, Laszig R, Rube C, Schafer U, Haase KD, Schilcher B, et al. Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy : randomised, double-blind, placebo-controlled trial. Lancet 2003 ; 362 : 1255-60.

20 Kaanders JH, Pop LA, Marres HA, Bruaset I, van den Hoogen FJ, Merkx MA, et al. ARCON : experience in 215 patients with advanced head-and-neck cancer. Int J Radiat Oncol Biol Phys 2002 ; 52 : 769-78.

21 Adams GE, Cooke MS. Electron-affinic sensitization. I. A structural basis for chemical radiosensitizers in bacteria. Int J Radiat Biol Relat Stud Phys Chem Med 1969 ; 15 : 457-71.

22 Dische S. Chemical sensitizers for hypoxic cells : a decade of experience in clinical radiotherapy. Radiother Oncol 1985 ; 3 : 97-115.

23 Overgaard J, Hansen HS, Andersen AP, Hjelm-Hansen M, Jorgensen K, Sandberg E, et al. Misonidazole combined with split-course radiotherapy in the treatment of invasive carcinoma of larynx and pharynx : report from the Dahanca 2 study. Int J Radiat Oncol Biol Phys 1989 ; 16 : 1065-8.

24 Kaanders JH, Wijffels KI, Marres HA, Ljungkvist AS, Pop LA, van den Hoogen FJ, et al. Pimonidazole binding and tumor vascularity predict for treatment outcome in head and neck cancer. Cancer Res 2002 ; 62 : 7066-74.

25 Evans SM, Judy KD, Dunphy I, Jenkins WT, Nelson PT, Collins R, et al. Comparative measurements of hypoxia in human brain tumors using needle electrodes and EF5 binding. Cancer Res 2004 ; 64 : 1886-92.

26 Evans SM, Judy KD, Dunphy I, Nelson PT, Margarelli D, Jenkins WT, et al. Hypoxia, as measured by EF5 binding, predicts recurrence in adult brain tumors. Int J Radiat Oncol Biol Phys 2003 ; 55 : 470-1.

27 Gregoire V. [Target-volume selection and delineation in the cervico-maxillo-facial region : beyond the concepts of the ICRU]. Cancer Radiother 2002 ; 6 : 29s-31s.

28 Van den Berge DL, De Ridder M, Storme GA. Imaging in radiotherapy. Eur J Radiol 2000 ; 36 : 41-8.

29 Mahy P, De Bast M, Leveque PH, Gillart J, Labar D, Marchand J, et al. Preclinical validation of the hypoxia tracer 2-(2-nitroimidazol-1-yl)-N-(3,3,3-[(18)F]trifluoropropyl)acetamide, [(18)F]EF3. Eur J Nucl Med Mol Imaging 2004 ; 31 : 1263-72.

30 Payne GS, Leach MO. Applications of magnetic resonance spectroscopy in radiotherapy treatment planning. Br J Radiol 2006 ; 79 : S16-S26.

31 McMillan KM, Rogers BP, Field AS, Laird AR, Fine JP, Meyerand ME. Physiologic characterisation of glioblastoma multiforme using MRI-based hypoxia mapping, chemical shift imaging, perfusion and diffusion maps. J Clin Neurosci 2006 ; 13 : 811-7.

32 Sartorelli AC, Hodnick WF, Belcourt MF, Tomasz M, Haffty B, Fischer JJ, et al. Mitomycin C : a prototype bioreductive agent. Oncol Res 1994 ; 6 : 501-8.

33 Rischin D, Peters L, Hicks R, Hughes P, Fisher R, Hart R, et al. Phase I trial of concurrent tirapazamine, cisplatin, and radiotherapy in patients with advanced head and neck cancer. J Clin Oncol 2001 ; 19 : 535-42.

34 Shulman LN, Buswell L, Riese N, Doherty N, Loeffler JS, von Roemeling RW, et al. Phase I trial of the hypoxic cell cytotoxin tirapazamine with concurrent radiation therapy in the treatment of refractory solid tumors. Int J Radiat Oncol Biol Phys 1999 ; 44 : 349-53.

35 Howard-Flanders P. Effect of nitric oxide on the radiosensitivity of bacteria. Nature 1957 ; 180 : 1191-2.

36 Mitchell JB, Wink DA, DeGraff W, Gamson J, Keefer LK, Krishna MC. Hypoxic mammalian cell radiosensitization by nitric oxide. Cancer Res 1993 ; 53 : 5845-8.

37 Mitchell JB, Cook JA, Krishna MC, DeGraff W, Gamson J, Fisher J, et al. Radiation sensitisation by nitric oxide releasing agents. Br J Cancer 1996 ; 27 : S181-S184.

38 Griffin RJ, Makepeace CM, Hur WJ, Song CW. Radiosensitization of hypoxic tumor cells in vitro by nitric oxide. Int J Radiat Oncol Biol Phys 1996 ; 36 : 377-83.

39 Janssens MY, Verovski VN, Van den Berge DL, Monsaert C, Storme GA. Radiosensitization of hypoxic tumour cells by S-nitroso-N-acetylpenicillamine implicates a bioreductive mechanism of nitric oxide generation. Br J Cancer 1999 ; 79 : 1085-9.

40 Janssens MY, Van den Berge DL, Verovski VN, Monsaert C, Storme GA. Activation of inducible nitric oxide synthase results in nitric oxide-mediated radiosensitization of hypoxic EMT-6 tumor cells. Cancer Res 1998 ; 58 : 5646-8.

41 Xie QW, Whisnant R, Nathan C. Promoter of the mouse gene encoding calcium-independent nitric oxide synthase confers inducibility by interferon gamma and bacterial lipopolysaccharide. J Exp Med 1993 ; 177 : 1779-84.

42 Melillo G, Musso T, Sica A, Taylor LS, Cox GW, Varesio L. A hypoxia-responsive element mediates a novel pathway of activation of the inducible nitric oxide synthase promoter. J Exp Med 1995 ; 182 : 1683-93.

43 Van den Berge DL, De Ridder M, Verovski VN, Janssens MY, Monsaert C, Storme GA. Chronic hypoxia modulates tumour cell radioresponse through cytokine-inducible nitric oxide synthase. Br J Cancer 2001 ; 84 : 1122-5.

44 Karin M, Cao Y, Greten FR, Li ZW. NF-kappaB in cancer : from innocent bystander to major culprit. Nature Rev Cancer 2002 ; 2 : 301-10.

45 Jung M, Zhang Y, Lee S, Dritschilo A. Correction of radiation sensitivity in ataxia telangiectasia cells by a truncated I kappa B-alpha. Science 1995 ; 268 : 1619-21.

46 Bradbury CM, Markovina S, Wei SJ, Rene LM, Zoberi I, Horikoshi N, et al. Indomethacin-induced radiosensitization and inhibition of ionizing radiation-induced NF-kappaB activation in HeLa cells occur via a mechanism involving p38 MAP kinase. Cancer Res 2001 ; 61 : 7689-96.

47 Russo SM, Tepper JE, Baldwin AS, Liu R, Adams J, Elliott P, et al. Enhancement of radiosensitivity by proteasome inhibition : implications for a role of NF-kappaB. Int J Radiat Oncol Biol Phys 2001 ; 50 : 183-93.

48 Wang CY, Mayo MW, Baldwin AS. TNFα and cancer therapy-induced apoptosis : potentiation by inhibition of NF-kappaB. Science 1996 ; 274 : 784-7.

49 De Ridder M, Van den Berge DL, Verovski VN, Monsaert C, Wauters N, Storme GA. NF-kappaB inhibition impairs the radioresponse of hypoxic EMT-6 tumour cells through downregulation of inducible nitric oxide synthase. Br J Cancer 2003 ; 88 : 120-4.

50 Raetz CR, Whitfield C. Lipopolysaccharide endotoxins. Annu Rev Biochem 2002 ; 71 : 635-700.

51 Engelhardt R, Mackensen A, Galanos C. Phase I trial of intravenously administered endotoxin (Salmonella abortus equi) in cancer patients. Cancer Res 1991 ; 51 : 2524-30.

52 Pance A, Reisser D, Jeannin JF. Antitumoral effects of lipid A : preclinical and clinical studies. J Investig Med 2002 ; 50 : 173-8.

53 De Ridder M, Verovski VN, Van den Berge DL, Sermeus AB, Monsaert C, Wauters N, et al. Lipid a radiosensitizes hypoxic EMT-6 tumor cells : role of the NF-kappaB signaling pathway. Int J Radiat Oncol Biol Phys 2003 ; 57 : 779-86.

54 de Bono JS, Dalgleish AG, Carmichael J, Diffley J, Lofts FJ, Fyffe D, et al. Phase I study of ONO-4007, a synthetic analogue of the lipid A moiety of bacterial lipopolysaccharide. Clin Cancer Res 2000 ; 6 : 397-405.

55 Xie QW, Kashiwabara Y, Nathan C. Role of transcription factor NF-kappa B/Rel in induction of nitric oxide synthase. J Biol Chem 1994 ; 269 : 4705-8.

56 Bingle L, Brown NJ, Lewis CE. The role of tumour-associated macrophages in tumour progression : implications for new anticancer therapies. J Pathol 2002 ; 196 : 254-65.

57 Sersa G, Willingham V, Milas L. Anti-tumor effects of tumor necrosis factor alone or combined with radiotherapy. Int J Cancer 1988 ; 42 : 129-34.

58 Kimura K, Bowen C, Spiegel S, Gelmann EP. Tumor necrosis factor-alpha sensitizes prostate cancer cells to gamma-irradiation-induced apoptosis. Cancer Res 1999 ; 59 : 1606-14.

59 De Ridder M, Verovski VN, Darville MI, Van den Berge DL, Monsaert C, Eizirik DL, et al. Macrophages enhance the radiosensitizing activity of lipid A : A novel role for immune cells in tumor cell radioresponse. Int J Radiat Oncol Biol Phys 2004 ; 60 : 598-606.

60 Cavaillon JM. Cytokines and macrophages. Biomed Pharmacother 1994 ; 48 : 445-53.

61 Trinchieri G. Interleukin-12 and the regulation of innate resistance and adaptive immunity. Nat Rev Immunol 2003 ; 3 : 133-46.

62 Matsuura M, Saito S, Hirai Y, Okamura H. A pathway through interferon-gamma is the main pathway for induction of nitric oxide upon stimulation with bacterial lipopolysaccharide in mouse peritoneal cells. Eur J Biochem 2003 ; 270 : 4016-25.

63 Thomsen LL, Miles DW, Happerfield L, Bobrow LG, Knowles RG, Moncada S. Nitric oxide synthase activity in human breast cancer. Br J Cancer 1995 ; 72 : 41-4.

64 Thomsen LL, Lawton FG, Knowles RG, Beesley JE, Riveros-Moreno V, Moncada S. Nitric oxide synthase activity in human gynecological cancer. Cancer Res 1994 ; 54 : 1352-4.

65 Ambs S, Merriam WG, Bennett WP, Felley-Bosco E, Ogunfusika MO, Oser SM, et al. Frequent nitric oxide synthase-2 expression in human colon adenomas : implication for tumor angiogenesis and colon cancer progression. Cancer Res 1998 ; 58 : 334-41.

66 Moncada S, Palmer RM, Higgs EA. Nitric oxide : physiology, pathophysiology, and pharmacology. Pharmacol Rev 1991 ; 43 : 109-42.

67 Rosbe KW, Prazma J, Petrusz P, Mims W, Ball SS, Weissler MC. Immunohistochemical characterization of nitric oxide synthase activity in squamous cell carcinoma of the head and neck. Otolaryngol Head Neck Surg 1995 ; 113 : 541-9.

68 Ambs S, Bennett WP, Merriam WG, Ogunfusika MO, Oser SM, Khan MA, et al. Vascular endothelial growth factor and nitric oxide synthase expression in human lung cancer and the relation to p53. Br J Cancer 1998 ; 78 : 233-9.

69 Klotz T, Bloch W, Volberg C, Engelmann U, Addicks K. Selective expression of inducible nitric oxide synthase in human prostate carcinoma. Cancer 1998 ; 82 : 1897-903.

70 Swana HS, Smith SD, Perrotta PL, Saito N, Wheeler MA, Weiss RM. Inducible nitric oxide synthase with transitional cell carcinoma of the bladder. J Urol 1999 ; 161 : 630-4.

71 Hajri A, Metzger E, Vallat F, Coffy S, Flatter E, Evrard S, et al. Role of nitric oxide in pancreatic tumour growth : in vivo and in vitro studies. Br J Cancer 1998 ; 78 : 841-9.

72 De Ridder M, Verovski VN, Chiavaroli C, Van den Berge DL, Monsaert C, Law K, et al. The radiosensitizing effect of immunoadjuvant om-174 requires cooperation between immune and tumor cells through interferon-gamma and inducible nitric oxide synthase. Int J Radiat Oncol Biol Phys 2006 ; 66(5) : 1473-80.

73 Hockel M, Knoop C, Schlenger K, Vorndran B, Baussmann E, Mitze M, et al. Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 1993 ; 26 : 45-50.

74 Fyles AW, Milosevic M, Wong R, Kavanagh MC, Pintilie M, Sun A, et al. Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 1998 ; 48 : 149-56.

75 Knocke TH, Weitmann HD, Feldmann HJ, Selzer E, Potter R. Intratumoral pO2-measurements as predictive assay in the treatment of carcinoma of the uterine cervix. Radiother Oncol 1999 ; 53 : 99-104.

76 Rofstad EK, Sundfor K, Lyng H, Trope CG. Hypoxia-induced treatment failure in advanced squamous cell carcinoma of the uterine cervix is primarily due to hypoxia-induced radiation resistance rather than hypoxia-induced metastasis. Br J Cancer 2000 ; 83 : 354-9.

77 Sundfor K, Lyng H, Trope CG, Rofstad EK. Treatment outcome in advanced squamous cell carcinoma of the uterine cervix : relationships to pretreatment tumor oxygenation and vascularization. Radiother Oncol 2000 ; 54 : 101-7.

78 Nordsmark M, Overgaard M, Overgaard J. Pretreatment oxygenation predicts radiation response in advanced squamous cell carcinoma of the head and neck. Radiother Oncol 1996 ; 41 : 31-9.

79 Nordsmark M, Overgaard J. A confirmatory prognostic study on oxygenation status and loco-regional control in advanced head and neck squamous cell carcinoma treated by radiation therapy. Radiother Oncol 2000 ; 57 : 39-43.

80 Brizel DM, Sibley GS, Prosnitz LR, Scher RL, Dewhirst MW. Tumor hypoxia adversely affects the prognosis of carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 1997 ; 38 : 285-9.

81 Stadler P, Becker A, Feldmann HJ, Hansgen G, Dunst J, Wurschmidt F, et al. Influence of the hypoxic subvolume on the survival of patients with head and neck cancer. Int J Radiat Oncol Biol Phys 1999 ; 44 : 749-54.

82 Rudat V, Stadler P, Becker A, Vanselow B, Dietz A, Wannenmacher M, et al. Predictive value of the tumor oxygenation by means of pO2 histography in patients with advanced head and neck cancer. Strahlenther Onkol 2001 ; 177 : 462-8.