Blood flow and associated pathophysiology of uterine cervix cancers: Characterisation and relevance for localised hyperthermia

Abstract

Cervical cancers exhibit substantial intra- and inter-tumour heterogeneities in blood flow prior to treatment, reflecting similar variability in vascularisation. When clinically relevant hyperthermia is applied as an adjuvant to established treatment modalities, blood flow may change in non-predictable directions, extents and durations, indicating subsequent variability in heat dissipation and in flow-associated parameters of the tumour microenvironment. Before heating, locally advanced cervical cancers are mostly hypoxic, acidic, exhibit substrate and energy deprivation and show lactate accumulation, which is spatially and temporally heterogeneous. Additionally a relatively homogeneous interstitial hypertension is observed. Most probably, metabolic parameters of the hostile microenvironment are able to greatly modulate the thermosensitivity of cancer cells. Adequate information concerning changes upon heat treatment is not available so far. Due to this lack of proven data for cervical cancers upon heat treatment, clinical studies are urgently needed in order to judge the possible impact of blood flow and the above-mentioned microenvironmental parameters.

• Cervical cancer
• blood flow
• oxygenation status
• hypoxia
• interstitial hypertension

Introduction

There is a clear interaction between the vascular and pathophysiological characteristics of solid tumours and localised hyperthermia (HT). Such an association includes both an effect of the tumour perfusion on heating and the effect of heat on these tumour characteristics [1–3]. Of the parameters involved in this association it is probably blood flow (perfusion) that plays the most crucial role [2], [3]. In this context it is important to recognise that perfusion responsible for heat transfer generally takes place through macrovessels with outer diameters greater than 100 µm (‘thermally significant vessels’), whereas nutrient delivery and exchange as well as drainage of metabolic waste products generally occurs in the microcirculation in vessels with outer diameters less than 100 µm. A poor microcirculation results in the development of large tumour areas that are nutrient- and oxygen-deprived (hypoxic), severely acidic and energy-depleted [4]. Tumour cells exposed to such hostile microenvironmental conditions are known to be more sensitive to the therapeutic effects of heat [2]. Due to the critical role of the tumour microcirculatory function (interstitial hypertension included), hypoxia, acidosis and the poor bioenergetic status in localised hyperthermia, this review provides basic information on these parameters which influence the heatability and thermosensitivity of cervical cancers, while at the same time being altered themselves by heat application.

Blood flow in cervical cancers upon hyperthermia

Although a range of techniques is available for tumour perfusion imaging (e.g. positron emission tomography with ¹⁵O-labelled water, dynamic contrast-enhanced magnetic resonance imaging, contrast-enhanced perfusion computed tomography and ultrasound [5]), pretreatment blood flow data for cancers of the uterine cervix are scarce and vary considerably, with mean values ranging from 0.08 to 1.03 mL/g/min (see Table I) [4], [7–11]. Some cervical cancers exhibit flow rates which are similar to those measured in organs with a high metabolic rate (e.g. liver) or in organs with special functions (e.g. spleen), whereas others show perfusion rates even lower than those found in tissues with a low metabolic turnover (e.g. skin at indifferent temperature levels, resting skeletal muscle or fat [4]). Flow data from multiple sites of measurement show substantial heterogeneity within individual cervical cancers. Heterogeneity within tumours is similar to that observed between tumours. In addition, there is no association between clinical size and blood flow (see Figure 1) [8]. These flow heterogeneities may reflect similar heterogeneities in vascularisation [12], [13]. Blood flow values reported for cancers of the uterine cervix in absolute terms may be biased by technological problems depending on the method used (see Table I), and by the analytical procedures employed [10].

Figure 1. Tumour blood flow (TBF) as a function of volume of cancers of the uterine cervix. Graph based on data published by Lyng et al. [7]. Values are means and ranges.

Table I.  Blood flow in normal uterus and in cancers of the uterine cervix.

	Blood flow (ml/g/min)^1	Method	References
Normal	0.10-0.25	Review of data	[6]
uterus	0.13	Review of data	[4]
Cervical	0.11 ± 0.03^2	Thermal washout	[7]
cancer	(0.05–0.18)^3
	0.08 [II/III]^4	CE–MRI	[8]
	(0.02–0.29)^3
	0.87 ± 0.38 [I]^4	DCE–CT	[9]
	0.99 ± 0.17 [II–IV]^4
	0.62 ± 0.36	H2^15O–PET	[11]
	0.72 (0.47–1.32)^3	CT (maximum slope)	[10]
	1.03 (0.59–1.43)^3	CT (1-compartment model)

¹Or (mL/mL/min).

²Recurrent cervical cancers.

³Range.

⁴Tumour stage.

As already mentioned earlier, pronounced flow heterogeneities assessed in cervical cancers most probably impact heat dissipation and heatability in some tumours (or at least in some tumour areas) through variabilities in the macrocirculation, and may affect factors of the tumour microenvironment (via heterogeneities in the microcirculation) which are known to modify heat sensitivity of cancer cells (for a recent review see Vaupel et al. [4]).

Perfusion changes in cervical cancers upon deep regional HT, a technique used for locally advanced cancers of the pelvis [14], have been investigated by Lüdemann et al. [11]. In this study the mean tumour temperature was 40.7° ± 0.6°C. Analysis of the H₂¹⁵O-PET data yielded a mean blood flow of 0.62 ± 0.36 mL/g/min before regional HT. Continuous heat application for 60 min resulted in a slight, insignificant increase in perfusion of 0.07 mL/g/min. Flow returned to initial values 20 min after heating. As a consequence of this clinical study, questions still exist regarding whether or not results obtained from fast-growing murine tumours are an appropriate reflection of what occurs clinically in human tumours, especially when sustained increases (over several hours) in blood flow of experimental tumours are considered (for a recent review see [2])

Changes in blood flow in cervical cancers upon thermoradiotherapy (TRT) have been communicated by Molls and Feldmann [7]. In this pilot study, locally recurrent cervical cancers were studied using thermal clearance rates. Flow rates estimated during a course of TRT did not change systematically as a function of the number of heat treatments, and alterations of flow (slight increases in 4 out of 17 treatments, minor decreases in 13 out of 17) were clinically not relevant.

From these (orientating) clinical studies it remains unclear whether or not the minimal flow increases observed may directly improve intratumour pharmacokinetics (e.g. in combined thermochemotherapy) or may facilitate heat dissipation (i.e. worsening of the heatability of tumours) in the clinical setting. Indirect effects of HT via modulations of the microenvironment in cervical cancers (e.g. modulation of thermo-, radio- and chemotherapy) also remain rather speculative in the clinical setting.

Oxygenation status of cervical cancers upon hyperthermia

The abnormal microcirculation, a characteristic feature of locally advanced solid tumours, obligatorily leads to the development of hypoxic subvolumes, which are heterogeneously (both spatially and temporally) distributed within the tumour mass. The pathogenesis of tumour hypoxia is multifactorial [2], [15]. Tumour hypoxia can drive malignant progression or can lead to acquired treatment resistance, and thus can act as an adverse prognostic factor [15–20].

Oxygen tension (pO₂) values measured in cervical cancers in the 1960s cannot be considered correct on an absolute scale [21–24]. Artefacts due to large tip electrodes, measurements in the very superficial tumour layers, problems with the calibration of the sensor [22] and the use of bare electrodes without membrane covers are obvious in the earlier studies. Assessment of the oxygenation status in cancers of the uterine cervix using computerised pO₂ histography started in the late 1980s and first reliable data were published in 1990 and 1991 [25], [26].

Our current knowledge concerning the oxygenation status of locally advanced cancers of the uterine cervix mostly refers to pre-therapeutic data. Mean and median pO₂ values were, on average, distinctly lower than in the normal cervix (see Tables II and III). Approximately 60% of locally advanced cancers investigated contained hypoxic tissue areas, independent of various patient demographics, clinical tumour size, stage, histological type, grade, and lymph node status. There was no characteristic topological distribution of pO₂ values within the tumour mass (e.g. tumour centre versus tumour periphery). Tumour oxygenation was weakly dependent on the pathological tumour stage (pT stage) only [26]. Transgression of cervical carcinomas into the urinary bladder muscle and mucosa significantly increased the oxygenation status of grade IV-A tumours [31].

Table II.  Oxygenation status of the uterine cervix.

Median pO2 (mmHg) [range]	n	HF 2.5 (%)	HF 5 (%)	HF 10 (%)	References
40[4–79]^a					[21], [22]
42[1–100]	7 ^b	8	13	20	[26]

^aEstimated from measurements in superficial tissue regions.

^b432 pO₂ measurements in the normal cervix of nullipara.

n, number of patients.

Table III.  Pre-therapeutic oxygenation status of primary, locally advanced cancer of the uterine cervix and prognostic significance of tumour hypoxia (n = number of patients; updated from Vaupel et al. [27]).

						Prognostic significance of tumour hypoxia
Centre	n	Median pO2 (mmHg) (range)	HF 2.5 (%)	HF 5 (%)	HF 10 (%)	Endpoint	Oxygenation parameter
Mainz/Leipzig	150	9 (2–34)	23	37	51	DFS, OS	pO2 < 10 mmHg
Toronto	156 ^a	5 (0–94)		50		DFS, PFS, DS^b	pO2 < 5 mmHg
Halle	87	17 (0–81)	16	26
Leipzig	86	5 (1–57)
Vienna	51	10 (0–60)	22	28	50	DFS, (LC)	pO2 < 10 mmHg
Oslo	49	4 (1–25)	47	64	76	DFS, OS, LC, DS	HSV^c
							pO2 < 5 mmHg
							pO2 < 10 mmHg
Manchester	43	3 (0–42)	39	58	75
	30	6 (0–41)
Paris	37	12	15	17	21
Aarhus	24	3 (0–19)	49	61	73
Vancouver	19	3 (0–34)	38	66	76
Vienna	10	17 (3–54)		42
Durham	9	5		61
Overall	751	9	28	44	59

^aData of Lim et al. [28] added.

^bIn node-negative patients.

^cHypoxic subvolume calculated as the product of the total tumour volume and the relative frequency of hypoxic pO₂ readings <5 mmHg.

There may be some overlaps in patients reported from the same institution. Empty boxes indicate lack of suitable information. Data of Nakamo et al. [29] and Sakata et al. [30] are not included (measurements in tumour periphery only). DFS, disease-free survival; OS, overall survival; PFS, progression-free survival; DS, distant spread; LC, local control.

Table IV.  Interstitial fluid pressure (IFP) in normal tissues and in cervical cancer.

Type of tissue	IFP (mmHg) (range)	References
Subcutis	−3	[43]
Fibrous tissue	−3	[44]
Submucosa (paravaginal)	1	[45]
Cervical cancers	16 (10 to 26)	[46]
	19 (−6 to 76)	[45]
	19 (3 to 48)	[47]
	15 (6 to 40)	[9]
	19 (−3 to 48)	[48]
	29 (4 to 94)	[49]

The oxygenation in cervical cancers was found to be extremely heterogeneous, was poorer in moderately/severely anaemic patients, and was found to be of prognostic significance (see Table III and [28–30]). The overall median pO₂ value was 9 mmHg in 751 patients (data reported from 12 oncology centres). Our own data on squamous cell carcinomas of the uterine cervix (13,596 pO₂ readings in 150 patients) are shown in Figure 2.

Figure 2. Frequency distribution (histogram) of oxygen partial pressures (pO₂) in squamous cell carcinomas (SCC) of the uterine cervix. N = number of patients, n = number of pO₂ values measured (modified from [64]).

Local recurrences of cervical cancers have a higher hypoxic fraction than the respective primary tumours [32]. An analysis of inter-group differences in the oxygenation status indicated that the greater the extent of hypoxia in the primary tumour, the higher the probability of local recurrence of cervical cancers. Hypoxia in cervical cancers has also been assessed by imaging techniques, which can provide clinically relevant information on tumour oxygenation that is predictive/prognostic of patient outcome [33–39].

Technical limitations of the direct O₂ sensing technique using invasive O₂ microsensors have prompted the use of surrogate markers for hypoxia in cervical cancers (and other tumour entities) such as hypoxia-related endogenous proteins (e.g. HIF-1α, the members of the HIF-cascade GLUT-1, CA IX, VEGF, EPO, and hypoxia-induced proteins independent of HIF (for a review see Vaupel and Mayer [18]). In many studies, endogenous markers showed prognostic significance for patient outcome. The prognostic relevance of exogenous markers, however, appears to be somewhat limited. A correlation with prognosis has never been demonstrated for cervical cancer [40]. When binding of the exogenous marker pimonidazole was correlated with endogenous hypoxia-related proteins in cervical cancers, co-localisation was greatest (but not optimal) for pimonidazole and CA IX [41]. In another study, a set of genes related to thermoradiosensitivity (TRT) of cervical carcinomas was identified by using gene-expression profiling. Evaluation of the data showed that HIF-1α and CA IX were up-regulated in the group in which cancer was the cause of death as compared to the ‘no-evidence of disease’ group, thus demonstrating the power to predict the outcome of TRT [42].

Since tumour hypoxia has been considered to be an indirect sensitiser for HT (most probably acting via a depletion of ATP), changes in tumour oxygenation have been evaluated upon tumour heating (for a recent review see Vaupel and Kelleher [2]). When the oxygen status is assessed in solid tumours at elevated temperatures in the preclinical setting, there is evidence that mild HT can improve tissue oxygenation, which parallels similar increases in blood flow (although the oxygen consumption rate increases at temperatures up to 42°C). At higher thermal doses, tissue oxygenation deteriorates, reflecting a breakdown of tumour microcirculation under these conditions [2]. There is some controversy concerning the duration of the improvement in tissue oxygenation upon mild HT, with a series of studies reporting a relatively rapid return to baseline conditions after the cessation of heating, i.e. the effects observed are rather short-lived. In contrast to these data, other studies have shown that the improvement in tumour oxygenation can actually last for 24–48 h following heating using low-dose HT [2]. These latter data may imply an enhanced response of tumours to radiotherapy and O₂-dependent chemotherapy/radiochemotherapy, which endure beyond the heating period.

There is however a clinical observation that seems to contradict the finding of a prolonged therapeutically relevant improvement in tumour oxygenation at mild HT temperatures: Significant changes in perfusion were not observed in cervical cancers upon regional HT (RHT), and minimal flow increases returned to initial values 20 min after heating at 40.7°C [11]. However, based on their PET-derived data, the latter authors speculate that an increase in tumour oxygenation induced by RHT lasts for more than 1 h after heating, provided HT is applied for at least 60 min. It is suggested that this might induce a subsequent increase in the efficacy of O₂-dependent therapeutic measures. To explain this obvious ‘uncoupling’ of changes in flow and tumour oxygenation, further investigations will be necessary if clearer insights into the relationship between thermal parameters, tumour perfusion and oxygenation upon HT are to be obtained. In this context, besides blood flow, the oxygen transport to the tissue may also be facilitated by a right-shift of the oxygen dissociation curve and an increase in the oxygen transport capacity due to a higher intravascular haematocrit.

Interstitial hypertension in cervical cancers

Microvessel leakiness, lack of functional lymphatics, interstitial fibrosis, contraction of the interstitial space mediated by stromal fibroblasts, and high oncotic pressures within the interstitium all contribute to the development of interstitial hypertension (i.e. increased interstitial fluid pressure) in solid tumours, cancers of the uterine cervix included (see Table IV). Whereas in normal tissues, which are not tightly encased (e.g. brain, kidneys), the interstitial fluid pressure (IFP) is slightly sub-atmospheric (‘negative’) or just above atmospheric values, an interstitial hypertension with values of up to 94 mmHg has been reported for cervical cancers, which forms a ‘physiological’ barrier to the delivery of diagnostic and therapeutic macromolecules [50]. In cervical cancer, IFP predicts disease recurrence and survival of patients following radiotherapy, independent of clinical prognostic factors and tumour hypoxia [47–49].

In an animal tumour model, Leunig et al. [50] reported a reduction in the IFP upon 43°C hyperthermia for 30 or 60 min, which resulted in an enhanced growth delay and partial response in 66% of tumours. In contrast, Hauck et al. [51] failed to show any effect of local HT on IFP in a glioma xenograft model. For the clinical situation, data on the impact of HT on IFP in cervical cancers are not available so far. Whether IFP is affected by local/locoregional HT in patients with cervical cancer – with a subsequent impact on therapeutic efficacy – is thus not clear.

Tumour pH upon heat treatment

Reduced pH values (acidosis) can sensitise tumour cells to single heat treatments and to fractionated HT [52]. Based on this information, a series of measurements has been performed on the effect of localised heating on intra- and extracellular pH values ([52–55], for a review see [2]). Summarising the relevant experimental data, it can be stated that tumour temperatures above 42.5°C and appropriate heating times (>30 min) can reduce both intra- and extracellular pH, which in turn may further sensitise tumour cells to HT in the sense of a positive feedback mechanism. Relevant pathogenetic mechanisms responsible for the intensified acidosis upon heat treatment (which is reversible after HT exposure) have been reviewed recently [2].

The significance of the baseline pH_e and pH_i values and changes occurring during HT in clinical studies still remains unclear, since – in contrast to in vitro and in vivo data derived from experimental tumours treated with HT alone – the results are equivocal so far. Contradictory results have been found in the clinical setting concerning the direction of heat-induced changes in pH (increase versus decrease), the prognostic power of pH_i/pH_e before treatment using HT alone or TRT, etc. (for a recent review see [2]. In the study of Lora-Michiels et al. [56], the pH_e seen in canine soft tissue sarcomas predicted for metastasis-free and overall survival. Further clarification of these effects is urgently needed, especially for cervical cancer since analogous data are not available so far.

Tumour glucose and lactate levels upon hyperthermia

Cancer cells intensively split glucose to lactic acid (lactate) and the increased capacity for glycolysis still remains one of the key features of tumour metabolism. Due to the high glucose consumption rate of cancer cells, median glucose concentrations in cancers of the uterine cervix are relatively low (1–2 mM). Glucose levels assessed in cryobiopsies of cervical cancers using single photon imaging and quantitative bioluminescence showed substantial intra-tumour heterogeneity with values ranging from 0.4 to 3.4 mM (see Figure 3). There was a positive linear correlation between the average glucose concentration in tumours and in blood [57]. Intra-tumoural FDG metabolic heterogeneity of cervical cancer has been described by Kidd and Grigsby [58] and was found to predict the risk of lymph node involvement at diagnosis, high risk of pelvic recurrence, poor response to therapy and progression-free survival.

Figure 3. Histogram of measured glucose concentrations in a cancer of the cervix (modified from [64]).

In animal tumour models HT caused no significant changes either in the microregional glucose distribution or in the mean and median glucose concentrations as compared to normothermic conditions [59]. Data on glucose distributions in the clinical setting (e.g. in cervical cancers) upon HT are not available to date.

Increased glucose uptake as described above, and lactate accumulation are characteristic features of cancers. Lactate accumulation is the net result of lactate production from glycolysis and glutaminolysis, cellular lactate export through MCT-4 transporters, cellular lactate import through MCT-1 transporters in the cell membrane of normoxic cancer cells with subsequent utilisation for oxidative phosphorylation instead of glucose [60], lactate diffusion from the interstitial space into nearby tumour microvessels, and (partially impaired) drainage with the tumour venous blood. The consequence of the intratumour lactate shuttling is a sparing of glucose by normoxic cells in the vicinity of blood vessels. Lactate may thus be the preferred substrate of normoxic cells and glucose the prominent energy substrate in hypoxic tumour cells. High tumour lactate levels can predict for distant metastasis, tumour recurrence and overall survival of patients with cervical cancer [61], [62]. In addition, radioresistance has been correlated with high lactate concentrations. Accumulation of lactate in cervical cancers is thus closely linked to the aggressiveness of this tumour entity [63]. Pre-therapeutic lactate concentrations in cervical cancer range from 3 to 40 mM with a median concentration of 14 mM [64] (see Figure 4). In patients without distant metastasis, lactate levels rarely exceeded 20 mM (‘low-lactate’ tumours), whereas in tumours with metastatic spread, concentrations often reached 40 mM (‘high-lactate’ tumours) [61].

Figure 4. Histogram of measured lactate concentrations in a cancer of the cervix (modified from [64]).

In experimental tumours lactate concentrations significantly increased upon localised HT (43.3°C, 60 min) [59], with a concomitant drop in the extra- and intracellular pH (for a review see [2]. Data on lactate accumulation in cervical cancers upon HT treatment have not been published so far, although the measurement of this parameter is technically feasible.

Bioenergetic status of tumours upon hyperthermia

When assessing the pre-therapeutic bioenergetic status of solid tumours, consistently found changes are: decreased ATP (see Figure 5) and PCr concentrations, increased inorganic phosphate (P_i), ADP, AMP and (extracellular) adenosine concentrations and lower PCr/P_i and ATP/P_i ratios as compared to their tissues of origin. For example, ATP content of soft tissue sarcomas was on average 0.5 mM compared to approximately 5 mM in normal skeletal muscle [4]. This situation has been repeatedly described for tumours with median pO₂ values <10 mmHg, mean glucose concentrations <2 mM and mean lactate levels in the tumour tissue >10 mM [65], [66].

Figure 5. Histogram of measured ATP concentrations in a cancer of the cervix (modified from [64]).

Cellular ATP depletion (most probably caused by substrate deprivation) is thought to sensitise tumour cells to hyperthermia [67–70]. Upon heating at temperatures >42.5°C and heating times >30 min, a further drop in the ATP and PCr concentrations has consistently been reported [54], [71], whereas the levels of P_i, ADP, AMP and adenosine increased further under HT conditions. HT-induced decrease in ATP was consistent with increased thermal cytotoxicity, and may be a measure of thermal treatment efficacy [72]. Data for cervical cancer reflecting the degree of direct cytotoxicity elicited from thermal therapy are lacking so far. ATP/P_i and PCr/P_i ratios substantially dropped at high thermal doses which would probably result in a sensitisation of tumours to heat treatment. At 41°C for 15 min, the ATP/P_i ratio only fell by approximately 30% for about 6.5 h [73]. Causes of the ATP decline observed upon HT have been discussed recently [2].

In many human tumours (other than brain tumours), ³¹P-NMR has often shown high concentrations of phosphomonoesters (PME) and phosphodiesters (PDE). The PME signal primarily includes membrane phospholipid precursors. The PDE peak was identified to be largely a result of membrane phospholipid decomposition products. Upon hyperthermia treatment (>42.5°C, ≥30 min) the PME/ATP ratio significantly increased, whereas the PME/P_i and PDE/P_i ratios decreased following heating [53].

It is of interest that PME/P_i and PDE/P_i ratios have been described as being significant predictive factors for treatment outcome upon thermoradiotherapy in addition to the oxygenation status (pathological cure rate), and the PCr/PDE, as well as the PME/PDE ratios (overall survival) in patients with high-grade soft tissue sarcomas. Again, data on cervical cancer describing changes in these ratios upon HT are unfortunately not available [74].

Conclusions

Cellular thermal sensitivity and heatability have repeatedly been shown to greatly depend on the efficacy of tumour blood flow, on critical parameters defining the metabolic microenvironment such as hypoxia, acidosis, substrate deprivation, lactate accumulation and cellular energy (ATP) depletion, and also on the development of interstitial hypertension. These factors ultimately (alone or in a cooperative manner) can modulate the thermosensitivity of cancer cells, at least in experimental in vitro and in vivo settings.

Some very few results obtained from patients with cancers of the uterine cervix treated with thermo-radiotherapy or thermo-radio-chemotherapy suggest that therapeutically relevant changes in blood flow and in the microenvironmental factors may not be as distinct as observed in animal tumour models (as far as extent, duration, and direction are concerned). In this context it should be remembered that thermal doses often differ between animal and human studies and that this should be considered when comparing the relatively small changes in blood flow and microenvironmental parameters that have so far been observed in the clinic as compared to pre-clinical, experimental models. In order to substantiate heat-induced effects on blood flow and on the flow-dependent hostile microenvironment of cervical cancers, clinical studies are urgently needed, since current data are scarce, sometimes biased by methodological problems and/or rather speculative. The observed heterogeneity between cervical cancers and within individual tumours might lead to a non-uniform response to heat treatment when used as an adjuvant to established treatment modalities which has been proven to be of unequivocal value in cervical cancers and other tumour modalities [75], [76].

Acknowledgements

The valuable editorial assistance of Anne Deutschmann-Fleck in preparing this manuscript is greatly appreciated.

Declaration of interest: The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.