Mild temperature hyperthermia and radiation therapy: Role of tumour vascular thermotolerance and relevant physiological factors

Abstract

Here we review the significance of changes in vascular thermotolerance on tumour physiology and the effects of multiple clinically relevant mild temperature hyperthermia (MTH) treatments on tumour oxygenation and corresponding radiation response. Thus far vascular thermotolerance referred to the observation of significantly greater blood flow response by the tumour to a second hyperthermia exposure than in response to a single thermal dose, even at temperatures that would normally cause vascular damage. New information suggests that although hyperthermia is a powerful modifier of tumour blood flow and oxygenation, sequencing and frequency are central parameters in the success of MTH enhancement of radiation therapy. We hypothesise that heat treatments every 2 to 3 days combined with traditional or accelerated radiation fractionation may be maximally effective in exploiting the improved perfusion and oxygenation induced by typical thermal doses given in the clinic.

• Mild temperature hyperthermia
• vascular thermal tolerance
• radiation

Introduction

The tumour microenvironment and physiology greatly affects the response of tumours to treatments such as hyperthermia [1]. Namely, blood flow is a crucial factor in the hyperthermic treatment of tumours since it controls heat dissipation from tumours during heating and subsequently heat-induced tissue damage [2]. Tumour cells die at an exponential rate when exposed to temperatures above approximately 42°C for 30 min, this has been observed in pre-clinical as well as clinical settings [3], [4]. Additionally, it has been shown that in animals tumour blood flow and oxygen delivery is significantly increased when hyperthermia is applied up to 42°C [5–8]. In human tumours, more interestingly, blood flow is maintained and stimulated in response to temperatures as high as 45°C [9]. This increase in blood flow, as well as the reduction of oxygen demand (due to hyperthermia-induced cell death and metabolic suppression at temperatures greater than 43°C) in the tumour, results in significantly increased tumour tissue oxygenation [6], [8], [21–25]. This might make hyperthermia the best hypoxic radiosensitiser available. It has been proposed that hyperthermia at low thermal doses causes reoxygenation of the tumour without direct thermal cytotoxicity [17]. Here we will address the possibilities and limitations of mild temperature hyperthermia to enhance tumour radiation sensitisation.

Tumour thermotolerance versus vascular thermotolerance

The phenomenon of vascular thermotolerance is a potentially powerful and clinically pertinent by-product of multiple hyperthermia treatments of the tumour [18]. The tentative definition of vascular thermotolerance can be summarised as: the blood flow response of the tumour to a second hyperthermia exposure is significantly greater than in response to a single thermal dose, even at temperatures that would normally cause vascular damage [19], [20]. Namely, the delivery of two heat treatments with an interval of around 12–48 h between them, causes a marked increase in tumour blood flow compared to a decrease in tumour perfusion from control level when the heatings are applied consecutively with no interval [8], [13]. Traditionally, the concept of thermotolerance in vitro has been a phenomenon where one heat treatment induces a temporary increase in resistance of cells against subsequent heatings. The working definition of vascular thermotolerance differs in that the measurable quantity, i.e. blood flow and oxygenation, actually increases above and beyond control level in many cases. Regardless, the general notion has been that individual cellular resistance must be at the root of the increased physiological reaction to a second heating.

In order to intelligently design clinical trials incorporating heat and radiation therapy, it is a necessity to understand the effects of a single or multiple hyperthermia treatments. Our studies have been designed to test whether certain sequences of fractionated hyperthermia and radiation may further increase radiation sensitivity via increased oxygenation due to a ‘tolerant’ state of the tumour vasculature. Previous studies were focused on temperatures generally above 42°–43°C, yet current hyperthermia studies employ lower temperatures in the range of 40°–42°C since the majority of clinical hyperthermia treatments do not exceed these temperatures. Therefore, we have been focusing on 41.5°C multi-treatment studies in line with clinically realistic temperatures and with our previous observations that this temperature maximally increase tumour oxygenation in various murine tumour models [8], [13].

The influence of mild temperature hyperthermia on tumour oxygenation

It has been shown that in both rodent as well as human tumours, varying fractions of clonogenic cells are chronically or acutely hypoxic [21–25]. The chronically hypoxic regions are allegedly caused by limited oxygen diffusion in tumour tissue due to insufficient vascular supply, and intermittent blood flow is held responsible for the acutely hypoxic regions caused by variations in interstitial pressure, clogging of vessels by immune cells or shed tumour cells and/or transient collapse of the lumens of immature tumour vessels [22], [26], [27].

The overall tumour oxygenation level depends on the balance of oxygen supply through blood flow and the oxygen consumption rate by tumour and stromal cells. It appears that the changes in tumour oxygenation generally parallel the changes in tumour blood flow [2], [18–20]. In physiological conditions, blood flow is tightly regulated by vasodilating factors that stimulate endothelial cells to release nitric oxide, which then causes relaxation of the vascular smooth muscle and subsequent increased blood flow. We have previously demonstrated that hyperthermia increased the amount of nitric oxide synthase and overall tumoural nitric oxide production [13], [31]. However, the tumour cell oxygen consumption rate has also been shown to be an important influence on tumour oxygenation levels [32]. Thermal exposure up to approximately 41°–42°C is able to transiently increase oxygen consumption, however above this threshold temperature the oxygen consumption rate declines [14], [33]. Due to these contrasting effects of hyperthermia on the tumour oxygenation, depending on the temperature and duration applied, it is essential to delineate the different effects and processes involved in order to rationally design the most optimal hyperthermia treatment regimens.

A single treatment of hyperthermia can have contrasting effects on tumour oxygenation; whereas mild temperature hyperthermia (i.e. 39°–41°C) improves the oxygenation levels, temperatures raised above 42°–43°C causes deterioration of the tumour oxygenation status [13], [34]. Interestingly, a single mild temperature hyperthermia treatment can induce long-lasting oxygenation improvement in murine tumours, depending on tumour type and duration of MTH [35]. Generally, multiple heatings, at various thermal dose combinations increase blood flow and oxygenation [18–20], [36], [37]. Figure 1 reiterates these findings and shows that in FSaII fibrosarcoma tumours [13], oxygenation remains elevated for 24–48 h post MTH at 41.5°C for 60 min. Figure 1 also illustrates the effect of multiple daily fractionated MTH on FSaII tumour oxygenation. After one MTH treatment, the median tumour oxygen tension increased from baseline 6.4 (± 0.5) mmHg to 16.6 (± 1.1) mmHg, which began a downward trend to 14.8 (±1.3) mmHg after two daily heatings, and dropped to 9.2 (± 1.2) mmHg after three daily MTH treatments (Figure 1B).

Figure 1. Tumour oxygenation as function of MTH and radiation. (A) A single application of MTH (41.5°C; 30 or 60 min) in FSaII murine fibrosarcoma and SCK murine breast carcinoma induces a lasting elevation of median oxygenation up to 48 hours [13]. (B) The extent of reoxygenation diminishes as a function of repeated daily MTH (41.5°C; 60 min) in FSaII tumours. Oxygenation was measured immediately after the treatments. (C) Radiation reduces MTH-induced improved tumour oxygenation. Radiation (3 Gy) reduces the MTH-induced improved tumour oxygenation measured 24 h after the treatments (middle bars), as well as when radiation (3 Gy) is combined with 2 daily MTH (41.5°C; 60 min) treatments and measured immediately after the treatment (bars on the right). The pO₂ values of all panels reported are the average of 120 to 700 individual readings derived from 9–15 mice per group per day, with four tracks per mouse and 10 values per track. *p < 0.02, Student's t-test. Tumours were heated by immersing the tumour-bearing legs of anaesthetised mice into preheated water for 60 min, as described previously [54]. The tumour pO₂ was measured with an Eppendorf pO₂ Histograph (Hamburg, Germany). Tumour-bearing animals were laid on a Plexiglas board and the tumour-bearing legs were gently stretched and affixed by taping the foot to the board. A pO₂ electrode (300 µm diameter, Eppendorf) was inserted by hand into the tumours through small incisions made in the skin over the distal side of the tumour. The electrode was then advanced by a computer-controlled system measuring pO₂ along the track: the electrode was advanced by 0.7 mm forward steps, immediately withdrawn by 0.3 mm to reduce the compression pressure and the pO₂ value was recorded [50], [38–43]. Tumours were locally irradiated with a Philips 250 Kv X-ray machine at a dose rate of 1.4 Gy/min. The body was shielded with lead and only the tumour and foot exposed to the X-ray beam, as described earlier [58].

In order to further understand the interaction of clinically relevant concomitant radiation and heating, tumour oxygenation was measured after combined treatment with MTH and radiation (Figure 1C). The median tumour oxygen tension (pO₂) at 24 h after 3 Gy radiation was 5.3 (±1.2) mmHg, slightly less than the average median pO₂ in untreated FSaII tumours which was 6.5 (±0.5) mmHg. Tumours heated with MTH (41.5°C; 60 min) had an average median pO₂ of 10.9 ± 1.3 mmHg 24 h post-heating. When tumours were treated with MTH and immediately exposed to 3 Gy, the median pO₂ 24 h later was only 7 ± 2.6 mmHg. In tumours that were treated with heat and radiation, and then heated again 24 h later immediately before measuring oxygenation, the median pO₂ was found to increase to 10 ± 1.9 mmHg. While this was still an elevated median pO₂ as compared to untreated control tumours, radiation subdued the MTH-induced tumour oxygenation enhancement, since it was significantly less than tumours treated with two treatments of MTH only (14.8 (±1.2) mmHg (p < 0.02)).

While it is uncertain whether everyday hyperthermia is even logistically and practically feasible in the radiation oncology clinic, these studies illustrate the possibility that when combining multiple MTH and radiation treatments an interval between the hyperthermia exposures of greater than 24 h may be optimal to allow the tumour physiological response to improve radiation response.

Tumour growth response after combination of MTH and fractionated radiation

Several tumour model experimental designs were tested to elucidate whether certain sequences of fractionated hyperthermia and radiation may further increase tumour radiation sensitivity. Initially the combination of multiple consecutive MTH and radiation treatments was applied and tumour growth response was monitored in the FSaII tumour mouse model. Whereas seven consecutive daily MTH (41.5°C; 60 min; schedule every day (q1d) for seven days (x7)) treatments inhibited tumour growth to a small extent, radiation (3 Gy for 7 days; q1dx7) caused a tumour growth delay of more than 15 days as compared to untreated control tumours (Figure 2A). The addition of MTH to each exposure of radiation caused a slight increase in tumour growth delay of approximately 2 days more, independent of whether MTH was administered prior or post radiation (Figure 2A).

Figure 2. The effect of MTH on FSaII tumour growth delay by radiation. (A) Daily exposures of radiation (3 Gy) and MTH (41.5°C; 60 min) for 7 consecutive days (schedule q1dx7) increased FSaII tumour growth delay to a modest extent compared to radiation alone. -▪- control; -•- MTH (41.5°C; 60 min.; q1dx7); -▾- radiation (3 Gy; q1dx7); -◂- radiation prior to MTH; -▸- MTH prior to radiation. Data points represent the means (n = 10 animals per group) ± SEM. Arrows indicate treatment schedule. (B) MTH (41.5°C; 60 min) applied every other day prior to daily irradiation (3 GY) increased FSaII tumour growth delay as compared to MTH given post radiation. -▪- control; -•- MTH (41.5°C; 60 min; q2dx4); -◂- radiation (3 Gy; q1dx7) prior to MTH; -▸- MTH prior to radiation. Data points represent the means (n = 9–10 animals per group) ± SEM. (C) MTH applied every other day before accelerated radiation therapy significantly improves tumour response to radiotherapy. -▪- control; -•- MTH (41.5°C; 60 min; q2dx3); -▾- radiation (daily 2 × 2 Gy with 6-h interval); -▸- MTH prior to radiation. Data points represent the means (n = 8–10 animals per group) ±SEM. Arrows alone indicate radiation schedule; arrows with a box-shaped addition indicate MTH and radiation treatment. The FSaII tumours were grown and handled as described earlier [13], with the slight modification that the mice were randomised and treatments were initiated when tumours reached at least an average of 200 mm³ in size. Tumour volume was measured with a caliper (Scienceware) and calculated according the equation: (a²× b) /2, where a is the width and b the length of the tumour [59]. Tumours were heated by immersing the tumour-bearing legs of anaesthetised mice into preheated water for 60 min, as described previously [54]. Tumours were locally irradiated with a Philips 250 Kv X-ray machine at a dose rate of 1.4 Gy/min. The body was shielded with lead and only the tumour and foot exposed to the X-ray beam, as described earlier [58].

In the next set of experiments intermittent MTH was tested as we hypothesised that fewer hyperthermia treatments may be more effective in reoxygenation, and thus increase the radiation response, of the tumour. MTH (41.5°C; 60 min) was applied every other day (q2dx4) in combination with multiple radiation exposures (3 Gy for 7 days; q1dx7). There was a noticeable increase in the tumour growth delay (∼5 days) when tumours were treated with MTH prior to radiation as compared to those treated post radiation (Figure 2B). This suggested that, at least qualitatively, MTH before every other radiation fraction improved radiation response to a greater degree than MTH after every other radiation fraction due in some part to improved tumour oxygenation. However, the application of heat after radiation every other day may have actually reduced the overall radiation response (as compared to the radiation alone treatment group in Figure 2A). This could be due to a low level induction of thermal tolerance in the tumour, which might have also affected the intrinsic radiation response and cancelled out the effects of any long-term improvement in tumour oxygenation upon subsequent radiation exposures.

We subsequently expanded these findings in FSaII tumours where intermittent MTH (41.5°C; 60 min; q2dx3) was combined with daily accelerated radiation (2 Gy, 6 hrs, 2 Gy) for 5 days. Interestingly, intermittent MTH given prior to the accelerated radiation was able to enhance the radiation-induced tumour growth delay (Figure 2C). The enhancement of radiation therapy response was more pronounced than when MTH was given on consecutive days (Figure 2A). Thus, a combination of multiple consecutive MTH treatments and fractionated radiation caused tumour growth delay independent of sequence while a combination of intermittent MTH and fractionated radiation caused differential tumour growth delay dependent on sequence.

One could speculate that heating every day at 41.5°C induces sufficient accumulated vascular damage or related cellular effects over time that the enhancement in radiation response due to the improved oxygenation begins to decrease as therapy continues. Another variable that must be considered is the fact that some degree of oedema is induced in tumours heated with water baths. It is conceivable that repeated oedema and some degree of vascular congestion occurring every day due to repeated heat treatments could be an unintended source of vascular inefficiency and may be a factor in our results. However, our recent results (data not shown) indicate that repeated mild hyperthermia treatments cause a normalisation of the murine tumour vasculature (i.e. destruction of immature and retention of mature vessels) leading to improved vascular efficiency, albeit a transient effect. Certainly, there is a great need for clinical measurements of tumour perfusion and oxygenation during the course of a thermoradiotherapy regimen involving multiple heat treatments from a non-water bath source to further develop our understanding of the tumour physiology involved. In addition, one might envision the potential inverted explanation of the concomitant radiation damage affecting hyperthermia performance. We found that the addition of 3 Gy irradiation in combination with MTH treatment suppressed the improvement in tumour oxygenation obtained by MTH alone (Figure 1C). This might be expected, since the radiation damage of both tumour parenchyma and stromal cells may reduce or inhibit normal cellular signalling that would govern the physiological reactions to hyperthermia or other treatments. While we have not investigated this exhaustively, the data presented here is a strong reminder that, clinically, tumour growth or regression is dependent on multiple factors. Therefore within a multi-modality regimen it is necessary to elucidate potential negating effects that particular treatments might inflict upon other treatment modalities. This is of particular importance since the majority of cancer patients are being treated with multi-modality regimens before, during or after thermoradiotherapy.

Regardless of the role of tumour oxygenation on the radiation response, it is also conceivable that a thermotolerant state of the tumour cells themselves could alter intrinsic radiosensitivity over the course of treatment. If true, the overall treatment response may be diminished as we observed in our study where hyperthermia and radiation were applied every day. However, most studies to date have revealed little to no direct effect of thermotolerance on the radiation responsiveness of cells, yet some reduction in thermal radiosensitisation can occur [38–43]. Taken together, it appears that the induction of cellular thermotolerance has a minimal effect on the results of combined radiation and hyperthermia treatment. However, depending on the sequence of treatments it has the potential to be a negating factor on the treatment response (Figure 2).

MTH considerations in multi-modality treatment

Since MTH improved tumour oxygenation as well as blood flow (Figure 1) [13], the delivery of various therapeutics as well as diagnostic agents could be significantly improved by proper MTH scheduling. For example, the combination of mild heating with carbogen breathing and/or nicotinamide (two agents used to improve experimental and human tumour oxygenation [23], [31], [44–49]) is markedly more effective in increasing tumour oxygenation and radiosensitivity than either treatment alone [31], [35], [50], [51].

Of course, the immediate and long-term effects of MTH alone on tumour perfusion and oxygenation may be part of the explanation for the positive outcomes of previous clinical trials that have combined hyperthermia and radiation [52]. In some trials, heating was applied several times per week and therefore it is conceivable that these patients’ tumours acquired vascular thermotolerance, which may have maximised oxygenation and blood flow status to allow radiation and chemotherapy to exert greater cell killing in the tumour. Unfortunately, there have been little to no formal studies to date exploring the mechanisms or translational significance of this clinically relevant topic. One study assessed the changes in nanoparticle distribution after one or two hyperthermia treatments [53]. Intriguingly, this study reported that pre-heating the tumour 6 to 8 hours in advance reduced the permeability and extravasation of the particles, presumably due to the development of some type of thermotolerance to the second heating [53]. Clearly the mechanisms by which repeated hyperthermia treatments may affect chemotherapy or drug delivery are far from understood and warrant further investigation. Elucidating the mechanisms and consequences of repeated hyperthermia holds great promise for improving clinical outcomes.

To conclude, MTH-induced reoxygenation of previously hypoxic tumour tissue has been unmistakably shown to have important implications on radiation sensitivity [11], [13], [28], [50]. Therefore, the induction of improved oxygenation via a single heating or via vascular thermotolerance induced by multiple heatings in human tumours may have marked effects on clinical outcomes. Our preclinical evidence suggests a threshold above which additional hyperthermia exposures may have little beneficial effect on tumour radiation response. Nonetheless, a very strong rationale for the combination of thermal therapy and cancer treatments such as hypofractionated high dose radiation therapy are supported by numerous pre-clinical studies. The potential to take advantage of substantial increases in radiation-induced cell killing in human tumours caused by thermal therapy is high, provided that we continue to characterise the tumour biology and mechanisms related to combined therapy strategies.

The data collected to date and our new observations using multiple heat and radiation exposures support the idea that two mild temperature hyperthermia (i.e. 41.5°C; 60 min) treatments may greatly improve tumour perfusion and oxygenation, while three or more hyperthermia exposures in a row may have a lessened influence. Moreover, that frequency and sequence are central parameters in the success of MTH enhancement of radiation therapy. These discoveries enhance our understanding of tumour thermotolerance in particular, and might challenge current multi-modality treatment dogma in general.

Acknowledgements

This work was supported by NCI CA44114 and Central Arkansas Radiation Therapy Institute (CARTI). We thank Nathan Koonce, MS for assistance in the figures and data analysis.

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