Hyperthermia, cisplatin and radiation trimodality treatment: A promising cancer treatment? A review from preclinical studies to clinical application

Abstract

This review discusses available clinical and experimental data and the underlying mechanisms involved in trimodality treatment consisting of hyperthermia, cisplatin and radiotherapy. The results of phase I/II clinical trials show that trimodality treatment is effective and feasible in various cancer types and sites with tolerable toxicity. Based on these results, phase III trials have been launched to investigate whether significant differences in treatment outcome exist between trimodality and standard treatment. In view of the clinical interest, it is surprising to find so few preclinical studies on trimodality treatment. Although little information is available on the doses of the modalities and the treatment sequence resulting in the largest degree of synergistic interaction, the results from in vivo and in vitro preclinical studies support the use of trimodality treatment for cancer patients. Animal studies show an improvement in treatment outcome after trimodality treatment compared with mono- and bimodality treatment. Studies in different human tumour cell lines show that a synergistic interaction can be obtained between hyperthermia, cisplatin and radiation and that this interaction is more likely to occur in cell lines which are more sensitive to cisplatin.

• Cisplatin
• hyperthermia
• radiotherapy
• trimodality treatment

Introduction

The beneficial effects of hyperthermia when added to radiotherapy have been proven by several randomized clinical trials [1–11]. Combined hyperthermia and radiotherapy regimens are regarded as safe for patients with minor acute side-effects [12]. Hyperthermia can also improve the effectiveness of several chemotherapeutic agents. A widely used agent often combined with hyperthermia treatment is cisplatin. Both in vitro and in vivo experiments have shown that its effectiveness is enhanced by the addition of concomitant hyperthermia [13–20]. In the clinic, combined cisplatin and hyperthermia treatment is successfully used for treatment of several cancer types [21–24]. Combined hyperthermia and cisplatin treatment is well tolerated by patients [24]. Good results have also been obtained combining cisplatin and radiotherapy particularly in cervical cancer patients [25–28]. In vitro, cisplatin has been shown to sensitize tumour cells to ionizing radiation [29–34]. Therefore, combining hyperthermia, cisplatin and radiation may result in even better treatment responses through interactions between the three modalities.

The aim of this review is to discuss available clinical and experimental data and the underlying mechanisms involved in trimodality treatment consisting of hyperthermia, cisplatin and radiotherapy.

Patient studies

Regimens using cisplatin as a part of trimodality treatment have been tested in phase I/II trials in several types of superficial tumours [35], breast carcinomas [36] head and neck and oesophageal carcinomas [37–44] and cervical carcinomas [45–47]. An overview of these studies is shown in Table I. This table shows treatment outcome as represented by the percentage of complete responses (CR) obtained after different trimodality treatment schedules.

Table I.  Phase II clinical studies for trimodality treatment consisting of hyperthermia, cisplatin and radiotherapy.

Tumor location	Total patients	Hyperthermia	Cisplatin	Radiation	Sequence	%CR	Reference
Superficialmalignancies	24	6 × (once per wk) for 60 min	6 × 20, 30 or 40 mg/m^2	18–24 × 2 Gy for fulldose tolerance	HT + cDDP → X	50	[35]
Breast	36	? × (once per wk) for 60 min	? × 40 mg/m^2	? × 4 Gy	HT + cDDP → X	53 for all treatments	[36]
Head and neck	18	2 × (once per wk) for 30–60 min	7 × 20 mg/m^2	35 × 2 Gy	cDDP → X → HT	72	[37]
Head and neck	36	>2 × (twice per wk) for 45 min	3–5 × 20 mg/m^2	15–39.9 Gy (implant)or 25.2–64 Gy(external beam)	cDDP → HT → X	93	[38]
Head and neckand sarcoma	26	4 × (once per wk) for 60 min	4 × 40 mg/m^2	15–35 × 2 Gy	X → HT + cDDP	19	[39]
Head and neck	21	3–8 × (twice per wk) for 60 min	2–6 × 30 mg/m^2	30–35 × 2 Gy	cDDP → X → HT	40	[40]
Head and neck	53	6 × (twice per wk) for 60 min	6 × 30 mg/m^2	35 × 2 Gy	cDDP → X → HT	82	[41]
Oesophagus	294	6 × (twice per wk) for 30 min	3 × 40 mg/m^2	15 × 2 Gy	X → HT + cDDP	35	[42]
Oesophagus	317	6 × (twice per wk) for 30 min	3 × 40 mg/m^2	15 × 2 Gy	X → HT + cDDP	30	[43]
Oesophagus	322	6 × (timing and duration unknown)	3 × 40 mg/m^2	Unknown	Unknown	35	[44]
Cervix	12	5 × (once per wk) for 60 min	5 × 40 mg/m^2	25–28 × 1.8 Gy	X → HT + cDDP	92	[45]
Cervix	68	5 × (once per wk) for 60min	5 × 40 mg/m^2	5 × 1.8–2.0 Gy+brachytherapyboost	HT + cDDP → X orX → HT + cDDP	90	[46]
Cervix	32	5 × (once per wk) for 60 min	5 × 40 mg/m^2	25 × 1.8 Gy	X → HT + cDDP	–	[47]

Note: ?= Unknown.

The first clinical trial was conducted by Herman et al. [35]. In this phase I/II trial the addition of systemic cisplatin to local hyperthermia and radiation was tested to determine the tolerable cisplatin dose and to estimate the therapeutic potential of this treatment in patients with locally advanced superficial malignancies. Tumour types included in this trial were breast adenocarcinoma, head and neck squamous cell carcinoma, non-Hodgkins lymphoma, soft tissue sarcomas and colon and lung cancer. Hyperthermia and cisplatin were given concurrently before irradiation. The treatment was feasible and effective and a cisplatin dose of 30 mg/m² injected intravenously weekly for 6 weeks appeared tolerable in patients heavily pre-treated with chemotherapy [35]. In patients who had had little or no prior chemotherapy, cisplatin at 50 mg/m² weekly was tolerable. It is notable, however, that pelvic radiation was not a component of therapy for any of these cases, and may have lowered the tolerable dose of cisplatin, since bone marrow toxicity was dose limiting (Herman, 2007, personal communication). Patients of this study were also included in a retrospective study of 29 patients with local-regional recurrence or advanced carcinoma of the breast [36]. In these patients hyperthermia and radiotherapy were combined with cisplatin alone or cisplatin with etanidazole or bleomycin. This resulted in an overall CR rate of 53% for all treatments. However significant more and severe complications were seen in patients who had a history of previous radiotherapy.

In patients with advanced head and neck/oesophageal carcinoma several phase I/II trials were conducted [37], [38], [40–44]. Various treatment schemes were investigated and all studies show that trimodality treatment with hyperthermia, cisplatin and radiotherapy is feasible in patients with acceptable toxicity. The highest CR rate for head and neck cancer patients was 93% after trimodality treatment compared with 86% in the no-chemotherapy group [38]. However, larger trials and longer follow-up are necessary to prove statistical significance. This was the only study in head and neck carcinomas in which concomitant hyperthermia and low-dose (20 mg/m²) cisplatin treatment preceded radiotherapy. In oesophageal cancer patients the CR was between 30% and 35% in all studies. In two of these studies, concomitant hyperthermia and cisplatin followed radiotherapy; the third study did not report the implemented treatment schedule.

Of all cancer sites, cervical carcinoma is an obvious choice for trimodality treatment because both cisplatin-based chemoradiation and thermoradiotherapy were reported to obtain large improvements in treatment response and overall survival compared with standard radiotherapy [5], [25–28], [48]. The feasibility of trimodality treatment in cervical carcinoma was tested in two phase II studies. In a phase II trial published by Jones et al. [45] concurrent radiotherapy, cisplatin and hyperthermia was tested in patients with locally advanced and locally recurrent cervical cancer. Treatment consisted of weekly cisplatin and hyperthermia after irradiation. The therapy resulted in an excellent clinical response (initial CR of 92% and local control rate of 83%) and was well tolerated in the 12 patients studied. Another publication analysed the data of three independent but similar trials in the USA, Norway and the Netherlands including the patients of the study by Jones et al. [46]. Sixty-eight patients with locally advanced cervical carcinoma received trimodality treatment. The patients received weekly cisplatin alone or concomitant with hyperthermia. Hyperthermia was applied before or after irradiation. The trimodality treatment was feasible and effective. Based on these results, a phase III study comparing the treatment with standard chemoradiation has been launched. Another trimodality study in cervical carcinoma was conducted by Sreenivasa et al. [47] In this study the treatment resulted in high response rates, enabling curative surgery in a high proportion of patients with FIGO IIB-IVA non-resectable cancer.

Trimodality treatment using a chemotherapy regimen consisting of weekly oxaliplatin concomitantly with hyperthermia and folinic acid and 5-fluorouracil was conducted in pre-irradiated patients with locally recurrent rectal cancer [49]. Treatment was performed for 6 consecutive weeks. Radiation was added to chemotherapy and hyperthermia in a subset of patients to elongate the remission interval. In this pilot study it was shown that this treatment was feasible with moderate overall toxicity. Oxaliplatin was used as the chemotherapeutic modality in this study because this is mostly regarded as the standard chemotherapy in rectal cancer patients.

According to the results of phase I/II clinical trials trimodality treatment is feasible and effective in several cancer types with only modest toxicity. Different sequencing could have an effect on the interaction of the three modalities and might therefore be important in determining treatment response. Because several in vitro studies showed that simultaneous hyperthermia and cisplatin result in increased cytotoxicity compared with consecutive treatment, most trials applied hyperthermia and cisplatin concomitantly before or after radiotherapy. Despite few pre-clinical studies, however, no conclusive evidence exists on the optimal regimen for trimodality treatment. This lack of evidence about the optimal treatment schedule and the effectiveness of trimodality treatment vs. standard treatment stressed the need for more phase III trials.

Animal studies

Only four papers were found on in vivo experiments concerning cisplatin-containing trimodality treatment; these are shown in Table II. The first in vivo study was conducted by Herman et al. [50]. In this study the effect of the sequence of the different modalities on tumour growth delay (TGD) and tumour cell kill was investigated to determine an optimal schedule for implementation in the clinic. All trimodality treatments resulted in a larger TGD compared with one-modality or two-modality treatment. The largest TGD was obtained by treating with cDDP just prior to hyperthermia with radiation after hyperthermia. This treatment scheme also had the largest cytotoxic effect as determined by excision of the tumours followed by plating of tumour cell suspensions for colony-forming assay. This suggests that the interaction of cDDP with hyperthermia prior to irradiation would improve tumour control, although this endpoint was not studied. Trimodality treatment using cisplatin/hyperthermia/radiation was shown to be significantly more effective (TGD of 25 days) than with etoposide/hyperthermia/radiation, which resulted in a TGD of only 14 days [51]. However, this is only a tentative conclusion since these results are from sequential studies; other treatment sequences might lead to different results.

Table II.  Trimodality treatment consisting of hyperthermia, cisplatin and radiotherapy in animals.

System	Hyperthermia	Cisplatin	Radiation	Sequence	Treatment outcome*			Reference
FSaIIC fibrocarcinoma in	Local	10 mg/kg	5 × 3 Gy	cDDP	TGD 8.0 days			[50], [51]
C3H/He mice	43°C, 30 min			HT	TGD 1.4 days
				cDDP → HT	TGD 5.9 days
				X	TGD 6.3 days
				HT → X	TGD 8.4 days
				cDDP → X	TGD 11.7 days
				X → cDDP + HT	TGD 13.9 days
				cDDP → X → HT	TGD 19.3 days
				cDDP + HT → X	TGD 25.2 days
Spontaneous mammary carcinoma from C3H/Tif injected in C3H/Tif/Bom x D2BA/Bom F1 mice	Local43.5°C, 60 min	6 mg/kg	10–65 Gy	X X → 4 h → cDDP	TCD50 54.4 Gy	TCD50 Air	TCD50 Clamped	[52]
			X → 4 h → HT	54.6 Gy	54.4 Gy	64.2 Gy
			X → 15 min → cDDP → 3.75 h → HT	30.3 Gy	54.6 Gy	57.0 Gy
				31.9 Gy	30.3 Gy	47.1 Gy
				32.5 Gy	31.9 Gy	40.4 Gy
FSaIIC fibrocarcinoma in C3H/He mice	Local	10 mg/kg	10 Gy	HT		Bright	Dim	[53]
43°C, 30 min			X		∼ 50%	∼ 15%
				cDDP		∼ 10%	∼ 20%
				HT → X		∼ 15%	∼ 40%
				cDDP → HT		∼ 5%	∼ 2%
				cDDP → X		∼ 0.5%	∼ 1%
				cDDP → HT → X		∼ 1.5%	∼ 7%
						∼ 0.01%	∼ 0.08%
Sarcoma-180 in BALB/c mice	Local	2.5 mg/kg	10 Gy	Untreated	0% CR			[54]
43°C, 30 min and42°C, 60 min				cDDP	19% CR
				X	38% CR
				HT (43°C, 30 min)	25% CR
				HT (42°C, 60 min)	45% CR
				cDDP → X	65% CR
				X → HT (43°C, 30 min)	59% CR
				cDDP → HT(42°C, 60 min)	65% CR
				cDDP → HT (43°C, 30 min)	70% CR
				cDDP → X → HT (42°C, 60 min)	Too toxic
				cDDP → X → HT (43°C, 30 min)	100% CR
SCCHN in nude mice	Local	2 mg/kg	5 × 2 Gy	X	CR 2/15			[55]
	41.0 and 41.8°C,60 min			X → HT (41°C)	CR 5/15
				X → HT (41.8°C)	CR 7/15
				X → cDDP + HT (41°C)	CR 6/15
				X → cDDP + HT (41.8°C)	CR 12/15

*Treatment outcome is determined by using different endpoints: TGD [50], [51]: the days taken by each individual tumour to reach 500 mm³ compared with the untreated controls. TCD₅₀ [52]: the radiation dose, which on average results in tumour control in half of the treated animals, with: TCD₅₀ air: TCD₅₀ in aerobic cells, TCD₅₀ clamped: TCD₅₀ in hypoxic cells. Bright: survival of oxygenated cells, Dim: survival of hypoxic cells [53]. CR, complete response [54]: complete regression with no regrowth during the 120 days of observation. CR, complete remission [55]: the absence of tumour up to day 60. Trimodality treatments and corresponding treatment outcomes are shown in bold.

Overgaard et al. investigated the interaction between cisplatin and hyperthermia when applied after irradiation [52]. To investigate the effect on radiosensitivity the endpoint TCD₅₀ was determined. The study design allowed evaluation of the individual effect of the modalities on the subfractions of hypoxic and oxygenated cells. The effect on the fraction of hypoxic cells was regarded as very important because it is this fraction of cells that is radioresistant and eventually determines treatment outcome. Therefore, especially hypoxic cells may benefit from combined treatment. In order to avoid direct interaction with radiation, they applied hyperthermia 4 h after irradiation either ‘simultaneously’ (hyperthermia 15 min after cisplatin) or ‘sequentially’ (hyperthermia 3.75 h after cisplatin) with cisplatin. The authors reasoned that a direct interaction of hyperthermia with radiation would not allow analysis of the extent of radiation modification. In addition to radiosensitivity, the effects of hyperthermia and cisplatin alone or in combination on clonogenic survival of hypoxic and aerobic cells were studied. While simultaneous hyperthermia and cisplatin resulted in significant additional cytotoxicity of well-oxygenated cells, trimodality treatment did not result in significant enhancement of the effect of combined radiation and hyperthermia. This was due to the lack of interaction between hyperthermia and cisplatin in the fraction of hypoxic cells as determined by clonogenic survival. A study by Herman et al. investigating the addition of 2-nitroimidazole radiosensitizers to cisplatin with radiation and with or without hyperthermia in the murine FSaIIC fibrosarcoma also showed a decrease in cell kill in the hypoxic cell fraction compared to the oxygenated cell fraction [53].

Another study by Rao et al. [54] investigated trimodality treatment in a sarcoma-180 mouse model. When compared with treatment with cisplatin, radiation or hyperthermia alone or a combination of two modalities, trimodality treatment had the largest benefit in overall survival and resulted in CR in 100% of the animals. In addition to trimodality treatment using hyperthermia at 43°C for 30 min, a combination with hyperthermia at 42°C for 60 min was investigated. This led to hyperthermia-induced trauma and some stress-induced death in the animals and is therefore too toxic in this experimental setting. The reason for this unexpected toxic effect was unknown.

Treatment with concomitant cisplatin and hyperthermia directly after irradiation was tested by Ressel et al. in nude mice carrying human-derived head and neck squamous cell carcinoma (SCCHN) xenografts [55]. In this study, treatment outcome with trimodality treatment using cisplatin chemotherapy was compared with ifosfamide as the chemotherapeutic agent. Two different hyperthermia regimens were used: one with a target temperature of 41°C and one with a target temperature of 41.8°C. The different temperatures were chosen because 41°C is often used as a mean temperature in clinically applied locoregional hyperthermia [56–59] and 41.8°C is used in whole body hyperthermia [60–72]. The most effective treatment regimen was trimodality treatment containing cisplatin and local hyperthermia at 41.8°C, which resulted in CR in 80% of the animals. The side effects were mild and not different to those associated with single and double treatment combinations.

From these studies it becomes evident that trimodality treatment has a beneficial effect on treatment outcome compared to bi- or monomodality treatment. The effect of trimodality treatment on radiosensitivity was investigated only by Overgaard et al. [52]. They found no beneficial effect on tumour radiosensitivity after addition of cisplatin to hyperthermia and radiation compared to hyperthermia and radiation alone. However, an additional effect may be observed by hyperthermia increasing cisplatin toxicity. No conclusion exists on the optimal timing of the modalities since all studies used different treatment schedules in various models. In addition, little is known about the toxicity of the treatments in animal experiments.

Studies using cultured tumour cell lines in vitro

Only a few studies exist on the interaction of hyperthermia, cisplatin and radiation in vitro. As stated earlier, an important factor determining the effectiveness of trimodality treatment is sequencing of the different modalities. In vitro studies have shown that simultaneous application of cisplatin and hyperthermia result in the largest cytotoxic effect [14], [20], [73], [74]. When cisplatin is combined with radiation, the addition of cisplatin just before irradiation results in the largest radiosensitization in some studies [75], [76]. However, other studies observed the largest degree of radiosensitization when cisplatin is applied after irradiatio [77], [78]. The effect of sequencing in trimodality treatment has been investigated in ovarian carcinoma cell lines with different cisplatin sensitivities [79], [80]. In these studies, treatment with 1 h 40°C hyperthermia and 1 h low concentration cisplatin combined with low dose rate (LDR) or high dose rate (HDR) irradiation was compared. An overview of the studies is shown in Table III. For HDR irradiation, all treatment sequences resulted in a synergistic interaction between the modalities in both the cisplatin sensitive and the resistant cell line. In the cisplatin sensitive cell line the largest effect on cytotoxicity was observed when cisplatin was applied before irradiation with hyperthermia afterwards. In the cisplatin resistant cell line, by contrast, the largest effect was obtained with concomitant hyperthermia and cisplatin after irradiation. However, in this study the sequence of cisplatin and hyperthermia before irradiation has not been tested. For LDR irradiation, a synergistic interaction between the three modalities was only obtained in the cisplatin sensitive parental cell line. No difference was observed between the different treatment sequences.

Table III.  Trimodality treatment consisting of hyperthermia, cisplatin and radiotherapy in cell lines.

				Cell line
Hyperthermia	Cisplatin	Radiation	Sequence	cDDP sensitive	Synergy?	Enhancement*	cDDP resistant	Synergy?	Enhancement*	Reference
40°C, long duration	1 µg/ml, long duration	LDR	Concomittant	Glioma	Yes	–	Glioma	Small	1.35 (DMF)	[84]
40°C, 1 h	1 and 3 µg/ml, 1 h	LDR	HT + cDDP → X	Ovariancarcinoma	Yes (3 µg/ml)	2.7 (RR)	Ovariancarcinoma	No	–	[79]
			X → HT + cDDP		Yes (3 µg/ml)	2.8 (RR)		No	–
40°C, 1 h	1 and 3 µg/ml, 1 h	HDR	CDDP → X → HT	Ovariancarcinoma	Yes	3.6 (RR)	Ovariancarcinoma	Yes	3 (RR)	[80]
			X → HT + cDDP		Yes (3 μg/ml)	4.6 (RR)		Yes	1.2 (RR)
40°C, long duration	0.5–3 µg/ml, long duration	LDR	Concomittant	Ovariancarcinoma	Yes	–	Ovariancarcinoma	Yes	–	[83]
41°C, 1 h	1–10 µM, 1 h	HDR	HT + cDDP → X	Cervicalcarcinoma	No	–	NSCLC	No	–	[81]
43°C, 1 h	1–10 µM, 1 h		HT + cDDP → X		No	–		No	–
41°C, 1 h	1 µM, continuously		HT + cDDP → X		Yes	2.0 (RER)		No	1.1 (RER)
43°C, 1 h	5 µM, continuously		HT + cDDP → X		Too cytotoxic	–		Yes	2.1 (RER)

*Enhancement is determined by using different ratios:

 DMF (Dose Modifying Factor at 10% survival): The dose required by one treatment to obtain 10% survival divided by the dose required to obtain the same survival level for combined treatments.  RR (Recovery Ratio): the survival after a specific incubation time divided by the survival for zero incubation time.  RER (Radiation Enhancement Ratio at 10% survival): The radiation dose necessary to decrease survival to 10% after radiation alone divided by the radiation dose necessary to decrease survival to  10% after combined treatment.

In our lab we recently studied the ability of hyperthermia at 41°C or at 43°C to increase cisplatin-induced radiosensitization in two cell lines with different sensitivities to cisplatin [81]. In this study, cisplatin and hyperthermia were applied concomitantly before irradiation. The two different temperatures were chosen because of a recent debate on the effectiveness of mild temperature hyperthermia [82]. The clinical goal to reach a tumour temperature of 43°C is often not accomplished and hyperthermia treatment in the range of 39–42°C can have powerful effects as well [82]. Our results showed that 41°C hyperthermia added to cisplatin and radiation only increased radiosensitivity in cisplatin-sensitive cells. As expected, hyperthermia at 43°C was able to increase radiosensitivity in both cell lines.

The role of cellular cisplatin sensitivity in the effect of trimodality treatment containing long duration mild hyperthermia (LDMH) was investigated by Raaphorst et al. [83]. In these studies LDMH of 40°C for 24 h was combined with LDR irradiation during which repair of sublethal damage takes place. This treatment resulted in a supra-additive effect on cell kill in both cisplatin resistant and sensitive ovarian carcinoma cell lines [83]. The largest synergistic effect was seen in the cisplatin resistant cell line, which had a greater capacity to repair sublethal damage. In glioma cells, on the other hand, hyperthermia treatment could not overcome the resistance to cisplatin-induced radiosensitization [84]. In the cisplatin sensitive glioma cell line trimodality treatment had a synergistic effect. Other studies on LDMH combined with radiation show that it can be an effective radiosensitizer in vitro and that the optimal treatment is to heat as long as practically possible [85–93].

The available in vitro data show that trimodality treatment can result in a synergistic interaction between the different modalities depending on the temperature, cisplatin concentration and duration and radiation dose rate. A synergistic interaction is more likely to occur in cell lines that are more sensitive to cisplatin. However, more studies are needed in order to determine which treatment schedule results in the largest degree of synergism.

Discussion

Because hyperthermia, cisplatin and radiation therapies have a variety of molecular targets, a combination of these three modalities might result in a synergistic interaction. Important factors determining the degree of interaction may include tumour temperature, cisplatin dose, radiation dose rate and the sequence of application.

Hyperthermia has been shown to enhance radiation effects by inhibition of DNA repair processes [94–107] or by tumour reoxygenatio [108], [109]. It is well known that an increase in tumour oxygenation results in an increased radiosensitivity of the tumour [108]. In addition, hyperthermia induces direct cell kill, specifically in insufficiently perfused parts of the tumour [110–115]. Hyperthermia has been shown significantly to enhance drug effectiveness. For cisplatin, enhancement takes place even at low temperatures without a threshold [115]. This suggests a positive interaction between the two modalities. Proposed mechanisms of interaction are increased drug uptake, increased DNA and protein damage and pharmacological changes other than drug uptake [115].

Increased drug uptake can be accomplished by hyperthermic damage to cell membranes [116]. When cisplatin has entered the cellular nucleus, it forms DNA adducts, which can lead to strand breaking and hampered DNA replicatio [33], [117]. Cisplatin-induced DNA damage can be increased by the inhibition of DNA repair proteins caused by hyperthermia. Hyperthermia (and radiation) can also decrease the cellular amount of free radical scavengers [116]. As a result, more DNA damage is induced by the production of free radicals by cisplatin [118]. Another mechanism of interaction is through induction of pharmacological changes. One possibility is a decrease in cisplatin-protein binding leading to an increase in the amount of the cytotoxic free form of the drug [116]. Another important pharmacological change is an increase in cisplatin delivery to the tumour by a hyperthermia-induced increase in blood flow [116].

These proposed mechanisms suggest that the interaction between hyperthermia and cisplatin is optimal when both modalities are applied simultaneously. This has been confirmed by several observations [14], [18], [20], [74]. The treatment-enhancing effects of cisplatin decrease when the time between hyperthermia and cisplatin treatment increases [14]. With hyperthermia increasing the effectiveness of cisplatin, one may expect that an even more effective treatment would consist of combining the two modalities with radiation. In addition to increasing the effectiveness of cisplatin, hyperthermia also increases radiosensitivity of cells by inhibition of DNA repair processes and tumour reoxygenation [85], [94–104, 106–109]. Hyperthermia-induced reoxygenation is caused by an increase in blood flow to the tumour, which also enhances delivery of cisplatin to tumour cells. The specific DNA repair pathways inhibited by hyperthermia are not completely known, but the non-homologous end-joining (NHEJ) [119–121], homologous recombination (HR) [121] and base excision repair (BER) pathways [98] have all been suggested. In addition, cisplatin also acts as a radiosensitizer through inhibition of DNA repair [33, 34, 122–124]. Nucleotide excision repair (NER) and NHEJ have both been reported to be inhibited by cisplatin [125].

A tumour temperature of 43°C for 1 h is the clinical goal for externally applied local or locoregional hyperthermia. Therefore, most in vitro studies on the effects of hyperthermia focus on temperatures above 42°C. However, these temperatures are rarely obtained for deep hyperthermia in clinical practice. Lower temperatures (below 42°C) may have a more beneficial effect on blood flow and therefore on drug delivery to the tumour [82]. In addition, cisplatin cytotoxicity is already enhanced at low temperatures [116]. Therefore, there is renewed interest in the effects of temperatures in the range of 40–41°C [82]. Results from Bergs et al. lead to the conclusion that the goal in trimodality treatment has to be to keep tumour temperatures as high as possible in order to radiosensitize tumours with different cisplatin sensitivities (see below). However, more pre-clinical studies are needed to investigate the importance of tumour temperature and its effects on trimodality treatment.

Different heating times at 40°C, cisplatin concentrations and radiation dose rates have been investigated by Raaphorst et al. [79], [80], [83], [84]. A synergistic interaction was obtained by all combined treatments in cisplatin-sensitive cells with cisplatin concentrations equal to or higher than 3 µg/ml. In patients a cisplatin plasma peak concentration of 49.8 µM/litre (or 16.6 µg/ml) can be achieved after administration of the maximal tolerated dose (MTD) of 100 mg/m² [126]. Heating to 41°C and 43°C in combination with radiation and various dosing regimens of cisplatin was studied by Bergs et al. [81]. The results of this study show that trimodality treatment using low concentration (1 µM) continuous cisplatin incubation and 41°C hyperthermia resulted only in a synergistic effect in a cisplatin-sensitive cell line. In a cisplatin-resistant cell line, trimodality treatment using hyperthermia to 43°C had to be applied in order to achieve radiosensitization.

A synergistic interaction between the three modalities can be obtained using both LDR and HDR irradiation [79], [80], [83], [84]. LDR was combined with LDMH in these studies. In the clinic LDR in the form of brachytherapy could be used together with LDMH and cisplatin in order to increase treatment effectiveness especially in cells resistant to cisplatin. However, no clinical studies on trimodality using LDMH were found and only very few clinical studies on radiotherapy combined with LDMH exist. One study comparing LDMH/LDR thermoradiotherapy with acute-hyperthermia/LDR observed a better tumour response in patients treated with LDMH/LDR therapy than in patients receiving LDR alone. A review on this topic is published by Armour and Raaphorst [85].

The mechanisms of action of each modality suggest that treatment sequence could be an important factor in determining the degree of interaction. The optimal treatment sequence leading to a synergistic interaction should be determined by in vitro experiments and verified in animal models. To date, no conclusive in vitro evidence exists on the optimal treatment sequence. Moreover, only one animal study by Herman et al. compared the effect of different treatment sequences on treatment outcome [50]. They observed the largest TGD after concomitant cisplatin and hyperthermia applied before irradiation. More studies are needed to compare different treatment sequences in different tumour types using relevant animal models and endpoints. Because no evidence exists on the treatment sequence resulting in the largest degree of (synergistic) interaction, there is no consensus about which sequence to use in the clinic. Based on the evidence of several in vitro studies that simultaneous hyperthermia and cisplatin result in increased cytotoxicity compared with consecutive treatment most phase I/II studies applied hyperthermia and cisplatin concomitantly before or after radiotherapy. Whether hyperthermia should be applied prior to or after radiotherapy has been subject to debate. According to van der Zee et al., the application of hyperthermia after radiotherapy would result in a maximal therapeutic gain with lower radiosensitization in the normal tissue than in the tumour [127]. In addition, hyperthermia before radiotherapy would increase distant metastases. However, Dewhirst et al. [128] argue that it is highly unlikely that hyperthermia applied before radiotherapy will lead to increased toxicity in normal tissue. Instead, treating with hyperthermia prior to radiation may take advantage of heat-induced tumour reoxygenation resulting in radiosensitization. Dewhirst et al. further believe that there is no strong evidence for concluding that administering hyperthermia prior to radiotherapy carries any increased risk for metastases [128].

Another important factor influencing the clinical response to trimodality treatment is the sensitivity of tumour cells to cisplatin. It has been shown that the addition of hyperthermia to cisplatin treatment could reduce cisplatin resistance in different cell lines [129–132]. Therefore one might expect that hyperthermia and cisplatin treatment lead to increased cellular radiosensitivity compared with cisplatin and radiation. However, the effect of combined hyperthermia and cisplatin on radiosensitivity has been shown to vary in different cisplatin-resistant cell lines [79], [80], [83], [84]. The influence of cisplatin-resistant tumour cells on the clinical response of patients is not known. However, a tumour generally consists of cell fractions with differing degrees of cisplatin sensitivity. Therefore, it might be difficult to use cellular cisplatin sensitivity as a predictor for clinical response.

A restriction is that, to date, only phase II clinical trials on trimodality treatment consisting of hyperthermia, cisplatin and radiation have been published. Phase III trials are needed to compare trimodality treatment with other treatment arms. In 2003 a phase III clinical trial has started to compare trimodality treatment with standard chemoradiation in advanced stage cervical carcinoma patients [46]. In this trial a total of 400 patients will be included in 4–5 years with a follow-up of 3 years. Patients are treated by external beam radiotherapy (45–50 Gy) for 5 weeks and brachytherapy with weekly hyperthermia and cisplatin (40 mg/m²). Cisplatin and hyperthermia are preferably administered concomitantly prior to radiotherapy.

Conclusion

In conclusion, trimodality treatment consisting of hyperthermia, cisplatin and radiation is effective and feasible in patients and seems to be promising. It is surprising that few preclinical studies exist on this subject. Several animal studies show that trimodality treatment has a beneficial effect on treatment outcome compared with mono- or dimodality treatment. This beneficial effect can be explained by the synergistic interaction between hyperthermia, cisplatin and radiation as shown by several in vitro studies. More pre-clinical studies are needed to investigate the mechanism of interaction, optimal doses, and sequencing of the modalities.

Acknowledgements

The authors thank Professor J. P. Medema for discussion in the work on trimodality treatment.