Could hyperthermia with proton therapy mimic carbon ion therapy? Exploring a thermo-radiobiological rationale

Abstract

Hyperthermia has been conventionally used in conjunction with photon beam irradiation. With a gradual increase in particle therapy facilities worldwide, this paper explores the physical, thermal and radiobiological implications of using a combination of hyperthermia with proton beam therapy. Hyperthermia is known to exhibit radiobiological features similar to those of high linear energy transfer radiation. Protons have many of the physical dose distribution properties of ¹²C ion therapy. Thus, the thermo-radiobiological advantages of hyperthermia coupled with the physical dose distribution advantages of proton beams could possibly mimic ¹²C ion therapy.

• Carbon ion therapy
• hyperthermia
• proton therapy
• radiobiology
• sarcoma

Introduction

Hyperthermia has been considered to be the ‘most powerful way of sensitizing ionizing radiation’ [1]. The effects of hyperthermia in association with radiation have been attributed to its direct cell kill, selective cytotoxicity of the radio-resistant hypoxic cells, ability to kill radio-resistant ‘S’ phase cells and inhibition of radiation-induced DNA damage repair [2–7]. Various in vitro experiments conducted with radio-resistant sarcoma and melanoma tumour cell lines, have demonstrated that thermal radio-sensitisation could be achieved by hyperthermia [8–13].

Ongoing research in the field of hyperthermia has been directed towards improving the therapeutic outcome with radiation therapy. These include a greater insight into the understanding of the thermo-radiobiological implications of sequential treatment with irradiation, the molecular mechanism behind the effects of heat shock proteins, hyperthermia-induced immunomodulation, changes in blood flow and tissue perfusion at elevated temperatures, hyperthermia treatment planning, non-invasive thermometry, technical innovations in treatment delivery, online monitoring and the use of nanoparticle-based hyperthermia [5,14–22]. Hyperthermia has also been shown to be a thermal sensitiser of chemotherapeutic agents and thus improves the therapeutic outcomes of certain adult and paediatric tumours [7,23–25].

There is currently a widespread interest in proton therapy. This is due primarily to the superior physical dose distribution over photons, which results in more precise dose delivery within the target and minimises dose to the normal structures in the vicinity of the tumour. An increasing number of centres worldwide are gradually acquiring proton therapy facilities [26], although the economic viability in some countries has been recently questioned [27] and additional ¹²C ion therapy units would only add to the costs.

As proton therapy gains a gradual momentum, it is appropriate to explore the possible thermo-radiobiological implications of a combination of protons with hyperthermia in the clinical setting. This paper examines the thermo-radiobiological implications of hyperthermia in conjunction with proton beam therapy.

Particle beam therapy: Protons and carbons

Particle beam therapy with proton and ¹²C ions has been considered as the next logical step in the improvement of radiation therapy [28]. Both these particles share similar physical dose distributions characterised by a low-dose entrance region, a homogeneous high-dose region at the target, followed by a steep fall-off to an almost negligible dose beyond the target. ¹²C ions exhibit a sharper Bragg peak with less scatter (i.e. a better penumbra) than protons but a more gradual dose fall-off in the fragment tail behind the peak [29,30] (Figure 1). This results in a slightly better physical dose distribution for ¹²C ions compared to the proton beams, but alone may not be clinically significant. As the mean linear energy transfer (LET) of the protons is slightly higher than for 250 kVp X-rays, protons could be considered as low-LET radiation similar to photons [31,32]. The relative biological effectiveness (RBE) would therefore be just slightly higher than photons and the average RBE at mid spread out Bragg peak (SOBP) is generally accepted as 1.1 for 150 to 250 MeV proton beams [31]. The ¹²C ions with a higher LET would result in a higher RBE. Thus, it is estimated that while the RBE for protons could vary from 1.02 at the proximal edge to 1.4 at the distal edge, the corresponding RBE for ¹²C ions could range from 1.5 at proximal edge to 6.7 at the distal edge [33]. The average RBE for ¹²C is therefore considered to vary between 2.5 and 3 [34]. Although physically both protons and ¹²C could reasonably conform to the dose distribution around the tumour, a RBE difference of two to three times in favour of ¹²C could be translated into a higher differential tumour cell kill. ¹²C ions may also be more effective against slow-growing tumours with a lower α/β ratio, e.g. prostate, melanoma, sarcoma, etc., as such tumours with lower α/β ratio are usually radio-resistant and are best treated with hypofractionated treatment regimens as used in most ¹²C ion therapy treatment protocols [30,32,33,35]. Moreover, the higher RBE of ¹²C ions could result in more complex DNA damage with a greater number of double strand DNA breaks refractory to post-irradiation repair effectively leading to a greater magnitude of lethal cell events [33].

Figure 1. Pristine and spread-out Bragg peaks of proton and ¹²C ion beams showing similar dose profiles except higher beam entrance dose and dose due to fragmentation tail behind the target with ¹²C ions. Reproduced with permission [29].

In addition, a higher LET would enable a lower oxygen enhancement ratio (OER) and could facilitate the eradication of tumours with larger hypoxic fraction [35]. Compared to an OER of around 2.8 to 3.0 for photons and protons, the OER for ¹²C ions would depend on the range of their LET and could decline to 1.2 at around 200 keV/μm [36]. With the OER nearly reaching unity with ¹²C ions, the differential radiation dose-modifying influence of oxygen between hypoxic and oxic regions in tumour tissue will be diminished. However, the benefit of a lower OER would be related to the mid SOBP depth of the ¹²C ions. This could be modest as the LET at mid SOBP (assuming SOBPs of 6–12 cm) has an OER greater than 2.2 [30,37].

The position of cells in the cell cycle is known to influence radio-sensitivity. ¹²C ions exhibit a lesser cell-cycle dependence than low LET radiation [30]. Thus, an increased cell kill would be expected irrespective of the temporal and spatial distribution of cells in the cell cycle. This could be advantageous to overcome accelerated repopulation without the need for accelerated fractionation to counter a greater quantum of ‘S’ phase cells in the cell cycle [36].

A potential key factor for tumour recurrence is the presence of slow growing cancer stem cells [36,38,39]. These are resistant to the conventional low LET radiation of photons and protons as they have an efficient repair mechanism [38,39]. Thus, cancer stem cells can remain as a viable but dormant nidus following treatment with low LET irradiation [35,38,40,41]. ¹²C ions are predicted to be particularly cytotoxic to these cancer stem cells and more so if they reside in hypoxic regions [33]. Moreover, ¹²C ions have also been shown to improve radio-responsiveness of photon-refractory tumours [35,40,42]. This was evident from various in vitro studies of intratumoural quiescent cells [43], glioblastoma cells [44], prostate cancer [45] and others [42]. Takahashi et al. [46] recently reported the impact of different LET radiation on the repair of the DNA double strands breaks and concluded that the greater radio-sensitivity observed with ¹²C ions, could be attributed to the predominant inhibition of non-homologous DNA repair, typical of high LET radiations. Thus, although protons offer physical dose distribution advantages over photons, both are low LET radiations and have nearly similar radiobiological properties. In contrast, ¹²C ions provide physical dose distribution superiority over photons, and being high LET, they offer distinct radiobiological superiority over protons.

Hyperthermia – does it share the high LET properties?

Hyperthermia displays selective cytotoxicity towards hypoxic cells and reduces radiation-induced DNA repair, radio-resistant ‘S’ phase cells. In intrinsically thermo-sensitive of tumours such as melanoma and sarcoma, hyperthermia exhibits features characteristic of ¹²C ions as described above. Even though the mechanism is distinct from the high LET-induced ionising events of ¹²C ions, the thermo-biological effects of hyperthermia on tumours between 41–43 °C are consistent with the radiobiological effects of ¹²C ions. Robinson et al. [47] conclusively demonstrated a marked reduction in OER at elevated temperatures by subjecting anoxic and oxic cells to radiation at temperatures between 37.5 °C and 43 °C. They reported a gradual reduction in the OER with increased temperature, and found that anoxic cells at 43 °C were more radio-sensitive by a factor of 5.07 and oxygenated cells by a factor of 2.83 than their controls at 37.5 °C. The thermal radio-sensitisation reduced the OER from 2.47 at 37.5 °C to 1.38 at 43 °C. The OER values with hyperthermia when compared with those of heavy particles (pi mesons: OER 1.5–2.2; fast neutrons OER 1.5–1.8; helium ions: OER 1.55–1.65), suggests that the OERs observed at 43 °C were comparable to those of the heavy ion particles. These data support Robinson’s description of hyperthermia as a ‘poor man’s high-LET radiation therapy’ [48].

Hyperthermia in conjunction with particle beams

Gerner and Leith studied the interaction of hyperthermia with radiation of different LETs and observed that hyperthermia and high LET radiation interacted additively rather than synergistically as is seen with low LET radiation [49]. They observed that with X-rays, hyperthermia at 43 °C reduced the D₀ by a factor of 2, leading to a major increase in the slopes of the cell survival curves from −0.00314 at 37 °C to −0.00571 at 43 °C (Figure 2). However, for ¹²C ions, the D₀ at 37 °C and 43 °C were the same at 85 cGy with almost similar slopes at −0.00502 and −0.00557 respectively. Furthermore, it was evident that survival responses for X-rays and ¹²C ions at 43 °C were alike. They concluded that hyperthermia converted the cytotoxic damage of low LET X-rays to that observed with high LET radiation. This could be due to the reduced impact of oxygenation or more probably the reduced repair of low LET radiation-induced DNA damage. Most of the damage inflicted by ¹²C ions would be lethal due to increased double strand DNA breaks; the addition of hyperthermia may not have resulted in additional cell kill. Chang et al. [50] looked further into the temporal sequence of hyperthermia and high LET radiation. They observed that a mere additive effect was observed if hyperthermia preceded irradiation as reported by Gerner and Leith [49]. However, they observed a synergy with hyperthermia and high LET radiation when heat was applied following irradiation, possibly due to modulation of post-irradiation protein synthesis. They therefore proposed that hyperthermia should follow high LET radiation. Ohnishi et al. [51] observed the reduction of tumour growth following hyperthermia at 42 °C with ¹²C ions, and suggested that high LET radiation-induced damage may not always be irreparable and could depend on the LET values. They suggested that ¹²C ions of energies less than 80 keV/μm could induce DNA strand breaks which could be repairable. Hyperthermia delivered following radiation could inhibit such repair, even following high LET radiation (if <80 keV/μm), thereby exhibiting a synergy with ¹²C ions.

Figure 2. Survival of asynchronous CHO cells exposed to either 4 MeV X-rays (circles) or ¹²C ions (squares) either at 37 °C (filled circles and squares) or 43 °C (open circles and squares). The marked thermal sensitisation at 43 °C is evident with X-rays while it was minimal with ¹²C ions (black arrow). Reproduced with permission [49].

Ultrasound-induced hyperthermia combined with a sub-therapeutic proton dose to treat experimentally induced choroidal melanoma has been reported in rabbits [52]. Hyperthermia potentiated the effects of protons even at lower doses and thus could be expected to produce less post-irradiation intraocular morbidity. However, to the best of our knowledge, no clinical studies have been reported using hyperthermia and proton beam therapy.

Extrapolating from data using photons, hyperthermia combined with protons could have effects analogous to high LET radiation. It could be predicted that proton beam thermo-radiotherapy could yield an amalgamation of the advantages of the physical dose distribution (Bragg peak and SOBP similar to ¹²C ions) and thermo-radiobiological (high LET as in ¹²C ions) (Figure 3). Thus, proton beam thermo-radiotherapy could theoretically mimic a ¹²C ion therapy.

Figure 3. Proton beam thermo-radiotherapy could yield a summation of the physical dose profiles of protons and the high LET advantages of hyperthermia. Thus, proton beam thermo-radiotherapy could mimic ¹²C ion therapy.

Proton thermo-radiotherapy: Clinical implications

Based on the numerous potential advantages of using proton beam therapy in conjunction with hyperthermia, a novel phase I/II trial (HYPROSAR) has been initiated to look at the feasibility and safety of this combination in primary and recurrent unresectable soft tissue sarcomas [53] (Figure 4). As soft tissue sarcomas are relatively radio-resistant to conventional photon beam therapy, these tumours could be expected to exhibit an intrinsic sensitivity to hyperthermia [12,13,54,55]. Both proton and ¹²C ion therapies have been reported to be individually efficacious and safe in soft tissue tumours [56–59]. However, there are several constraints on the interpretation of the available outcome data. These have been very effectively summarised by Suit et al. [30]. One of the key aspects relates to the relatively short follow-up for evaluation and assessment of the delayed normal tissue injuries. Thus, a uniform end point score with longer follow-up would be needed to answer questions pertaining to the safety and efficacy of these therapies along with the risk of developing second malignancies.

Figure 4. Schematic outline of the ‘HYPROSAR’ protocol.

The present trial proposes to treat patients of primary and recurrent unresectable soft tissue sarcomas once a week with local hyperthermia to a temperature of 41.5–42.5 °C and proton beam therapy to a pre-operative dose of 55–60 Gy (RBE) at 1.8–2.0 Gy (RBE)/fraction for 5 days a week. The primary goal would be to downstage these unresectable tumours for later surgical excision. Patients who could fail to achieve more than 30% tumour regression following 55–60 Gy (RBE), would continue to receive a radical dose of 72–76 Gy (RBE) with hyperthermia. The trial would establish a proof of the concept of proton thermo-radiotherapy to mimic ¹²C ion therapy. However, additional pre-clinical studies are warranted to confirm this hypothesis. These may be followed by a randomised trial comparing proton thermo-radiotherapy with ¹²C ion therapy.

According to the latest global estimates of the status of particle therapy in 2014, 52 of the 64 (81.2%) particle therapy facilities have protons while only seven use ¹²C ions, the latter predominantly located in Japan [26]. Proton therapy has thus been offered to 86.3% of the patients treated with particles. With the anticipated rise in treatment with proton beams, the thermo-radiobiological implications of a combination of proton beam and hyperthermia deserves to be explored in future clinical trials as this may pave the way for a cost-effective and viable alternative to ¹²C ion therapy.

Declaration of interest

This study has been supported by a partial grant from the Research Council, Kantonsspital Aarau (Forschungsrat KSA) and Paul Scherrer Institut, Villigen to N.R.D. The authors alone are responsible for the content and writing of the paper.

Appendix

Brief explanation of the radiobiology terms used in the text. For more detail, please see the excellent textbooks listed [60,61].

1. Oxygen enhancement ratio (OER): The OER is the ratio of the radiation dose required to give the same biological effect in the absence or presence of oxygen. For most cell types, the OER for photons (X-rays) is around 3.0, while for high LET radiation (greater than 200 keV/µm), it is around 1.0. Thus high LET irradiation offsets the adverse effects of hypoxia.

2. Linear energy transfer (LET): The LET is the average energy (in keV) deposited per unit length (µm) of the radiation particle track. It is expressed as keV/µm. γ-rays have a LET of about 0.3 keV/µm and are described as low LET radiation while α-particles have a LET of about 100 keV/µm and are an example of high LET radiation.

3. Relative biological effectiveness (RBE): RBE is defined as the ratio of the dose in Gy of 250 kVp X-rays or γ-rays that is required to produce the same biological effect as that of a test radiation. For most high LET radiations at therapeutic doses, RBE ranges from 2 to 10.

4. Interplay of LET, RBE and OER: RBE increases as the LET increases, reaching a maximum at around 100 keV/µm. Beyond this, the RBE decreases because of cellular overkill effect. OER also reduces rapidly to 1.0 as the LET rises to 100 keV/µm.