Hyperbaric oxygen therapy and cancer—a review

Solid tumors often contain areas subjected to acute or chronic hypoxia [1], though with variable severity in patients both within and among different tumor types [2]. Although severe or prolonged hypoxia is deleterious, adaptation to the hypoxic microenvironment has allowed cancer cells to survive and proliferate in this hostile milieu [3]. Tumor hypoxia develops due to the structural and functional abnormalities of the tumor vasculature since cancer growth often overrides the ability of the cancer vasculature to adapt to the increasing oxygen demand.

Traditionally, hypoxia was thought of as a factor limiting cancer growth by reducing the ability of cells to divide. However, more recently, hypoxia has proven to be a causative factor in many pathophysiological events, including cancer progression. Multiple reports have demonstrated that decreased oxygen tension selects for more malignant cells and induces multiple cellular adaptations, which again sustains and fosters cancer progression and thereby induces cancer growth (Fig. 1). Hypoxia is reported to result in cellular responses which improve oxygenation and viability through induction of angiogenesis, an alteration in metabolism by increased glycolysis and upregulation of genes involved in cell survival/apoptosis [4]. Hypoxia has also been shown to increase genetic instability, activate invasive growth, and preserve the undifferentiated cell state [1, 3]. Studies have demonstrated that hypoxia is implicated in the resistance to conventional therapy [5]. Oxygen concentration has an especially crucial role in radiation oncology and radiation resistance [6, 7]. The epithelial-to-mesenchymal transition in cancer has been shown to be induced by hypoxic conditions [8], leading to cancers with an invasive or metastatic phenotype [9]. Given its important role as a negative prognostics and predictive factor, hypoxia is considered as one of the best targets in cancer treatment.

The dual role of oxygen leads to the question: will lack of oxygen inhibit cancer progression, or is hyperoxygenating the tumor tissue the way to go in order to prevent cancer growth and development?

Hyperbaric oxygen can be used to overcome hypoxia. HBO is based on administration of 100 % oxygen at higher than normal atmospheric pressure. HBO treatment enhances the amount of dissolved oxygen in the plasma, thereby increasing O₂ tissue delivery independent of hemoglobin [10]. As in normal tissue, the pO₂ in cancer tissue increases significantly during HBO exposure [11]. Thus, elevation of the tumor oxygen pressure has been shown to be preserved clinically for approximately 30 min after HBO exposure [12, 13]. HBO therapy is today accepted and routinely used for many disorders, related to both ischemia and/or hypoxia [10]. HBO is considered safe and complications are rare using today’s standard treatment protocols. The Undersea and Hyperbaric Medical Society has a list of approved indications for HBO therapy, including decompression sickness, severe carbon monoxide poisoning, nonhealing wounds, and late radiation injury.

As oxygen is believed to be required for all the major processes involved in wound healing, including resistance to infection, activation of fibroblasts, collagen deposition, angiogenesis, and epithelization [14], it has been feared that HBO would have a proliferative effect in cancers. Thus, for many decades, the focus has been to elucidate if HBO promotes cancer growth. In the early 2000s, both Feldmeier et al. [15] and Daruwalla et al. [16] reviewed the literature concerning HBO and cancer. The reviews included both experimental and clinical studies using different types of cancers, with and without additional therapy, and the results showed varied responses. Nevertheless, the conclusion in both reviews was that HBO did not promote cancer growth, and that the use of HBO in patients with malignancies was considered safe.

There are extensive studies on the effect of HBO on normal tissue and wounds. Interestingly, evidence implies that cancer tissue might differ in response from normal tissue. The studies performed on HBO and cancer are complex due to a wide range of experimental designs and treatment regimes. Nevertheless, in an attempt to clarify the differences in response to oxygenation, we have summarized the literature concerning the effect of HBO on crucial hallmarks of cancer, the effect of HBO on chemo- and radiation therapy, and in addition we have clustered the different cancer type responses.

Studies of prolonged hyperoxia have shown that elevated levels of reactive oxygen species (ROS) overwhelm the antioxidant defense and lead to cellular damage and possible organ dysfunction [17]. The tissue damage is found to be dependent on the cell type, concentration of oxygen, and the duration of the exposure. Gore et al. [17] have summarized the molecular mechanisms behind hyperoxia-induced cell death, revealing a complex signaling system including protein kinases and receptors such as RAGE, CXCR2, TLR3, and TLR4.

Studies of apoptosis in neoplasms treated with HBO are limited. Two in vitro studies on mammary and oral cancer cells, respectively, showed no change in apoptosis after HBO [18, 19]. On the other hand, Chen et al. [20] observed activation of the pro-apoptotic pathway MAPK and downregulation of the anti-apoptotic ERK pathway in hematopoetic cells after HBO. Additionally, a study of HBO using osteosarcoma cells also demonstrated induction of apoptosis [21]. In two different animal models, gliomas and mammary tumors, respectively, our group has demonstrated induction of cell death after HBO treatment [22‐24].

Furthermore, reduced cell proliferation, together with a significant change in histology, has also been shown after HBO treatment in DMBA-induced mammary tumors in vivo [22, 24]. Granowitz et al. [18] observed the same reduction in cell proliferation in their mammary in vitro study. In addition, two recent studies on osteosarcoma cells [21] and nasopharyngeal carcinoma [25] support inhibition of cell division after HBO treatment.

Together, this might imply that changes in oxygen concentration influence antioxidant pathways [26], leading to a change in cell survival signaling. However, the picture is complex, and mechanistic studies are required before any final conclusions can be drawn.

Today, angiogenesis is proposed to be a key factor for cancer growth and metastasis. Thus, large experimental studies and clinical trials have investigated the effect of antiangiogenic therapies in the treatment of cancers. Since HBO in general has been shown to promote cellular and vascular proliferation in normal tissue and wounds (although the mechanisms are not fully understood), it was assumed that it would also induce angiogenesis in cancers. In contrary to what is expected and addressed in the literature, HBO has been shown to induce an antiangiogenic effect in two mammary tumor models [22, 24, 37], in addition to one glioma model [23]. Furthermore, multiple studies showed no change in angiogenesis after HBO treatment [27‐32]. In his review, Feldmeier et al. [15] thoroughly discussed oxygen and tumor angiogenesis, underlining the difference between cancer tissue and wounds and concluded that HBO is not likely to enhance tumor angiogenesis. Thom [31] commented on the fact that the influence HBO has on hypoxia-induced factor isoform expression appears to vary with different tissues and possibly with chronology (e.g., looking early or late after wounding or an ischemic insult). We believe it to be important to distinguish between normal or injured tissue and tumor tissue when it comes to the effect of HBO and angiogenesis since there is no evidence for enhanced angiogenesis in cancerous tissue.

In 1966, Johnson and Lauchlan first raised concern that HBO might have metastatic potential [33]. However, it was not possible to show a statistically significant increase in the number of patients with distant metastasis, as the number of patients in the series was too small. Nevertheless, special attention was given to metastatic growth because the first reports suggested that HBO might be affecting this part of tumor progression [34]. Metastasis is a complex process requiring multiple steps, including local tumor cell invasion, entry into the blood or lymph vessels, and re-penetration and colonization at a distant site [35]. Eventually, angiogenesis is also required for distant metastasis to form.

So far, only observational studies have been performed, and studies of the effect of HBO on the individual steps of the metastatic process are still lacking [34]. None of the studies reviewed showed induced metastasis after HBO [21, 36‐39]. Furthermore, a recent study found HBO to induce a mesenchymal-to-epithelial transition (MET) in DMBA-induced mammary tumors, leading to a less aggressive tumor type [24], thus indicating that oxygen might be a key factor in MET [40]. This transition should lead to cancers with a less invasive and metastatic phenotype.

Hypoxia has been described as an important factor for chemotherapeutic resistance [5]. Teicher [41] underlined that the importance of hypoxia on the response to chemotherapy is highly drug dependent. However, hypoxia-mediated chemoresistance has been ascribed to: (1) altered cellular metabolism reducing drug cytotoxicity; (2) the redox state, meaning that oxygen is required to generate ROS to be maximally cytotoxic; and (3) genetic instability, which can lead to more rapid development of drug-resistant cells. In addition to the cytotoxicity, availability of the chemotherapeutic drug in high enough dose is important to obtain a maximal effect. Tumor tissue anatomy influences transport of intravenously injected substances to the cancer cells, and thus determines the efficacy of the drug.

Al-Waili et al. [42] summarized the potential role of HBO in combination with conventional therapies. They hypothesized that HBO could improve and help overcome chemotherapeutic resistance by increasing both tumor perfusion and cellular sensitivity [42].

Studies on HBO as a chemotherapeutic adjuvant have shown augmented effects both in vitro [18, 21, 25, 43] and in vivo [21, 44‐47], although the mechanism(s) are not known. Heys et al. [28] studied the effect of HBO on chemotherapy in a clinical setting, using HBO as a pretreatment to improve vascularity, and thereby improve the effect of chemotherapy. However, HBO did not increase the neovascularity, and they correlated the lack of chemotherapeutic potentiation to this. In a mammary tumor model, Moen et al. [48] found that the uptake of chemotherapy is increased for the duration of, and immediately after, HBO treatment. Based on this study, potentiation of chemotherapy can probably not occur unless the chemotherapeutic agent is administered during or immediately after the HBO session, when the pO₂ is elevated. Another study by Moen et al. [24], on the same mammary tumor model, found altered genetic expression after HBO indicating a change to less tumorigenic metabolism, possibly influencing the chemotherapeutic response. Many have ascribed the enhanced chemotherapeutic effect after HBO to increased levels of ROS. Moen et al. [48], however, found no change in MDA levels after HBO, indicating that in this study ROS levels cannot be the main determinant of an increased chemotherapeutic effect. Microarray studies have made it possible to classify breast cancers at the molecular level [49, 50] and correlate their signatures with metastatic behavior and clinical outcome, and thereby making it easier to develop targeted therapy. Underlining the importance of breast cancer subtyping, it is important to comment on the differences between different tumor models: Moen et al. [48] found an increased uptake of the chemotherapeutic drug 5-FU into DMBA-induced tumors after HBO, while Jevne et al. [51] failed to find the same correlation in the 4T1 mammary tumor model.

The combination of HBO and chemotherapy has also been tried in other cancer types. Suzuki et al. [44] suggest that HBO therapy prolongs the biological residence time of carboplatin in glioma patients. However, there are still uncertainties concerning the mechanisms of action of HBO on the efficacy of carboplatin. The same group found that HBO enhanced transendothelial permeability in rat brains and HBO might therefore be favorable for the uptake and therefore also the effect of carboplatin [52]. Preliminary results from a small, clinical study, on nonsmall cell lung cancer, show promising results when combining hyperthermia and HBO with paclitaxel and carboplatin [45]. However, they emphasize that the study lacks proper controls, and thereby the additional value of HBO to the chemohyperthermia response cannot be made. Kawasoe et al. [21] found, both in vitro and in vivo, that HBO enhanced the chemotherapeutic effect of carboplatin in osteosarcomas. Furthermore, combining HBO and cisplatin significantly reduced tumor volume in a human ovarian cancer xenograft model [46].

It is, however, important to underline that Mayer et al. [53] list up five chemotherapeutic agents (doxorubicin, bleomycin, disulfiram, cisplatin, and mafenide acetate); all of which are strongly contradictory in combination with HBO due to potential potentiation of toxicity [54]. Of the reviewed papers, only Heys et al. [28] and Selvendiran et al. [46] have utilized the listed chemotherapeutics in combination with HBO.

Relating the knowledge on the different chemotherapeutics in relation to cancer subtypes will be important for further studies and for development of therapies and adjuvant therapies. In addition, proper randomized studies are necessary in order to be able to make any final conclusions regarding the effect of HBO in combination with chemotherapy.

Radiotherapy in combination with HBO has been used clinically in two different applications: (1) as a therapeutic agent for treating late radiation injury and (2) as a radiosensitizer, aiming to increase the effect of radiotherapy [53]. In this review, we focus only on the latter application of HBO.

Gray et al. [6] proved in the 1950s that the oxygen concentration influences the effect of radiotherapy and the influence of hypoxic modification in relation to radiotherapy has been extensively studied since then. In 2011, Overgaard published a meta-analysis reviewing the influence of hypoxic modification of radiotherapy in head and neck carcinoma [7]. Overall, Overgaard found that out of the various hypoxic modification techniques, HBO showed the most pronounced effect [7], and thus will improve the results of radiotherapy. Nevertheless, in a recent and extensive review by Bennett et al. [55], the authors have also reviewed the effect of radiotherapy in combination with HBO. They concluded that there is some evidence that HBO improves local tumor control and mortality in tumors of the head and neck; however, the outcomes seem to be related to the use of unusual fractionation schemes, and Bennett et al. [55] thereby conclude that the benefits of HBO should be interpreted with caution.

It has also been shown that adverse side effects like oxygen poisoning and severe tissue radiation injury is associated with the use of HBO in combination with radiotherapy [55]. However, it is important to emphasize the importance of timing of HBO exposure in relation to the radiation [53]. Kohshi et al. [56] found that to avoid hazardous side effects, irradiation should be administered immediately after and not concurrently to HBO treatment. It has been shown that euoxic conditions persist for some time after HBO exposure due to postponed oxygen saturation and washout kinetics [12, 13]. Thus, a change in protocols could possibly reduce or prevent serious side effects, and thereby justify the use of HBO in radiosensitization [53]. A conclusion regarding the use of HBO in combination with radiotherapy still remains unclear.