Reoxygenation during cancer treatment is one of the pinnacles of the four “Rs” of radiobiology, along with repair, repopulation, and reassortment (or redistribution) (24).

The presence or absence of molecular oxygen dramatically influences the biological effect of radiation in biological tissues. To generate an effect, molecular oxygen must be present during radiation exposure, or at least for the lifetime of the radiation-generated free radicals (24–26).

The primary mechanism of RT stems from the creation of ROS, consequently inducing cell death by apoptosis, necrosis, autophagy, and senescence. As oxygen is required for ROS generation, hypoxic tumors are resistant to the cytotoxic effects of RT. Hence, oxygen is responsible for the permanent damage induced by free radicals. Consequently, in its absence, these RT-induced lesions can be repaired.

For radiosensitization, only a small amount of oxygen is required. It is estimated that 0.5% oxygen (pO₂ of about 3 mmHg) results in an intermediate radiosensitivity between hypoxia and total oxygenation. Several tumors implanted in animals have been shown to contain hypoxic cells that limit curability by single doses of X-rays. Hypoxic fractions range from 0 to 50%, with an average of about 15%.

There is strong evidence that human tumors contain hypoxic cells, including histological appearance, oxygen probe measurements, nitroimidazole binding, positron emission tomography (PET)/SPECT studies, and pretreatment hemoglobin levels. Oxygen probes with fast response times, implanted in a tumor and under computer control, may be used to obtain an oxygen profile. In addition, hypoxia in tumors can also be visualized with hypoxia markers such as pimonidazole or hypoxia-inducible factors (HIF—which play a role in radioresistance by upregulating downstream genes involved in apoptosis, metabolism, proliferation, and neovascularization) (27–32).

As a dynamic environment, an irradiated tissue can reoxygenate itself when hypoxic cells become, once more, oxygenated. Though the extent of reoxygenation and its rate varies widely for different experimental animal tumors, if it is rapid and complete, hypoxic cells will have little influence on the outcome of a fractionated radiation schedule.

While acutely hypoxic cells rapidly reoxygenate as the tumor's blood vessels open and close, chronically hypoxic cells respond slowly as the tumor shrinks. It has also been described that HBOT has a positive effect on radiosensitization in low-oxygenated tumor regions, as it helps to surpass the constraints of oxygen deficiency. Albeit reoxygenation cannot yet be measured in in vivo human tumors, it is presumed to occur.

Relying on anaerobic metabolism, gliomas thrive under hypoxic microenvironments, contributing to their poor prognosis. In fact, high lactic acid concentrations correlate with a worse prognosis (33).

Hypoxia plays a critical role in maintaining glioma malignancy and resistance, as it induces HIF pathways responsible for regulating important cell cycle genes (34). Consequently, hypoxia will also contribute to cell stemness maintenance by inhibiting cell cycle progression (blocking cells in the G1 phase).

Oxygen starvation mechanisms help explain GBM resistance to CT, promoting tumor growth, angiogenesis, and invasion. Angiogenesis is largely induced and driven by HIF-1α (35). However, the aberrant neovascularization that follows creates immature and flawed vessels that are unable to perfuse the rapidly growing tumor, leading to a hypoperfused neoplasm. The latter results in increasingly numerous necrotic areas (Figure 2).

After demonstrating that glioma cells submitted to hyperbaric oxygenation grow, Wang et al. (36) also studied the effects of HBOT on the expression of the two main molecules that contribute to tumor progression: HIF-1α and HIF-2α. Both are also responsible for regulating sex-determining region Y-box 2 (SOX2—a stem cell marker responsible for blocking the cell cycle, inducing and maintaining cell stemness). In the same way, HIF-1α promotes the multidrug resistance mutation 1 (MDR1) transcription, thus enhancing multidrug resistance to cytotoxic agents. The authors hypothesize that a more oxygen-rich microenvironment not only promotes cell cycle progression but also decreases cell stemness (correlated with chemoresistance) by inhibiting HIF-1α, HIF-2α, and SOX2. In their experiment, mice submitted to HBOT alone had increased tumor volumes than control mice (e.g., normoxia). Yet, when HBOT was used adjunctively with TMZ, the mice had smaller tumors and prolonged survival than those treated with TMZ alone or given only HBOT. Therefore, despite promoting cell cycle progression (and inevitably, tumor growth), HBOT as an adjuvant therapy contributes to chemosensitisation (36). Even with this seemingly contradictory result, it is relevant to emphasize that the available literature does not support the hypothesis of HBOT as a promoter of malignant cell proliferation. Malignancy is not a contraindication for HBOT (37).

In 2004, Evans et al. had already established an association between hypoxia and tumor aggressiveness in 18 patients with supratentorial glial neoplasms. By intravenous administration of the agent 2-nitroimidazole EF5 [2-(2-nitro-1-H-imidazole-1-yl)-N-(2,2,3,3,3-pen-tafluoropropyl) acetamide] in the 24 to 48 h before the tumor biopsy, they measured hypoxia levels in vivo. Their findings support that severe hypoxia correlates with more aggressive clinical behaviors, with early recurrence being one of the consequences (38).

Hadanny and Efrati (39) also proposed an interesting concept: the “hyperoxic–hypoxic paradox.” This is based on the notion that the metabolic changes experienced at the cellular level triggered by hypoxic environments can also be induced by intermittent exposure to hyperoxia. In these conditions, tissue regeneration is activated by stimulating sirtuin 1 (SIRT1) and mitochondrial transfer. This results in a neuroprotective effect and the induction of neuro-recovery mechanisms (39).

The first documented use of hyperbaric medical therapy was in 1662 by Henshaw, a British physician who placed patients in a pressurized air container. Interestingly, it was carried out before the formulation of the Boyle–Mariotte Law, which described the relationship between pressure and volume of a gas, and before John Priestly discovered oxygen more than 100 years later.

Presently, the treatment of a patient resorting to HBOT occurs at elevated atmospheric pressures (higher than 1 ATA), with 100% oxygen. Thus, the tissue's oxygen partial pressure increases, resulting in a myriad of benefits, including improved oxygen supply, reduced inflammation and oedema, and even inhibition of infection.

HBOT employs several physiological principles of gas' diffusion and solubility, especially to predict the behavior of oxygen under pressure. Based on its solubility under pressure, when the oxygen concentration in a solution increases, the diffusion gradient for its delivery deeper into tissues also increases. Ultimately, the increase in dissolved oxygen generated by HBOT has several physiological effects capable of altering tissue responses to disease and injury. In a hyperbaric environment, it is possible to dissolve sufficient oxygen in the plasma to meet the body's usual needs. As physically dissolved oxygen will be more available than the one bound to hemoglobin, oxyhemoglobin will be able to pass unchanged from the arterial to the venous side (40).

The brain is the organ where HBOT has the most important metabolic effects. These are more pronounced in hypoxic/ischaemic states, where hyperbaric oxygenation can reduce cerebral oedema and improve the function of neurons rendered inactive by these lesion-inducing mechanisms. An enhanced electrical brain activity reflects this improvement in brain function. HBOT can also improve microcirculation, relieve cerebral hypoxia, partially preserve damaged tissue, and improve cerebral metabolism.

During treatment, the starting optimal pressure for patients with brain injury is 1.5 ATA, as the cerebral glucose metabolism is balanced at this pressure. Consequently, raising the pressure, even if only to 2 ATA, may have unfavorable effects. At high pressures, oxygen can have acute cytotoxic effects. These were first described in 1878 and usually manifest with seizures—the Paul-Bert effect—whose incidence ranges from 0.002 to 0.061% (41–43). Individual factors (e.g., previous seizures, carbon dioxide retention, alcohol, few drugs, fever, hypothermia, anxiety, cerebral trauma, metabolic acidosis, shock, increased pressure, and oxygen exposure) (44–46) can alter a person's sensibility and reduce their seizure threshold. The first-line approach for hyperoxic seizures is immediate treatment interruption.

Generally, HBOT therapy is safe and well-tolerated by humans at 1.5–2 ATA (11, 12).

Although rare and usually non-severe, HBOT has known potential side effects. In a retrospective analysis of 1.5 million treatments, only 0.68% were associated with an adverse event, with barotrauma and confinement anxiety being the most reported events (42). When patients are previously studied and the correct protocol is applied, side effects are scarce.

HBOT acts by increasing pressure (via pure oxygen administration) on the patient. The most common adverse effects will occur by disturbing air cavities, including ear, paranasal sinus, pathologic dental spaces, and emphysematous bulla (47).

Middle ear barotrauma (MEB) is this therapy's most common side effect. Its incidence varies according to the series and ranges from 0.37 to 84% in non-ventilated patients vs. 94% in ventilated patients (42, 48). Clinical presentation of MEB usually includes ear pain, hearing loss, tinnitus, and, more rarely, vertigo. Diagnosis is performed by otoscopy, and treatment will depend on its severity, generally consisting of decongestants, nasal and systemic steroids, and antibiotics.

To correctly equalize the air in the middle ear, the eustachian tube (which connects the middle ear to the nasopharynx) needs to be opened by the patient using Valsalva maneuvers (e.g., swallowing or chewing). Nevertheless, in patients who have difficulties equalizing air pressure, even medicated and trained, in some cases, a myringotomy with ventilation tubes may be necessary.

Sinus barotrauma (SB) is the second most frequent adverse effect, usually associated with upper airway infection. SB is rare with controlled pressurization and depressurization (49). The manifestations will depend on the affected sinus. Clinical presentation may include sensations of pressure, epistaxis, and pain. Treatment includes nasal decongestants, antihistamines, and/or steroid nasal spray, which should be prescribed prior to HBOT for SB prevention (50).

Despite being a highly vascularized tumor, GBM is characterized by extensive necrotic areas with surrounding hypoxia (6). Hypoxia has been associated with an increase in glioma stem-like cell properties, glioma growth, changes in cancer cell metabolism, and treatment resistance.

HBOT favorably improves oxygen transport to the hypoxic tumor tissues, thereby potentially increasing the sensitivity of tumor cells to antineoplastic treatment. However, while used alone, it does not inhibit tumor growth (33). In contrast, when used as an adjuvant treatment with RT and/or CT, it leads to a significant decrease in tumor size. Initial HBOT can modify the hypoxic microenvironment and impair the stemness-associated properties of cancer cells (51).

Regarding its benefit when used as an adjuvant treatment with TMZ, tumor tissue hypoxia tends to decrease with HBOT, and some hypoxic areas even return to regular oxygen supply. With an increase in oxygen concentration, the tumor tissue becomes more sensitive to CT drugs. In addition, hyperbaric oxygenation also amplifies the cell cycle arresting effect of CT, which might result in an increased affinity of tumor cells for TMZ (33), indicating that for the same drug dose, there is a higher rate of apoptosis. By combining nimustine and HBOT, the expression of inflammatory factors is reduced, and tumor growth is inhibited, indicating that inflammation plays a key role in tumor development (33).

HBOT has been used to treat late-onset RT-related injuries since 1975 (52), being CNS lesions a type 3 recommendation. In 1953, Gray et al. postulated that the primary source of radioresistance was oxygen deficiency (53). As RT's efficacy is based on the tumor tissue's oxygen content, which is why the combination with HBOT amplifies its therapeutic effect (functioning as a radiosensitizer), improving local tumor control and potentially extending survival time.

Using a mouse model of GBM, Wang et al. (54) investigated the effects of HBOT on ROS signaling from transplanted glioma cells, finding that under hyperbaric oxygenation, ROS signaling from glioma and brain cells was diminished. However, their findings raised a new question regarding the pro-oncogenic effects of HBOT on non-cancerous cells, as this therapy appears to produce systemic immunosuppression by inhibiting thymic T-cell maturation (54).

By transplanting gliomas into nude rats, Stuhr et al. (10) concluded that the tumor's hypoxic microenvironment could be altered through hyperoxia. Both normobaric and hyperbaric oxygenation (1.02 ATA vs. 2.04 ATA) inhibit tumor growth after only 4.5 h of treatment (approximately three exposures) (10). Nevertheless, HBOT significantly improves oxygen perfusion, damaging the tumor by inducing apoptosis and reducing its central vascular density.

The anticancer effects of hyperoxia and its action mechanisms are mostly identified in mouse lung cancer (55, 56). To detect the potential mechanism of HBOT in lung tumors, Chen et al. (57) selected, as an in vivo model, non-small cell human lung carcinoma A549-cell-transferred severe-combined immunodeficiency mice (SCID). Their results revealed that HBOT improves tissue angiogenesis and tumor hypoxia and increases tumor apoptosis by modifying the tumor's hypoxic microenvironment. The treatment also suppressed tumor growth in murine xenograft tumor models (57). This was the first study to demonstrate that HBOT immediately downregulated endogenous p53 protein, which rebounded to baseline after 20 h of treatment.

Zembrzuska et al. (58) studied the combination of isothiourea derivatives and HBOT, reporting that this regimen appears to be a promising therapeutic approach for malignant glioma, as it reduces the proliferation and viability of in vitro GBM cells compared to cells under normoxia and hypoxia.

Brizel et al. (59) also performed in vivo studies on rats with implanted mammary adenocarcinoma. These were divided into five groups, posteriorly treated with inhaled normobaric or hyperbaric oxygenation or normobaric or hyperbaric carbogen. Positive results were observed in the groups treated with high-pressure gases (59). Thews and Vaupel (60) conducted a similar study, which tested the influence of normal and hyperbaric oxygenation (pure oxygen vs. carbogen), local hypoxic regions occurrence, and spatial distribution of pO₂, using the subcutaneously injected DS-sarcoma cell line in rats. Under the pressure of 2.07 ATA, the median pO₂ was 5-fold higher. Following HBOT, the spatial distribution of pO₂ profiles showed a nearly absolute elimination of hypoxic regions.

However, isolated HBOT has a limited curative effect and is typically not applied by itself (60).

Stuhr et al. (10) elucidated the effects of hyperbaric oxygenation alone on BT4C rat glioma xenografts in vivo. In his study, hyperoxia (both normal and hyperbaric) resulted in a 60% retardation of tumor growth compared to the control group. Although the proliferation rate of cells remained unchanged in the two groups, there were more necrotic areas, 20% more apoptotic cells in control tumors, and no damage to normal tissues (10). More inconsistent results were found with isolated HBOT, probably due to different HBOT procedures and intracranial measurement methods (61, 62).

Regarding the pO₂ of normal brain tissue, in animal experiments, it has been noted to decrease quickly after HBOT, whereas the pO₂ in high-grade gliomas falls more slowly after decompression due to the lower rate of oxygen consumption and the reduced blood flow to the tumor (63). Therefore, the pO₂ within the tumor may remain elevated after decompression for a substantial time, suggesting that HBOT may increase the sensitivity of hypoxic tumor cells, even without increasing injury to the normal brain tissue (64).

Kunugita et al. (65) examined the effect of RT after HBOT on spontaneously arising murine squamous cell carcinomas (SCCVII) (with a radiobiological hypoxic fraction of ~10%) that were subcutaneously transplanted into C3H/He mice using a growth delay assay. They noted a significant SCCVII tumor growth delay in the treated animals within 30 min after HBOT, with the delay time being 1.61 times longer than that after RT alone (66).

Xie et al. (9), studied the combination of HBOT and nanotemozolomide in glioma using nude mice. Their findings confirm that HBOT is a suitable adjuvant treatment to CT, as the proliferation of tumor tissue was significantly lower when the two therapies were combined (9). Moreover, significant inhibition of tumor growth, weight, and size was also observed, thus supporting the pivotal role HBOT can have in enhancing antitumor agents' efficacy. A decrease in hypoxic fields after HBOT was also verified, adding up to 50% of hypoxic regions modified. Potentiation of the cell cycle arrest effect of the agent was also observed. The authors postulate that the increased tumor oxygen tension improves the therapeutical response to CT.

Accordingly, Lu et al. (67) found that pO₂ in mice who underwent HBOT combined with nimustine therapy was significantly higher than standard, not only on the healthy brain tissue itself but also on the glioma tumor cells, which were approaching normal brain tissue pressures.

In a mouse model of aggressive metastatic cancer, Poff et al. (68) evaluated the efficacy of metabolic therapies, including the ketogenic diet, ketone supplementation, and hyperbaric oxygenation. In vitro, each therapy inhibited the proliferation and viability of tumor cells and decelerated disease progression. Nonetheless, when combined, they potentiated powerful antineoplastic effects by inhibiting metastasis proliferation and doubling the survival time of mice with systemic metastatic disease (69).

There are scarce data on prospective clinical studies evaluating HBOT in combination with RT or chemoradiation for GBM. Most of the evidence available consists of single-centered trials evaluating RT and HBOT. Nevertheless, findings have shown promising results in patients with high-grade glioma (69, 70).

In 1977, Chang et al. published their findings regarding whether HBOT during RT could improve results in glioma treatment (71). The median OS of the patients in the HBOT group was 38 weeks, compared with 31 weeks for patients in the control group, and survival rates after 18 months were 28 and 10%, respectively.

Kohshi et al. (72), conducted a non-randomized trial that evaluated the feasibility of associating RT with HBOT for 29 high-grade glioma patients. The median OS for patients with and without HBOT was 24 and 12 months, respectively (p < 0.05). All patients (n = 4) who received RT more than 30 min after decompression had worse outcomes. This constituted the early perception that the timing between modalities was critical to achieving the best outcomes. As no serious side effects were observed in HBOT patients, the authors concluded that irradiation after HBOT appeared to be a valuable treatment for high-grade gliomas. However, irradiation should be administered immediately after decompression (72).

Beppu et al. (69) published their phase II study integrating HBOT and beta-interferon, nimustine hydrochloride (ACNU), and RT (IAR, Interferon-ACNU-Radiotherapy) for the treatment of malignant supratentorial gliomas. Daily RT was completed within 15 min after HBOT. Of the 39 patients, 35 underwent a complete HBOT and IAR therapy schedule. Thirty patients (76.9%) either maintained or increased their Karnofsky Performance Status Scale (KPS) during HBOT/IAR, with a mean duration of 68 days. The combined complete and partial response rates for GBM, anaplastic astrocytoma, and overall general histologies were 50, 30, and 43%, respectively. Median time to progression (TTP) for GBM, anaplastic astrocytoma, and overall patients was 38, 56, and 43 weeks, respectively. The authors suggested that HBOT/IAR therapy could be applied to selected patients with poor prognostic factors due to its short treatment period, acceptable toxicity profile, and identical response when compared to patients with good prognostic factors (69).

Ogawa et al. reported the long-term results of a phase II trial for daily conventionally fractionated RT (60 Gy/30 fractions) 15 min after HBOT with multiagent CT (procarbazine, nimustine, and vincristine) in patients with high-grade gliomas. All patients were able to complete the RT immediately after HBOT. The median OS of 39 patients with GBM was 17.2 months. However, acute toxicity developed in almost 48 % of patients (70, 72).

Kohshi et al. (73) also investigated the combination of HBOT with fractionated stereotactic RT (FSRT) in a small cohort of previously irradiated patients with recurrent gliomas. The median OS of patients treated with this protocol was 19 and 11 months for anaplastic astrocytoma and GBM patients, respectively (73).

In another Japanese study, Yahara et al. (74) studied IMRT boosts after HBOT concurrent with TMZ-based CT. Of note, the authors estimated a biologically effective dose (BED) of 85.8 Gy10 for the high-dose clinical target volume (CTV). Previous studies have reported that dose escalation of RT for GBM above 60 Gy using conventional RT had limited success. The total BED of 85.8 Gy10 was relatively low compared to other trials. The first site of progression was locally in 58% of the patients with GBM, and grade ≥3 toxicity, such as brain radionecrosis, was not observed. The median OS was 22.1 months, which was promising considering a cohort of patients with GBM (75).

Arpa et al. (6) published their experience with HBOT and RT in a pilot study for recurrent high-grade glioma patients. In their small series of nine patients, RT was administered in daily 5-Gy fractions for 3–5 consecutive days. Each fraction was delivered within 1 h after HBOT. The disease control rate 3 months after HBOT-RT was 55.5% (5 patients), the median PFS for all patients was 5.2 months (95% CI: 1.34-NE), while 3- and 6-month PFS was 55.5% (95% CI: 20.4–80.4) and 27.7% (95% CI: 4.4–59.1), respectively. The median OS was 10.7 months (95% CI: 7.7-NE). No acute or late neurologic toxicity grade >2 was observed in 88.88% of patients, although one developed grade 3 radionecrosis. Despite the small cohort of selected patients, the paper poses a convenient short course alternative to systemic therapies for patients who cannot or refuse to undergo such treatments in this late stage of disease. To our knowledge, this is the only recruiting trial in this setting on ClinicalTrials.gov (6).