The experimental studies using intracranially implanted GBM cells and intra-tumoral bacterial lipopolysaccharide (LPS), a potent PAMP administration, proved increased survival time and tumor regression in animal models (12–14).

Despite promising experimental studies, comparing animal experiments with clinical studies remains problematic. Several factors may play a role in the severity of surgical wound infection. These factors include the time course and site of infection or the type of microorganism responsible for the infection. Therefore, there is considerable variation in conclusions between some studies. Regarding the early surgical wound complications, a study comprising 3748 patients after GBM resection did not show any profit from the bacterial site infection that occurred till a month after surgery. Such a complication was not associated with better outcomes and longer PFS (8).

Similarly, unplanned readmissions within 30 days after primary GBM resection for postoperative complications, including infectious, neurological, and thromboembolic, were associated with worse outcomes. In these patients, the median OS was shorter by 9 months compared to patients without postoperative infectious complications (15). A multicenter retrospective study from Salle et al. supported the assumption that surgical site infection after initial GBM resection leads to shorter OS. In these cases, the mean time between surgery for infection was 55 days (16). Moreover, Bohman et al. found in a single-center study with 382 patients that postoperative infection did not confer any survival advantage in GBM patients. However, subgroup analyses showed a non-significant advantage in mean survival in patients whose infection occurred late after HGG resection, as well as in patients with deep S. aureus infections (7). Based on these studies mentioned above, it can be assumed that the time between surgery and the onset of infection may play a role in OS in patients with HGG. However, it seems that etiologic agents may also play a particular role in the impact of infection on OS. In 2010, Bonis et al. had a significantly longer median survival time (30 months) in patients with bacterial infection when compared with patients without infection (15 months) with a high prevalence of S. aureus (60%). This was a single-center study with 197 patients with about 5 percent of infectious complications after primary surgery for malignant brain tumors, including grade 3 astrocytomas and GBMs (10).

In addition to the studies mentioned above, individual cases presenting the impact of bacterial infection on outcomes in patients after HGG resection have been described. A case report describing the infection of the Ommaya reservoir with S. aureus 3 years after primary GBM resection showed no recurrence of the tumor for 6 years after the onset of the infection (17). Bowles et al. referred to long-term survival in 4 cases after glioma resection followed by bacterial infection. In these cases, the authors suggested that tumor suppression may be potentiated by the immune response as well as the direct oncolytic effect of bacteria (18). This suggestion is supported by the finding of non-recurrence of GBM 4 years after tumor resection and treatment in a patient with subdural fluid collection positive for S. epidermidis (19).

Another proposed mechanism is the stimulation patient’s immune response within or near the tumor bed while avoiding the systemic response (6, 11). It is well known that GBM induces tumor-associated immune suppression both within its microenvironment as well as systematically (20). Glioma cells, along with microglia, are potent to produce immunosuppressive factors to inhibit T-cell proliferation and stimulate T-regs (21, 22). At the same time, GBM induces systemic immunosuppression by reducing T-cell activity with preserved B-cell activity (23). GBM cancer stem cells are characterized by weak expression of MHC, and co-stimulatory molecules contribute to defective immunogenicity (14). Moreover, the expression of anti-inflammatory molecules like IL-10, PDL1, and FAS ligands by glioma-infiltrating cells probably reduces tumor immunogenicity (24).

A growing body of literature has been describing both experimental as well as clinical evidence for complete or partial reduction of tumor cell growth by microorganisms in their close vicinity. During the microbial infection, the host cells recognize so-called pathogen-associated molecular patterns (PAMPs) present in microorganisms and alert the innate immune system (25). In animal models, inactivated bacteria experimentally inoculated into the tumor triggered an influx of macrophages, CD4+, and CD8+ T-cells, thus, stimulating effective anti-tumor response (26, 27). Another study described the effect of directly invading bacteria via activating the stimulator of IFN genes (STING), contributing to anti-tumor immunity by re-educating tumor-supportive M2 macrophages towards proinflammatory M1 phenotype and stimulating T-cell influx and action (28, 29). The phenotypic changes from M2 to M1 proinflammatory macrophages are supported by the activation of Toll-like receptor 9 (TLR9), which recognizes microbial products and initiates a complex immune response leading to the elimination of invading microorganisms. These changes include the secretion of proinflammatory molecules such as IL‐1, TNF‐α, ROS, and NO, which are able to affect tumor cells’ proliferation, migration, and invasiveness (29). This antitumoral effect was probably caused by the activation of toll-like receptor 4 (TLR4) as well as TLR9, which resulted in the activation of microglia and inflammatory cells at the site of the tumor process (13, 29). Moreover, depending on TLR4 signaling, in vitro LPS stimulation of GBM cells for 6 hours resulted in increased expression of proinflammatory molecules like MHC-I, MHC-II, CD80, CD86, CXCL10, TNF-α, IL-6, and down-regulation of the anti-inflammatory cytokine IL-10 (30).

Interestingly, in vitro study using GBM cells treated with S. aureus enterotoxin B (SEB) suggests another anti-tumor mechanism of bacterial infection. It was found that SEB can decrease smad2/3 expression in GBM cells leading to down-regulated TGF-β signaling and the reduction of tumor cell proliferation (31). Interestingly, there is some evidence that SEB has the ability to induce FasL/Fas (CD95L/CD95) mediated cytolysis through the CD8+ cytolytic T lymphocytes (32).

Apart from the possible impact of the bacterial infection on survival in both presented cases of HGG patients, some other related factors should be mentioned, such as antibiotic treatment, smoking, and the metabolic demands of bacteria.

There is some evidence that antibiotics targeting mitochondria can effectively eradicate cancer stem cells across multiple tumor types, including GBM (33). In this regard, the antibiotics with anti-tumor activity represent the erythromycins, tetracyclines, glycylcyclines, chloramphenicol, and an anti-parasitic drug-like pyrvinium pamoate. However, none of these antibiotics have been used in treating our patients.

Another considered etiopathogenic factor was nicotine consumption in our patient suffering from GBM because she was a heavy smoker. However, the recent review dealing with nicotine-containing products in patients treated with GBM suggested that nicotine has the potential not only to promote tumor growth but also to reduce the effectiveness of chemotherapeutic agents used in the treatment of HGG (34). On the other hand, in vitro studies on human glioma and glioblastoma cell lines showed nicotine dose-dependently cytotoxicity, which was probably caused by a rapid increase in the intracellular calcium concentration (19). Tobacco consumption was also associated with the loss of MGMT gene expression (35). The MGMT gene encodes a DNA-repair protein that removes alkyl groups from the O6 position of guanine, an important site of DNA alkylation, and contributes to the drug-resistant phenotype of GBM cells (36).

Several theories have been proposed to explain how bacteria affect the microenvironment around the GBM cells. One of the possible mechanisms is the sequestration of nutrients needed for bacterial cells leading to a limitation of resources for the high metabolic demands of GBM cells (6).

Despite these encouraging case reports, postoperative wound infection is still considered a serious condition, and randomized trials cannot be performed. However, bacterial toxins in different variants are laboratory tested as a new strategy for the treatment of different brain pathologies, including brain tumors. Our findings, therefore, justify the need for experimental studies focused on the interaction of bacterial components or toxins, the immune system, and glioma cells.