Based on the efficacy of radiotherapy for achieving at least temporary disease regression in children with brainstem gliomas, investigators designed an alternative strategy involving administration of radiosensitizing agents concurrently with irradiation, with the goal of potentiating the efficacy of this treatment modality. Radiosensitization is thought to synergistically potentiate the efficacy of irradiation within rapidly dividing tumor cells. Multiple studies have been conducted to evaluate the response of patients with brainstem gliomas to treatment with chemotherapy during radiotherapy [22, 32–35]. One of the earliest studies was conducted by Jakacki et al. which investigated the concurrent radiotherapy and dose-intensive chemotherapy with procarbazine, lomustine and vincristine. Six patients with diffuse brainstem glioma treated according to this regimen. Patients underwent PBSC harvesting after mobilization with G-CSF. Chemotherapy consisted of CCNU 130 mg/m² on day 0, vincristine 1.5 mg/m² on days 0 and 7, and procarbazine 150 mg/m² on days 1–7. PBSCs were reinfused on day 9 of each course. Four courses of chemotherapy were administered 28 days apart or when counts recovered. Involved field radiation was administered to newly diagnosed high-grade glioma patients following recovery from chemotherapy. The reported median overall survival time was 11 months [28]. Packer et al. assessed the concurrent administration of radiotherapy and chemotherapy consisting of carboplatin as a radiosensitizer and a bradykinin analog RMP-7 as a selective-blood-brain-barrier permeability increasing agent [33]. Carboplatin dose was 35 mg/m², and RMP-7 was given at a dose of 0.3 mg/kg of ideal body weight as a 10-min intravenous infusion beginning 5 min before the end of the carboplatin infusion. These two drugs were given for 5 successive days during radiotherapy. Thirteen patients were treated whose median age was 7 years. The estimated median survival was reported as 328 days with only one patient remaining disease progression-free more than 400 days from the initiation of treatment. Nine patients were treated with hyperfractionated radiotherapy and chemotherapy with carboplatin and etoposide [35] in a phase I/II study setting. Carboplatin, given in combination with fixed-dose etoposide, was escalated in successive cohorts to determine its maximum tolerated systemic exposure. Eight of the nine children on this study died of their disease at a median of 44 weeks, essentially the same survival as those treated on a previous Pediatric Oncology Group study using hyperfractionated radiation therapy alone. Allen and his colleagues [22] evaluated the usefulness of administration of carboplatin as a single agent for the purpose of radiosensitization along with radiotherapy. Carboplatin was infused twice-weekly in a dose-escalating manner with a maximum tolerated dose of 110/mg/m² or a total cumulative dose of 1,540 mg/m² over 7 weeks. A total of 38 patients were enrolled with a median age of 7.8 years. Median progression-free survival was 8 months, and the overall survival was 12 months. Five long-term survivors remained in continuous remission after a mean follow-up period of 79 months. In a Brazilian cooperative study, Broniscer et al. used tamoxifen concurrent with radiation treatment [36]. Tamoxifen was administered orally with a dose of 200 mg/m² per day along with conventional radiotherapy and for 52 additional weeks. Of 29 patients treated, 27 completed radiotherapy. Eleven of the 22 assessable patients had an objective radiologic response. Only three patients completed the entire course of treatment without tumoral progression or significant toxicity. Median survival was 10.3 months, and the 1-year survival rate was 37 ± 9.5 %. As a result, the authors recommended alternative treatments be tested in patients with diffuse brainstem glioma. Bernier-Chastagner et al. [23] conducted a phase II study to evaluate the survival and toxicity in children with diffuse brainstem glioma treated with daily radiotherapy with topotecan as a radiosensitizer. Topotecan was administered intravenously at a dose of 0.4 mg/m² per day within 1 h before irradiation. All 32 patients who were included in the study completed the treatment. Only partial responses were observed occurring in 40 % of the patients. The 9-month and the 12-month survival rates were 34.4 and 25.5 %, respectively. The median duration of survival was 8.3 months. The results were not encouraging, and this treatment at this schedule and doses was not recommended for the treatment of patients with brainstem glioma. A recent study by Turner et al. [37] aimed at assessing the efficacy of administering daily thalidomide concomitantly with radiation and continuing for up to 1 year following radiation. Thirteen children, 12 with brainstem glioma, were enrolled. All patients received focal radiotherapy to a total dose of 5,580 cGy. Thalidomide was administered once daily beginning on the first day of radiation and continued for 12 months or until the patient came off study. The starting dose was 12 mg/kg (rounded down to the nearest 50 mg) and was increased by 20 % weekly, if tolerated, to 24 mg/kg or 1,000 mg. No patients completed the planned 12 months of thalidomide therapy, and all have since died of disease progression. The median duration of therapy was 5 months (range 2–11 months). The median time to progression was 5 months (range 2–11 months), and the median time to death was 9 months. With these results, adding thalidomide to radiation was not considered to improve survival comparing to historical controls; however, toxicity appeared to be increased. In another phase I study done by Greenberg et al. [24], investigators enrolled seven brainstem glioma patients in order to be treated as an induction therapy with oral etoposide, continuous infusion cyclosporine A given with escalating doses of vincristine, and concomitant standard-dose irradiation. Maintenance therapy was to be administered after induction comprising six 28-day cycles with the same drugs as in induction. All seven patients completed radiotherapy, while only three completed more than 1 month of chemotherapy because of disease recurrence. The study was closed before completion because of dose-limiting neurotoxicity. One patient had tumor necrosis at 6 weeks suggesting some antitumor effect. Median survival for the whole group was 11 months. Cyclosporine A was utilized in this study with the intension of modifying the P-gp, competitively binding to and inhibiting the efflux function of the protein, but actually no clear benefit was achieved from it but it rather increased the drug levels of neurotoxic chemotherapeutic agents in the central nervous system. Considering the fact that the use of concurrent radiotherapy with temozolomide has become the standard care for adult patients with malignant gliomas, Sirachainan et al. [38] decided to investigate the usefulness of the same agent in pediatric patients with diffuse intrinsic brainstem glioma. Twelve children were treated with concurrent radiotherapy with temozolomide followed by adjuvant temozolomide and cis-retinoic acid. Radiotherapy was administered at a conventional dose at the tumor site with temozolomide (75 mg/m²/day) for 6 weeks followed by temozolomide 200 mg/m²/day for 5 days with cis-retinoic acid (100 mg/m²/day) for 21 days with a 28-day cycle after concurrent radiotherapy. Ten of the 12 patients experienced clinical response after radiotherapy. Seven patients had partial response, four had stable disease, and one had progressive disease. At the time of the publication, 9 of the 12 patients had died of tumor progression, 1 was alive with progression, and 2 patients were alive with clinical improvement. One-year progression-free survival was 41.7 %, and the median survival was 13.5 months. One-year overall survival was 58 %. In a Children’s Oncology Group phase II study, Korones et al. [39] evaluated the efficacy of vincristine and oral etoposide with radiotherapy in children with diffuse brainstem glioma. Patients received local radiotherapy of 54 Gy. Chemotherapy consisted of two 28-day cycles of vincristine 1.5 mg/m² on days 1 and 8 and oral etoposide 50 mg/m² days 1–21 administered concurrent with radiation and continuing for ten cycles following radiation. Of the 30 children enrolled, 7 had a partial response, 18 had stable disease, and 2 had progressive disease. Response in three patients could not be evaluated. All 30 children died. Overall survival at 1 and 2 years were 27 and 3 %, respectively. Median survival was 9 months (Table 37.2).

The role of chemotherapy administration after radiotherapy has been investigated in several studies [25–27]. In one of them, Bouffet et al. [40] assessed the benefit of high-dose chemotherapy (HDC) after radiotherapy in 35 children with newly diagnosed diffuse pontine gliomas. Two to three months after radiotherapy, all patients received busulfan (150 mg/m² on days 8, 7, 6, and 5) and thiotepa (300 mg/m² on days 4, 3, and 2) followed by autologous bone marrow transplantation. Eleven patients could not receive HDC because of early progression [9] or parental refusal [2]. Three patients died of HDC-related complications. All 21 patients who survived HDC eventually died of disease progression. The median survival time was 10 months. Statistical analysis did not suggest any evidence of survival benefit. A prospective randomized Children’s Cancer Study Group trial [25] included 74 children who received either a combination of carmustine, vincristine, and prednisone or no treatment after conventional radiotherapy. The overall 5-year survival rate was 20 %, and was not prolonged with the adjuvant chemotherapy regimen compared to radiotherapy alone. Wolff et al. [27] assessed the efficacy of a combination of trofosfamide and etoposide during and after conventional radiation treatment in 20 patients with diffuse pontine gliomas. None of the 12 evaluable patients showed complete response to therapy. In three patients partial response was achieved; stable disease and progressive disease were observed in four and five patients, respectively. All tumors progressed locally, and all patients died. The overall median survival was 8 months. The authors concluded that oral trofosfamide in combination with etoposide did not prolong survival of pontine glioma patients. On the other hand, a study done by Benesch et al. [41] included 11 children with brainstem glioma who were treated with external beam radiation simultaneous with an intensive chemotherapy consisting of two cycles of ifosfamide, etoposide, methotrexate, cisplatin, and cytarabine separated by a 3-week interval. Maintenance chemotherapy with carmustine, carboplatin, and vincristine (eight cycles over a 1-year period) was given to those patients who showed clinical or radiographic response to induction chemotherapy. Six of 11 patients showed an objective reduction in tumor size on magnetic resonance imaging, and 4 of 11 were reported to be alive at the time of report >22, >22, >90, and >92 months, respectively, after diagnosis without radiographic evidence of tumor progression (one complete remission, two partial remissions, one stable disease) but had suffered from moderate to severe long-term side effects. Three patients died due to disease progression, and the tumors in four patients showed short-term stabilization, and these patients died within 1 year after diagnosis. This intensive combined modality treatment yielded objective responses in more than 50 % and long-term survival in one third of the patients that was considered encouraging when compared with the historical cases. Wagner et al. [42] reviewed the data of 153 patients with pontine gliomas treated in different prospective multicenter studies that were registered in HIT-GBM (Hirntumor glioblastoma multiforme) database. Ninety children received chemotherapy according to the HIT-GBM protocols. Conventional fractionated radiotherapy with a total dose of 54–60 Gy was a part of each induction regimen. Radiotherapy was combined with oral trofosfamide and etoposide in the HIT-GBM A protocol, with simultaneously given intensive chemotherapy comprising carboplatin, etoposide, and ifosfamide in the HIT-GBM B protocol and with further intensified chemotherapy (addition of vincristine) in HIT-GBM C protocol. The 1-year overall survival rate of all patients was 39.9 %. Favorable prognostic factors were found to be age less than 4 years, low-grade histology, and smaller tumor. As a result of the statistical analysis, chemotherapy was found beneficial in achieving a better overall survival. A group of investigators from Mexico has initiated several chemotherapy trials for the treatment of brainstem gliomas in children. Their first experience was with a combination chemotherapy consisting of BCNU (120 mgm² on day 1 every 6 weeks), procarbazine (100 mg/m² for 14 days every month for 12 months), and vincristine (2 mg/m² every 6 weeks for 12 months) given subsequent to surgery and conventional dose radiotherapy. Patients had a 5-year survival of 29 % (unpublished data). It was then followed by a new protocol in 1994 including four courses of neoadjuvant ifosfamide, carboplatin, and etoposide (ICE) followed by hyperfractionated radiotherapy and four more courses of ICE. Initial tumor reduction was observed after two courses of chemotherapy, but progressive disease was found after the fourth course. Survival was only 20 % at 18 months [43]. More recently the same group tried four courses of ICE associated with temozolomide (200 mg/m² for 5 days) every 4 weeks followed by 54 Gy hyperfractionated radiotherapy. Four high-grade brainstem tumors were treated with 30 % survival at 18 months (unpublished data) (Table 37.3).

Recently investigators in the field of cancer have introduced new therapeutic approaches such as the administration of antiangiogenic drug. Angiogenesis is a fundamental process of blood vessel growth that is a hallmark of cancer. Survival of the tumor depends on the generation of new blood vessels which assure the distribution of oxygen and the nutrients as well as the excretion of toxins. Angiogenesis is tightly regulated by proangiogenic and antiangiogenic factors. Among proangiogenic ones vascular endothelial growth factor (VEGF) is the most studied and serve as a key regulator of endothelial cell proliferation and migration. Angiopoietin, FGFs, PDGFs, and hepatocyte growth factor/scatter factor (HGF/SF) are other known proangiogenic factors. Angiostatin, endostatin, thrombospondin, and interleukin-12 are antiangiogenic factors [44]. Pathologic angiogenesis is caused by the imbalance between proangiogenic and antiangiogenic forces, a process called “angiogenic switch.” Hypoxia, acidosis, low blood glucose, mechanical stress from proliferating tumor cells, inflammatory responses, and genetic alterations of angiogenic regulators can trigger this switch [45]. Malignant glioma is one of the most vascularized tumors with expression of VEGF and high microvessel density. Increased microvessel density has been associated with poor prognosis in patients with malignant gliomas [46]. Inhibition of angiogenesis may contribute to the prevention of angiogenic switch intervening in the rapid expansion of small tumors as well as inducing the regression of large cancers [47]. More than 75 antiangiogenic compounds have entered clinical trials. Preliminary results of these trials suggest that these drugs as single agents are poorly active in advanced tumors, and responses can only be created if they are used in association with chemotherapy. Bevacizumab (Avastin) is a recombinant human neutralizing monoclonal antibody of VEGF and is the first FDA-approved antiangiogenic agent in cancer treatment [48]. It has been shown to decrease tumor vascularity, enhance tumor apoptosis, and prolong survival in a rat intracranial C6 glioma model [49]. The antitumor mechanism of bevacizumab includes decreasing vessel diameter, density, and permeability; reducing the pressure of the interstitial fluid; and in some studies increasing the intratumoral uptake of chemotherapy [50, 51]. Bevacizumab has demonstrated a promising antitumor activity when used with a topoisomerase I inhibitor, irinotecan in an anecdotal series [52]. Vredenburgh et al. has treated 32 adult patients with GBM and WHO grade III gliomas with bevacizumab 10 mg/kg every 2 weeks along with irinotecan 125 mg/m² for patients not on enzyme-inducing anticonvulsants (EIACs) or 340 mg/m² for patients on EIACs. This drug combination induced a radiographic response rate of 61 % for recurrent anaplastic gliomas. The median progression-free survival was 23 weeks for all patients. The 6-month progression-free survival probability was 38 %, and the 6-month overall survival probability was 72 %. Few patients developed venous thromboembolism, and one patient had an arterial ischemic stroke [53]. A case report on an adult patient with progressive diffuse brainstem glioma treated with bevacizumab and irinotecan demonstrated remarkable improvement in the clinical condition of the patient as well as significant radiographic response [54]. Bevacizumab has been used in children with pontine gliomas as therapy for radiation necrosis. Four children treated with irradiation later developed radiation necrosis in the brainstem. They then received bevacizumab as a treatment for the radiation necrosis. After bevacizumab therapy, three children had significant clinical improvement and were able to discontinue steroid use. One child continued to decline and, in retrospect, had disease progression, not radiation necrosis. In all cases, bevacizumab was well tolerated [55]. Another agent that inhibits VEGF by blocking ligand-receptor binding is VEGF-trap (regeneron). A clinical trial of VEGF-trap in recurrent malignant gliomas is ongoing. Recently there has been a shift in thinking towards the view that more fractionated schedules of drug administration using smaller doses than the maximum tolerated dose would be as or perhaps even more effective than the classic chemotherapy administration. Such schedules known as metronomic schedules increase the antiangiogenic activity of certain drugs [56]. Metronomic chemotherapy may delay the onset of acquired drug resistance as the target of the therapy is the endothelial cell rather than the cancer cells [56, 57]. In addition, antiangiogenic drugs do not need to cross the blood–brain barrier as they target endothelial cells. A number of agents such as etoposide, temozolomide, cyclophosphamide, and vinblastine are currently used in metronomic schedules against pediatric brain tumors [58–60]. Recently Aguilar et al. [61] described the results of their phase II study of metronomic chemotherapy with thalidomide, carboplatin, vincristine, and fluvastatin in the treatment of brainstem tumors in children. Their objective was to determine tumor response to chemotherapy combined with an antiangiogenic drug thalidomide and fluvastatin that decreases the cholesterol substrate necessary for cancer cell growth by inhibiting biosynthesis of cholesterol. Nine recently diagnosed pediatric brainstem glioma patients were included. Patients received four continuous courses of chemotherapy consisting of carboplatin on day 1, vincristine on day 1, fluvastatin orally every 24 h on days 1–14, carboplatin and vincristine same dosage on day 15, and thalidomide orally on days 15–28. One month after finishing the last course of chemotherapy, patients underwent hyperfractionated radiotherapy followed by the administration of four more courses of the same chemotherapy. There was a significant reduction in the tumor volume, and overall survival was 71.5 % after 24 months. In this trial the antitumor effect of fluvastatin could not be attributed only to ability to inhibit cholesterol biosynthesis but also its potential to inhibit Ras farnesylation. More recently fluvastatin has been reported to activate caspase-1 and induce a small secretion of IL-18 in human peripheral blood monunuclear cells. This interleukin is known to exhibit antitumor activities by activation of cytotoxic T lymphocytes and natural killer cells by production of IFN-gamma and by inhibiting angiogenesis [62]. Thalidomide was investigated as an antineoplastic drug because of its known anti-inflammatory and immunomodulatory activities through degradation of mRNA encoding tumor necrosis factor-alpha in monocytes [63]. Thalidomide has also been shown to be antiangiogenic in the rabbit cornea micropocket assay [64] involving the anti-hypoxia-inducible factor pathway [65]. Results show a poor response when used as monotherapy but potential benefit when used in combination with other drugs [65] (Table 37.4).

Despite advances in surgical-, radiation-, and chemotherapy-based strategies, malignant gliomas continue to be associated with a poor prognosis. Immunostimulants offer a novel treatment approach. Early attempts at glioma therapy based on immunostimulants failed to demonstrate effectiveness. Current immunostimulant therapies have shifted to a more multifaceted approach combining two or more different immunotherapeutic strategies. Immunotherapy, and in particular immunostimulants (also known as biologic modifiers), is an example of an area of research into novel therapy for use in high- and low-grade gliomas. Immunotherapy, or treatment that uses the body’s immune system to combat tumors, is attractive for cancer therapy for several reasons, including the conviction that it would be less toxic than the traditional cytotoxic therapies and may lead to sustained responses through immunologic memory. A group of immunostimulants commonly used are interferons. INFs are glycoproteins, which are cell-signaling molecules produced by the cells of the immune system in response to challenges such as viruses and tumor cells. IFNs assist the immune response by inhibiting viral replication within host cells, activating natural killer cells and macrophages, increasing antigen presentation to lymphocytes, and inducing the resistance of host cells to viral infection. A member of the group is interferon-beta (IFN-beta). IFN-beta exerts its antitumor effect by inhibiting glioma cells in the S phase, enhancing natural killer cell and cytotoxic T-cell activity, and perhaps synergizing with cytotoxic chemotherapies [66]. An earlier example of immunostimulatory therapy was carried out by Packer et al. who attempted to treat newly diagnosed children with brainstem glioma with recombinant beta-interferon combined with hyperfractionated radiation therapy [19]. Thirty-two children with diffuse intrinsic brainstem gliomas were included in the study who were treated with 72 Gy of hyperfractionated radiotherapy along with beta-interferon that was initially begun at 12.5 × 10⁶ IU/m² and escalated up to 400 × 10⁶ IU/m². The safe starting dose was determined to be 100 × 10⁶ IU/m². Due to unacceptable toxicity, the maintenance dose was reduced to 200 × 10⁶ IU/m². Interferon-beta was continued for 8 weeks following radiotherapy. Unfortunately 30 of the 32 patients have developed progressive disease. The median time to progression from study entry was 5 months, and the median time to death was 9 months. This therapy did not result in a higher rate of disease control. Ohno et al. [67] evaluated the effects of treatment with IFN-Beta, MCNU, and radiotherapy (IMR therapy). Another aim of the study was to determine patient response to IMR therapy by evaluating O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation in serum DNA. Fifteen newly diagnosed children with brainstem tumors were administered IFN-beta (1–2 MIU/day, days 1–7, 0.5–1 MIU/day, days 8–14) and MCNU (80 mg/m² on day 2) concurrent with conventional radiotherapy. Of the 15 patients, partial response, stable disease, and progressive disease were noted in 5 patients each. The median overall survival and the median progression-free survival were 14.7 and 4.6 months, respectively. The MGMT promoter methylation status in the serum appeared to correlate with a positive response to IMR therapy. Another IFN that was used for the treatment of glioma is IFN-gamma. IFN-gamma causes increased major histocompatibility complex expression on brain cells, class 1 antigen on local endothelial and ependymal cells, and class 2 antigen on microglial, ependymal, and perivascular cells. It also recruits lymphocytes and other inflammatory cells. These effects were accompanied by significant upregulation of tumor MHC-1 and MHC-2 expression. In addition, therapy downregulated the expression of endothelial Fas ligand, a cell membrane protein implicated as a contributor to immune privilege in cancer. It also promotes Th1 development by enhancing IL-12 secretion from macrophages and maintaining the expression of IL-12R on CD4+ cells making them more responsive to IL-12. IFN-gamma and IL-12 production appear to correlate with the induction of antitumor activity [68–71]. The proliferation, adhesion to hyaluronic acid, and the migratory capacity of the glioma cells that express IFN-gamma receptor are inhibited by IFN-gamma [72, 73]. Interferon-gamma also has significant antiangiogenic effects [74]. Wolff et al. investigated the usefulness of interferon-gamma and low-dose cyclophosphamide as a maintenance treatment in pediatric high-grade glioma patients [75]. Mitsuoka et al. first showed the immunomodulatory effect of cyclophosphamide, when given in lower doses [76]. A specific sensitivity of immunosuppressive T cells to very low concentration of cyclophosphamide has been shown in vitro [77]. The protocol in Wolf’s study [75] used two cycles of chemotherapy consisting cisplatin and etoposide in the first and cisplatin, etoposide, and ifosfamide in the second cycle. Following induction, the maintenance therapy started with IFN-gamma with a starting dose of 25 mcg/m²/day which increased by 25 mcg/m² per week up to a maximum of 175 mcg/m². When the maximum dose of IFN was reached, cyclophosphamide was initiated as a 1 h infusion of 300 mg/m² to be repeated every 21 days. Forty pediatric high-grade glioma patients were enrolled, 24 of whom had brainstem tumors. The median overall survival of 1 year was not statistically different from the historical controls. The pontine glioma patients fared even worse, and this treatment was found to have no sufficient benefit for the treatment of such patients (Table 37.5).