Aggressive and recurrent oligodendroglial tumors are challenging to treat. Although extensive surgical resection has been shown, in two prospective studies, to be associated with increased survival (Cairncross et al. 2006; van den Bent et al. 2006), further adjuvant treatment (radiation or chemotherapy) is required in patients with these higher-grade tumors to prevent recurrence (van den Bent 2007).

Since oligodendroglial tumors were found to be sensitive to PCV chemotherapy in the mid 1990s (Cairncross et al. 1994; van den Bent et al. 1998), there has been increased interest in optimizing adjuvant chemotherapy regimens for these tumors. Most prospective uncontrolled single arm studies evaluating chemotherapy for recurrent oligodendroglial tumors have involved PCV or TMZ chemotherapy. While grouping of oligodendroglial tumors with other gliomas in clinical trials makes it difficult to draw conclusions regarding oligodendroglial tumors unequivocally, two trials specifically evaluated therapies for aggressive oligodendroglial tumors. A Radiation Therapy Oncology Group (RTOG) Phase III trial for grade III oligodendroglioma randomized 289 patients to receive either radiation alone or neoadjuvant dose intense PCV chemotherapy followed by radiation. With 3-year follow-up the median survival times were similar (4.9 years in the group receiving PCV + radiation therapy versus 4.7 years in the group receiving radiation therapy alone). However, PFS time was increased in the group receiving PCV (2.6 years) compared to radiation therapy alone (1.7 years, P = 0.004). Unfortunately, 65% of patients receiving chemotherapy experienced grade 3 or 4 toxicity and one patient died (Cairncross et al. 2006). In another similar multicenter randomized controlled trial, 368 patients with anaplastic oligodendroglioma were randomized to receive either radiation therapy alone, or radiation followed by standard PCV chemotherapy. Results in this trial were similar, showing no significant difference in median survival, but increased PFS in the group receiving PCV (van den Bent et al. 2006). These studies suggest that, although chemotherapy in addition to radiation can prolong PFS in patients with aggressive oligodendroglial tumors, it is associated with significant toxicity that limits dose escalation.

Because the PCV regimen is associated with significant hematologic and GI toxicity, and most patients do not tolerate the six cycles intended, TMZ has developed a role as therapy for aggressive oligodendroglial tumors. Compared to PCV chemotherapy, TMZ is a better-tolerated oral agent with modest myelosuppression and easily controlled nausea and vomiting. The reported response rate of recurrent oligodendroglioma to TMZ after failure of radiation therapy is up to 50% with a median PFS of 10–12 months (van den Bent et al. 2003). Although no formal comparison between PCV and TMZ in recurrent oligodendroglial tumors is yet available, TMZ is often employed in practice because it is better tolerated by patients and more easily administered (van den Bent 2007).

For cases of aggressive oligodendroglial tumors that are resistant to PCV or TMZ chemotherapy, few treatment options are available. These patients are often poor candidates for additional surgery due to tumor size and location, often involving eloquent brain tissue. And, as mentioned earlier, simple dose escalations are limited by systemic toxicities. Improved innovative therapy with low toxicity is required for treatment of these refractory tumors.

Many innovative approaches have been investigated with the goal of increasing direct tumor cytotoxic effects while avoiding non-specific systemic toxicities. Myeloablative chemotherapy with autologous peripheral blood progenitor cell rescue is one approach that was developed in order to achieve higher dose intensity and to avoid the toxicities associated with radiation therapy. For treatment of aggressive oligodendroglial tumors, this strategy was investigated several years ago (Mohile et al. 2008). In this study, 20 patients with aggressive oligodendroglial tumors were treated with four cycles of PCV chemotherapy every 6 weeks. Of the 20 treated patients, 15 demonstrated a response. The 15 responders were treated with myeloablative chemotherapy followed by autologous peripheral blood progenitor cell rescue. Fourteen patients underwent transplantation, with median disease-free and overall survival of at least 36 weeks (not yet reached at time of report). Although this approach allowed deferral of radiation for at least 3 years, a major limitation was the acute toxicity associated with both the induction and consolidation regimens, making this a poor option in most cases.

Administration of a chemotherapeutic agent directly into an artery was first investigated as a method to treat brain tumors more than 50 years ago (Newton 2006). The aim of this approach is to achieve higher dose intensity of chemotherapeutic agent directly to tumor cells while achieving a low systemic exposure, avoiding many of the systemic toxicities that occur with dose escalations. This strategy increases the intra-tumoral cellular concentration of chemotherapeutic drug and subsequently can improve tumor cell death. Because primary and metastatic brain tumors are localized within brain tissue and receive an arterial blood supply that can be accessed using standard interventional techniques, the regional strategy of IA therapy is ideally suited.

The benefits of IA over IV administration of a chemotherapeutic agent, from a pharmacology viewpoint, occurs with the “first pass” of drug through the brain and tumor circulation, resulting in a higher concentration of drug delivered to brain and tumor tissue (Eckman et al. 1974). After the first pass, the drug continues to circulate through the blood until it is cleared. Certain drugs with a high extraction fraction, or high lipid solubility, maximize the first pass effect when delivered using this approach. Chemotherapeutic drugs with this favorable pharmacologic profile also possess rapid systemic metabolism and/or excretion. Evidence from several studies suggests that nitrosourea drugs, such as carmustine, are ideal for IA administration. In one early study, up to fivefold delivery of 14C-labeled carmustine was demonstrated with IA administration compared to IV in monkeys (Levin et al. 1978). In another study, super selective IA infusion of 11C-labeled carmustine into the middle cerebral artery of patients with recurrent gliomas resulted in up to a 50-fold increase in drug delivered compared to IV administration (Tyler et al. 1986).

Chemotherapeutic agents possessing a rapid total body clearance are more appropriate for IA delivery than those with slower clearance (Eckman et al. 1974). In this regard, the drugs most appropriate for IA chemotherapy include BCNU, cisplatin, carboplatin, and etoposide. Unfortunately, the two drugs most effective for IA delivery, BCNU and cisplatin, also have the most significant neuro-toxicity. Although there are pharmacological differences making one agent more appropriate for IA delivery than another, in general, administration of chemotherapy via the IA route for the treatment of brain tumors in both animal studies and clinical trials has demonstrated a higher concentration of drug to the tumor and brain surrounding tumor (Dropcho et al. 1992; Cloughesy et al. 1997) with less associated systemic toxicity compared with IV administration. Several investigators have assessed the toxicity and/or efficacy of specific antineoplastic agents administered IA to treat many types of brain tumors (Dropcho et al. 1992; Mortimer et al. 1992; Shapiro et al. 1992; Stewart et al. 1992). Experience with IA chemotherapy in patients with high-grade gliomas is extensive (Newton 2006). In studies treating patients with newly diagnosed gliomas using IA chemotherapy, administration has typically occurred just prior to or in combination with radiation. Single agents that have been used in this context include carmustine, nimustine, HeCNU, cisplatin and 5-fluorouracil. In a review of 15 studies evaluating a total of 395 patients, the median time to progression (TTP) of newly diagnosed patients receiving single-agent IA chemotherapy ranged from 12 to 32 weeks, with a median survival of approximately 1 year (range of 32–73 weeks) (Newton 2006). Combination IA chemotherapy in patients with newly diagnosed gliomas, in which one or more drugs were administered via the IA route has yielded similar results. In a review of eight studies, evaluating a total of 542 patients, the median TTP of combination IA regimens ranged from 33 to greater than 50 weeks and the median survival for combination regimens is approximately 1 year, with a range of 40–228 weeks (Newton 2006).

Naturally, in addition to treatments for newly diagnosed gliomas, single agents administered by the IA route have been studied as therapy for previously treated gliomas that were refractory to standard therapies. The majority of published studies evaluated either a nitrosourea derivative or a platinum analog. In a review of data from 18 studies, involving a total of 348 patients, median TTP ranged from 13 to 40 weeks in responders, and median survival ranged from 8 to 50 weeks in responders. Percent response ranged from no response to 80%. IA chemotherapy approaches utilizing multiple IA agents or the combination of IA agent with oral or IV drug has also been extensively evaluated for recurrent gliomas. Most regimens focused on the use of IA carmustine in combination with other drugs. Data from 11 studies, with a total of 291 patients showed percent response ranging from NR to 68%, with median TTP of 14 to 24 weeks, and median survival of 14–56 weeks. Taken together, these extensive studies in patients with high-grade gliomas provide hope for the use of IA delivery in general for treatment of glial tumors. None of these studies evaluated IA therapy specifically for oligodendroglial tumors.