Dexamethasone as an antiemetic is either contraindicated or strongly discouraged in children with brain tumors due to the theoretical concern of decreased penetration of chemotherapy agents into the central nervous system. On the other hand, dexamethasone is considered to be the drug of choice to provide temporary relief of symptoms due to an increase in intracranial pressure and edema in children with a brain tumor. It stabilizes the blood–brain barrier (BBB), leading to attenuation of vasogenic brain edema. The mechanism of action of glucocorticoids for control of vasogenic edema is not fully understood. Dexamethasone, most frequently used glucocorticoid, has recently been shown to upregulate Ang-1, which is a strong BBB-stabilizing factor, whereas it downregulates VEGF, a strong permeabilizing factor in astrocytes and pericytes [19]. Dexamethasone may also increase the clearance of peritumoral edema by facilitating the transport of fluid into the ventricular system from which it is cleared by cerebrospinal fluid (CSF). Generally, the dose of dexamethasone required to treat brain edema is much lower than the dexamethasone antiemetic dose.

Treatment of brain tumors is limited by the ability of therapeutic drugs to cross the blood–brain barrier (BBB). The BBB is damaged in patients with a brain tumor, thus facilitating penetration of cytotoxic drugs into tumor tissue. Corticosteroids, by repairing damage to the BBB, may reduce the delivery of chemotherapeutic agents to the brain. This restoration of the BBB may decrease uptake of cytotoxic agents into the brain tissue, reducing the efficacy of chemotherapy in the treatment of a child with a brain tumor [20].

Brain around tumor (BAT) is a transitional zone where the brain tissue is infiltrated with tumor cells. Chemotherapy drugs that are unable to cross the intact BBB show increased penetration into BAT when compared to normal brain tissue though decreased compared to tumor thereby demonstrating a partially damaged BBB in BAT. In a rat model, the concentration of platinum was significantly lower in BAT treated with cisplatin and dexamethasone compared with the rat treated with cisplatin alone, indicating that the efficacy of cisplatin in patients with a brain tumor may be impaired with commitment administration of dexamethasone [21].

Steroids have also shown to decrease the concentration of intravenous contrast media in the tumor and the area around the tumor on imaging studies including computed tomography and MRI, reflecting stabilization of blood tumor and peritumoral BBB [22]. Also, in several carcinoma cell lines, dexamethasone has been shown to enhance resistance to radiotherapy and chemotherapy [13]. This induction of resistance to chemotherapy and radiotherapy by dexamethasone may result in suboptimal clinical responses in children with brain tumors. In view of the abovementioned evidence, dexamethasone use as an antiemetic should be avoided, if possible, in children being treated for a brain tumor.

The child being treated for a brain tumor will often receive craniospinal radiation, surgery, and chemotherapy. Radiation to the brain often causes nausea and vomiting that can be managed with the use of a 5-HT3 antagonist given prior to radiation. The chemotherapy regimens usually used to treat brain tumors are often platinum based and hence are highly emetogenic. As dexamethasone is discouraged, it is very important to maximize alternative antiemetics.

The development of the 5-hydroxytryptamine (5-HT₃) receptor antagonists has greatly improved the control of chemotherapy-induced emesis and these agents are considered the gold standard. However, 5-HT₃ receptor antagonist as single agent leads to poor CINV control in patients receiving moderately emetogenic chemotherapy (MEC) and highly emetogenic chemotherapy (HEC) (Table 22.1) [17]. The recently published POGO guidelines recommend 5-HT₃ receptor antagonist administered together with nabilone or chlorpromazine in whom corticosteroids are contraindicated [16].

Ondansetron is the most studied 5-HT₃ antagonist and is considered an effective first-line antiemetic in children undergoing chemotherapy, radiotherapy, and surgery. Ondansetron is generally well tolerated in children with minimal toxicity. Ondansetron, granisetron, dolasetron, and tropisetron are all 1st-generation 5-HT₃ antagonists and have been compared in various studies. First-generation 5-HT₃ antagonists (ondansetron, granisetron, or tropisetron) have comparable efficacy in preventing CINV in cisplatin-based chemotherapy when administered with dexamethasone [20]. The efficacy of the intravenous and oral routes of ondansetron is equivalent, and recent studies have demonstrated that the oral formulations are safe, effective, and probably less expensive than the intravenous formulations during non-cisplatin-containing moderately and highly emetogenic chemotherapy [23]. The recommended dose of ondansetron for children receiving moderately or highly emetogenic chemotherapy is 5 mg/m²/dose (0.15 mg/kg/dose) IV/PO pre-therapy x 1 dose and then q8h for highly emetogenic chemotherapy and every 12 hourly for moderately emetogenic chemotherapy. The ondansetron dose for children receiving chemotherapy with low emetogenicity is 10 mg/m²/dose (0.3 mg/kg/dose; maximum 16 mg/dose IV or 24 mg/dose PO) pre-therapy × 1 dose [16].

Granisetron is a specific 5-HT₃ antagonist, which is well tolerated with the common adverse effects reported being headache, constipation, diarrhea, asymptomatic arrhythmia, and transient elevation of serum transaminase. The intravenous and oral routes are equally efficacious. The usual intravenous dose is 40ug/kg as a single daily dose for chemotherapy of any emetogenicity [24].

A genetic polymorphism has been identified that affects patients’ responses to antiemetic therapy following moderate to highly emetogenic chemotherapy. The 5-HT₃ receptor antagonists are metabolized by the cytochrome P450 (CYP) enzymes, and patients who have been identified as ultrarapid metabolizers of the isoenzyme CYP2D6 have a significantly higher frequency of vomiting within the first 24 h after chemotherapy with the administration of ondansetron. Alternatively, granisetron is metabolized exclusively by the CYP3A subfamily and so would not be expected to have diminished effect in rapid metabolizers of CYP2D6 and would potentially be an alternative agent in these patients. Conversion to an alternate 5-HT₃ antagonist is a simple intervention in patients who do not respond to initial 5-HT₃ antagonist and benefit may be observed in 40–50 % of patients; however, genetic polymorphism may not be the only factor responsible [13].

Palonosetron is a newer 2nd-generation 5-HT₃ antagonist with a 40-h half-life in adults compared to the 4–8 h for the 1st-generation antagonist. With its longer half-life, this agent shows promise in the control of both acute and delayed nausea and vomiting [20].

Nabilone is a synthetic cannabinoid used in control of CINV. Preliminary animal studies showed that nabilone had significant antiemetic activity against drugs potent in causing emesis, such as cisplatin and carmustine with its primary action on the medulla oblongata and some secondary anxiolytic activity medicated through the forebrain. Nabilone has been reviewed in a meta-analysis concluding that cannabinoids were slightly better at controlling CINV than prochlorperazine, metoclopramide, domperidone, or haloperidol [25]. The antianxiety properties of nabilone may contribute to its antiemetic effect. In a randomized, double-blind, crossover trial, nabilone was compared to prochlorperazine for control of CINV in children aged between 3.5 and 17.8 years. Major side effects commonly experienced during nabilone administration were dizziness, drowsiness, and mood alteration. Dose reduction improved the symptoms without interfering with the efficacy of nabilone. A higher proportion of patients experienced control of CINV during nabilone treatment compared to prochlorperazine and more patients preferred nabilone [26].

Despite the lack of significant supporting evidence, the use of nabilone is recommended in combination with a 5-HT₃ antagonist in patients in whom corticosteroids are discouraged. Nabilone is only available in an oral formulation which may restrict its use in the child unable to swallow. The safety and efficacy of nabilone in children less than 4 years has not been established. Generally, nabilone is more effective in adolescent and older children who are refractory to standard prophylaxis [13].

Metoclopramide was one of the first antiemetics used in clinical practice. It was initially thought to be a dopamine antagonist acting centrally in the chemoreceptor trigger zone and peripherally promoting gut motility but was later also found to be a week 5-HT₃ receptor antagonist [27]. Prospective trials have evaluated the efficacy and side effects of metoclopramide versus 5-HT₃ receptor antagonists administered to control CINV in children receiving antineoplastic therapy. The ability of metoclopramide is marginal in the setting of highly emetogenic antineoplastic therapy. Extrapyramidal reactions (EPR) have been noted in patients treated with metoclopramide. Diphenhydramine is an anticholinergic drug that acts on the cholinergic receptors to prevent or lessen the dystonia caused by metoclopramide [27]. Metoclopramide may have a role in children receiving moderately emetogenic antineoplastic therapy for whom corticosteroids need to be avoided [16].

Chlorpromazine has been evaluated in a single randomized trial and reported to have only moderate success in providing complete vomiting control in children receiving highly emetogenic antineoplastic therapy [16]. There is a limited published experience with chlorpromazine for CINV; however there is extensive general pediatric experience with use of chlorpromazine. As an antiemetic, its efficacy has not been evaluated in combination of 5-HT₃ antagonist. Despite lack of evidence, the use of chlorpromazine in combination with a 5-HT₃ antagonist may be considered for children who cannot receive dexamethsoane [16]. Sedation and hypotensive effects should be given consideration while using in the outpatient setting.

For the treatment of anticipatory nausea or in children who experience breakthrough nausea and vomiting, benzodiazepines can be prescribed. It should be noted that benzodiazepines, like lorazepam, cause sedation and should be used cautiously with other agents that cause sedation, e.g., chlorpromazine. The antiemetic effect of lorazepam is thought to be independent of its sedative effect and it is believed to act by inhibiting input to the vomiting center [13].

For breakthrough nausea and vomiting, there is a lack of evidence to give specific directions regarding management. The first step will be maximization of the current antiemetic regimen. For example, if the patient is on low-dose 5-HT₃ antagonist prophylaxis, the 5-HT₃ antagonist dose should be maximized. If the patient is already on highest antiemetic dose, then addition of other antiemetics should be considered. The choice of other antiemetics will depend on need and circumstances of individual patients [13].