Epileptic seizures are common in patients with brain tumors and affect between 30–80 % of all patients with brain tumors [13]. Seizures can be potentially life threatening and can cause significant morbidity especially in patients with otherwise controlled tumors [14]. The seizure frequency in patients with glioma is generally higher in low-grade (WHO grade II) tumors (60–80 %) and less prevalent in glioblastoma (WHO grade IV) (29–49 %) [13]. Generally, patients with low-grade glioma and location in the temporal lobe, parietal lobe, or cortex are at greater risk for seizures than those with tumors in the infratentorial or white matter location. About 40 % of brain tumor patients present with a seizure as a first symptom and these patients remain at an increased risk for recurrent seizures despite treatment with antiepileptic drugs (AEDs) [15]. Despite good seizure control during treatment, up to 86 % of patients suffer from seizures at the end of life [16].

The use of antiepileptic prophylaxis for glioma patients has been investigated in multiple, mostly retrospective studies [17–20] and is currently not recommended for patients who have never had a seizure [15]. AED should be tapered off 1 week after surgery, if a patient has never had a seizure [15].

Once a patient with a brain tumor has had a seizure, long-term treatment with an AED should be strongly considered as these patients are at increased risk for seizure recurrence. It is important to consider the pharmacokinetic interactions of AEDs with anticancer therapeutics. A number of older AEDs such as phenobarbital, phenytoin, primidone, carbamazepine, and oxcarbazepine induce the cytochrome 450 (CYP450) -dependent hepatic enzymes and consequently increase their own metabolism and influence the metabolism and efficacy of many commonly used cytotoxic agents. In addition, these CYP450 inducers might affect the effectiveness of dexamethasone, which utilizes the same metabolic mechanism [21]. Valproic acid, however, is a CYP450 inhibitor that might increase therapeutic drug levels of antineoplastic agents and might have other antineoplastic properties. There have been several reports of the in vitro as well as in vivo antineoplastic activity of this histone deacetylase inhibiting antiseizure medication (reviewed in Weller et al.) [22]. A retrospective (and unfortunately insufficiently powered) analysis of patients participating in the EORTC/NCIC temozolomide trial for glioblastoma revealed prolonged survival with use of valproic acid [23]. On the other hand, another retrospective analysis of patients in the North Central Cancer Treatment Group trial came to the opposite conclusion. In that study, patients taking enzyme-inducing AEDs had increased survival [24]. Currently, the true impact of valproic acid on brain tumor survival remains unclear.

Newer AEDs such as levetiracetam, lacosamide, and zonisamide are not influenced by CYP450 and other metabolic pathways and therefore do not interact with other agents utilizing these pathways. Side effects are more frequent and pronounced in brain tumor patients when compared to the general epilepsy population [15, 25]. In the setting of brain tumor patients undergoing active treatment, levetiracetam is the most studied of the newer generation antiepileptic medications. According to several studies it has been well tolerated and safe, but possible side effects include neurocognitive deficits and psychiatric effects [26, 27].

Seizure management in the end-of-life phase of brain tumor patients is crucial given the high frequency and the fact that epileptic events have been associated with nonpeaceful death [28]. The optimization in this setting is challenging due to the fact that many brain tumor patients are unable to swallow or to take oral medications due to changes in mental status [8]. If oral or intravenous application is not warranted, a number of seizure medications can be supplied intramuscularly, subcutaneously, rectally, or via buccal or intranasal application (See Anderson et al. for details) [29].

Diazepam especially can be delivered as rectal suppository and the buccal or intranasal application of midazolam has also been shown to stop seizures [30]. All these recommendations are either extrapolated from other studies or follow general guidelines for the treatment of seizures as there are no prospective trials investigating seizure management in the end-of-life phase of glioma patients.

Fatigue is defined as a persistent sensation of physical, cognitive, or emotional tiredness that is not linked to any recent activity and that impairs normal function [31]. Fatigue is a well-recognized problem in oncology and is a prevalent as well as impairing symptom in brain tumor patients [32]. Patients report impairment due to fatigue throughout the course of the disease, but it is most prevalent at the time of radiation therapy. More than 80 % of patients with primary brain tumors report fatigue at the time of radiation therapy [33]. Symptoms especially appear to increase toward the end of the 6-week radiation treatment and can last after radiation has been discontinued. Fatigue is not strictly related to tumor grade and one study showed that 39 % of patients with a low-grade glioma continued to report severe fatigue more than 8 years after finishing therapy [34]. While the exact mechanism for cancer-associated fatigue remains unclear, factors associated with a higher risk for fatigue include older age, female sex, decreased performance status, and treatment-related factors [32, 35–37].

All patients with fatigue should undergo screening for depression, which can affect between 16–39 % of patients with brain tumors [38]. Prior to initiating any fatigue-specific treatment, factors such as pain, anemia, sleep issues, metabolic problems such as thyroid dysfunction and low Vitamin D and Vitamin B12 levels, and malnutrition should be ruled out. A critical review of the patient’s medication list should be performed as some frequently used medications such as antiseizure medications and corticosteroid taper are also known to cause symptoms similar to cancer-induced fatigue. After addressing these factors, general strategies to manage fatigue should be implemented (please see Armstrong TS et al. for an excellent review) [35]. Patients are encouraged to implement energy conservation techniques such as prioritization of activities and delegation of more energy intensive activities. Extensive naps during the day should be avoided to maintain the normal sleep cycle. A variety of nonpharmacological interventions have been evaluated in other solid tumors such as breast cancer. Unfortunately, there is a lack of validated interventions for brain tumor-associated fatigue and especially exercise-based regimens are often of limited value due to focal neurological deficits. Two large meta-analyses of randomized controlled trials evaluating exercise for cancer-related fatigue came to the conclusion that exercise resulted in clinically relevant improvement and might be effective in treating fatigue [39, 40]. Successful exercise programs included a focus on walking or resistance training which can only be applied to a select group of brain tumor patients without paralysis or weakness. Other nonpharmacologic interventions might include psychosocial programs focusing on education, stress management, or cognitive-behavioral interventions [40].

Integrative medicine approaches such as yoga have been successfully employed in lowering self-reported levels of fatigue [41] and in another study had a positive impact on inflammation markers, fatigue and vitality [42]. The use of acupuncture is also frequently mentioned by patients, but conclusive data describing the impact on cancer-related fatigue remains lacking [43, 44].

Many pharmacologic interventions such as hemopoietic growth factors, corticosteroids, antidepressants, and psychostimulants have been evaluated in the general solid tumor population, but only the use of stimulants such as methylphenidate and modafinil have been evaluated specifically in glioma patients [35, 45]. While earlier open-label studies with methylphenidate and modafanil showed promising cognitive and functional improvement [46, 47], later randomized-blinded trials were not able to duplicate these findings [48, 49].

Fatigue is a prevalent and disabling symptom in primary brain tumor patients with high impact on function and quality of life. Due to the lack of conclusive data for brain tumor patients, interventions must be chosen individually with consideration of possible side effects.

Another communication challenge that clinicians who care for patients with glioma face is navigating discussions about goals of care when the disease has progressed and the burdens of further anticancer therapy are beginning to outweigh the benefits. When oncologists discuss the possibility of discontinuing chemotherapy, they often have feelings of guilt and failure that they were unable to rescue the patient from impending death [70]. In addition many clinicians worry that discussing end-of-life care will take away hope; however, most patients want detailed information about their illness and what to expect [71]. Patients and families do want clinicians to convey empathy, support, and hope [72]. It is helpful to realize that an entire spectrum of hope can exist and evolve over time, including hope for cure, hope for living longer, hope for having time with loved ones, and hope for having a peaceful death [73]. Patients want to establish relationships with clinicians who see them as individuals [74], and want to trust clinicians with whom they discuss end-of-life concerns [75]. A useful roadmap to meet these patient and family needs in discussing goals of care late in the disease course is outlined by the acronym “REMAP” [77]. (Table 3) For patients with glioma, these later conversations will likely often occur with surrogate decision makers. A key skill is to ask questions to elicit the patient’s big-picture values prior to discussing specific treatment plans like stopping chemotherapy or enrolling in hospice care [70], e.g., “Did your wife ever talk about what we should do if she got a lot sicker and it looked like time was short?” Eliciting the patient’s values will then allow clinicians to recommend a care plan that is tailored to meet the patient’s goals.