Abstract
Patients with primary brain tumors may present with neuropsychiatric symptoms such as behavioral and personality changes at any point during the disease course. Symptoms may be due to tumor treatment, the disease itself, or due to therapy for symptoms such as seizures. Levetiracetam is an attractive choice of antiepileptic medication because of relatively fewer drug interactions and may be administered through the subcutaneous route. Aggressive behavior in advanced brain tumors can have a profound negative impact on patients and their families and carers, particularly when deteriorating toward death. We report a case of neuropsychiatric symptoms in progressive glioblastoma multiforme while on stable levetiracetam doses, which improved with reduction and cessation of the drug. We review the existing literature on seizures in primary brain tumors, their antiepileptic drug management, and the risk of consequent neuropsychiatric adverse events.
Background
Neuropsychiatric symptoms such as cognitive dysfunction, behavioral, and personality changes are common problems for people with primary malignant brain tumors. 1 These issues may be the result of the tumor itself or a complication of treatment including medications. For example, this group of people are very regularly prescribed medications with the aim of improving their condition such as dexamethasone that paradoxically may worsen some behavioral problems. 2
Another issue that very commonly requires attention in primary brain tumors is seizures. Many will require antiepileptic medications either prophylactically to prevent seizures or to maintain seizure control. However, these medications may also be associated with unwanted adverse effects that detrimentally affect quality of life. 3 Because of its perceived relatively benign side effect profile, levetiracetam is increasingly recommended both as an antiseizure agent for these patients 4 and as an agent to manage seizures at the end of life. 5
Like any distressing symptoms, the development or worsening of neuropsychiatric symptoms in patients with primary brain tumors requires the underlying cause to be considered. Although such changes are often the result of the primary disease, there may be other contributing factors such as medications or new seizures. We report a case of a man with an advanced primary brain tumor who developed behavioral and personality changes while on regular levetiracetam.
Case Report
A 57-year-old man with incurable progressive glioblastoma multiforme (GBM) was referred to the hospital consultation specialist palliative care service for symptom control. He had previously been diagnosed with a left temporal lobe GBM after presenting with new onset generalized seizures. His initial management included incomplete surgical debulking followed by combination radiotherapy and chemotherapy. He then underwent a trial of immunotherapy. Approximately seven months after his initial diagnosis, staging scans demonstrated tumor progression and he was recommended again for surgery. The second craniotomy was complicated by right-sided weakness and expressive dysphasia. Throughout this time, he was considered at high risk of seizures and hence maintained on stable doses of levetiracetam.
Approximately six weeks after his second surgical resection, he developed new personality changes characterized by intermittent agitation and aggression. These behaviors escalated over one week to the point where his wife and two young adult sons could no longer manage him at home necessitating acute hospital admission. Cerebral computerized tomography and magnetic resonance imaging scans confirmed extensive left cerebral changes with associated mass effect. Compared with previous scans, despite the postoperative changes, the scans were reported as consistent with significant disease progression. Lengthy discussions with the family resulted in consensus that the goal of care would be to optimize symptom control. This included increasing dexamethasone from 4 to 16 mg per day. All agreed the patient's life expectancy was short, and he was unlikely to benefit from any further treatments. Although he had not experienced any further seizures from the time of his diagnosis, he continued to be considered at high risk of problems and levetiracetam was continued.
During the initial palliative care consultation, he was observed to be agitated and impulsive. Despite this, he was able to communicate with the team denying headaches and other distressing physical symptoms. Over the next five days, he became more agitated. His wakefulness fluctuated consistent with a delirium; repeated clinical assessment did not reveal a reversible cause. Because of this and his overt behaviors, regular olanzapine was prescribed but despite dose titrations, it failed to settle him. His family expressed their angst as they witnessed dramatic changes in his personality coupled with aggressive outbursts. His family found his behavior was so distressing and out of character that, despite his limited prognosis, they started to limit their visits. Psychiatry services were consulted to assist with his management and their initial recommendation was to wean and cease levetiracetam. Over the next few days, a marked improvement was seen in his behaviors, allowing olanzapine to be ceased and his family to re-engage. He was subsequently commenced on low-dose clonazepam as an anticonvulsant and remained settled until the time of his death one week later.
Discussion
There are numerous reasons that this patient may have been agitated including disease progression and concomitant use of dexamethasone. However, there is little doubt that temporally, his agitation settled when levetiracetam was ceased. This is an important observation as levetiracetam is now routinely recommended to control seizures in brain tumors,6,7 and increasingly recommended as an appropriate antiseizure agent at the end of life.5,8
There are a number of reasons for these recommendations. Levetiracetam is one of the least sedating antiepileptics and it does not require drug monitoring. 9 Case reports suggest it is well tolerated when administered through subcutaneous infusion. 10 Furthermore, levetiracetam has few reported drug interactions with no effect on CYP450 enzymes. 11 Although these characteristics make it attractive for use in palliative care and end-of-life care, there is still insufficient data to support this as routine practice. This is especially important when there have been reports of distressing adverse effects of behavioral problems associated with levetiracetam in the management of both primary epilepsy and brain tumor-related epilepsy.
In more detail, levetiracetam in primary epilepsy has resulted in a number of people with pre-existing behavioral problems developing nervousness, hostility, and anxiety. 12 Furthermore, pharmacovigilance work has compared the real-life tolerability of four different antiepileptic agents. Of the four agents studied, levetiracetam (n = 151), lamotrigine (n = 167), valproic acid (n = 73), or controlled-release carbamazepine (n = 47), levetiracetam was reported as independently associated with anger/aggression, nervousness/agitation, upset stomach, depression, and sleep disturbance. Minor adverse effects were noted in 76.8% (n = 116) and more significant adverse effects in 23.2% (n = 35). 13 There is limited information on the timing of behavioral changes although a case series in epilepsy reported a delay in months from starting levetriacetam to the recognition of adverse symptoms. 14
The data on primary epilepsy regarding levetiracetam's adverse effects must give pause when the high rates of behavioral problems experienced by people with gliobastomas are considered. Up to 40% of this cohort will develop seizures at diagnosis and a further 15% develop seizures with progressive disease. 15 Nearly 50% are at risk of developing behavioral problems such as anger and inappropriate actions. 1 These figures highlight this group's need for effective seizure management as poorly managed seizures both negatively affect quality of life 13 and may result in more significant behavioral problems. 1 Seizure management in this group is challenging as the precise nature of epileptogenesis in brain tumors remains unclear. Numerous factors are believed to contribute including specific tumor types, biological factors, and tumor location. As previously mentioned, ∼50% of GBMs present with seizures with a further spike linked with progressive disease. Other issues include, but are not limited to, changes in the permeability of the blood–brain barrier and the peritumor environment. 15 Lastly, structural and metabolic changes are common in this group including tumor recurrence; disease progression, as a consequence of cancer treatments including radionecrosis; and other factors such as infections or metabolic disturbances. 16 Brain tumor-related epilepsy is a distinct and separate entity to primary epilepsy made even more complicated as it is suggested that these patients are at risk of abnormal mechanisms that bypass seizure control medications. 17 Compared with primary epilepsy, those with brain tumors are more likely to develop adverse effects from antiepileptic medications. Approximately 25% of patients with brain tumors will require a medication switch due to adverse effects compared with 10% of patients with primary epilepsy. 16 Focusing again on levetriracetam, a higher prevalence and magnitude of neuropsychiatric dose-independent adverse effects have been observed with its use compared with other agents. 18 Notably, the effects did not seem to be related to tumor localization in the levetriacetman group compared with other agents wherein problems were most likely to occur when people had frontal lobe tumors.
Collectively, the observations briefly summarized here support our own clinical experience that although an antiepileptic agent was strongly recommended for our patient, continuing levetiracetam may have contributed to his behavioral problems. More research is required to confirm this association and further discussions are welcome. Levetiracetam, like many medications, will have initially been evaluated in more robust participants compared with this man. This reinforces the benefits of postmarketing phase IV studies to enable clinicians to recommend interventions for specific patients, especially those who are less robust. This is particularly important when caring for people at the end of life with this cohort increasingly recognized as a distinct group at high risk of adverse drug effects. 19 Above all else, such work improves the quality of care when quality care is defined as evidence based, patient centered, and low risk.
Footnotes
Funding Information
No funding was received.
Author Disclosure Statement
No competing financial interests exist.
