Impact of Disease and Treatment on Health-related Quality of Life

Both the disease and its treatment can affect HRQOL. Common disease-specific symptoms that may negatively influence HRQOL include paresis, sensory loss, visual-perceptual deficits, cognitive deficits, and symptoms associated with increased intracranial pressure, such as nausea and headache. Moreover, fatigue and mood issues can contribute to poorer HRQOL. Patients with newly diagnosed GBM experience significantly lower levels of HRQOL than healthy controls, shortly after surgery. Compared with other neurologic patient groups, HRQOL among patients with stable high-grade glioma is similar, but patients with GBM report less positive affect, more depression, and more illness intrusiveness than other patients with cancer.

The impact of antitumor treatment (surgery followed by radiotherapy and/or chemotherapy) on HRQOL can be both negative and positive. Surgery can either improve HRQOL by reducing tumor mass and the disease-specific symptoms associated with increased intracranial pressure, or it can decrease HRQOL by damaging functional tissue. In patients with low-grade glioma, treatment with radiotherapy has generally been associated with a negative impact on HRQOL, because increased fatigue and cognitive deficits can hinder everyday functioning. These late toxic effects of treatment play a less prominent role in patients with high-grade glioma, who have a poorer prognosis. On the group level, the HRQOL of patients with high-grade glioma does not seem to be negatively influenced by a treatment regimen consisting of radiotherapy and chemotherapy. However, specific aspects of HRQOL, such as social and cognitive functioning, may be affected temporarily. In long-term survivors (>2.5 years) of anaplastic glioma treated with radiotherapy and chemotherapy, lower levels of motor functioning and social, cognitive, and emotional aspects of HRQOL have been reported. The side effects that occur with chemotherapy, such as nausea or vomiting, appetite loss, and drowsiness, can also negatively influence patients’ functioning and HRQOL.

Adding bevacizumab, an antiangiogenic agent that may benefit progression-free survival in a subgroup of patients with recurrent GBM, does not seem to affect HRQOL, although one study reports a decrease in HRQOL. Other medications commonly administered to manage symptoms, such as antiepileptic drugs (AEDs) and corticosteroids, can influence HRQOL. Decreasing seizure frequency may improve HRQOL, but side effects or drug-drug interactions may decrease HRQOL. However, second-generation AEDs such as levetiracetam and oxcarbazepine cause fewer side effects and do not seem to significantly influence HRQOL in the longer term. Similarly, corticosteroids may improve HRQOL by relieving symptoms of increased intracranial pressure but can also cause adverse effects that may harm HRQOL. Therefore, it is advised to administer corticosteroids in the lowest effective dose to minimize side effects.

Of note, the HRQOL results of studies including patients with GBM are often biased, because those patients participating in trials are often young and have a good performance status. In addition, patients who show clinical deterioration are more frequently lost to follow-up. Thus, HRQOL could be overestimated, especially in studies with long follow-up. As GBM recurs or the disease progresses, HRQOL seems to become worse. However, in long-term survivors of GBM (ie, survival ≥16 months) HRQOL can return to a level comparable with normal controls. Despite the poor prognosis, patients often long to return to a normal everyday life, including return to work and social activities after primary treatment. It is especially during this phase of life that HRQOL may be correlated with cognitive functioning, because even subtle cognitive deficits might hamper patients’ autonomy and professional life ( Fig. 21.1 ).

The relationships between cognition, health-related quality of life, and quality of survival in patients with GBM.

The Association Between Health-related Quality of Life and Neurocognitive Functioning

Almost all patients with GBM experience cognitive deficits, which can include dysfunction in the domains of information processing, attention, psychomotor speed, executive functioning, and verbal and working memory. Shortly after diagnosis, memory storage and retrieval, categorical word fluency and also information processing capacity and more complex executive functioning tasks seem to be most commonly affected. In general, cognitive functioning declines as the disease progresses, although long-term survivors (>3 years) may only show mild cognitive impairment.

Cognitive functioning is part of the concept of HRQOL, because even mild deficits may affect patients’ ability to function independently, affecting their social and emotional functioning to a great extent. When severe, cognitive deficits can also affect patients’ ability to accurately report on their HRQOL, complicating the association between cognition and HRQOL. Studies on the relationships between HRQOL measures and objective cognitive functioning throughout the GBM disease trajectory are rare, although some evidence exists. A recent longitudinal study showed preoperative cognitive symptoms to be predictive of HRQOL during the first year of GBM. In patients with newly diagnosed GBM, information processing speed was associated with mental health. Moreover, in patients with temporal lobe glioma, verbal learning, processing speed, and executive functioning were shown to be related to general, social, and functional well-being, respectively. Furthermore, self-reported cognitive complaints are related to HRQOL in patients with GBM, although self-reported measures of cognition are only moderately correlated with the performance on formal neuropsychological assessments. It is important to keep track of cognitive functioning throughout the disease trajectory, not only because of the link between HRQOL and cognitive functioning but also because it is known to predict functional decline and even survival.

Neurocognitive Issues Associated with Glioblastoma Treatment

Cognitive functioning is influenced by the patient’s premorbid level of cognitive functioning and psychological distress, but cognitive deficits can also be the result of the brain tumor, its symptoms, and its treatment (eg, surgery, radiotherapy, chemotherapy, epilepsy, use of AEDs and corticosteroids). The invasive nature of GBM can cause damage to functional brain tissue, disrupting neural networks and causing cognitive problems. Depending on the size and location of the tumor, this may lead to specific (eg, language deficits) or more global cognitive dysfunction (eg, attentional deficits). Before resection, larger tumors have been shown to result in more cognitive deficits, especially in the areas of verbal and visual memory, verbal fluency, shifting between different tasks, and visuospatial abilities. Cognitive deficits caused by increased intracranial pressure or disruption of functional brain networks may be alleviated after surgical resection of the tumor, although worsening of symptoms may also be observed shortly after surgery. Removal of functional tissue in an attempt to dissect the tumor may cause specific cognitive deficits depending on the location in the brain. For example, lesions of the supplementary motor area can lead to agraphia.

Radiotherapy has long been associated with cognitive effects, which may be transient (acute or early-delayed toxicity), or lasting and progressive (late-delayed toxicity). Vascular abnormalities, demyelination, radionecrosis, and cerebral atrophy are recognized as a late-delayed effects (>6 months after irradiation). However, cognitive deficits may occur independently from radiographic or clinical evidence of injury to the brain. It is estimated that 50% to 90% of patients with brain tumors treated with fractionated whole-brain radiotherapy experience cognitive impairment 6 months later. Verbal and spatial memory, attention, and problem-solving seem to be particularly affected. Cognitive deficits may remain present even many years after low-dose radiotherapy, because patients with irradiated low-grade gliomas experienced a decline in attentional functioning compared with nonirradiated patients as much as 12 years after treatment. In long-term survivors (>2.5 years) of anaplastic glioma treated with radiotherapy and chemotherapy, scores on tests of working memory, attention, psychomotor speed, information processing, and executive functioning were significantly worse than in healthy controls matched for age, sex, and educational level. Although most studies investigating radiation effects have been performed in patients with low-grade glioma, note that many patients with glioblastoma also survive long enough (>6 months) to be affected by cognitive sequelae.

Chemotherapy is also known to affect cognitive functioning. Risk factors for developing neurotoxicity after chemotherapy include higher doses, treatment with multiple chemotherapeutic agents, treatment with both chemotherapy and radiotherapy either concurrently or subsequently, intra-arterial administration with disruption of the blood-brain barrier, and intrathecal administration. Across different cancer groups, attention, concentration, memory, and visual functioning are vulnerable after chemotherapy. In patients with high-grade glioma specifically, diminished learning ability, information processing, and attentional functioning and concentration seem to be linked to chemotherapy treatment. However, other studies show mild cognitive effects, or even no cognitive effects of chemotherapy. For example, in elderly patients with GBM with poor performance status, treatment with chemotherapy alone did not seem to influence HRQOL or cognitive functioning before disease progression occurred. Moreover, a small study assessing progression-free patients with GBM during their treatment with radiotherapy plus concomitant and adjuvant chemotherapy did not show a decline in cognitive functioning. Similarly, the cognitive effects of the antiangiogenic agent bevacizumab in patients with GBM remain ambiguous. Some studies find no clear negative cognitive effects of bevacizumab treatment, whereas a large randomized controlled trial shows a decline in both HRQOL and cognitive functioning in patients with GBM, especially after longer treatment with bevacizumab.

Apart from antitumor treatment, medications for symptom management can lead to cognitive dysfunction. Both epileptic seizures and AEDs have been associated with cognitive deterioration. First-generation AEDs (eg, carbamazepine, phenytoin, and valproic acid) are known to affect cognitive functioning. Cognitive slowing and attention deficits can result in subsequent poor performance on other cognitive tests. Fatigue associated with AED use can also hamper neuropsychological test performance. In patients with glioma specifically, seizures and AEDs seem to be related to deficits in processing speed, attention, and executive functioning. Corticosteroids can lead to cognitive improvement with decreased intracranial pressure, but long-term use may cause behavioral problems as well as attention and memory disturbances. These issues largely depend on the steroid dose and treatment duration, and may be reversible after treatment cessation. Of note, the treatment regimen in GBM almost invariably includes multimodal treatments, making it difficult to decipher to what degree cognitive deficits are the result of any specific treatment.

Instruments to Evaluate and Monitor Cognitive Deficits

To screen for cognitive deficits in clinical practice, the Mini-Mental State Examination (MMSE) is often used. This instrument is scored on a 0 to 30 scale, with higher scores indicating better functioning. A score less than 25 is considered cognitively impaired, but this may be too insensitive to detect more subtle cognitive problems, so a score less than 26 or 27 has also been used as a cutoff point in clinical trials (eg, Refs. ). It touches on 7 cognitive areas: orientation in time, orientation in place, working memory, verbal memory, concentration and calculation, language, and visuoperceptual functioning. Other screening measures are the Montreal Cognitive Assessment (MoCA) and the Addenbrooke Cognitive Examination–Revised (ACE-R). ACE-R incorporates the MMSE and has additional questions, making it a more comprehensive screening instrument. MoCA consists of 30 items and can be administered in about 10 minutes. MoCA taps into the domains of orientation in time and place, attention, working memory and verbal memory, visuospatial functioning, executive functioning, language, and fluency. All 3 screening instruments have been translated into various languages and are widely used. However, the usefulness of cognitive screening measures per se can be debated, because these are crude measures and subtle cognitive dysfunction may go unnoticed. However, major cognitive problems are likely to be detected.

After cognitive screening, a comprehensive neuropsychological assessment by a trained neuropsychologist may be warranted. Because of their background and training, neuropsychologists are especially suitable to provide insight and interpret neuropsychological test results. For example, a poor performance on a test of visual memory can be the result of a deficit in 1 or more components of memory (eg, capacity to learn new material, encoding, retrieval, consolidation) but also may be influenced by problems in attention, visual-perceptual deficits, or executive functioning (eg, planning a learning strategy). Motivation, fatigue, and mood play crucial roles as well. Neuropsychologists make an estimation of the contribution of each of these factors in their reports.

A full neuropsychological assessment can be burdensome for patients with GBM. Depending on the degree of cognitive dysfunction and the patient’s insight into problems, it can be uncomfortable for patients to be confronted with their cognitive issues in such a straightforward way. Therefore, neuropsychologists generally aim to limit the number of tests performed in areas in which, based on the patient’s history and test performance, there does not seem to be a problem. In cognitive domains in which a patient’s test scores lag behind, more than 1 neuropsychological test may be necessary to determine the exact nature of the problem. In clinical practice, differences between institutes exist, but in general about 1 to 1.5 hours is considered sufficient to obtain a cognitive profile.

However, in a research setting, a test battery covering a wide range of cognitive domains, but not in depth, is more appropriate. The test battery should be short and not take more than approximately 30 minutes to administer, it should be suitable for repeated assessments over time (ie, alternate versions should be available to limit practice effects), and it should have good psychometric properties. Naturally, it must also be possible to detect change in functioning over time. Using often-used tests makes it easier to compare across different studies and populations. Table 21.1 presents an overview of neuropsychological tests that are often used or are suggested by leading investigators, including the International Cancer and Cognition Task Force. Of note, for the working memory domain no specific tests were suggested because none of the tests met all requirements in terms of psychometric properties. Moreover, neuropsychological tests often tap into multiple cognitive domains. For example, fluency is sometimes considered executive functioning because it requires the patient to use strategies to retrieve words from memory, or language functioning, because of the heavy verbal component, but can also be scored as a separate domain. Domains and tests displayed in the table are merely examples. Deciding on any particular set of tests remains dependent on the goal of the research study.

When interpreting published results on cognition in patients with GBM, note that estimates of the prevalence of cognitive deficits may vary because of differences in the patient populations studied, the neuropsychological tests administered, or the normative data and cutoff scores used. Similar to the overestimation bias reported in HRQOL, results on cognitive functioning as an outcome measure in clinical trials should be interpreted with caution because patients who deteriorate (and are thus likely to have cognitive decline) may drop out of the study, whereas patients who perform well remain. Moreover, the nuanced report of a neuropsychologist is lost in clinical trials, in which factors such as fatigue, motivation, and mood may be disregarded and only pure test performance is taken into consideration.

Treatment of Cognitive Deficits in Patients with Glioblastoma Multiforme

After neuropsychological assessment, an estimation of the need for treatment or support in coping with cognitive deficits can be made. Here, it is important to keep in mind that cognitive performance can be influenced by the antitumor treatment and symptom management, as described earlier. These effects can be reversible (eg, in the case of corticosteroid use or AEDs) or irreversible, as is the case in surgery. However, the brain’s plasticity should not be underestimated because, even in adults, spontaneous recovery of cognitive functions can occur after acquired brain damage. This regenerative process mostly occurs within 6 months after brain injury, but may continue for a longer period. Although starting treatment of cognitive deficits early in the disease process can make it difficult to determine whether recovery occurred as a result of the therapy or because of spontaneous neuronal regeneration, it may help speed up the recovery process and could have a positive influence on HRQOL. It is especially relevant to achieve recovery early in patients with GBM, who are faced with a prognosis that is usually measured in months, rather than years.

In determining the need for treatment or supportive care, not only the patient’s cognitive profile plays a role but also their motivation to improve cognitive functioning, and the burden of cognitive dysfunction experienced by the patient and their family members, are equally important to consider. Therefore, a participatory approach should be used, taking into account the opinions of the patients and their families. Discussion in multidisciplinary meetings in which physicians, (neuro)psychologists, nurse specialists, and social workers are present may further assist in determining what kind of support is most suitable for individual patients with GBM.

Nonpharmacologic treatment strategies for cognitive deficits include cognitive rehabilitation and psychoeducation or coaching. Programs available for cognitive rehabilitation vary and can be intense, with patients undergoing therapy for several hours a day, over the course of multiple weeks. This schedule may be too demanding for a subset of patients with GBM, especially those on active treatment. Moreover, as described earlier, cognitive deficits generally improve gradually over the course of weeks to months following surgery. Perhaps because of this, those patients referred to cognitive rehabilitation have generally finished their primary treatment and are in a period of stable disease. However, there is increasing attention on the early treatment of cognitive deficits. As early inpatient rehabilitation focused on physical functioning proved feasible and effective in patients with newly diagnosed GBM, cognition may also be improved early in the disease trajectory. It would be interesting to perform a study comparing early cognitive rehabilitation with cognitive rehabilitation provided during stable disease, but, to our knowledge, no such trial has been initiated so far.

Several studies have been performed that show some beneficial effects of cognitive rehabilitation in patients with glioma. Furthermore, a multidisciplinary rehabilitation intervention found subjective improvement in cognitive functioning, although formal neuropsychological assessments were not performed. Of note, study populations are often small, and study designs generally lack a control condition. One randomized controlled trial has been performed in patients with glioma with favorable prognosis. After 6 months, attention, verbal memory, and mental fatigue improved compared with a care-as-usual control group. However, this was a very intensive rehabilitation program and patients with GBM were not included, thereby limiting the ability to generalize to this patient group. In a meta-analysis, Langenbahn and colleagues conclude that there is still too little evidence for the effectiveness of cognitive rehabilitation studies in adult patients with brain tumors.

As an example of less intensive treatment, psychoeducation in the form of feedback from neuropsychological assessment early in the disease trajectory can help patients and their families understand the cognitive difficulties that patients experience, which may substantially improve the home and family situation. Patients and family caregivers indicate that this kind of support is currently often lacking in health care. The feedback could help health care professionals and families in assessing what kind of formal or informal support the patients need at home, and it might be helpful in determining whether and when return to work may be feasible. Other interventions include restructuring of the home environment to aid patients to rely less on their impaired functions, providing advice on using external aids and technology, teaching strategies to cope with their cognitive problems, and retraining specific cognitive skills. This support may also lead to better mental adjustment as well as functional improvement and can help patients maintain independence.

Other nonpharmacologic alternatives are also being investigated in research studies. An online application of the rehabilitation program investigated in a randomized controlled setting is now under development, allowing patients to access training at a time of their choosing. Interventions based on physical exercise also show promising results in improving cognitive functioning. Although potentially useful for some patients, here, too, subsets of patients with GBM are excluded: those with difficulty using computers and those whose physical status prevents them from exercising.

Pharmacologic treatment of cognitive deficits is also possible. Sometimes, psychostimulants are prescribed to try to improve cognitive functioning. Studies evaluating the use of methylphenidate, modafinil, armodafinil, memantine, and donepezil have been performed and show modest effects on cognitive functioning at best. Most studies have included patients with GBM, although these made up most of the sample in only 1 report. Several studies included patients who were undergoing radiotherapy or chemotherapy, although some focused on patients in a stable disease phase. After psychostimulant use, performance on cognitive screening measures or tests of working memory, attention, processing speed, information processing, executive functioning, and sometimes even verbal and visual memory may improve. However, these may be attributable to a placebo effect because there are often no reported differences in cognitive performance when comparing 2 psychostimulants, or when comparing the psychostimulant with placebo. In a recent systematic review, Day and colleagues discussed treatment interventions for fatigued patients during cranial irradiation, and similarly concluded that, although pharmacologic interventions such as memantine or donepezil may have beneficial effects, studies were limited by small sample sizes and high risk of bias.