Prevalence of changes in personality and behavior in adult glioma patients: a systematic review

Abstract

Background

Gliomas are rare, with a dismal outcome and an obvious impact on quality of life, because of neurological, physical and cognitive problems, as well as personality and behavioral changes. These latter changes may affect the lives of both patients and their relatives in a profound way, but it is unclear how often this occurs and to what extent.

Methods

We performed a systematic review to determine the prevalence of changes in personality and behavior in glioma patients. Searches were conducted in PubMed/Medline, PsycINFO, Cochrane, CINAHL and Embase. Based on predetermined in- and exclusion criteria, papers were screened for eligibility. Information on the topics of interest were extracted from the full-text papers.

Results

The search yielded 9895 papers, of which 18 were found to be eligible; 9 qualitative and 9 quantitative studies. The reported prevalence rates of changes in personality and/or behavior varied from 8%–67% in glioma patients, and was 100% in a case series with bilateral gliomas. Moreover, these changes were associated with distress and a lower quality of life of patients as well as informal caregivers. Methods of measurement of personality and behavioral changes differed considerably, as did the description of these changes.

Conclusion

To determine the true prevalence of changes in behavior and personality, present but poorly labeled in the reported studies, prospective studies are needed using proper definitions of personality and behavioral changes and validated measurement tools. Ultimately, these findings may result in improved supportive care of both patients and caregivers, during the disease trajectory.

Gliomas are the most common primary malignant brain tumors in adults, and although rare—a yearly incidence of 6 cases per 100 000 persons¹—these tumors have a disproportionate share in morbidity. Glioma patients suffer from both a terminal cancer and from a progressive neurological disease. Apart from headaches, seizures, focal and/or cognitive deficits, these patients may also present with, or develop, changes in personality and behavior,² which occurs uniquely in the brain tumor population and not in patients with systemic malignancies without central nervous system involvement.

A single definition of personality is lacking. Descriptions like (i) “a dynamic and organized set of characteristics possessed by a person that uniquely influences his or her cognitions, motivations, and behaviors in various situations”³ and (ii) “a pattern of relatively permanent traits and unique characteristics that give both consistency and individuality to a person's behavior”,⁴ suggest that personality and behavior are strongly interlinked.

Both the tumor and its treatment may cause damage to the brain, and although it is commonly assumed that the frontal lobe plays an important role in behavior, tumor location in other areas might also result in behavioral and personality changes (BPC).⁵ Symptoms that are associated with BPC in brain tumor patients cover a broad spectrum, with on one end apathy, loss of initiative, and indifference and on the other end egocentrism and decreased empathy, loss of emotional control, disinhibition, and socially unadjusted or childish behavior.^2,6–10 BPC may also mimic and partly overlap with delirium or dementia, with features that touch upon deficits in memory, thinking and judgment, as well as clinical depression with symptoms such as apathy and loss of initiative.¹¹ In slowly growing lesions like low-grade gliomas, plasticity of the brain may lead to more insidious BPC, often only recognized as such at a later stage of the disease, whereas fast growing lesions may cause such changes within a few weeks.¹²

During the disease process BPC may progress and require specific supportive care. BPC may negatively influence quality of life for both patients and their informal caregivers, particularly if these specific supportive care needs are unmet.^13,14

As the disease progresses and time passes, the impact of patients' BPC usually becomes more obvious,¹⁵ and more challenging to cope with. Both patients and caregivers report that dealing with such changes can be distressing and overwhelming,¹⁶ resulting in psychosocial problems such as depression, anxiety and isolation. Also, brain tumor patients and their caregivers experience a greater need of everyday life support compared with other cancer patients.¹⁷ Typically patients experience loss of relational closeness and social inactivity,¹⁸ feel as if they are no longer the same person as before,⁶ and are bothered by changes in their personality.^14,19 Moreover, partners notice that family members and friends are less willing to be involved in sharing care responsibilities when patients develop behavioral problems.²⁰ Thus, BPC have a large impact on the social environment of the patients; they cause a distortion of family life and are associated with poor quality of life among caregivers.^9,21

Although BPC often might disrupt the lives of both brain tumor patients and their relatives in a profound way, with a large impact on social, emotional, and psychological well-being, it is unclear how common this problem is, at which moment in the disease trajectory they occur, and to what extent they might influence patients' medical decision-making competence and adherence to treatment. Therefore, the primary aim of this systematic review was to assess the prevalence of BPC in glioma patients. In addition, we sought to explore when these changes occurred and how they are defined and measured in the identified studies.

Methods

Search Strategy

A literature search was conducted using the following databases up to June 2014: PubMed/Medline, PsycINFO, Cochrane, CINAHL and Embase. The search strategy consisted of two search strings, one related to “changes in personality and behavior” and one related to “primary brain tumors”. The complete search strings can be found in the Supplementary file.

All retrieved titles and abstracts were screened by two reviewers (HZ and LD) and full texts of potentially relevant articles were read. The reference lists of the selected full text articles were screened to identify additional relevant studies. Any uncertainty about the relevance of a particular study was resolved in consensus.

PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines were followed.²²

Inclusion and Exclusion Criteria

Original observational studies or those using a validated checklist (eg, FACT-BR) reporting on BPC (condition), were included, with specification of point or period prevalence (context) in adult glioma patients at any time point during their disease course (population). Exclusion criteria were: depression, anxiety or mood disorders, and personality disorders as classified by the Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV-TR).²³

Data Extracted

From the selected articles, the following data were extracted: study design/methods, sample size/setting, purpose of study, prevalence of BPC, time point of BPC, other major findings related to BPC, definition of BPC, and tools used.

Results

The search yielded 9895 unique titles, of which 104 titles/abstracts were classified as eligible. Conference abstracts and those for which no full text could be retrieved were excluded (n = 30), resulting in a total of 74 papers. After screening the eligible articles full-text, another 56 articles were excluded because they comprised case studies (n = 12), reviews (n = 8), editorials (n = 1), did not report the items of FACT-brain (n = 5) or tumor type (n = 11) separately, had another focus (n = 18), or prevalence rates were unclear (n = 1). This resulted in a total of 18 articles that were included in this review (Fig. 1). The main characteristics of these studies are described in Table 1 (qualitative studies, n = 9) and Table 2 (quantitative studies, n = 9). The prevalence of BPCs in these studies ranged from 8% to 67%, with one case series reporting a prevalence rate of 100% in bilateral thalamic glioma.⁸

Fig. 1.

Flowchart of methods used to identify studies for inclusion.

Table 1.

Qualitative studies with prevalence rates for changes in behavior and personality

Author	Sample/setting	Design/methods	Purpose	Prevalence BPC	Time point BPC#	Other major findings*	Tool used for CB/CP	Changes in
Arber, 201325	N = 22 caregivers of primary malignant brain tumor patients	Cross-sectional study; interviews, grounded-theory approach	To explore the support needs of family caregivers	–	2	Some caregivers reported difficulty in handling changes in personality and behavior	None	Personality and behavior
Cavers, 201226	N = 21 glioma patients, N = 17 caregivers, N = 19 GPs	Longitudinal study; in-depth interviews at key stages of the illness; grounded-theory approach	To explore the experience of and coping with a glioma in relation to transition toward death in an effort to understand the need of support	–	2	For caregivers, dealing with personality changes was most distressing	None	Personality and behavior
Davies, 200327	N = 12 adult, 2-year survivors of malignant glioma and their caregivers (N = 10)	Cross-sectional study; semi- structured interviews	To describe patient out-comes and contact with rehabilitation services	8%–17%	2	2/12 patients reported irritability (17%), 1 relative reported in 1 of the patients a personality change (8%)	Comprehensive Psychopathological Rating Scale	Personality and behavior
Davies, 200521	N = 56 bereaved relatives of adult glioma patients	Longitudinal study; semi-structured interviews	Exploration of views of bereaved relatives about quality of survival after RT	23%	1	Cognitive/personality changes in 6/26 (23%) patients, but no differentiation between cognitive and personality changes	None	Cognition and personality
Salander, 1996a9	N = 24 spouses of malignant glioma patients	Longitudinal study; Structured interviews at different points in the disease course	To examine the psychosocial situation of spouses in different crisis trajectories	16%	1, 2	During disease trajectory personality changes in glioma patients became a danger to mutuality within the relationship	None	Personality
Salander, 1996b15	N = 30 patients with malignant glioma and their spouses	Longitudinal study; structured interviews (n = 2); grounded theory approach	To generate new insights into how the patient constructs a new sense of reality	23%	1	7/30 patients were excluded due to personality change (23%)	None	Personality
Salander, 199928	N = 28 patients with malignant glioma and their spouses (same cohort as Salander 1996b)	Longitudinal study; structured interviews (n = 2)	Descriptive study of symptom development and obstacles to early diagnosis	32%	1	Pathway to medical care was longer in cases of personality change. 9/28 patients had a change in behavior (32%), and in 1/28 (4%) a change in behavior was a trigger symptom to diagnosis	None	Personality and behavior
Sherwood, 201129	N = 10 family caregivers of malignant glioma patients	Longitudinal study; telephone interviews (n = 2)	To examine how caregivers transition into the caregiver role and their perception of changes over time	–	1	It took time for caregivers to accommodate to patients’ behavioral changes	None	Behavior
Whisenant, 201124	N = 20 caregivers of grade 3 and 4 glioma patients	Cross-sectional descriptive study; interviews, Story Theory	To explore the experience of caregiving	–	2	Caregivers wanted to endure the dramatic change in personality and behavior	None	Personality and behavior

*with respect to changes in personality and behavior.

^#1=at diagnosis, 2=during treatment and follow-up, 3=in the end of life phase.

Abbreviations: CB, changes in behavior; CP, changes in personality; N, number of subjects; GPs, general practitioners; RT, radiotherapy.

Table 2.

Quantitative studies with prevalence rates for changes in behavior and personality

Author	Sample/setting	Design/methods	Prevalence BPC	Timepoint BPC#	Purpose	Other major findings*	Tool used for CB/CP	Changes in
Allam, 200030	N = 37 oligodendroglioma patients (grades 1-4)	Retrospective analysis of medical charts	16%	1	Analysis of prognostic factors affecting survival	Behavioral changes as presenting symptoms in 16%, not analyzed as prognostic factor	None	Behavior
Collins, 201433	N = 482 adult glioma patients	Retrospective cohort study, three groups: (1) died during admission, (2) died <120 days, (3) >120 days (reference group)	12%–41%	3	(1) Description of clinical presentation of short-term survivors of glioma; (2) map their hospital utilization; (3) identify factors to serve as a prompt for palliative-care referral	Cognitive/behavioral difficulties in 41% (group 1), 19% (group 2), and 12% (group 3), but no differentiation between cognition and behavior. BPC was the strongest predictor of death during admission (RR = 1.3).	None	Cognition and behavior
Flechl, 201334	N = 52 caregivers of GBM patients	Retrospective study using questionnaire	50%–67%	3	To describe caregivers’ perspective on the end-of-life phase (last 3 months of life)	Behavioral changes last 3 months 50%, last week 67%	Questionnaire developed by VUmc on end-of-life phase	Behavior
Gofton, 201235	N = 168 brain tumor patients, of which N = 110 were glioma patients	Retrospective descriptive assessment based on electronic patient records	33%–66%	LGG 22% = 3 78% = 2 HGG 43% = 3 57% = 2	Assessment of symptom burden and palliative care needs of patients with primary and metastatic brain tumors requiring inpatient hospital care	33% of patients with LGG and 66% with HGG had cognitive/personality changes (memory impairment, altered personality, impaired ADL for non-physical reasons), but no differentiation in specific impairment	None	Cognition and personality
Gundersen, 199631	N = 495 GBM patients	Retrospective cohort-study, N = 384 (main group) and N = 111 (to test reliability of prognostic index)	53%–57%	1	To describe prognostic factors for survival in GBM patients	Clinical characteristics of behavioral changes in 53% (main group) and 57% (test group) appeared to be prognostic discriminators for survival	None	Behavior
Lai, 201432	N = 50 patients with grade 3/4 brain tumors, N = 10 physicians	Prospective study, including an interview and questionnaires	16%	2	(1) Identifying highest priority symptoms, (2) comparing patient priority ratings with oncology experts, and (3) developing symptom checklist (NFBrSI-24)	16% of patients reported changes in personality in top 5 symptoms to be assessed	Patient and physician survey	Personality
Partlow, 19928	N = 4 adult bilateral thalamic glioma patients	Retrospective case series	100%	1	To describe clinical, radiographic, and neuropathologic features	All patients displayed personality changes and/or mental deterioration (100%)	None	Personality
Riggs, 195837	N = 86 glioma patients	Retrospective study	15%–53%	1	Clinico-anatomic study of personality and mood disturbances	N = 13 (15%) of patients had behavioral alterations, N = 46 (53%) personality alterations, and N = 18 (21%) alterations in affect	None	Personality and behavior
Sundararajan, 201436	N = 678 malignant glioma patients	Retrospective cohort study using hospital and emergency department databases	12%–20%	1	Quantification of association of symptoms and receipt of supportive and palliative care and site of death	Increase in cognitive/behavioral difficulties observed at 2 time points: at diagnosis (12%) and at admission, during which patient died (20%)	None	Cognition and behavior

*with respect to changes in personality and behavior.

^#1=at diagnosis, 2=during treatment and follow-up, 3=in the end of life phase.

Abbreviations: CB, changes in behavior; CP, changes in personality; N, number of subjects; ADL, Activities of Daily Living; NFBrSI-24, National Comprehensive Cancer Network/Functional Assessment of Cancer Therapy Brain-Symptom Index; GBM, glioblastoma multiforme; LGG, low grade glioma; HGG, high grade glioma; VUmc, VUniversity Medical center, Amsterdam.

Qualitative Studies

We found 9 qualitative studies that were all, except one,²⁴ semi-structured interviews.^{9,15,21,24–29} The main objectives of these studies were the assessment of symptoms and clinical features (n = 1),²⁸ psychological well-being (n = 2),^15,26 and experiences (n = 6)^{9,21,24,25,27,29} of patients as well as informal caregivers. The sample sizes ranged from 10 to 56 patients. In 1 of 9 studies²⁷ a tool (Comprehensive Psychopathological Rating Scale) was used to measure the relevance of psychiatric symptoms (responses were rated on a 4-point Likert scale ranging from “no problem” to “severe problem”), among other symptoms. Various definitions of BPC were used including changes in “personality and behavior” (n = 4), “behavior” (n = 1), “personality” (n = 3), and “cognition/personality” (n = 1). Prevalence—if possible to conclude from the reported results—of BPC in these qualitative studies ranged from 8% to 32%. Of note, the largest study (n = 56) from Davies et al²¹ reported a prevalence of 23%, but the authors did not differentiate between cognition and personality. Finally, one prevalence rating is based on the observation that 7 of 30 (23%) patients were excluded from the study population due to personality changes.¹⁵ In 5 studies, BPC were described at diagnosis,^{9,15,20,21,28} but they were not evaluated during the course of the disease. In 5 other studies BPC occurred during the disease trajectory,^9,24–27 and Salander specifically described 2 patients with BPC at the moment of recurrence.⁹ The location of the tumor was reported on in 2 of 9 studies; Davies et al described a group of inactive (eg, disinhibition, cognitive and communication deficits) patients of which 5 of 7 patients had a frontal lobe tumor,²⁷ and Salander described 7 patients of which 3 had a glioma located in the frontal lobe and 4 in other locations.¹⁵

Other major findings in these qualitative studies were that dealing with BPC was difficult and patients' progressive personality changes were most distressing for the informal caregivers,^25,26 even becoming a danger to mutuality within the relationship during the disease trajectory.⁹ Patients themselves reported irritability²⁷ and other changes in behavior such as loss of initiative. One patient described a change in his behavior as a trigger symptom of his later diagnosed glioma;²⁸ thus, BPC may even result in delayed medical care, in case it is the first symptom.²⁸ During bereavement, caregivers mentioned that the personality change of the patient was a burden they were unprepared for and unable to deal with, resulting in a lower quality of life.^15,21,27 Although informal caregivers expressed the wish to endure the dramatic change in personality of the patient,²⁴ it may take time to accommodate to patients' behavioral changes.²⁹

Quantitative Studies

We found 9 quantitative studies, all with a retrospective study design. The sample size of patients and/or caregivers in these studies ranged from 4 to 678. Although all studies reported the rate of BPC, the main focus of these studies was on clinical characteristics and caregiver burden in relation to prognostic factors,^30,31 development of a symptom checklist including items on BPC,³² supportive care or palliative care needs,^33–36 or on the relation between the location of the tumor and BPC.^8,37

In 2 of 9 studies a newly developed questionnaire was used to measure BPC,^32,34 among other symptoms, while in the remaining studies BPC were on occasion reported by patients and/or caregivers as presenting symptoms or as symptoms occurring during the disease trajectory. Several terms to define BPC were used and included changes in “personality” (n = 2), “cognition/personality” (n = 1), “behavior” (n = 4), and “cognition/behavior” (n = 2). Prevalence of BPC in these quantitative studies ranged from 12% to 67%, with the exception of a case series of patients (n = 4) with bilateral thalamic gliomas, with a prevalence of BPC and/or cognitive deterioration of 100%,⁸ which severely limits the value of this included study in terms of prevalence. Five studies described BPC at diagnosis,^{8,30,31,36,37} 2 during disease trajectory,^32,35 while the remaining studies focused on BPC in the end-of-life phase.^33–35 The study by Lai et al³² reported on the development of a symptom checklist, based on ratings of symptoms by both patients and physicians, revealing that 16% of the patients rated a change in personality in the top 5 of most important symptoms, similar to the rate of BPC as presenting symptoms in the study of Allam et al³⁰ and of Sundararajan et al³⁶ In assessing symptom burden and palliative care needs of brain tumor patients requiring inpatient hospital care, a change in cognition/personality was found in 33% of patients with low-grade glioma and in 66% of patients with high-grade glioma.³⁵ However, no distinction was made between changes in cognition and personality. Sundararajan et al³⁶ retrospectively analyzed clinical data of glioma patients and described a prevalence of BPC at diagnosis of 12%, which increased to 20% during hospital admission during which the patient died. Of note, the authors did not discriminate between cognitive and behavioral difficulties. Flechl et al³⁴ also described BPC during the end-of-life phase from the caregiver's perspective. Fifty percent of glioblastoma patients were reported to experience a change in behavior in the last 3 months of life, increasing to 67% in the last week of life.

For patients with tumors located in bilateral thalamic regions, a 100% prevalence of BPC was reported in a case series.⁸ In a clinicoanatomic study distinguishing emotional circuits of the brain, BPC was associated with several tumor locations.³⁷ Behavioral alterations occurred in 13 of 86 (15%) and personality alterations in 46 of 86 (53%) glioma patients, but the tumor locations were variable and often outside the presumed region of control of emotion in the hypothalamus/thalamus. In 2 other quantitative studies, the location of the glioma at diagnosis was reported but no association was found between location and the occurrence of BPC.^33,36

Regarding BPC as a prognostic factor, two studies assessed BPC affecting treatment outcome. In a study of 495 patients with glioblastoma, Gundersen et al reported that BPC were present in more than half of the patients, which appeared to be a prognostic discriminator for the length of survival. Behavioral change was independently associated with an increased relative risk (RR) of death (RR = 1.3; 95% CI, 1.0-1.6, P <.05) next to Karnofsky Performance Status, steroid dependency, age, and extent of surgery.³¹ A study of glioblastoma patients that aimed to identify prognostic factors affecting treatment outcome in short-term survivors revealed that the prevalence of cognitive and behavioral difficulties was highest in short-term survivors; 41% of patients who died during the diagnostic hospital admission reported cognitive or behavioral difficulties, compared with 19% of patients who survived admission but lived shorter than 120 days and 12% of patients who survived more than 120 days.³³ Moreover, cognitive or behavioral difficulties were the strongest predictor of death during admission, and patients with these difficulties were 3.1 times more likely to die.

Tools

In 8 of 9 qualitative studies semi-structured interviews were used to identify BPC, from both the patient and caregiver perspective. Only one study used a validated instrument, the Comprehensive Psychopathological Rating Scale, to assess the psychiatric state of the patient.²⁷ The instrument covered the symptoms of depression, anxiety, fatigue, irritability, psychotic experience and BPC, with responses ranging from “no problem” to “severe problem.” In the quantitative studies, 2 different newly developed symptom questionnaires^32,38 were used, both including a question about the occurrence of BPC. In the other quantitative studies, no specific tools were used.

Discussion

Data on prevalence and severity of BPC in glioma patients are scarce, which is underscored with this review. The unclear definition of BPC, with terms such as “personality change,” “behavioral change,” and “cognitive or behavioral change” used by different research groups,^21,33,35,36 and the difficulties distinguishing delirium, dementia, or depression hampered our aim to assess the prevalence of BPC in glioma patients accurately.¹¹ The wide prevalence range of BPC that we observed be due to the different characteristics and criteria of the included studies. First, the differences in study design, with quantitative studies using a more systematic approach than qualitative studies to assess BPC, might have contributed to the wide prevalence range. Other factors influencing prevalence are the heterogeneity in the study populations, the retrospective design that is prone to recall bias, and the limited use of validated instruments to assess BPC. To determine the true prevalence of BPC, studies with large sample sizes should be regarded as more valuable because they provide a more robust estimate. However, not only sample size seems important to determine the prevalence rate, but also the definition of BPC and different time points BPC are measured. For example, in 2 studies with large sample sizes the difference in prevalence rate is remarkable: Gundersen et al analyzed prognostic factors at diagnosis in glioblastoma patients and found a prevalence of 53% to 57%, while Collins et al looked for predictors of death in the end-of-life phase of glioblastoma patients and found a prevalence of 12% to 41%. Moreover, Gundersen et al referred to “mental changes” while Collins referred to “changes in cognition/behavior.” In most studies, semi-structured interviews were applied, which do reveal problems of patients and caregivers with BPC but do not quantify the extent of the problem. Also, different terms such as “cognition,” “personality,” and “behavior” were used in combination and interchangeably. Still, even studies with a clear description in changes in behavior or personality reported varying prevalences: 8% to 23% for personality changes,^9,15,27,32 16% to 67% for behavioral changes,^27,30,31,34 and 15% to 53% for a combination of these changes.^28,37 Another limitation is that several aspects of cognitive functioning were not included in our definition of BPC and search terms. It is therefore possible that not all studies comprising BPC were identified. Our decision not to include personality disorders as classified by the DSM-IV may also have contributed to this.

Even though it is difficult to determine the exact prevalence of BPC in glioma patients, this review shows that BPC are quite common and may affect the lives of both patients and their relatives in a profound way. Especially the qualitative studies reveal the difficulties patients and their caregivers experience in handling BPC. However, we may have missed important information on the impact of BPC on the well-being of the patient and their caregivers due to our stringent inclusion and exclusion criteria focusing on prevalence data only. Although most of the identified studies comprise small sample sizes, they suggest that BPC might delay the diagnostic process,²⁸ and that the severity of the impact of BPC on everyday life of the patient as well as the caregiver is obvious, not only during the disease trajectory but also during bereavement after the patient has died.^13,21 BPC are described at diagnosis,^{8,9,15,21,28–31,36,37} during the course of the disease,^{9,24–27,32,35} and in the end-of-life phase,^33–35 but it remains unclear how BPC are related to disease recurrence or the end-of-life phase. In addition, it was found that the patients' behavioral problems result in lower caregiver mastery; ie, the perception of the sense of worth as caregiver and the ability to meet the demands of providing care.²⁰ The occurrence of BPC is probably associated with poor prognosis; glioblastoma patients with BPC³¹ tend to have a more dismal prognosis than those without BPC, but it remains unclear whether the influence of BPC on patients' medical decision-making competence and treatment adherence contributed to this. This is supported by the finding that the prevalence of cognitive and behavioral difficulties was highest in short-term malignant glioma survivors.³³ In addition, over the disease course, towards death, the prevalence of behavioral changes increases.³⁴ Although we chose to limit our search for BPC to glioma patients, our findings are similar to those reported from studies including all types of primary brain tumor patients (glioma, brain metastases, meningioma and other benign tumors). In 2 small studies, 13% to 16% of brain tumor patients had behavioral problems,^10,40 while in a large study, 34% of the patients experienced behavioral changes.³⁹ Moreover, 2 other studies of patients with various primary brain tumors found that 56% of caregivers had problems coping with the changes in patients' behavior,¹⁴ and also that personality changes exacerbated stress of caregivers.⁴¹ Both Gregg et al⁷ and Whiting et al¹⁶ aimed to quantify BPC using the Frontal Systems Behavior Scale, the Overt Behavior Scale, and the Emotional and Social Dysfunction Questionnaire, addressing BPC in relation to executive functioning. Their systematic approach would allow comparison between these studies. Gregg et al⁷ measured BPC by specifically addressing apathy, disinhibition, and executive dysfunction. They found that clinical levels of disinhibition were significantly higher in patients with frontal tumors than in those with nonfrontal tumors, while apathy and executive function difficulties were reported by at least 40% of the patients regardless of tumor location. Although it is commonly assumed that the frontal lobe plays an essential role in behavior,⁵ it seems difficult to establish a causal link between BPC and tumor location.³⁷ Whiting et al published a report¹⁶ (although not eligible according to our inclusion criteria [not peer-reviewed]) that systematically addressed the behavioral and cognitive sequelae of primary brain tumors, and revealed that 20% to 35% of patients (n = 54) experienced behavioral changes, including inertia, anger, and inappropriate behavior. Moreover, this study found similar rates of behavioral changes across patients with malignant and benign brain tumors, suggesting that tumor histology is not specifically related to BPC.

Conclusion

Changes in behavior and personality are present in a substantial number of glioma patients, and are associated with distress and a lower quality of life for both patients and informal caregivers. To determine a more accurate prevalence of BPC, their severity, and their relation to tumor location, prospective studies are needed that use proper definitions of BPC and validated measurement tools. Moreover, future research should measure such changes in relation to executive functioning and depression, to help more clearly define BPC. To gain a better understanding of BPC in glioma patients, the research community should agree on a uniform definition of BPC and together they should develop a proper measurement tool to assess BPC. This will result in a tool that is supported by the whole research community and will subsequently lead to a better assessment of the true prevalence of BPC in glioma patients. Ultimately, these findings may result in improved supportive care for both patients and caregivers, during the disease trajectory until the end of life.

Funding

There has been no funding supporting the research.

Conflict of interest statement. All authors have declared no conflict of interest.