Pharmacological Treatments for Frontotemporal Dementias

Abstract

The objective of this review is to summarize the current data on the pharmacological treatments for frontotemporal dementias from randomized controlled trials. A systematic search of 4 major databases, PubMed, Medline, PsychINFO and Cochrane, found a total of 9 randomized controlled, double-blinded clinical trials. Of these, 2 trials used the selective serotonin reuptake inhibitor (SSRI), paroxetine; 1 trial used trazodone; 2 trials used stimulants (methylphenidate and dextroamphetamine); 1 trial used the acetylcholinesterase inhibitor, galantamine; 2 trials used the N-methyl-d-aspartate antagonist, memantine; and 1 trial used the neuropeptide oxytocin. The analysis of the available data indicates that SSRIs, trazodone, and the amphetamines may be effective in reducing some behavioral symptoms, but none of these medications had an impact on cognition. Available data indicate that these medications were well tolerated in all the trials.

Introduction

The term frontemporal dementia (FTD) is used to describe a group of related neurodegenerative conditions characterized by disturbances in behavior or language as well as by cognitive and neurological changes. ¹ Over a 100 years ago, Arnold Pick described a series of cases that distinguished focal atrophies of the frontal cortex of the brain from what was then called senile atrophy. ¹ Later, “Pick’s disease” came to describe a condition in patients with frontal brain degeneration and personality change, along with the characteristic pathological silver staining (argyrophilic) round inclusions called Pick’s bodies in the neurons and ballooned neurons. ² Cases with similar clinical findings but without this pathological marker were renamed Frontotemporal dementias. ³

Behavioral alterations are the predominant and usually the first clinical presentation of FTD, although sometimes they may occur later on in the clinical course of the disease process. ¹ Patients are often without appropriate social emotions, some with overactivity and disinhibition and others with apathy and blunted affect. ¹ Cognitively, these patients present with significant deficiencies in executive function and working memory. ⁴ Frequently, there is perseveration, poor attention, poor abstraction, and difficulty shifting mental set noted in these patients. ⁵ However, spatial skills seem to be unaffected in these patients. ⁶ Neurological signs usually present later on in the disease state where patients develop Parkinsonian signs of akinesia, tremors, and muscular rigidity. ⁶

There are 3 main subgroups of FTD; frontal variant or behavioral variant (fvFTD or bvFTD), progressive nonfluent aphasia (PNFA), and semantic dementias (SDs). ¹ The symptoms of each subtype reflect the distribution of pathological changes in the brain rather than a precise histological subtype. ¹ Patients may even initially present with symptoms suggestive of 1 FTD syndrome and later progress on to a different subtype. ⁷ The FTD has a heterogenous pathology involving gliosis, neuronal loss, and superficial spongiform degeneration in the frontal and/or temporal cortexes. Tau-inclusions have also been confirmed in some cases. ^8,9

Frontal variant FTD presents with personality changes including, disinhibition, impulsivity, stereotypies, apathy, and hyperorality. ¹ Many of these behavioral changes are due to orbitobasal cortical involvement, while apathy is linked with medial frontal anterior cingulated involvement. ¹⁰ The most common cognitive deficit in fvFTD is an impairment of executive function or working memory, reflecting frontal and prefrontal cortical involvement. ⁴ Other common cognitive deficits are seen in attention, abstraction, shifting mental set, and perseveration. ⁵ As with all the subtypes of FTD, heterogeneity in clinical presentation is evident, with social behavior more disrupted in patients with predominantly right cerebral hemisphere pathology. ⁴

The PNFA principally involves expressive language with severe difficulty in retrieving words while maintaining word comprehension. ¹ These patients do not present with behavioral problems until later in the disease. ⁴ The PNFA is associated with asymmetric atrophy of the left cerebral hemisphere. ⁶

The SD, also known as temporal FTD, usually presents with language abnormalities, but in contrast to PNFA, there is loss of word meaning in addition to loss of memory for words. ⁴ Patients are often unaware of their comprehension difficulties and speak fluently, using substitute phrases like “thing” and “that.” Nonverbal deficits in visual and auditory modalities are also present, indicating problems in conceptual knowledge more so than a specific problem with language. ¹ Behavioral symptoms may present early or late in the clinical course, often with more compulsivity and less apathy than in patients with fvFTD. ⁴

In an excellent review by Ghosh and Lippa, the authors found that FTDs are more rapidly progressing than other types of dementias. They found that the mean survival from the time of diagnosis in patients with FTD is between 3 and 4 years. ¹¹ The diagnosis of FTD is often delayed with estimated 3.6 years elapsing from the onset of symptoms to the actual diagnosis of the disease being made. This delay in diagnosis probably leads to worse outcomes in patients with FTD. The authors also found that up to 40% of FTDs are familial. In up to a third of these familial FTDs, there is mutation in the microtubule-associated protein τ (MAPT) gene that is located on chromosome 17. Mutations in the C9ORF72, a hexanucleotide repeat expansion present on chromosome 9 has also been identified as the genetic abnormality in cases of familial FTD.

Currently, there are no curative or disease-modifying treatments for FTD, let alone any Food and Drug Administration (FDA)-approved treatments for symptoms of FTD. ¹ As FTD is increasingly recognized and diagnosed, addressing the unmet need for treatment of symptoms is of the utmost importance for patients with FTD and their families. ¹

Search Strategy

The purpose of this review is to summarize the data from published randomized controlled double-blind clinical trials regarding the use of all classes of medications for FTDs. We performed a literature search of 4 major databases, PubMed, Medline, PsychINFO and Cochrane. Search terms were “frontotemporal dementia” and “frontotemporal lobar degeneration,” with “Randomized Controlled Trial” as a filter in each case. The databases were searched till July 30, 2013. There was no language restrictions set in order to be as inclusive as possible with the search strategy. The relevant articles (based on the title and abstract) were retrieved for more detailed evaluation. The bibliographies of relevant articles were searched for additional references.

Results

Pubmed generated 20 articles of which 9 comprised the group that are included in our review. Medline generated 3 articles of which 2 were included in the 9 selected from the PubMed search. No articles were found on PsychINFO and 1 was found on Cochrane database, also among the 9 selected from the PubMed search. Among those articles not included were those that involved aspects of FTD aside from pharmacological treatment (ie, regarding pathology or genetics) and those not about FTD (ie, about other dementias).

We found a total of 9 randomized controlled double-blinded clinical trials. Of these, 2 trials used the selective serotonin reuptake inhibitor (SSRI) paroxetine, 1 trial used trazodone, 2 trials used stimulants (methylphenidate and dextroamphetamine), 1 trial used the acetylcholinesterase inhibitor (AChEI) galantamine, 2 trials used the N-methyl-d-aspartate (NMDA) antagonist memantime, and 1 trial used the neuropeptide oxytocin.

To make the evidence easier to evaluate, we have divided the data based on each individual drug class. Tables have been included to aid in this process.

Selective Serotonin Reuptake Inhibitors

In a study by Moretti et al, ¹² 16 patients, 10 women and 6 men, aged 64 to 68 years with a diagnosis of FTD (according to the Lung and Manchester criteria) were included. In all, 8 patients were randomized to receive paroxetine 20 mg/d (group A) and 8 to receive piracetam (γ-aminobutyric acid [GABA] analog) 1200 mg twice a day (group B). Participants receiving paroxetine began treatment on a lower dose of 10 mg/d and were titrated to the higher dose of 20 mg/d after 2 weeks. Patients in both the groups were allowed to continue previous therapy (eg, antihypertensive, antidiabetic, antidyslipidemic agents). All participants were followed for 14 months with neurological and neuropsychological examinations at months 1, 4, 8, and 12, and a complete neuropsychological examination at baseline and after 14 months. At all visits, caregiver’s stress was evaluated via the relative stress scale for a more complete clinical overview. Global cognitive function was assessed using the Mini-Mental State Examination (MMSE). Executive and planning function were assessed using the Ten Point Clock (TPC) Test, Proverb Interpretation Tasks, and the Stroop Test. Behavioral performance was measured using the Neuropsychiatric Inventory (NPI), the Clinical Insight Rating (CIR) Scale, and the Cornell Scale for Depression in Dementia (CSDD). In addition to these tests completed at each visit, the Behavioral Pathology in Alzheimer’s Disease Rating Scale (BEHAVE-AD) was completed at baseline and at 14 months. The incidence of potential side effects was monitored throughout the study period, and blood pressure and heart rates were measured at each visit. Laboratory tests including liver function tests and a complete blood count were measured every 6 months. At baseline both the groups presented with personality changes, behavioral withdrawal, lack of insight, and disinhibition, with no statistical differences between the 2 groups on any of the rating scales. At 14 months, patients in group A showed statistically significant improvements in the BEHAVE-AD score (by 13.075 points over group B), P < .01, the NPI score (by 8.25 points over baseline), P < .01, and on the CSDD (by 3.15 points over group B), P < .01. Baseline scores of the CIR stayed at very low levels in both the groups. Caregiver stress decreased in group A as measured by a mean reduction of 8.54 points in the RSS compared with baseline (P < .05) and 14.12 points when compared to group B (P < .01). The MMSE and TPC, measuring cognition and planning, remained stable in group A, while there was a decrease in performance on the Stroop and Proverb Tests (reduction of 5.3 and 4.8 points; P < .05), measuring attention, and abstract reasoning. There was no increase in benzodiazepine or neuroleptic use in this group. In group B, there were no improvements in any domain, and a more significant deterioration in planning capacities as shown by a mean reduction of 2.5 points in the TPC (P < .001) was noted. There was a progressive and constant intake of neuroleptics in this group to reduce behavioral changes. At follow-up, 2 patients in group B demonstrated drastic dietary changes with evident hyperphagia and carbohydrate craving. A review of side effects showed no serious adverse events, drug–drug interactions, cardiac effects, or blood pressure modifications in either arm of the study. Three patients experienced transient nausea in group A, and 4 patients in group B reported increasing agitation and aggression. The authors concluded that long-term treatment with paroxetine may influence noncognitive aspects of FTD with improvements in behavior and social conduct as evidenced by a reduction in aggressiveness, agitation, weeping and depressed mood, social conduct, eating problems, and sleep disorders. However, they acknowledged that their results could be considered limited due to the small number of patients enrolled in the study.

The study by Deakin et al ¹³ proposed to subject paroxetine to a more rigorous test of its efficacy in patients with FTD when compared to the trial done by Moretti et al ¹² using a placebo arm instead of an active drug comparator group. Ten patients meeting the consensus criteria for fvFTD were enrolled in the trial. The participants were then randomized to receive 40 mg/d dosing of paroxetine or an identical placebo capsule. The treatment period was 9 weeks. Paroxetine was started at 20 mg/d dose and increased to 30 mg in week 2 and to 40 mg in week 3. The dose of paroxetine remained at 40 mg/d for the next 4 weeks. Then, the dose was reduced by 10 mg/d each week to avoid an SSRI discontinuation syndrome. The treatment assessment was done in the sixth or seventh week of the trial. The tests were taken from the Cambridge Neuropsychological Test Automated Battery (CANTAB) and the CANTAB extensions and given in the following order: immediate and delayed pattern recognition, spatial recognition, spatial span, spatial working memory, visual discrimination learning/attentional set shifting, decision-making “gamble,” and paired associates learning. Behavioral data were collected from caregivers using the NPI and Cambridge Behavioral Inventory (CBI). Verbal fluency and digit span was also assessed. Results indicated that paroxetine did not significantly improve scores on the subjective measures of the NPI and CBI. Paroxetine however did increase errors on performance on several neuropsychological tests, including the reversal component of the visual discrimination task (P = .050), the errors in the paired associates learning task, (P = .056), and delayed pattern recognition memory accuracy (P = .020). Spatial recognition memory, spatial working memory, and immediate pattern recognition were not significantly affected by paroxetine therapy, although there were some nonsignificant trends toward impairment seen in the extradimensional shift and spatial recognition memory tasks. There was no effect of paroxetine relative to placebo on verbal fluency (P = .659) or on forward or backward digit span (P = .266, P = .333, respectively). In this study, no mention was made of the side effects or adverse events from the drug treatment. The authors concluded that their results offer no evidence to support the use of paroxetine as a treatment for fvFTD.

Trazodone

In a study by Lebert et al, ¹⁴ 16 women and 15 men with a mean age of 61.7 years were enrolled from 3 centers. Of the 31 participants, 5 withdrew from the trial. A total of 26 participants completed the two 6-week treatment periods and achieved the primary end point. Participants were randomly assigned to 1 of the 2 treatment groups, placebo-trazodone or trazodone-placebo. Since trazodone has a relatively short half-life, no washout period was provided. There was a 6-day titration period with trazodone 50 and 100 mg/d or placebo, after which point participants received 150 mg/d of trazodone or matching placebo for 3 weeks. Trazodone was increased to 300 mg/d for the last 3 weeks if the patient had no side effects (hypotension, fatigue) reported during the telephone interview. No other psychotropic agents were permitted for use except for occasional meprobamate for at most 2 days. Patients and caregivers were questioned by telephone at midpoint and during the visit on the last day of the study period, at which point assessments were performed. Total NPI score was used to measure the severity of the behavioral disturbances and was the primary outcome variable. Secondary variables included the clinical global impression of improvement (CGI-I) and the MMSE scores. Side effects were assessed by standardized open questioning of patients and recorded blindly. Results showed a clear and significant decrease in the NPI total score after trazodone treatment when compared to placebo, 18 patients in the trazodone group versus 8 in the placebo group, P = .028. In 10 patients on trazodone, a decrease in more than 50% in the NPI score was observed. A 25% to 50% improvement was seen in 6 participants, and less than a 25% improvement was seen in 10 patients. This improvement was noted on eating disorder, agitation, irritability, and depression/dysphoria. A possible effect on disinhibition could have been underestimated due to a significant period (P = .02) and a sequence effect (P = .04). Based on a multivariate analysis, patients with more severe eating disorders at baseline derived the greatest benefit from trazodone and were correlated with “responders” (P = .04). Mild to moderate improvement in clinical severity was seen after the trazodone period, assessed by the CGI-I scale, but the difference with placebo was not significant (P = .08). No statistical difference was found between the 2 treatments with regard to the variation in the MMSE score (P = .10). A total of 5 participants withdrew from the study, 3 in the first patient period (in the trazodone-placebo sequence), 1 for nonelective surgery, 1 for agitation after the trazodone sequence, and 1 for symptomatic hypotension with syncope. Two participants dropped out in the second treatment period in the same sequence, 1 for nonelective surgery and 1 for a motor deficit related. Adverse events were more frequent with trazodone group than with the placebo group. A total of 11 patients in the trazodone period reported a treatment emergency adverse event, including fatigue, dizziness, hypotension, and cold extremities. Three participants manifested fatigue and dizziness in the placebo period. All adverse events were rated as being mild, and only 1 patient left the trial. The authors concluded that trazodone can reduce the behavioral symptoms of FTD that were assessed by the NPI score, especially eating disorder, irritability, agitation, and depressive symptoms.

Stimulants

In a study by Rahman et al, ¹⁵ 8 patients, 1 female and 7 male, with a mean age of 62 years were recruited with a clinical diagnosis of fvFTD (according to the Lund and Manchester Groups criteria) and a MMSE ≥ 20 to allow for computerized neuropsychological assessment. Of the 11 patients originally screened for the study, 3 were excluded due to concomitant illnesses that contraindicated with use of methylphenidate (glaucoma, unstable hypertension and stable/unstable angina). The participants were randomized into 2 groups matched for age, MMSE, and National Adult Reading Test Intelligence Quotient (NART IQ) estimate with the treatment order fully counterbalanced across patients. One group took methylphenidate 40 mg first, then placebo, and 1 took placebo first then methylphenidate, with the 2 test sessions for each patient separated by 1 to 2 weeks. Cognitive testing lasted about 2 hours and commenced at 90 minutes after administration of the drug or placebo at approximately 1430 hours, with all test session in the afternoon. Peak plasma concentrations of methylphenidate are reached at approximately 2 hours after ingestion. Adverse effects during the session were monitored and recorded with particular care to identify any nervousness, headache, and gastrointestinal disturbances. Blood pressure was measured using a traditional sphygmomanometer just prior to ingestion of the tablet and to cognitive testing. Visual analog scales were administered before tablet ingestion and before cognitive testing for the scales alert–drowsy, calm–excited, strong–feeble, muzzy–clear headed, well coordinated–clumsy, lethargic–energetic, contented–discontented, troubled–tranquil, mentally slow–quick witted, tense–relaxed, attentive–dreamy, incompetent–proficient, happy–sad, antagonistic–amicable, interested–bored, and withdrawn–gregarious. In both the test session, the participants were given the same assessment in a fixed order from the CANTAB battery: pattern recognition memory, spatial recognition memory, spatial span, spatial working memory, and intradimensional (ID)/extradimensional (ED) attentional-set shifting, and Tower of London test of spatial planning. The Cambridge Gamble Task was used to assess decision-making cognition. Results showed no effects of methylphenidate on pattern recognition memory, spatial recognition memory, spatial working memory, ID/ED attentional set-shifting, and one-touch Tower of London tasks. In the Cambridge Gamble Task, there were no effects of methylphenidate on the average deliberation time (P = .779) or choice of the likely outcome (P = .865). However, there was a statistically significant difference in betting behavior with patients demonstrating reduced betting on methylphenidate (P = .017), consistently seen across all 8 patients. Further statistical analysis with a mixed-model analysis of variance was performed, showing a significant main effect of drug (P = .032), supporting the effect in the nonparametric test. Baseline measures of pulse (P = .528) and systolic blood pressure (P = .231) did not differ significantly between drug and placebo sessions, although the baseline measurements of diastolic blood pressure approached significance (P = .061). There was a general increase in blood pressure after methylphenidate treatment. The differences in systolic and diastolic blood pressures for the drug condition compared to the placebo condition approached significance (P = .075) but were nonsignificant when the patient receiving atenolol was excluded from the analysis (P = .138). There was no effect of the drug on visual analog scales. The authors concluded that there is a significant effect of methylphendiate on risk taking in patients with fvFTD associated with the behavioral disturbance induced by the dementia. Methylphenidate acted to normalize risk-taking behavior bringing patients with fvFTD toward the typical performance of healthy older adults.

In a study by Huey et al, ¹⁶ investigators contrasted the effects of dextroamphetamine and quetiapine in 8 patients with fvFTD in a randomized double-blind crossover trial. Over 1 week, medication daily dosage was gradually increased to either 20 mg of dextroamphetamine or 150 mg of quetiapine in divided doses. The patients were on the target dose for 3 weeks before returning to the clinic for reevaluation, at which point the patients were tapered to half the study medication for 2 days before discontinuation. They then underwent washout for 1 week, and the process was repeated with the other medication. The primary measure of behavioral symptoms was the NPI, and the primary cognitive measure was the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS). All patients were able to tolerate the full dose of dextroamphetamine. One patient experienced sedation on quetiapine treatment and was unable to tolerate the full dose. The most common adverse effect of both the medications was sleep disturbance. Using nonparametric methods, results showed a significant effect of treatment on the total NPI (P = .05). Post hoc Wilcoxon signed rank tests showed that the total NPI was significantly lower than pretreatment baseline on dextroamphetamine (P = .02), but there was no significant difference between baseline and quetiapine nor between quetiapine and dextromphetamine. The NPI subscales that decreased the most on dextroamphetamine were apathy (2.8 points) and disinhibition (2.4 points). There was no significant overall effect of treatment on the RBANS. The authors reported that treatment with dextroamphetamine improved behavioral symptoms, including disinhibition, in patients with FTD, likely through its action of augmenting brain dopamine and norepinephrine.

Galantamine

In a study by Kertesz et al, ¹⁷ 36 patients with fvFTD and primary progressive aphasia (PPA) were enrolled in a trial to investigate the efficacy and safety of galantamine on behavioral symptoms, aphasia, and their global clinical status. The study design consisted of an 18-week, open-label, galantamine treatment (phase 1), followed by an 8-week, double-blind, placebo-controlled, randomized withdrawal (phase 2). The participants were stratified by diagnostic types (fvFTD vs PPA) and then randomized to placebo or galantamine. The open-label study medication was provided in bottles, while the double-blind study medications were provided in numbered blister cards with placebo and galantamine (8, 12 mg) in identical format. The patients were on 8 mg daily for the first 4 weeks of open-label phase and received 16 to 24 mg for the rest of the 18 weeks of the trial. The participants were then randomized in equal numbers to an additional 8 weeks of galantamine or 8 weeks of placebo treatment. The primary efficacy measures were the changes from week 18 to week 26 (double-blind withdrawal) in the Frontal Behavioral Inventory (FBI), the Aphasia Quotient of the Western Aphasia Battery (WAB), the CGI of severity (CGI-S), and the CGI-I. Secondary objectives were to assess the safety of galantamine withdrawal treatment in patients with fvFTD and PPA and explore the efficacy of galantamine treatment in other cognitive domains. Secondary efficacy measures were the MMSE, Mattis Dementia Rating Scale (MDRS), Frontal Assessment Battery (FAB), NPI, Alzheimer’s Disease Cooperative Study—Activities of Daily Living Scale, and the subscales of the WAB and the FBI for the same time periods as the primary efficacy assessments. Safety assessments were conducted at screening, baseline, and weeks 12, 18, and 26 as well as telephone contacts for safety at weeks 4, 8, and 22. These safety assessments included incidence rates of adverse events, electrocardiograms, physical and neurological examinations, vital signs, and laboratory tests of biochemistry, hematology, and urinalysis. Results showed no significant differences in behavior or language for the total group. A treatment effect (P = .009) in a subgroup of patients with PPA in the global severity score favored galantamine when compared to placebo during the placebo-controlled withdrawal phase, but it was not considered significant after correction for multiple comparisons. The language scores for the treated PPA group also remained stable compared to the placebo group, which showed deterioration. Galantamine appeared safe in both patients with fvFTD and patients with PPA, although there was mild agitation in 5 participants during the open-label phase. All serious adverse events were deemed unrelated to study medication. The authors concluded that galantamine is not effective in the treatment of fvFTD, but a trend of efficacy was shown in the aphasic subgroup.

Memantine

In a study by Vercelletto et al, ¹⁸ 49 patients with fvFTD aged 45 to 75 years with a MMSE score ≥19 were enrolled in a national, randomized, double-blind, placebo-controlled Phase II trial to test the efficacy and tolerability of 1 year of treatment with memantine 10 mg twice a day. Patients with PPA, SD, motor neuron disease involvement, and those treated with an AChEI were excluded. The primary end point was the Clinician’s Interview-Based Impression of Change Plus Caregiver Input (CIBIC-Plus). The secondary end points included NPI, FBI, MDRS, MMSE, Disability Assessment for Dementia (DAD), and the Zarit Burden Inventory (ZBI). The screening visit was performed between 7 and a maximum of 21 days before randomization. Participants had 1 visit per month during the first 3 months, then every 3 months. Memantine safety was assessed by monitoring at each visit by recording adverse events, physical and neurological examination results, vital signs, and routine laboratory values (creatinine, CBC, platelets, and aspartate aminotransferase/alanine aminotransferase) were taken at inclusion and after 3 and 12 months of treatment. Of the 49 patients in the intent-to-treat population, 8 patients dropped out of the study. Of these, 5 were in the memantine group; 1 patient had worsening of SD (2 at months 4 and 5, 1 withdrew consent, and 1 was withdrawn by decision of the investigator at month 6); and 3 were in the placebo group (2 were withdrawn by the decision of the investigator at months 6 and 11 and 1 patient withdrew due to severe renal insufficiency). Results indicated that 52 weeks after baseline assessment, there were no significant differences on the CIBIC-Plus scores between the memantine group (n = 23) and the placebo group (n = 26), P = .4458. In the qualitative analysis, it was noted that 10 patients worsened in the memantine group versus 17 in the placebo group, although this difference was not statistically significant (P = .142). For secondary end points there were no differences in the scores between the memantine group and the placebo groups, (MMSE, P = .63), (MDRS, P = .95), (NPI, P = .25), (ZBI, P = .43), and (DAD, P = .10) However, the FBI score was lower in the memantine group compared to the placebo group (P = .0417). There were no medically significant changes in laboratory parameters during the study, and no differences were found between the 2 groups. The authors concluded that memantine appeared safe and well tolerated in patients with fvFTD but did not demonstrate efficacy for cognitive or behavioral symptoms in these patients.

Boxer et al conducted a randomized, parallel group, double-blind, placebo-controlled trials in patients with FTD. Participants met the Neary criteria for behavioral variant FTD (bvFTD) or SD, and they had the characteristic brain atrophy. ¹⁹ The use of AChEIs was prohibited in this study. Individuals were randomly assigned to receive either memantine 20 mg/d or matching placebo tablets (1:1) in blocks of 2 and 4 patients. The study period was 26 weeks. All patients and study personnel were masked to the treatment assignment. The primary end points were the change in total NPI score and the CGI of change (CGIC) score after 26 weeks and were analyzed by intention to treat. A total of 81 patients were randomly assigned to receive either memantine (39 patients) or placebo (42 patients). A total of 5 (6%) patients discontinued the study. Seventy-six patients completed the 26-week treatment period. The investigators found that treatment with memantine had no effect on either the NPI score, mean difference 2.2, 95% CI: 3.9 to 8.3, P = .47, or the CGIC, mean difference 0.0, −0.4 to 0.4, P = .90, after 26 weeks of treatment. The memantine group displayed worse neuropsychological performance compared to the placebo group on the tests of naming and processing speed. Memantine was generally well tolerated but patients in the memantine group had more frequent cognitive adverse events (confusion, memory loss, language disorders) when compared to the placebo group; 6 patients versus 1 patient, P = .056.

Oxytocin

In a study by Jesso et al, ²⁰ 20 patients with fvFTD received 1 dose of 24 international units of intranasal oxytocin or placebo and then returned 2 weeks after the first treatment visit and received the alternate medication, each time followed by the same assessments. The primary outcome measure was emotion recognition, operationalized using a facial expression recognition task, Facial Expression Recognition, and Intensity. Other measures of emotion processing and social cognition included Vocal Affect (an auditory emotion recognition task) and the Mind in the Eyes Theory of Mind task. Task order was randomized across patients, but for each individual patient, tasks were administered in the same order for their placebo and oxytocin treatment visits. Secondary outcome measures assessed whether a single dose of oxytocin improved FTD-related behaviors as indexed by caregiver reports on the NPI and the FBI at 8 hours and 1 week following drug administrations. Vital signs were recorded during each experimental visit prior to nasal solution administration and after completion of the emotion processing tasks. Adverse event and side effect information was collected via follow-up telephone calls by a research nurse to caregivers on the evening of treatment visits and 1 week after each treatment visit. A significant improvement in scores in the NPI was observed on the evening of oxytocin administration compared with placebo and the baseline ratings. Oxytocin was also associated with reduced recognition of angry facial expressions by patients with fvFTD. There were no significant adverse events attributable to the study medication, and there were no dropouts in the study. The authors opined that these findings suggest that oxytocin is a potentially promising, novel symptomatic treatment candidate for patients with fvFTD and that further study of this neuropeptide in FTD is warranted.

Discussion

The treatments for FTD thus far have involved modulating different neurotransmitters and neuropeptides that are involved in the disease process. ²¹ The neurotransmitters and neuropeptides that have been implicated in the disease process include serotonin, dopamine, glutamine, acetylcholine, and oxytocin. Serotonin is a prominent neurotransmitter involved in the frontosubcortical neural pathway. ^11,12 The FTD is associated with marked losses of brain serotonin, which most likely contributes to the symptoms of hyperorality, weight gain, altered preference for carbohydrates, and compulsions. ^{12 –14} Dopamine has also been postulated to play a role in fvFTD via aberrant function in the orbitofrontal cortex, particularly in the area of reward-based deficits. ^15,16 There is conflicting evidence for the role of cholinergic neurotransmission in FTD, with some reviews favoring normal cholinergic function. ¹⁷ Glutaminergic transmission has also been considered in patients with FTD with alterations of 2-amino-3-[3-(carboxymethoxy)-5-methylisoxazol-4-yl]propionate) and NMDA receptor activity. ¹⁸ Finally, the neuropeptide oxytocin, produced in the supraoptic and paraventricular nuclei of the hypothalamus and delivered to the pituitary for systemic release, is an important mediator of social cognition and emotion processing. ²⁰ Given these deficits in patients with FTD, it has been hypothesized that oxytocin could serve as a symptomatic treatment for social and behavioral deficits in these patients.

In our systematic review of the literature, we found a total of 9 randomized controlled double-blinded clinical trials for the treatment of patients with FTD. Of these, 2 trials used the SSRI paroxetine, with 1 trial using trazodone, 2 trials using stimulants (methylphenidate and dextroamphetamine), 1 trial using the AChEI galantamine, 2 trials using the NMDA antagonist memantime, and 1 trial using the neuropeptide oxytocin. ^{12 –19,20} Of the 9 trials, 5 trials focused exclusively on the fvFTD subtype of FTD, 1 trial focused on the fvFTD and the PPA subgroups, and 1 trial focused on the fvFTD and SDs. ^{13,15,16 –19}

In the SSRI trials, 1 trial compared paroxetine (up to 20 mg/d) to piracetam (GABA analog) in patients with FTD and 1 trial compared paroxetine to placebo in patients with fvFTD. ^12,13 The first trial demonstrated an improvement in behavioral symptoms with paroxetine as demonstrated by a reduction in caregiver stress, whereas the second showed no improvement in behavioral symptoms and a selective impairment on certain types of memory function. Paroxetine appeared to be well tolerated in both these trials. In the trazodone trial, trazodone was compared to placebo in patients with FTD. ¹⁴ There was improvement in behavioral symptoms on trazodone as measured by a >50% decrease in NPI scores in 10 participants. There were more adverse events in the trazodone group compared to the placebo group, but they were rated as mild and only 1 patient left the trial. In the methylphenidate trial, methylphenidate was compared to placebo in patients with fvFTD. ¹⁵ Improvement was seen in “normalizing” the decision making of participants in becoming less risk taking, although no significant effects on over all cognitive function or on subjective mood was noted. Methylphenidate was well tolerated in this trial. In the dextroamphetamine trial, dextroamphetamine was compared to quietapine in patients with FTD. ¹⁶ There were improvements in the behavioral symptoms in the dextroamphetamine group including disinhibition and apathy. Both medications were well tolerated in this trial. In the galantamine trial, galantamine was compared to placebo in patients with both fvFTD and PPA. ¹⁷ No significant improvements in behavior or language were found for the total group, but a trend of efficacy (in global severity and language) was seen in the PPA subgroup. Galantamine was well tolerated in this trial. In the more recent studies, memantine was compared to placebo in 2 trials, and in 1 trial, oxytocin was compared to placebo in patients with fvFTD. ^18,19,20 In the memantine trials, there were no improvements in behaviors in either trail, and in the second trial, cognition appeared a little worse in the memantine group when compared to the placebo group. ^18,19 In the oxytocin trial, there were improvements in behavioral scores and a reduced recognition of anger and fear in the oxytocin group when compared to the placebo group. ²⁰ In all 3 trials, the medications were well tolerated when compared to placebo. ^18,19,20

All trials were rated as being of good quality based on Centre for Evidence Based Medicine randomized controlled trial evaluation criteria (Table 1). The major limitations of these studies include the small number of participants, the use of inconsistent outcome variables that did not consistently target specific behaviors and the use of rating scales that are not sensitive to the symptoms of FTD like the MMSE.

Table 1.

Quality of studies reviewed Based on Centre for Evidence Based Medicine (CEBM) Randomized-Controlled Trial Evaluation Criteria (http://www.cebm.net/index.aspx?o=1913).

Name of Study	Randomization	Similar Groups Initially?	Equal Treatments?	All Participants Accounted For?	Analyzed in Groups to Which They Were Randomized?	Objective/“Blind” Treatments?
Moretti et al 12	Yes	Yes	Yes	Yes	Yes	No
Deaklin et al 13	Yes	Yes	Yes	Yes	Yes	Yes
Lebert et al 14	Yes	Yes	Yes	Yes	Yes	Yes
Rahman et al 15	Not specified	Yes	Yes	Yes	Yes	Yes
Huey et al 16	Yes	Not specified	Yes	Yes	Not specified	Yes
Kertesz et al 17	Yes	Yes	Yes	Yes	Yes	No
Vercelletto et al 18	Yes	Yes	Yes	Yes	Yes	Yes
Boxer et al 19	Yes	Yes	Yes	Yes	Yes	Yes
Jesso et al 20	Yes	Yes	Yes	Yes	Yes	Yes

The summary of evidence from this review indicates that of the 9 randomized controlled double-blinded trials discussed, 6 showed some degree of improvement with treatment (Table 2). With the SSRI trials, there was conflicting evidence regarding improvement in behavioral symptoms, as initial data showed efficacy for paroxetine in the treatment of fvFTD whereas a subsequent trial did not. In the trazodone trial, there was an improvement in behavioral symptoms in patients with FTD. Behavioral symptoms were also noted to be improved in the methylphenidate, dextroamphetamine, and oxytocin trials. The memantine trials did not show any benefit for behaviors. No improvements were noted for cognition or function in any of the trials. In the second memantine trial, cognition appeared a little worse in the drug-treated group compared to the placebo group. These medications were noted to be well tolerated in the trials.

Table 2.

Summary of Studies.

Name of the Study	Diagnosis	Total Number of Participants	Duration of Trial	Comparators	Maximum Dosage of Active Drugs	Outcomes	Tolerability
Moretti et al 12	FTD	16	14 months	Paroxetine or piracetam	Paroxetine up to 20 mg/d, piracetam up to 1200 mg/d	Paroxetine was associated with improvements in behavioral symptoms as reflected by a reduction in caregiver stress.	Medications were well-tolerated in both groups.
Deakin et al 13	fvFTD	10	9 weeks	Paroxetine or placebo	Paroxetine up to 40 mg/d	There was no improvement in cognition or behavior.	Paroxetine appeared to be well-tolerated in this trial as there was no mention of side-effects.
Lebert et al 14	FTD	26	12 weeks	Trazodone or placebo	Trazodone up to 150 mg/d	There were improvements in behavioral symptoms as measured by a >50% decrease in NPI scores in 10 patients.	Side effects were more frequent with trazodone than with placebo, but were rated as mild.
Rahman et al 15	fvFTD	8	1-2 weeks	Methyl-phenidate or placebo	Single-dose methyl-phenidate 40 mg	There were improvements in “normalizing” the decision-making ability of patients where they were becoming less risk taking, although no significant effects were noted on cognition or on subjective mood state.	Methyl-phenidate was well tolerated with no significant cardiovascular effects noted.
Huey et al 16	FTD	8	9 weeks	Dextro-amphetamine or quetiapine	Dextro-amphetamine up to 20 mg/d, quetiapine up to 150 mg/d	Improvements in behavioral symptoms were noted for the dextroamphetamine group, especially with disinhibition and apathy.	One participant had with sedation on quetiapine; otherwise both medications were well tolerated in this trial.
Kertesz et al 17	fvFTD and PPA	36	26 weeks	Galantamine or placebo	Galantamine 8-24 mg/d	No significant improvements were noted in behavior or language for the total group but a trend of efficacy was noted in the PPA subgroup.	Galantamine was well tolerated in this trial.
Vercelletto et al 18	fvFTD	49	1 year	Memantine or placebo	Memantine 10 mg twice a day	There were no improvements in symptoms in the memantine group.	Memantine was well tolerated in the trial.
Boxer et al 19	fvFTD/bvFTD or semantic dementia	81	26 weeks	Memantine or placebo	Memantine 10 mg twice a day	There were no improvements in symptoms in the memantine group.	Memantine was well-tolerated in the trial but cognitive side effects were more in the memantine group.
Jesso et al 20	fvFTD	20	3 weeks	Oxytocin and placebo	Single dose 24 IU intranasal oxytocin	Improvement in behavioral scores on NPI and reduced recognition of anger and fear were noted in the oxytocin group.	Oxytocin was well tolerated in the trial.

Abbreviations: bvFTD, behavioral-variant frontemporal dementia; fvFTD, frontal-variant frontemporal dementia; NPI, Neuropsychiatric Inventory; PPA, primary progressive aphasia.

How does the evidence from these randomized controlled trials compared to those of recent nonrandomized trials? In an excellent review, Vossel and Miller reviewed recent treatment studies in patients with FTD. ²² The investigators found a total of 5 studies published after January, 2007. There were 2 open label trials (memantine and galantamine), 1 case series of memantine and 2 case reports, 1 of topiramate and 1 of sertraline. The memantine study showed a worsening in cognition in patients with FTD and no improvement in their behavior. ²³ The galantamine study showed no benefit for the drug in patients with bvFTD, while there was a trend for efficacy for patients with PPA. ²⁴ The memantine case series of 3 patients with bvFTD showed improvement in the NPI scores, especially those for apathy, anxiety, and agitation. ²⁵ The topiramate case report indicated that in 1 case, the medication was beneficial for alcohol abuse but not for other behaviors. ²⁶ In the sertraline case report, the drug was beneficial in treating inappropriate sexual behaviors in a patient with FTD. ²⁷ In an open-label trial, 43 patients who met clinical criteria for Frontal Lobar Degeneration received 26 weeks of open-label treatment with memantine at a target dose of 20 mg/d. ²⁸ In this study, no concurrent treatment with AChEIs was allowed. The investigators noted transient improvements in the total NPI score, especially in the FTD group. Variable declines were noted on the Alzheimer's Disease Assessment Scale-Cognitive (ADAS-cog), Executive Interview (EXIT25), FBI, NPI, Texas Functional Living Scale (TFLS), and the Unified Parkinson's Disease Rating Scale (UPDRS) scores. The investigators found that the FTD and SD groups had declined the most of the cognitive and behavioral outcome measures but had remained stable on the UPDRS. The PNFA group remained relatively stable on the ADAS-cog, NPI, and TFLS but declined on the UPDRS. Memantine was well tolerated in these patients at the target dose.

As there are no FDA-approved medications to treat FTD what is the prescription pattern for individuals with FTD? This issue was evaluated in the study by Hu et al. ²⁹ The investigators reviewed the use of anti-Alzheimer’s Disease medications, antipsychotics, mood stabilizers, antidepressants, and anxiolytics in patients with bvFTD when compared to patients with AD from a cohort that was derived from the Alzheimer’s Research Centers of California (ARCC) database and a multicenter, natural history study (NHS) of FTD. They found that the overall use of such medications was as common in patients with bvFTD as compared to patients with AD in specialized dementia clinical and research centers. Approximately 41% of the patients with bvFTD took an AChEI. The use of memantine was similar between the 2 groups at about 29%. The investigators also found that the prescription of most psychiatric medications was similar between patients with AD and patients with bvFTD. They noted that the use of antidepressants was higher in the bvFTD groups than the AD groups and attributed this difference to the younger age in the patients with bvFTD, as younger age was more strongly associated with antidepressant use than diagnosis when the data from both groups were combined.

Are there concerns with the use of any of these medications in patients with FTD? In a study involving 12 patients with FTD who had received donepezil for 6 months who were compared to 12 FTD controls on behavioral measures, the investigators found that the 2 groups did not differ on most variables at baseline or at 6 months but the donepezil group had greater worsening on the FTD Inventory with 4 patients having increased disinhibited or compulsive acts that resolved on discontinuation of the medication. ³⁰ In a retrospective chart review by Irwin et al, the investigators found that in 21 patients who met the criteria for behavioral or aphasic variants of frontotemporal lobar degeneration (FTLD), treatment with AChEIs was associated with a decrease in the MMSE scores by 1.6 points (P = .02) and an increase in total behavior scores by 1 of 3 points (P = .03). ³¹ The investigators found that the mean MMSE and behavioral scores were improved in patients who received SSRIs when compared to AChEIs (P = .04). The 2 studies indicate that when patients with FTD are treated with AChEIs, there may be worsening of cognitive and behavioral symptoms in these patients.

Can pharmacotherapeutic trails for patients with FTD be standardized? Current evidence indicates that studies use different diagnostic criteria for the diagnosis of FTD, and the rating scales used in these studies are also variable. ³² Some of the scales that are used in these studies are not sensitive to the symptom presentation in patients with FTD. A good example is the MMSE which is often used as a standard scale for assessing clinical severity in patients with dementia but is relatively insensitive to symptoms of early FTD. Rating scales used in these studies should not only assess cognition but also frontal lobe functions, behaviors, the severity of the illness, and the level caregiver burden. Examples of standardized scales that are appropriate for studies in patients with FTD include the Clinical Dementia Rating (CDR) Scale, the FAB, the EXIT, Wisconsin Card Sorting Test, Clock Drawing Test, FBI, Clinician's Interview-Based Impression of Change plus (CIBIC-Plus), and so on. Global measures like the frontotemporal lobar degeneration (FTLD) modified CDR (FTLD-modified CDR), and the CGIC have also been found to have suitable properties for clinical trials in these patients. ³³

What are the future therapies that are in the pipeline for treating patients with FTD? In an extensive review of possible therapeutic modalities for FTD, Boxer et al indicated that molecular and animal models are being used to develop new treatments for FTD. ³⁴ These trials include drugs that target tau-proteins including lithium, riluzole, rasagiline, Co-Q10, and chloroquine to name a few. Drugs that target progranulin gene are also being trialed including chloroquine, amiodarone, and resveratrol. Preclinical models are targeting the MAPT gene that codes for tau-proteins, transactive response DNA-binding protein 43 encoded by the TARDBP gene, valosin-containing protein, and the charged multivesicular body protein 2b (CHMP2B) gene. Although some of these therapies have shown promise, they are not yet ready for routine clinical use.

Conclusions

This systematic review of randomized controlled trials indicates that paroxetine, trazodone, methylphenidate, dextroamphetamine, and oxytocin, improve behavioral symptoms of FTD while being well tolerated. However, none of the trials showed an improvement in cognition or function in patients with FTD. Clinical trials using standardized diagnostic criteria with well-defined outcomes that are measured by certain specific rating scales sensitive to FTD symptoms are needed before it can be concluded that these medications are definitively indicated for the treatment of patients with FTD. Future therapies will include drugs that target certain proteins and genes that lead to the pathological brain changes and symptoms associated with FTDs.