Safety and Tolerability of GRF6019 Infusions in Severe Alzheimer’s Disease: A Phase II Double-Blind Placebo-Controlled Trial

Abstract

Background:

The plasma fraction GRF6019 shows multiple benefits on brain aging in mice, including enhanced cognition, neurogenesis, and synaptic density, as well as reduced neuroinflammation.

Objective:

To evaluate the safety, tolerability, and preliminary efficacy of GRF6019 in patients with severe Alzheimer’s disease (AD).

Methods:

A phase II, double-blind, placebo-controlled study in patients with severe AD (Mini-Mental State Examination score 0–10). Patients were randomized 2 : 1 to GRF6019 (N = 18) or placebo (N = 8) and received daily 250 mL intravenous infusions over 5 days. The primary endpoints were the rates of adverse events (AEs) and the tolerability of GRF6019 as assessed by the number of patients completing the study. Change from baseline in cognitive and functional assessments was also evaluated.

Results:

All patients completed 100%of study visits and infusions. The rate of AEs was similar in the GRF6019 (8/18 patients [44.4%]) and placebo (3/8 patients [37.5%]) groups, and there were no deaths or serious AEs. The most common AEs considered related to treatment were mild, transient changes in blood pressure in the GRF6019 group (hypotension: 2 patients [11.1%]; hypertension: 1 patient [5.6%]); there were no related AEs in the placebo group. The trial was not powered to detect statistically significant differences between treatment groups. At the end of the study, patients in both treatment groups remained stable or improved on all cognitive and functional endpoints.

Conclusion:

GRF6019 demonstrated excellent safety, feasibility, and tolerability. Future trials designed to characterize the potential functional benefits of GRF6019 and related plasma fractions in severe AD are warranted.

Keywords

Aging Alzheimer’s disease blood proteins dementia plasma randomized controlled trial

INTRODUCTION

Alzheimer’s disease (AD) is a progressive neurodegenerative disease of which cognitive decline is the primary manifestation. Age is the most important risk factor for developing the disease. Approximately 5.8 million Americans aged 65 and older have AD, of whom 80%are aged 75 and older, and it is predicted that this number will rise to 13.8 million by the year 2050 [1, 2]. The initial symptoms typically include an inability to retain recently acquired information, whereas late-stage disease is characterized by behavioral changes, loss of functional abilities, impaired communication, and disorientation [2]. Cholinesterase inhibitors (e.g., donepezil) and memantine are approved for the symptomatic treatment of AD; however, the effects are modest [3 –5]. No drugs are currently approved to treat the behavioral symptoms that develop in the later stages of the disease. There is a substantial unmet need for treatments to slow or stop AD progression in mild and moderate AD, as well as to alleviate its symptoms and improve function in later stages of disease. Given the association between age and the development and progression of AD, therapeutics that target the aging process itself, specifically age-related changes in the plasma proteome, may be a novel approach to the treatment of AD.

Although mice do not develop AD, the brains of old mice are characterized by reduced neurogene-sis, reduced synaptic density and function, increased astrogliosis and microgliosis, decreased c-Fos ex-pression in the dentate gyrus (indicative of decreased neuronal activity), and impaired spatial and associative learning [6]. These age-related changes in mice are similar to what occurs in aging human brains [7, 8]. Moreover, there is substantial overlap between these age-related changes and the pathological hallmarks of AD such as microgliosis and synaptic and neuronal loss [9].

Data from heterochronic parabiosis in mice suggest that rejuvenating factors from the plasma of young animals may ameliorate the effects of aging [10]. Subsequent nonclinical studies confirmed that exposure of old mice to young plasma counteracts age-related impairment at the molecular and structural levels in the hippocampus, and there is a beneficial effect of young human plasma and plasma protein fractions in age-related cognitive decline and histopathological endpoints in mouse models [11, 12]. Together, these studies provide the foundation for the hypothesis that soluble circulating factors from young plasma may prove beneficial against brain aging and ameliorate memory impairment in people experiencing cognitive decline from age-associated neurodegenerative diseases.

The safety and feasibility of whole plasma administration in individuals with AD was first evaluated in the Plasma for Alzheimer Symptom Amelioration (PLASMA) study [13]. While it was not powered to determine efficacy, patients demonstrated improvement on the AD Cooperative Study Activities of Daily Living (ADCS-ADL) scale and the Functional Activities Questionnaire. However, there are inherent risks in whole plasma administration, such as the potential transfer of unknown pathogens to recipients, allergic reactions to proteins such as clotting factors and immunoglobulins, and other acute transfusion reactions [14, 15]. Safer products have been developed by pooling plasma from multiple donations, fractionating them into more defined products, and adding additional processing steps to reduce the potential for pathogen transmission. GRF6019 is a human plasma fraction that leverages this fractionation technology. It is depleted of coagulation factors and gamma globulins, does not require refrigeration or cross-matching, and serves as a viable source of infusible plasma proteins from young, healthy donors. In a recent study conducted in patients with mild-to-moderate AD, daily infusions of either 100 mL or 250 mL of GRF6019 for 5 consecutive days over 2 dosing periods were safe and well tolerated, and patients in the study experienced no cognitive decline and minimal functional decline [16].

Having observed safety, tolerability, and preliminary efficacy of GRF6019 infusions in patients with mild-to-moderate AD, the current trial was designed to evaluate the safety and tolerability of a similar dosing regimen in severe AD and explore the feasibility of daily intravenous (IV) dosing in this challenging population with late-stage disease.

MATERIALS AND METHODS

Study design

This was a randomized, double-blind, placebo-controlled phase II study conducted at 4 sites in the United States. Patients received a 250 mL IV infusion of GRF6019 or placebo each day over 5 consecutive days (“pulsed dosing”). During the 5-day dosing period, patients who did not already reside in a long-term care or skilled nursing facility were admitted to the research facility to enable safety monitoring and evaluation. The timing of the infusions each day was optimized to minimize the risk of sundowning or other behaviors from potentially interfering with the infusion. For example, infusions were often administered after a meal, and patients would occasionally fall asleep during the procedure. The 250 mL GRF6019 dose level was selected based on safety data from prior clinical experience and isometric scaling from preclinical studies in mice [16]. The total study duration was approximately 9 weeks for each patient, including a 4-week screening period and 4 weeks of follow-up after the 5-day infusion period.

The study was conducted in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice and was approved by appropriate Institutional Review Boards. Participating patients’ legally authorized representatives provided written informed consent for the patient to participate in the study, and each patient provided assent. The study is registered on ClinicalTrials.gov (identifier NCT03765762). There were no major changes to trial design or outcomes after the trial commenced.

Participants

Patients living at home or in skilled nursing facilities who were 60–95 years old with a diagnosis of AD based on the National Institute of Aging–Alzheimer’s Association criteria and a Mini-Mental State Examination (MMSE) score of 0–10 (inclusive) were eligible to participate in the study. Since patients with severe AD are often treated with cholinesterase inhibitors (rivastigmine, galantamine, donepezil) and/or memantine, use of these medications was allowed to ensure that the sample was representative of this population. If patients were taking any of these medications for cognition or antipsychotics, antidepressants, and/or benzodiazepines, they had to have been on a stable dose for at least 8 weeks prior to baseline. Patients were excluded if they had a level of agitation that, in the opinion of the investigator, could interfere with study procedures, or if they had evidence of a clinically relevant neurological disorder other than AD or other conditions that may have interfered with the safety of the patient or interpretation of the study data (see Supplementary Methods for full inclusion and exclusion criteria). Given that a volume of 250 mL of GRF6019 or placebo was administered daily over 5 consecutive days in euvolemic patients, one potential concern was volume overload in patients with undiagnosed compromised cardiac function; therefore, transthoracic echocardiogram was performed at screening to exclude patients with impaired cardiac function. During screening, patients also underwent a medical and neurologic evaluation, laboratory tests, and electrocardiographic evaluations. A detailed medical and social history was obtained through an interview with the patient’s caregiver or review of prior medical records.

Randomization and blinding

Patients were randomized in a 2:1 ratio to re-ceive GRF6019 or placebo using block randomization stratified by sex. The 2:1 randomization ratio was intended to increase the number of patients exposed to active GRF6019 and thus provide sufficient safety data while preserving the principles of blinded treatment assessment. A statistician who was not otherwise involved in the study provided the computer-generated randomization codes that were manually assigned to patients sequentially within the specified strata. Unblinded site personnel contacted the unblinded clinical research organization staff to obtain the next sequential randomization number and associated treatment assignment. The unblinded site personnel assigned a kit number from the site study drug supply based on the randomized treatment allocation. All outcome measures were assessed by blinded outcomes assessors or other blinded raters. An unblinded pharmacist or other qualified staff member responsible for drug accountability dispensed the vials containing 250 mL of GRF6019 or placebo to an unblinded infusion nurse who administered the infusion. GRF6019 has a pale-yellow color and colorless saline was used as the placebo control; therefore, blinding of the vial was necessary. Robust measures were taken to mask the GRF6019/placebo containers and IV setup to ensure that outcomes assessors, raters, the principal investigator, other study personnel, and patients and their caregivers were unaware of the allocation. Most patients received their infusions in a designated area, and a curtain, drape, blinding bag, or equivalent barrier was used to shield the infusion administration setup and/or vial from view from everyone but the unblinded infusion nurse, who was also responsible for concealing and returning the used containers of GRF6019 or placebo to the pharmacy. When infused through a standard IV line, the color of GRF6019 is indistinguishable from normal saline; therefore, covering the IV lines was not necessary. During infusions, the unblinded infusion nurse was instructed to limit communication with blinded study personnel to only that which was required to ensure the immediate safety of patients.

Outcome measures

The primary endpoints were the rates of treatment-emergent adverse events (AEs) and serious AEs (SAEs) and the number of patients completing 4 weeks of observation after receiving 5 infusions.

The secondary safety endpoints were change from baseline in clinical laboratory parameters, vital sign measurements, body weight, physical examination findings, and electrocardiogram recordings. Secondary feasibility endpoints included the success of blinding and patient compliance with the study visit schedule, procedures, infusions, and adaptations necessary for study completion.

Cognitive and functional secondary endpoints in-cluded change from baseline on the MMSE [17], Severe Impairment Battery (SIB) [18], ADCS-ADL for Severe AD (ADCS-ADL-Severe) [19], Alzhei-mer’s Disease Cooperative Study –Clinical Global Impression of Change (ADCS-CGIC) [20, 21], and the Neuropsychiatric Inventory (NPI; or Neuropsychiatry Inventory –Nursing Home version [NPI-NH] for patients residing in nursing homes or other long-term care settings [22, 23] (see schedule of assessments in Table 1). For assessments that required an informant (e.g., caregiver), the informant was kept consistent throughout the duration of the study unless a change was unavoidable. Caregivers were required to have had frequent contact with the patient (approximately 10 hours per week).

Table 1

Cognitive and functional assessment timepoints

	Screening/Baseline		Treatment	Follow-up		EOS
Week	0		1		2	5
Day	– 28 to – 8	– 7 to – 1	1 to 5	6±1– 3^a	12±3	33±7
Visit	1	2	3 to 7	8	9	10
MMSE	X			X		X
SIB		X		X	X	X
ADCS-ADL-Severe		X				X
ADCS-CGIC		X		X	X	X
NPI or NPI-NH^b		X			X	X

ADCS-ADL-Severe, Alzheimer’s Disease Cooperative Study – Activities of Daily Living Severe; ADCS-CGIC, Alzheimer’s Disease Cooperative Study – Clinical Global Impression of Change; EOS, end of study; MMSE, Mini-Mental State Examination; NPI, Neuropsychiatric Inventory; NPI-NH, Neuropsychiatric Inventory – Nursing Home; SIB, Severe Impairment Battery. ^aThe treatment period was 5 days. Visit 8 could occur on Day 6 + 1 to 3 days following the treatment period to allow flexibility, if needed. ^bThe NPI was used for patients residing in the community and the NPI-NH was used for patients residing in care facilities.

Statistical analysis

The sample size was chosen based on clinical considerations of an exploratory study designed to evaluate the safety and tolerability of the GRF6019 dosing regimen in patients with severe AD. Safety and tolerability were evaluated by comparing the rates and severity of AEs, including treatment-emergent AEs and AEs leading to discontinuation from the study, in the active and placebo groups. All safety analyses were performed using the Safety Set, which included all patients who received at least 1 dose of the study agent. Adverse events were coded using the Medical Dictionary for Regulatory Activities (MedDRA) Preferred Term (PT) and grouped by System Organ Class (SOC). Patients experiencing more than 1 AE within a given MedDRA PT or SOC were counted once within that PT or SOC at the highest severity and strongest attribution of relatedness to the study agent.

The study was not powered to detect differences between GRF6019 and placebo on secondary efficacy endpoints; therefore, the statistical approach was primarily descriptive and no adjustments for multiplicity were employed. Within-patient changes from baseline for each dosing group and their distribution around a null value of zero, as well as between-group differences, were evaluated to identify any trends in differences between patients randomized to GRF6019 and placebo. Because the analyses were performed on the change from baseline scores, no additional adjustment was made for baseline in this small study. Analyses of efficacy endpoints were performed on both the Evaluable and Per Protocol Sets. The Evaluable Set included all patients who received at least 4 of the 5 planned doses, and the Per Protocol Set was a subset of the Evaluable population who did not have any protocol deviations that could have affected the evaluation of efficacy. All analyses were performed using SAS® version 9.4 or higher. No imputation was performed for missing data, and no interim analyses were conducted.

RESULTS

Disposition and baseline characteristics

From January 2019 to December 2019, 43 patients were screened for eligibility, of whom 26 were randomized to receive either GRF6019 (N = 18) or placebo (N = 8). All 26 enrolled patients received all 5 infusions and completed the study (Fig. 1). Therefore, the Safety Set consisted of 26 patients (18 who received GRF6019 and 8 who received placebo). During monitoring, it was discovered that 3 patients had errors in their screening MMSE scores (when recalculated, these patients had screening MMSE scores of 11 or 12). Another patient had severe aphasia due to AD and, while the level of functioning and other tests was consistent with severe AD, it was not possible to obtain a reliable MMSE score. As a result, these 4 patients were not included in the Per Protocol Set. All 4 were in the GRF6019 group; therefore, the GRF6019 Evaluable Set had 18 patients while the GRF6019 Per Protocol Set had 14 patients. The placebo arm had 8 patients in both the Evaluable and Per Protocol Sets.

Fig. 1

Study flow diagram. ^aFour patients in the GRF6019 treatment group did not meet the Mini-Mental State Examination score inclusion criteria at screening, which was not discovered until after the patients had completed all infusions and study visits. As a result, these patients were included in the safety analyses but excluded from the Per Protocol efficacy analyses.

Overall, patients in the study were predominantly female (61.5%) and white (96.2%) with a mean (standard deviation [SD]) age of 73.5 (8.5) and a duration of AD of 5.3 (2.5) years (Table 2). The groups were well balanced on most demographic characteristics, including the distribution of apolipoprotein E (APOE) genotypes. Seventy-seven percent of the patients were APOE4 gene carriers. There were two important baseline differences between the groups: a greater proportion of patients randomized to GRF6019 were taking a cholinesterase inhibitor (77.8%) than patients randomized to placebo (37.5%), and patients receiving GRF6019 were older than those receiving placebo (mean [SD] age of 75.2 [8.3] years and 69.5 [8.1] years, respectively).

Table 2

Patient demographics and baseline characteristics (Safety Set)

Characteristic	GRF6019 (N = 18)	Placebo (N = 8)	Overall (N = 26)
Age (y), mean (SD)	75.2 (8.25)	69.5 (8.12)	73.5 (8.49)
Sex, n (%)
Male	7 (38.9)	3 (37.5)	10 (38.5)
Female	11 (61.1)	5 (62.5)	16 (61.5)
Race, n (%)
White	17 (94.4)	8 (100)	25 (96.2)
Asian	1 (5.6)	0	1 (3.8)
Ethnicity, n (%)
Hispanic or Latino	0	1 (12.5)	1 (3.8)
Not Hispanic or Latino	18 (100)	7 (87.5)	25 (96.2)
Duration of AD (y), mean (SD)	5.0 (1.81)	6.0 (3.59)	5.3 (2.46)
APOE genotype, n (%)
E2/E3	2 (11.1)	0	2 (7.7)
E2/E4	1 (5.6)	0	1 (3.8)
E3/E3	2 (11.1)	2 (25.0)	4 (15.4)
E3/E4	10 (55.6)	5 (62.5)	15 (57.7)
E4/E4	3 (16.7)	1 (12.5)	4 (15.4)
E4 Carriers	14 (77.8)	6 (75.0)	20 (76.9)
Concomitant medications, n (%)	17 (94.4)	8 (100)	25 (96.2)
Cholinesterase inhibitors	14 (77.8)	3 (37.5)	17 (65.4)
Memantine	9 (50.0)	4 (50.0)	13 (50.0)
Antidepressants	9 (50.0)	3 (37.5)	12 (46.2)
Antipsychotics	1 (5.6)	4 (50.0)	5 (19.2)
Benzodiazepines	0	2 (25.0)	2 (7.7)
Trazodone	2 (11.1)	0	2 (7.7)
Valproate	2 (11.1)	2 (25.0)	4 (15.4)
Current residence, n (%)
Home	14 (77.8)	6 (75.0)	20 (76.9)
Care facility	4 (22.2)	2 (25.0)	6 (23.1)

AD, Alzheimer’s disease; APOE, apolipoprotein E; SD, standard deviation.

Safety

The rate of treatment-emergent AEs that occurred in the study was similar in the GRF6019 (8/18, 44.4%) and placebo (3/8, 37.5%) groups (Table 3). The most common AEs (occurring in > 1 patient) were diarrhea (2/18, 11.1%in GRF6019 group; 1/8, 12.5%in placebo group), hypotension (2/18, 11.1%in GRF6019 group; 1/8,12.5%in placebo), edema (1/18, 5.6%in GRF6019; 1/8, 12.5%in placebo), and nausea (1/18, 5.6%in GRF6019 group; 1/8,12.5%in placebo). All AEs in the placebo group were mild in intensity and considered unrelated to study treatment. The majority of AEs in the GRF6019 group were mild in intensity, with 3 patients experiencing moderate-intensity AEs (1 with nausea and 2 with increased blood pressure). Five patients receiving GRF6019 experienced AEs considered possibly related to treatment; of these, changes in blood pressure (hypotension: 2/18, 11.1%; hypertension: 1/18, 5.6%) were the most common. Based on data from a previous study with GRF6019 in mild-to-moderate AD, it was expected that mild, transient changes in blood pressure would occur [16]. Therefore, the occurrence of blood pressure values above or below pre-specified thresholds was evaluated and found to have a similar occurrence across both treatment groups (Table 4). Transient changes in blood pressure normalized by 2 hours after the end of the infusions in most patients.

Table 3

Treatment-emergent adverse events^a

	GRF6019 (N = 18)	Placebo (N = 8)
	Number (%) of Patients

Patients with any AE	8 (44.4)	3 (37.5)
Mild	5 (27.8)	3 (37.5)
Moderate	3 (16.7)	0
Severe	0	0
Patients with any study drug-related AE^b	5 (27.8)	0
Diarrhea	1 (5.6)	0
Hypotension^c,d	2 (11.1)	0
Hypertension^c	1 (5.6)	0
Edema^c	1 (5.6)	0
Patients with any SAE	0	0
Patients with any AE leading to discontinuation	0	0
Patients with an AE leading to death	0	0
Most common adverse events (occurring in > 1 patient)
Diarrhea	2 (11.1)	1 (12.5)
Hypotension	2 (11.1)	1 (12.5)
Edema	1 (5.6)	1 (12.5)
Nausea	1 (5.6)	1 (12.5)

AE, adverse event; MedDRA, Medical Dictionary for Regulatory Activities; PT, Preferred Term; SAE, serious adverse event. ^aAEs were coded based on MedDRA PT. Patients experiencing more than 1 AE within a given MedDRA PT were counted once within that PT at the highest severity and strongest attribution of relatedness to the study agent. ^bAssessed by investigator as possibly related to the study drug. ^cAlso considered an AE of special interest. ^dTwo patients experienced hypotension; of these, 1 patient experienced 2 episodes.

Table 4

Blood pressure measurements of special interest

Blood Pressure, mmHg	GRF6019 (N = 18)^a	Placebo (N = 8)
	Number (%) of Patients
Systolic blood pressure
> 200	0	0
> 180	0	0
< 90	0	0
Diastolic blood pressure
> 120	0	0
> 110	0	0
< 50	2 (11.1)	2 (25.0)
> 30%change in systolic/diastolic blood pressure^b	8 (44.4)	4 (50.0)

^aTwo of the patients receiving GRF6019 experienced AEs of hypotension, both of which were considered related to the study drug. None of the other blood pressure measurements of special interest were considered AEs. ^bAn absolute change from Day 1 baseline in systolic and/or diastolic blood pressure. This change includes positive and negative percent changes.

There were no SAEs or deaths reported, and no patients discontinued treatment due to AEs. In addition, there were no clinically meaningful changes in electrocardiograms or physical examination findings, and only one patient had clinically significant abnormal laboratory results: a mild AE of increased blood creatine phosphokinase that was deemed unrelated to study treatment and resolved by the end of the study.

Feasibility and tolerability

All patients were 100%compliant with the visit schedule and completed all 5 infusions. No blinded study staff believed they knew which treatment any patient received. In addition, no patients, caregivers, or other study personnel were intentionally or unintentionally unblinded.

Efficacy

At baseline, patients receiving GRF6019, in addition to being older, had lower mean scores (indicating more cognitive impairment) on the MMSE and SIB than patients in the placebo group, with mean (SD) MMSE total scores of 5.9 (3.5) in the GRF6019 Evaluable patients, 5.0 (3.1) in the GRF6019 Per Protocol patients, and 7.3 (3.4) in the placebo patients, and mean (SD) SIB total scores of 51.6 (29.2) in the GRF6019 Evaluable patients, 44.1 (26.4) in the GRF6019 Per Protocol patients, and 61.7 (26.4) in the placebo patients (Table 5). While these baseline differences were not statistically significant due to the small sample size, they may have affected the response to GRF6019. The baseline score on the ADCS-ADL-Severe was comparable in the two groups. Different versions of the NPI were used depending on whether the patients lived at home or in a nursing facility. A total of 21 patients (n = 15 GRF6019 Evaluable; n = 6 placebo) resided in the community and were assessed using the NPI, while 5 patients (n = 3 GRF6019 Evaluable; n = 2 placebo) resided in nursing homes and were assessed using the NPI-NH. Patients receiving GRF6019 demonstrated a lower severity of behavioral symptoms at baseline than patients in the placebo group, with mean (SD) NPI/NPI-NH Total Severity scores of 19.9 (15.4) in GRF6019 Evaluable patients and 17.9 (16.9) in GRF6019 Per Protocol patients versus 38.5 (31.7) in placebo patients, and baseline NPI Distress scores of 9.1 (8.1) in GRF6019 Evaluable patients and 7.7 (8.3) in GRF6019 Per Protocol patients versus 15.8 (16.8) in placebo patients.

Table 5

Cognitive and functional assessment scores^a

Total Score, mean (SD)	GRF6019 Evaluable Population (N = 18)			GRF6019 Per Protocol Population (N = 14)			Placebo (N = 8)^b			Between-Group Difference in LSM^c (95% CI)
	Baseline	End of Study	Change from Baseline^d	Baseline	End of Study	Change from Baseline^d	Baseline	End of Study	Change from Baseline^d	Evaluable Population	Per Protocol Population
MMSE	(n = 17)	(n = 17)		(n = 14)	(n = 14)		(n = 8)	(n = 8)
	5.9 (3.5)	7.2 (6.0)	1.3 (3.5)	5.0 (3.1)	5.4 (4.7)	0.4 (3.1)	7.3 (3.4)	9.0 (5.2)	1.8 (3.2)	– 0.1 (– 3.0, 2.8)	– 0.9 (– 4.0, 2.2)
SIB	(n = 17)	(n = 17)		(n = 14)	(n = 14)		(n = 6)	(n = 6)
	51.6 (29.2)	58.4 (31.4)	6.7 (7.6)^**	44.1 (26.4)	51.0 (29.7)	6.9 (8.3)^**	61.7 (26.4)	72.0 (26.5)	10.4 (4.9)^**	– 2.5 (– 11.0, 6.0)	– 1.3 (– 11.3, 8.8)
ADCS-ADL-Severe	(n = 18)	(n = 18)		(n = 14)	(n = 14)		(n = 8)	(n = 8)
	22.7 (14.0)	24.3 (14.3)	1.6 (5.1)	19.9 (13.8)	22.8 (14.0)	2.9 (4.3)^*	20.0 (13.6)	20.5 (15.3)	0.5 (4.9)	1.1 (– 3.4, 5.7)	2.6 (– 1.9, 7.1)
ADCS-CGIC,^e n (%)	–	(n = 18)		–	(n = 14)		–	(n = 8)
2-Moderate Improvement	–	2 (11.1)	–	–	2 (14.3)	–	–	0	–	–	–
3-Minimal Improvement	–	1 (5.6)	–	–	1 (7.1)	–	–	2 (25.0)	–	–	–
4-No change	–	12 (66.7)	–	–	9 (64.3)	–	–	4 (50.0)	–	–	–
5-Minimal worsening	–	3 (16.7)	–	–	2 (14.3)	–	–	2 (25.0)	–	–	–
Mean score (SD)	–	3.9 (0.8)	–	–	3.8 (0.9)	–	–	4.0 (0.8)	–	– 0.12 (– 0.8, 0.6)	– 0.15 (– 1.0, 0.7)
NPI/NPI-NH Total Severity^f	(n = 18)	(n = 18)		(n = 14)	(n = 14)		(n = 8)	(n = 8)
	19.9 (15.4)	15.7 (13.9)	– 4.2 (12.5)	17.9 (16.9)	14.0 (11.6)	– 3.9 (8.6)	38.5 (31.7)	34.6 (33.6)	– 3.9 (10.8)	– 3.1 (– 14.7, 8.4)	– 2.7 (– 12.7, 7.2)
NPI Distress^g	(n = 15)	(n = 15)		(n = 11)	(n = 11)		(n = 6)	(n = 6)
	9.1 (8.1)	5.9 (6.2)	– 3.3 (6.6)	7.7 (8.3)	4.8 (6.2)	– 2.9 (4.3)	15.8 (16.8)	14.2 (16.1)	– 1.7 (1.4)^*	– 3.2 (– 8.9, 2.5)	– 2.3 (– 6.3, 1.7)
NPI-NH Occupational Disruptiveness^h	(n = 3)	(n = 3)		(n = 3)	(n = 3)		(n = 2)	(n = 2)
10.0 (11.3)	9.0 (5.6)	– 1.0 (7.6)	10.0 (11.3)	9.0 (5.6)	– 1.0 (7.6)	16.0 (2.8)	7.0 (1.4)	– 9.0 (4.2)	–	–

ADCS-ADL-Severe, Alzheimer’s Disease Cooperative Study – Activities of Daily Living – Severe; ADCS-CGIC, Alzheimer’s Disease Cooperative Study – Clinical Global Impression of Change; CI, confidence interval; LSM = Least Squares Mean; MMSE, Mini-Mental State Examination; NPI, Neuropsychiatric Inventory; NPI-NH, Neuropsychiatric Inventory – Nursing Home; SIB, Severe Impairment Battery. ^aHigher scores on the MMSE, SIB, and ADCS-ADL-Severe indicate improvement; a positive difference in LSM favors GRF6019. Lower scores on the NPI/NPI-NH and ADCS-CGIC indicate improvement; a negative difference in LSM favors GRF6019. ^bThe Placebo Evaluable and Per Protocol populations contained the same patients (N = 8). ^cBased on Analysis of Covariance analysis including baseline and sex as the covariates. ^dp value based on Student’s t statistics: ^*p < 0.05, ^**p < 0.01. ^eThe ADCS-CGIC focuses on clinicians’ observations of change in the patient’s cognitive, functional, and behavioral performance since the beginning of the trial, and therefore baseline scores are not obtained. At the end of the study, no patients were rated as 1-marked improvement, 6-moderate worsening, or 7-marked worsening. ^fNPI/NPI-NH Total Severity was calculated by adding the first 10 domain scores together (frequency×severity). ^gEvaluated only for patients residing in the community. ^hEvaluated only for patients residing in care facilities.

At the end of the study (Day 33), patients in both the GRF6019 and placebo groups remained stable or improved compared to baseline on all cognitive and functional endpoints (Table 5; see Supplementary Table 1 for efficacy results by sex). Although the study was not powered to detect statistically significant between-group differences on any of the secondary endpoints, there were numerical differences of interest.

The GRF6019 Evaluable patients had a larger mean (SD) increase (improvement) on the ADCS-ADL-Severe than the placebo patients: + 1.6 (5.1; 95%confidence interval [CI]: –0.9, 4.1) versus + 0.5 (4.9; 95%CI, –3.6, 4.6), respectively, with a Least Squares Mean (LSM) difference of + 1.1 point (95%CI: –3.4, 5.7) (Fig. 2). This difference was larger in the GRF6019 Per Protocol patients: + 2.9 (4.3; 95%CI: 0.5, 5.4) versus + 0.5 (4.9; 95%CI: –3.6, 4.6), with an LSM difference of + 2.6 points (95%CI: –1.9, 7.1) favoring GRF6019. A cutoff of 3.5 points was shown in early studies with the ADCS-ADL-Severe to indicate a meaningful change [24]. Five (5/14, 35.7%) GRF6019 Per Protocol patients showed clinically significant improvement using this 3.5-point threshold, and in the placebo group, 2 (2/8, 25%) patients exceeded this threshold.

Fig. 2

Mean change over time in cognitive and functional assessments. Mean change from baseline in total score with standard error of the mean: A) Mini-Mental State Examination. B) Alzheimer’s Disease Cooperative Study –Activities of Daily Living Inventory for Severe Alzheimer’s Disease. C) Severe Impairment Battery. D) Neuropsychiatric Inventory Questionnaire Total Score (frequency×severity). The baseline Mini-Mental State Examination scores were from screening (Day –28 to –8) and the other assessments were conducted at baseline (Day –7 to –1).

The GRF6019 group had larger reductions (improvement) from baseline in NPI/NPI-NH Total Severity score, with a mean (SD) change of –4.2 (12.5; 95%CI: –10.4, 2.0) in GRF6019 Evaluable patients, –3.9 (8.6; 95%CI: –8.9, 1.1) in GRF6019 Per Protocol patients, and –3.9 (10.8; 95%CI: –12.9, 5.2) points in placebo patients, and LSM differences of –3.1 (95%CI: –14.7, 8.4) and –2.7 (95%CI: –12.7, 7.2) points, favoring GRF6019 Evaluable and GRF6019 Per Protocol, respectively.

The mean (SD) increase from baseline (improvement) on the MMSE was larger in placebo patients (+ 1.8 [3.2]; 95%CI: –0.9, 4.4) than GRF6019 Evaluable patients (+ 1.3 [3.5]; 95%CI: –0.5, 3.1) and GRF6019 Per Protocol patients (+ 0.4 [3.1]; 95%CI: –1.4, 2.1), with LSM differences of –0.1 (95%CI: –3.0, 2.8) and –0.9 (95%CI: –4.0, 2.2) favoring placebo over GRF6019 Evaluable and GRF6019 Per Protocol, respectively. A similar pattern was seen for the SIB. Only 6 out of 8 placebo patients had evaluable SIB scores at the endpoint. In these 6 patients, the mean (SD) improvement on the SIB was larger (10.4 [4.9] points; 95%CI: 4.3, 16.5) than for GRF6019 Evaluable patients (6.7 [7.6]; 95%CI: 2.8, 10.6) and GRF6019 Per Protocol patients (6.9 [8.3]; 95%CI: 2.1, 11.7), with LSM differences of –2.5 (95%CI: –11.0, 6.0) and –1.3 (95%CI: –11.3, 8.8) points favoring placebo over GRF6019 Evaluable and GRF6019 Per Protocol, respectively.

Clinical improvement from baseline was assessed by the same rater at baseline and at end of study using the ADCS-CGIC. Results appeared favorable for GRF6019: at the end of the study, 15 of the 18 (83.3%) GRF6019 Evaluable patients and 12 of the 14 (85.7%) GRF6019 Per Protocol patients were rated as 2-moderately improved, 3-minimally improved, or 4-unchanged, compared to 6 of the 8 (75.0%) placebo patients. Among these patients, 2 of the 14 (14.3%) patients in the GRF6019 Per Protocol population were described as 2-moderately improved, whereas in the placebo group, no patients were rated as 2-moderately improved.

Across all assessments and treatment groups, patients who were more impaired at baseline (i.e., in the bottom half of the range of baseline scores) tended to stay the same or worsen, whereas patients with less impairment at baseline tended to improve (Fig. 3).

Fig. 3

Change from baseline in cognitive and functional assessments by patient. Change in total score from baseline to end of study by patient: A) Mini-Mental State Examination (MMSE). B) Alzheimer’s Disease Cooperative Study –Activities of Daily Living Inventory for Severe Alzheimer’s Disease (ADCS-ADL-Severe). C) Severe Impairment Battery (SIB). D) Neuropsychiatric Inventory Questionnaire Total Score (frequency×severity) (NPI/NPI-NH Severity). Green = GRF6019 Per Protocol patients; Blue = GRF6019 Evaluable patients who were not included in the Per Protocol dataset; Red = Placebo patients. The baseline MMSE scores were from screening (Day –28 to –8) and the other assessments were conducted at baseline (Day –7 to –1).

DISCUSSION

With approximately 1 in 10 adults over the age of 65 suffering from AD and a significant increase expected in the both the number and proportion of older adults with AD over the next several decades, there is an urgent unmet need for treatments that address the symptoms and cognitive decline associated with late-onset AD [1, 2]. In the dementia phase of AD, approximately 40%of the time is spent with severe AD [2], attesting to the importance of finding therapies for patients with this degree of impairment. Advances in our understanding of the relationship between aging and the plasma proteome has opened up a novel avenue for the treatment of age-related neurologic diseases. Building on preclinical studies that demonstrated treatment with plasma from young mice reduces tau and amyloid beta pathologies, restores synaptic plasticity, reduces neuroinflammation, and improves neurogenesis and cognition in mice [10–12 , 26], recent clinical trials have evaluated the effects of plasma administration in patients with neurodegenerative diseases such as AD and Parkinson’s disease [13, 27]. In the pilot PLASMA study in AD, infusions of whole plasma resulted in improvements in activities of daily living and function; however, the study was not powered to detect statistically significant changes from baseline [13]. Similarly, a recent phase II study of the plasma fraction GRF6019 in mild-to-moderate AD demonstrated that patients experienced minimal functional decline and no cognitive decline over 24 weeks, a timeframe in which patients would be expected to cognitively worsen [16]. The Alzheimer’s Management By Albumin Replacement (AMBAR) study examined the effects of plasma exchange with albumin replacement in a larger cohort and found that cognitive and functional decline was slowed over a period of 14 months [28]. Further research is necessary to extend the promising results of these studies and fully characterize the potential benefits of plasma and plasma fractions in AD.

To our knowledge, this is the first study evaluating infusions of a plasma fraction in patients with severe AD. In this phase II study, we demonstrated that recruitment of patients with severe AD—a population that can be challenging to enroll in clinical trials [29]—for daily 2-hour IV infusions over 5 consecutive days was feasible, and 100%of patients completed all infusions and study visits. While the study showed excellent compliance with the dosing regimen and study retention, a larger study would be needed to assess whether the level of compliance observed is generalizable to the wider severe AD population. To enable administration of GRF6019 and saline placebo, the infusion nurse was unblinded in this study. For future trials, the unblinded pharmacist will place the vials of active or placebo drug in an opaque shell that we have developed, and then seal the shell and dispense it to the infusion nurse. The use of this shell in future studies will maintain the blind of all non-pharmacy staff, including the infusion nurse, as well as the patients and caregivers. There is the potential for false positives when relying solely on a clinical diagnosis of AD [30], and imaging or cerebrospinal fluid biomarker confirmation of AD was not required for inclusion in this study. However, the high rate of APOE4 carriers among the patients (76.9%) is comparable to the rate observed in studies that require positive amyloid imaging consistent with AD [31 –33].

GRF6019 was safe and well tolerated in both nursing home residents and home-dwelling patients, and there were no new safety signals identified when compared to those observed in a prior study of GRF6019 in approximately 50 patients with mild-to-moderate AD [16]. There were no deaths, SAEs, or withdrawals due to AEs. Changes in blood pressure were the most frequent AEs considered related to treatment with similar frequencies across the GRF6019 and placebo groups; consistent with the prior study in mild-to-moderate AD, these blood pressure changes were typically mild and transient.

While the goal of treatment in mild and moderate AD is to improve cognition and slow disease progression, patients with severe AD are often agitated, may have difficulty recognizing family members, and gradually lose the ability to speak and perform basic activities of daily living [2]. As a result, the treatment goals in severe AD are more focused on alleviating symptoms and improving functional abilities and behavior in the short term, with a focus on the well-being of both the patient and the caregiver [34 –36]. Approved therapies such as acetylcholinesterase inhibitors (donepezil, rivastigmine, galantamine) and memantine can ameliorate symptoms, but the magnitude of this symptomatic effect is modest, and there remains a need for therapies that preserve functional abilities in the later stages of the disease [3–5 , 37–39]. Although this study was not powered to detect statistically significant differences between groups in cognitive and functional endpoints, preliminary results show a potential benefit in the GRF6019 treatment group on the ADCS-ADL-Severe, an important assessment in patients with severe AD who typically have advanced functional decline [40]. As a measure of severity and frequency of behavioral disturbances, the NPI/NPI-NH is also a key assessment in severe AD [41]. Patients in both treatment groups had similar absolute improvements in NPI/NPI-NH scores, while the LSM analyses favored GRF6019 over placebo in both total severity and caregiver distress. Clinician ratings of disease severity, as demonstrated by ADCS-CGIC scores, were also lower (improved) in the GRF6019 group than placebo at the end of the study.

The cognitive assessment results for both the GRF6019 and placebo groups reflected either stable or improved performance compared to baseline. There was an improvement in SIB and MMSE scores in both the GRF6019 and placebo groups, which we interpret as a placebo response given that patients receiving placebo in moderate-to-severe AD studies of longer duration have exhibited marked declines in SIB scores and, to a lesser extent, MMSE scores [42 –44]. This higher than anticipated placebo response on the MMSE and SIB assessments in this study may be due the shorter observation period and/or other factors, such as rater expectations and procedural learning [45 –47], which could have a larger impact in a shorter duration trial. Additionally, these cognitive assessment instruments are performance measures administered by raters directly to the patients. In contrast, the ADCS-ADL-Severe, NPI/NPI-NH, and ADCS-CGIC utilize information provided by caregivers and clinicians. The difference between the results for the cognitive testing and rating scales appears to reflect greater weighting of functional abilities and neurobehavioral functioning by clinicians and caregivers than changes in cognitive functioning. Even with the improvement seen in cognitive test scores at the end of the study, the patients’ level of performance still shows severe cognitive impairment. Future studies of longer duration should further investigate the relationship between cognitive endpoints (MMSE and SIB) and rating scale endpoints (i.e., ADCS-ADL-Severe, NPI/NPI-NH, and ADCS-CGIC) in patients with severe AD.

This study has important limitations. There was an imbalance in several important parameters at baseline that may have limited the ability to interpret the efficacy results, and the small sample size prevented statistical correction for these differences. Patients randomized to GRF6019 were older and had lower mean MMSE and SIB scores, but lower severity of behavioral symptoms (i.e., higher NPI/NPI-NH scores) than patients in the placebo group at baseline. Additionally, a larger proportion of patients receiving GRF6019 than placebo was being treated with a cholinesterase inhibitor, and there were site-to-site differences in baseline severity.

The number of patients in this study was sufficient to evaluate the primary objectives of safety and tolerability. The sample was not statistically powered to determine differences in efficacy, and the sample size was limited further as 4 patients were excluded from the GRF6019 Per Protocol population due to incorrectly scored MMSE for inclusion at screening, and 2 patients in the placebo group did not have end-of-study SIB scores. These factors resulted in a limited number of patients for whom to evaluate change from baseline on this assessment.

This preliminary study established the feasibility of treating patients with advanced AD with a paradigm of IV infusions of a plasma fraction. The safety and tolerability of GRF6019 was supported, and the study yielded encouraging signs of improvement, particularly with regards to clinical disease severity, ADLs, behavioral disturbances, and caregiver distress. Combined with the excellent safety, feasibility, and tolerability of GRF6019, these results lay the foundation for larger trials designed and powered to characterize the potential benefits of such plasma fractions in severe AD and other neurodegenerative disorders typified by cognitive dysfunction. Future studies may be designed to evaluate endpoints closer to dosing, stratify results by baseline severity, include patients with a wider range of MMSE scores, and be longer in duration.

Footnotes

ACKNOWLEDGMENTS

The authors wish to thank the patients and care-givers who participated in this trial, and the investigators and site staff who conducted the trial. Manuscript writing and formatting support was provided by Colleen Brown (Mark Consulting, Inc.) and Shireen Dunwoody (Dunwoody Consulting). This work was supported by Alkahest, Inc., San Carlos, CA and Grifols, S.A., Barcelona, Spain. Alkahest was responsible for study oversight, conduct, data collection, monitoring, statistical analyses, and interpretation of the data. Grifols was responsible for manufacturing investigational product. Both companies participated in the study design and reviewed the manuscript. All final content decisions were made by the authors.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Hebert

, Weuve

, Scherr

, Evans

(2013) Alzheimer disease in the United States (2010-2050) estimated using the 2010 census. Neurology 80, 1778–1783.

Alzheimer’s Association (2020) 2020 Alzheimer’s disease facts and figures. Alzheimers Dement 16, 391–460.

Adlimoghaddam

, Neuendorff

, Roy

, Albensi

(2018) A review of clinical treatment considerations of donepezil in severe Alzheimer’s disease. CNS Neurosci Ther 24, 876–888.

McShane

, Westby

, Roberts

, Minakaran

, Schneider

, Farrimond

, Maayan

, Ware

, Debarros

(2019) Memantine for dementia. Cochrane Database Syst Rev 3, CD003154.

Raina

, Santaguida

, Ismaila

, Patterson

, Cowan

, Levine

, Booker

, Oremus

(2008) Effectiveness of cholinesterase inhibitors and memantine for treating dementia: Evidence review for a clinical practice guideline. Ann Intern Med 148, 379–397.

Weber

, Wu

, Hanson

, Alam

, Solanoy

, Ngu

, Lauffer

, Lin

, Dominguez

, Reeder

, Tom

, Steiner

, Foreman

, Prusky

, Scearce-Levie

(2015) Cognitive deficits, changes in synaptic function, and brain pathology in a mouse model of normal aging. eNeuro 2, ENEURO.0047-15.2015.

Angelova

, Brown

(2019)Microglia and the aging brain: Are senescent microglia the key to neurodegeneration? J Neurochem 151, 676–688.

Koellhoffer

, McCullough

, Ritzel

(2017) Old maids: Aging and its impact on microglia function. Int J Mol Sci 18, 769.

Gabandé-Rodríguez

, Keane

, Capasso

(2020) Microglial phagocytosis in aging and Alzheimer’s disease. J Neurosci Res 98, 284–298.

10.

Villeda

, Luo

, Mosher

, Zou

, Britschgi

, Bieri

, Stan

, Fainberg

, Ding

, Eggel

, Lucin

, Czirr

, Park

, Couillard-Despres

, Aigner

, Li

, Peskind

, Kaye

, Quinn

, Galasko

, Xie

, Rando

, Wyss-Coray

(2011) The ageing systemic milieu negatively regulates neurogenesis and cognitive function. Nature 477, 90–94.

11.

Villeda

, Plambeck

, Middeldorp

, Castellano

, Mosher

, Luo

, Smith

, Bieri

, Lin

, Berdnik

, Wabl

, Udeochu

, Wheatley

, Zou

, Simmons

, Xie

, Longo

, Wyss-Coray

(2014) Young blood reverses age-related impairments in cognitive function and synaptic plasticity in mice. Nat Med 20, 659–663.

12.

Kheifets

, Braithwaite

(2019) Plasma-based strategies for therapeutic modulation of brain aging. Neurotherapeutics 16, 675–684.

13.

Sha

, Deutsch

, Tian

, Richardson

, Coburn

, Gaudioso

, Marcal

, Solomon

, Boumis

, Bet

, Mennes

, Van Oort

, Beckmann

, Braithwaite

, Jackson

, Nikolich

, Stephens

, Kerchner

, Wyss-Coray

(2019) Safety, tolerability, and feasibility of young plasma infusion in the Plasma for Alzheimer Symptom Amelioration Study: A randomized clinical trial. JAMA Neurol 76, 35–40.

14.

Kato

, Uruma

, Okuyama

, Fujita

, Handa

, Tomiyama

, Shimodaira

, Kurata

, Takamoto

(2013) Incidence of transfusion-related adverse reactions per patient reflects the potential risk of transfusion therapy in Japan. Am J Clin Pathol 140, 219–224.

15.

Hatayama

, Matsumoto

, Hamada

, Kojima

, Hara

, Hino

, Motokura

(2018) Analysis of acute transfusion reactions and their occurrence times. Yonago Acta Med 61, 87–90.

16.

Hannestad

, Koborsi

, Klutzaritz

, Chao

, Ray

, Páez

, Jackson

, Lohr

, Cummings

, Kay

, Nikolich

, Braithwaite

(2020) Safety and tolerability of GRF6019 in mild-to-moderate Alzheimer’s disease dementia. Alzheimers Dement (N Y) 6, e12115.

17.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

18.

Panisset

, Roudier

, Boiler

, Saxton

(1994) Severe Impairment Battery: A neuropsychological test for severely demented patients. Arch Neurol 51, 41–45.

19.

Galasko

, Schmitt

, Thomas

, Jin

, Bennett

, Ferris

, Alzheimer’s Disease Cooperative Study (2005) Detailed assessment of activities of daily living in moderate to severe Alzheimer’s disease. J Int Neuropsychol Soc 11, 446–453.

20.

Ferris

, Mackell

, Mohs

, Schneider

, Galasko

, Whitehouse

, Schmitt

, Sano

, Thomas

, Ernesto

, Grundman

, Schafer

, Thal

(1997) A multicenter evaluation of new treatment efficacy instruments for Alzheimer’s disease clinical trials: Overview and general results. Alzheimer Dis Assoc Disord 11, S1–S12.

21.

Schneider

, Olin

, Doody

, Clark

, Morris

, Reisberg

, Schmitt

, Grundman

, Thomas

, Ferris

(1997) Validity and reliability of the Alzheimer’s Disease Cooperative Study-Clinical Global Impression of Change. Alzheimer Dis Assoc Disord 11, S22–S32.

22.

Cummings

, Mega

, Gray

, Rosenberg-Thompson

, Carusi

, Gornbein

(1994) The Neuropsychiatric Inventory: Comprehensive assessment of psychopathology in dementia. Neurology 44, 2308–2314.

23.

Wood

, Cummings

, Hsu

, Barclay

, Wheatley

, Yarema

, Schnelle

(2000) The use of the Neuropsychiatric Inventory in nursing home residents: Characterization and measurement. Am J Geriatr Psychiatry 8, 75–83.

24.

Winblad

, Kilander

, Eriksson

, Minthon

, Båtsman

, Wetterholm

, Jansson-Blixt

, Haglund

, Severe Alzheimer’s Disease Study Group (2006) Donepezil in patients with severe Alzheimer’s disease: Double-blind, parallel-group, placebo-controlled study. Lancet 367, 1057–1065.

25.

Middeldorp

, Lehallier

, Villeda

, Miedema

SSM

, Evans

, Czirr

, Zhang

, Luo

, Stan

, Mosher

, Masliah

, Wyss-Coray

(2016) Preclinical assessment of young blood plasma for Alzheimer disease. JAMA Neurol 73, 1325–1333.

26.

Zhao

, Qian

, Zhang

, Liu

, Iqbal

, Dai

, Gong

(2020) Young blood plasma reduces Alzheimer’s disease-like brain pathologies and ameliorates cognitive impairment in 3×Tg-AD mice. Alzheimers Res Ther 12, 70.

27.

Parker

, Martinez

, Deutsch

, Prabhakar

, Lising

, Kapphahn

, Anidi

, Neuville

, Coburn

, Shah

, Bronte-Stewart

(2020) Safety of plasma infusions in Parkinson’s disease. Mov Disord 35, 1905–1913.

28.

Boada

, López

, Olazarán

, Núñez

, Pfeffer

, Paricio

, Lorites

, Piñol-Ripoll

, Gámez

, Anaya

, Kiprov

, Lima

, Grifols

, Torres

, Costa

, Bozzo

, Szczepiorkowski

, Hendrix

, Páez

(2020) A randomized, controlled clinical trial of plasma exchange with albumin replacement for Alzheimer’s disease: Primary results of the AMBAR Study. Alzheimers Dement 16, 1412–1425.

29.

Grill

, Karlawish

(2010) Addressing the challenges to successful recruitment and retention in Alzheimer’s disease clinical trials. Alzheimers Res Ther 2, 34.

30.

Harris

, Thompson

, Gall

, Richardson

AMT

, Neary

, du Plessis

, Pal

, Mann

DMA

, Snowden

, Jones

(2015) Do NIA-AA criteria distinguish Alzheimer’s disease from frontotemporal dementia? Alzheimers Dement 11, 207–215.

31.

Cummings

, Zhong

, Kinney

, Heaney

, Moll-Tudla

, Joshi

, Pontecorvo

, Devous

, Tang

, Bena

(2016) Double-blind, placebo-controlled, proof-of-concept trial of bexarotene in moderate Alzheimer’s disease. Alzheimers Res Ther 8, 4.

32.

Sevigny

, Suhy

, Chiao

, Chen

, Klein

, Purcell

, Oh

, Verma

, Sampat

, Barakos

(2016) Amyloid PET screening for enrichment of early-stage Alzheimer disease clinical trials: Experience in a phase 1b clinical trial. Alzheimer Dis Assoc Disord 30, 1–7.

33.

Egan

, Kost

, Voss

, Mukai

, Aisen

, Cummings

, Tariot

, Vellas

, van Dyck

, Boada

, Zhang

, Li

, Furtek

, Mahoney

, Harper Mozley

, Mo

, Sur

, Michelson

(2019) Randomized trial of verubecestat for prodromal Alzheimer’s disease. N Engl J Med 380, 1408–1420.

34.

Voisin

, Vellas

(2009) Diagnosis and treatment of patients with severe Alzheimer’s disease. Drugs Aging 26, 135–144.

35.

Atri

(2011) Effective pharmacological management of Alzheimer’s disease. Am J Manag Care 17, S346–S355.

36.

Schmitt

, Wichems

(2006) A systematic review of assessment and treatment of moderate to severe Alzheimer’s disease. Prim Care Companion J Clin Psychiatry 8, 158–159.

37.

Birks

, Harvey

(2018) Donepezil for dementia due to Alzheimer’s disease. Cochrane Database Syst Rev 6, CD001190.

38.

Joe

, Ringman

(2019) Cognitive symptoms of Alzheimer’s disease: Clinical management and prevention. BMJ 367, l6217.

39.

Qaseem

, Snow

, Cross

, Forciea

, Hopkins

, Shekelle

, Adelman

, Mehr

, Schellhase

, Campos-Outcalt

, Santaguida

, Owens

, Joint American College of Physicians/American Academy of Family Physicians Panel on Dementia (2008) Current pharmacologic treatment of dementia: A clinical practice guideline from the American College of Physicians and the American Academy of Family Physicians. Ann Intern Med 148, 370–378.

40.

Kahle-Wrobleski

, Coley

, Lepage

, Cantet

, Vellas

, Andrieu

, PLASA/DSA Group (2014) Understanding the complexities of functional ability in Alzheimer’s disease: More than just basic and instrumental factors. Curr Alzheimer Res 11, 357–366.

41.

Soto

, Abushakra

, Cummings

, Siffert

, Robert

, Vellas

, Lyketsos

(2015) Progress in treatment development for neuropsychiatric symptoms in Alzheimer’s disease: Focus on agitation and aggression. A report from the EU/US/CTAD Task Force. J Prev Alzheimers Dis 2, 184–188.

42.

Black

, Doody

, Li

, McRae

, Jambor

, Xu

, Sun

, Perdomo

, Richardson

(2007) Donepezil preserves cognition and global function in patients with severe Alzheimer disease. Neurology 69, 459–69.

43.

Reisberg

, Doody

, Stöffler

, Schmitt

, Ferris

, Möbius

, Memantine Study Group (2003) Memantine in moderate-to-severe Alzheimer’s disease. N Engl J Med 348, 1333–1341.

44.

Feldman

, Gauthier

, Hecker

, Vellas

, Subbiah

, Whalen

, Donepezil MSAD Study Investigators Group (2001) A 24-week, randomized, double-blind study of donepezil in moderate to severe Alzheimer’s disease. Neurology 57, 613–620.

45.

Popp

, Williams

JBW

, Kobak

, Boehm

, Detke

(2012) The role of expectation bias in neurocognition clinical trials. Poster presentation at: Alzheimer’s Association International Conference, Vancouver, British Columbia, Canada. July 8, 2012.

46.

Ito

, Corrigan

, Romero

, Anziano

, Neville

, Stephenson

, Lalonde

(2013) Understanding placebo responses in Alzheimer’s disease clinical trials from the literature meta-data and CAMD database. J Alzheimers Dis 37, 173–183.

47.

Kobak

(2010) Inaccuracy in clinical trials: Effects and methods to control inaccuracy. Curr Alzheimer Res 7, 637–641.