Two phase 1 randomised studies investigating the safety and pharmacokinetics of bepranemab in healthy participants of different ethnicities

Abstract

Background

Bepranemab is a recombinant, humanised, full-length IgG4 monoclonal antibody targeting a mid-region tau epitope. Two phase 1 studies assessed the safety, tolerability, pharmacokinetics (PK) and pharmacodynamics (PD) of bepranemab.

Methods

UP0047 (NCT03464227) and UP0065 (NCT03605082) were phase 1, double-blind, placebo-controlled, single-dose, dose-escalation studies of intravenous bepranemab in healthy participants (Caucasian and Japanese descent, respectively). Primary endpoint: safety and tolerability of single ascending doses of bepranemab. PK were assessed in serum and cerebrospinal fluid (CSF); PD (levels of free tau) in CSF. A physiologically based PK/PD (PBPK/PD) model was developed to predict dose response.

Results

UP0047: Caucasian participants (N=52) were randomised to bepranemab 0.3 mg/kg, n=2; 1 mg/kg, n=6; 3 mg/kg, n=6; 10 mg/kg, n=6; 30 mg/kg, n=6; 60 mg/kg, n=6; 120 mg/kg, n=6; placebo, n=14). UP0065: participants of Japanese descent (N=24) were randomised to bepranemab 30 mg/kg, n=6; 60 mg/kg, n=6; 120 mg/kg, n=6; placebo, n=6). No serious treatment-emergent adverse events (TEAEs) or discontinuations due to TEAEs were observed. One participant (bepranemab 60 mg/kg) experienced treatment-related TEAEs of headache (moderate intensity), nausea and vomiting (mild intensity). There was no effect of ethnicity on PK parameters. A dose-response effect of bepranemab on free tau levels was observed. Data applied to a PBPK/PD model supported a dose of 90 mg/kg of bepranemab every 4 weeks to achieve a reduction in mean change from baseline in free tau levels of up to 90%.

Conclusions

Safety, PK and PD data support continued investigation of bepranemab for the treatment of tauopathies.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Targeting tau using bepranemab, a recombinant, humanised, IgG4 monoclonal antibody targeting a mid-region tau epitope, has been validated in preclinical work. From in vitro experiments, bepranemab effectively blocked seeding of human Alzheimer’s disease and progressive supranuclear palsy tau in cell-based assays. In preclinical mouse models, bepranemab prevented induction of tau pathology and blocked tau spread throughout the brain, indicating that bepranemab may have the potential to slow progression of tauopathies.

WHAT THIS STUDY ADDS

• Bepranemab was well tolerated and had an acceptable safety profile in healthy participants following single intravenous doses of up to 120 mg/kg. There were no effects of ethnicity on serum pharmacokinetics (PK) parameters, and dose-proportional PK was observed; bepranemab reduced levels of free tau in cerebrospinal fluid in a dose-proportional manner.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• These data were applied to a physiologically based PK/pharmacodynamics model that supported a bepranemab dose of 90 mg/kg, administered every 4 weeks, which is predicted to reduce free tau levels by up to 90% from baseline. These results support continued clinical investigation of bepranemab as a potential disease-modifying therapy for tauopathies.

Introduction

Bepranemab is a recombinant, humanised, full-length IgG4P monoclonal antibody (mAb) in development as a potential treatment for tauopathies.¹ Bepranemab specifically targets a mid-region epitope of human tau (amino acids 235–250), within the central proline-rich region, adjacent to the microtubule binding region (MTBR),² which is thought to be essential for tau aggregation.³

Tauopathies are progressive neurodegenerative disorders that result from the misfolding and aggregation of tau protein.⁴ The spread of tau pathology in the brain is closely linked to neuron dysfunction and loss, and disease progression.⁴ Tauopathies include Alzheimer’s disease (AD), corticobasal degeneration, behavioural variant frontotemporal dementia, non-fluent variant frontotemporal dementia and progressive supranuclear palsy (PSP).⁴ AD is characterised by two pathological hallmarks: extracellular deposition of amyloid-beta plaques and intracellular accumulation of tau (neurofibrillary tangles).⁵ Progression of tauopathies, such as AD and PSP, is associated with increasing loss of function and cognition, exerting a high burden on individuals, caregivers, families and society.5,7 Despite recent approvals of anti-amyloid disease-modifying therapies,⁸ there remains a critical need for drugs with improved safety and efficacy profiles, including the ability to further slow disease progression in AD.⁹ The intrinsic role of tau in disease pathology makes it a valid target for disease modification.

Targeting tau using bepranemab has also been validated in preclinical work: during in vitro experiments, bepranemab more effectively blocked the seeding of human AD and PSP tau in cell-based assays.^{1 2} Furthermore, in preclinical mouse models of disease, bepranemab prevented the induction of tau pathology and blocked tau spread throughout the brain.¹ Taken together, these data indicate that bepranemab may have the potential to slow progression of tauopathies.

Here, we report findings from the first-in-human study of bepranemab in healthy Caucasian participants (UP0047; NCT03464227) and a phase 1 study in healthy participants of Japanese descent (UP0065; NCT03605082). The safety, tolerability, pharmacokinetics (PK) and pharmacodynamics (PD; reduction in free tau levels) were assessed in these studies. These data were then used to predict doses of bepranemab for further use in clinical development, including in a phase 1b study in people with PSP (NCT04185415) and a phase 2 study in AD (NCT04867616).

Methods

Trial design and population

Both the UP0047 and UP0065 studies were conducted in accordance with the International Council for Harmonisation Good Clinical Practice guidelines and local regulatory requirements.

Based on preclinical toxicology, PK and PD data, a starting dose of 0.3 mg/kg bepranemab was administered, with the highest dose being 120 mg/kg (online supplemental table 1). The predicted exposure margins versus the no observed adverse effect level (NOAEL) in toxicological species and predicted reduction in cerebrospinal fluid (CSF) tau (target occupancy (TO)) are presented in online supplemental table 1 and illustrate the measured exposures in toxicological species, predicted exposures in human and the exposure margins at the proposed starting and escalating doses.

UP0047

UP0047 was a phase 1, first-in-human, placebo-controlled, participant-blind, investigator-blind, single-dose, dose-escalation study of bepranemab in healthy participants (NCT03464227). The study was conducted at a phase 1 clinical study unit in Germany.

Male participants aged ≥18 and ≤75 years, with a body mass index (BMI) ≥18.0 kg/m² and <30.0 kg/m², body weight of 50–100 kg and in good physical and mental health were recruited. All participants were required to have laboratory test results within normal reference ranges, blood pressure (BP) within normal range (systolic 90–140 mm Hg, diastolic 50–90 mm Hg) and a normal ECG (or any abnormalities considered clinically non-significant by the investigator).

Key exclusion criteria were: previous exposure to any biological compound targeting the same mechanism of action as bepranemab; participation in a study of another investigational medicinal product (IMP) in the previous 3 months or within five half-lives of study drug administration; major surgery in the 6 months before, or planned surgery within 6 months after, study drug administration; current infection or serious infection (requiring hospitalisation or parenteral antibiotics) within 4 weeks prior to dosing; clinically relevant abnormal findings on physical examination, laboratory tests, vital signs, ECG or brain MRI; history of recurrent headaches, including migraines; lifetime history of suicide attempt or suicidal ideation within the previous 2 years. For additional eligibility criteria, refer to the online supplemental material.

Participants were randomised to bepranemab or placebo in a 3:1 ratio within each of the seven cohorts (except cohort 1, where a ratio of 1:1 was used), as shown in figure 1A. All members of the main study team, including Sponsor/Contract Research Organisation and Investigator site personnel, involved in the study were blinded to the randomised IMP. Each cohort included two sentinel participants, one receiving bepranemab and one receiving placebo, who were monitored for 48 hours post dose; the remaining participants in the cohort were dosed only after review of the safety data for the sentinel participants. If approved by the blinded safety review group, dose escalation for the next cohort proceeded as planned. Bepranemab was administered by intravenous infusion over ≥120 min. Placebo treatment (0.9% sodium chloride) was prepared similarly to the dosing solution for bepranemab. Bepranemab and the placebo were not identical in appearance, and to maintain blinding, both bepranemab and the placebo were covered with opaque film.

Figure 1. CONSORT diagram. (A) UP0047: cohort 1 (bepranemab 0.3 mg/kg: n=2); cohort 2 (bepranemab 1 mg/kg: n=6); cohort 3 (bepranemab 3 mg/kg: n=6); cohort 4 (bepranemab 10 mg/kg: n=6); cohort 5 (bepranemab 30 mg/kg: n=6); cohort 6 (bepranemab 60 mg/kg: n=6); cohort 7 (bepranemab 120 mg/kg: n=6); placebo: n=14; (B) UP0065: cohort 1 (bepranemab 30 mg/kg: n=6); cohort 2 (bepranemab 60 mg/kg: n=6); cohort 3 (bepranemab 120 mg/kg: n=6); placebo n=6. CONSORT, Consolidated Standards of Reporting Trials.

UP0065

UP0065 was a phase 1, placebo-controlled, participant-blind, investigator-blind, single-dose, dose-escalation study of bepranemab in healthy participants of Japanese descent (NCT03605082). The study was conducted at a phase 1 clinical study unit in the UK.

Participants were male or female, aged ≥20 years and ≤75 years, of Japanese descent (all four grandparents were born in Japan), had a BMI ≥18.0 kg/m² and <30.0 kg/m², with a body weight of 50–100 kg for males and 45–100 kg for females and were in good physical and mental health. All participants were required to have laboratory test results within normal reference ranges, BP within normal range (systolic 90–140 mm Hg, diastolic 40–90 mm Hg) and a normal ECG (or any abnormalities considered clinically nonsignificant by investigator).

Key exclusion criteria were prior exposure to any biological compound targeting the same mechanism of action as bepranemab; participation in a study of another IMP in the previous 3 months or within five half-lives of study drug administration; major surgery in the 6 months before, or planned surgery within 6 months after, study drug administration; current infection or serious infection (requiring hospitalisation or parenteral antibiotics) within 4 weeks prior to dosing; clinically relevant abnormal findings on physical examination, laboratory tests, vital signs, ECG or brain MRI; history of recurrent headaches, including migraines; lifetime history of suicide attempt or suicidal ideation within the previous 2 years. Additional criteria are listed in the online supplemental material.

Participants were randomised to bepranemab or placebo in a 3:1 ratio, within one of three cohorts, as shown in figure 1B. All members of the main study team, including Sponsor/Contract Research Organisation and Investigator site personnel, involved in the study were blinded to the randomised IMP. The procedure for unblinded personnel was the same as in UP0047. The maximum dose level was confirmed based on toleration of 120 mg/kg of bepranemab in the highest dosing cohort in UP0047. Bepranemab was administered by intravenous infusion over ≥120 min, or 135 min for participants in cohort 3. This allowed for a lower infusion rate of 0.5 mg/kg/min for the first 30 min of infusion of the highest bepranemab dose, to mitigate the risk of infusion-related reactions. Bepranemab and the placebo solutions were not identical in appearance. To ensure blinding was maintained, suitable ancillaries were supplied by the Contract Research Organisation.

Endpoints 

UP0047

The primary objective of UP0047 was to assess the safety and tolerability of single ascending intravenous doses of bepranemab in healthy participants. The secondary objectives were to evaluate the serum and CSF concentrations of bepranemab (PK). The exploratory objectives of the study were:

• To determine the proportion of participants developing anti-drug antibodies (ADAs (anti-bepranemab)) in serum (immunogenicity).

• To assess free tau levels in CSF following various doses of bepranemab (PD).

UP0065

The primary objective of UP0065 was to assess the safety, tolerability and serum PK of bepranemab in healthy participants of Japanese descent, to determine whether these differed from the safety, tolerability and PK assessments in Caucasian participants in UP0047. An exploratory objective of the study was to assess the incidence and emergence of ADAs in serum (immunogenicity).

Measurements and evaluations

UP0047

Following an initial screening period of approximately 4 weeks (day –28 to day –2), all participants were admitted to the site on day –1, for 8 consecutive days (maximum). Participants received a single administration of intravenous bepranemab or placebo on day 1 according to the dose specified for that cohort. After day 7, participants were discharged and followed up for 19 weeks, attending the site for clinical study visits on days 14, 28, 42, 56, 84 and 140.

Safety variables were measured according to the schedule of study assessments. PK was also assessed in CSF samples obtained by lumbar puncture from participants in cohorts 3–7, on days –7, 7, 28 and 84. CSF was not taken from cohorts 1 and 2 as these were initial starting doses and predicted to result in low TO (free tau reduction; online supplemental table 1).

For details on study assessments, PK, ADAs and CSF sample assessments, refer to the online supplemental material.

UP0065

Following an initial screening period of approximately 4 weeks (day –28 to day –2), all participants were admitted to the site from day –1, for 6 days, and received a single administration of bepranemab or placebo on day 1. Participants remained at the clinic until day 5 when they were discharged, provided the investigator was satisfied that the participants were in good health. The safety follow-up period was 19 weeks (135 days), from day 6 until the end of the study (day 140) with participants attending the clinical site for study visits (days 7, 14, 28, 42, 56, 84 and 140). Safety variables were measured and collected according to the schedule of study assessments.

PK and ADAs were assessed in serum samples obtained from participants in all cohorts, at the same time points described for the UP0047 study in the online supplemental material.

Statistical analysis 

All statistical analyses were performed using SAS V.9.3 or higher (SAS Institute). Continuous variables were summarised by treatment group (bepranemab dose level or pooled placebo within each study; note, there was no pooling of data across studies), visit and timepoint (where applicable) including number of participants, mean, SD, median, minimum, maximum and CIs. Categorical variables were summarised by treatment group, visit and time point (where applicable) with frequency counts and percentages.

Geometric coefficient of variation, geometric mean and 95% CI for the geometric mean were also presented for PK concentration data. Confidence limits were restricted to the possible values that the variable could have taken.

No formal statistical sample size calculations were performed, as the aim of the studies was to assess the safety and tolerability of bepranemab while limiting exposure to a minimum number of participants.

Results

Participants

In the UP0047 study, a total of 258 participants were screened between 16 February 2018 and 12 June 2018. Of these, 52 participants were randomised to receive either bepranemab or placebo and completed the study (bepranemab 0.3 mg/kg: n=2; 1 mg/kg: n=6; 3 mg/kg: n=6; 10 mg/kg: n=6; 30 mg/kg: n=6; 60 mg/kg: n=6; 120 mg/kg: n=6 or placebo: n=14; figure 1A); 206 participants (79.8%) failed screening. The main reason reported for screen failure was ineligibility (139 participants (53.9%); figure 1A). Randomised participants were followed up until 1 December 2018. Participants were all male, with a mean age of 49.7 years. The study population was predominantly Caucasian, with one participant of Hispanic/Latino ethnicity.

In the UP0065 study, a total of 49 participants were screened between 10 July 2018 and 27 September 2018. Of these, 24 were randomised to receive either bepranemab or placebo and completed the study (bepranemab 30 mg/kg: n=6; 60 mg/kg: n=6; 120 mg/kg: n=6 or placebo n=6; figure 1B); 25 participants (51.0%) failed screening. Randomised participants were followed up until 2 March 2019. Of the 24 participants randomised, 42% were male and the mean age was 35.5 years. Concomitant medications were permitted for both studies; for additional information, refer to the online supplemental material.

A summary of participant demographics for both studies is given in table 1 and online supplemental table 2.

Table 1. Participant demographics for the UP0065 study (FAS).

Category	Placebo (n=6)	Bepranemab 30 mg/kg (n=6)	Bepranemab 60 mg/kg (n=6)	Bepranemab 120 mg/kg (n=6)	All participants (N=24)
Age (years)
Mean (SD)	28.7 (5.2)	34.5 (10.0)	43.8 (11.1)	35.2 (9.6)	35.5 (10.2)
Median (min, max)	28.0 (22, 36)	30.0 (24, 49)	45.5 (26, 59)	30.5 (28, 51)	31.0 (22, 59)
Age, n (%)*
18 to <65 years	6 (100)	6 (100)	6 (100)	6 (100)	24 (100)
65 to <85 years	0	0	0	0	0
Age, n (%)†
19 to <65 years	6 (100)	6 (100)	6 (100)	6 (100)	24 (100)
≥65 years	0	0	0	0	0
Gender, n (%)
Male	4 (66.7)	3 (50.0)	2 (33.3)	1 (16.7)	10 (41.7)
Female	2 (33.3)	3 (50.0)	4 (66.7)	5 (83.3)	14 (58.3)
Weight (kg)
Mean (SD)	61.85 (5.78)	64.52 (7.51)	57.62 (8.33)	56.85 (5.99)	60.21 (7.25)
Median (min, max)	60.5 (54.5, 71.9)	62.55 (57.4, 78.5)	57.2 (49.3, 71.1)	56.6 (50.9, 62.7)	60.45 (49.3, 78.5)
Height (cm)
Mean (SD)	169.8 (8.6)	165.8 (12.3)	164.3 (7.9)	162.3 (5.1)	165.6 (8.7)
Median (min, max)	167.5 (160, 183)	165.0 (150, 185)	166.0 (154, 176)	163.0 (155, 168)	166.0 (150, 185)
BMI (kg/m2)
Mean (SD)	21.48 (1.83)	23.48 (1.47)	21.3 (2.32)	21.58 (2.14)	21.96 (2.04)
Median (min, max)	21.65 (19.1, 23.8)	23.4 (21.8, 25.5)	20.4 (19.7, 25.8)	21.75 (18.3, 24.5)	21.95 (18.3, 25.8)
Racial group, n (%)
White	0	0	0	0	0
Other/mixed	6 (100)	6 (100)	6 (100)	6 (100)	24 (100)
Ethnicity, n (%)
Hispanic or Latino	0	0	0	0	0
Not Hispanic or Latino	6 (100)	6 (100)	6 (100)	6 (100)	24 (100)
Country
UK	6 (100)	6 (100)	6 (100)	6 (100)	24 (100)

^*EudraCT age categories.

^†ClinicalTrials.gov age categories.

BMI, body mass index; EudraCT, European Union Drug Regulating Authorities Clinical Trials Database; FAS, full analysis set; max, maximum; min, minimum; n, number of participants.

Safety and tolerability of bepranemab in healthy participants

UP0047

Of the participants receiving at least one dose of bepranemab or placebo, 65.8% (25/38) and 64.3% (9/14), respectively, experienced at least one treatment-emergent adverse event (TEAE). There were no serious TEAEs, discontinuations due to TEAEs or drug-related TEAEs (online supplemental table 3). One participant with a medical history of varicose veins and varicose vein stripping, who received bepranemab 120 mg/kg, experienced a severe TEAE of varicose ulceration, reported on day 46, which was not considered related to study treatment per the investigator. The most common TEAEs by preferred term are shown in online supplemental table 3.

UP0065

In UP0065, 61% (11/18) of participants receiving at least one dose of bepranemab experienced at least one TEAE. No TEAEs were reported in the placebo group. There were no serious or severe TEAEs, or discontinuations due to TEAEs (table 2). One participant in the bepranemab 60 mg/kg group experienced TEAEs within a few hours of drug infusion that were assessed by the investigator as treatment related. The participant reported a headache (moderate intensity), nausea and vomiting (both mild intensity).

Table 2. UP0065 study—incidence of TEAEs: overview (FAS) and TEAEs by PT (reported by ≥2 participants in any treatment group; FAS).

Category n* (%) (#)	Placebo (n=6)	Bepranemab 30 mg/kg (n=6)	Bepranemab 60 mg/kg (n=6)	Bepranemab 120 mg/kg (n=6)	Total bepranemab (n=18)
Any TEAEs	0	3 (50.0) (8)	5 (83.3) (15)	3 (50.0) (5)	11 (61.1) (28)
Serious TEAEs	0	0	0	0	0
Participant discontinuations due to TEAEs	0	0	0	0	0
Drug-related TEAEs	0	0	1 (16.7) (3)	0	1 (5.6) (3)
Severe TEAEs	0	0	0	0	0
MedDRA (V.21.1) SOC
PT
Gastrointestinal disorders	0	2 (33.3) (2)	1 (16.7) (2)	1 (16.7) (1)	4 (22.2) (5)
Infections and infestations	0	2 (33.3) (2)	3 (50.0) (5)	2 (33.3) (2)	7 (38.9) (9)
Nasopharyngitis	0	2 (33.3) (2)	3 (50.0) (5)	1 (16.7) (1)	6 (33.3) (8)
Nervous system disorders	0	1 (16.7) (1)	3 (50.0) (4)	1 (16.7) (1)	5 (27.8) (6)
Headache	0	1 (16.7) (1)	3 (50.0) (4)	1 (16.7) (1)	5 (27.8) (6)

Note: (#) was the number of individual occurrences of the TEAE in that category.

^*n was the number of participants reporting at least one TEAE in that category. Percentages were based on the total number of participants in each group.

FAS, full analysis set; MedDRA, Medical Dictionary for Regulatory Activities; n, number of participants; PT, preferred term; SOC, system organ class; TEAE, treatment-emergent adverse event.

The most common TEAEs, listed by preferred term, are shown in table 2.

PK of bepranemab in healthy participants

Serum concentrations of bepranemab in studies UP0047 and UP0065

Concentration-time profiles for bepranemab in serum in the UP0047 and UP0065 studies are shown in online supplemental figure 1A,B, respectively, and the comparison of the PK parameters (C_max and area under the curve (AUC)) between the studies is shown in figure 2A,B.

Figure 2. Comparison of serum PK parameters between studies UP0047 and UP0065. Box and whisker plots (median; IQR; 1.5× IQR). Values outside the box and whisker plots represented by red crosses are considered as outliers. (A) AUC was analysed using non-compartmental analysis and stratified per dose level in healthy Caucasian participants and healthy participants of Japanese descent. (B) C_max was stratified by cohort (dose levels) in healthy Caucasian participants and healthy participants of Japanese descent. Despite the limited sample size, ethnicity did not impact PK in the investigated dose range. AUC, area under the curve from time 0 to infinity; C_max, maximum observed concentration; PK, pharmacokinetics.

In UP0047 and UP0065, the geometric mean serum concentration-time profiles for bepranemab increased proportionally with increasing doses, following the administration of single ascending doses (UP0047: 1, 3, 10, 30, 60 and 120 mg/kg; UP0065: 30, 60 and 120 mg/kg). The terminal phase of the concentration-time profiles declined in a parallel manner, indicating linearity in PK in the elimination phases.

In UP0047, no participants exceeded the C_max and AUC_tau observed at the NOAEL in the most sensitive species, cynomolgus monkeys (C_max=9640 µg/mL; C_avg=6057 µg/mL). The time taken to achieve C_max (t_max) ranged from 2 to 8 hours and infusions lasted 2 hours. This variability in t_max, and the fact that t_max did not always correspond with the end of the intravenous infusion, may result from measurement error in the concentrations, which should be approximately constant from 2 to 12 hours post dose.

PK parameters of bepranemab in serum for the UP0047 and UP0065 studies are shown in online supplemental tables 4 and 5, respectively.

Dose proportionality was observed for C_max in UP0047 (online supplemental figure 2A). Bepranemab exposure (AUC) also increased with increasing dose (online supplemental figure 2B) but did not meet the criteria for dose proportionality in the power analysis, as the 90% CI did not include the numerical value 1 (90% CI 0.896 to 0.955), potentially due to limited sample size. Dose proportionality was confirmed for both C_max and AUC across the dose range 30–120 mg/kg in UP0065 (online supplemental figure 3A,B); the slight deviation from dose proportionality in the exposure (AUC) is likely due to variability between participants and the limited sample size. Body weight has an effect on PK; therefore, body weight adjusted dose would guarantee the same exposure for every participant. Half-life geometric mean values ranged from 23.5 to 33.1 days (564.0 to 794.7 hours) in UP0047 (online supplemental table 4) and 25.6 to 27.9 days in UP0065 (online supplemental table 5), consistent with a linear PK.

Comparison between Caucasian participants and participants of Japanese descent showed no difference in PK parameters in the range of doses investigated (figure 2). The PK of bepranemab was characterised with low variability for all PK parameters across all dose groups (with one outlier in the 30 mg/kg group in UP0065; online supplemental table 5).

CSF concentrations of bepranemab in study UP0047

In UP0047, bepranemab concentrations in the CSF increased with increasing dose, across the dose range 3–120 mg/kg. The geometric mean of the CSF:serum ratio of bepranemab was comparable across the dose range (table 3). Bepranemab concentrations in CSF and serum at day 28 (table 3) suggested that this ratio would be maintained immediately prior to the estimated time of a second dose of bepranemab in future multiple-dosing studies.

Table 3. UP0047 study—PK of bepranemab in CSF (PK-PPS).

Parameter (unit)	Statistic	Bepranemab 3 mg/kg (n=6)	Bepranemab 10 mg/kg (n=6)	Bepranemab 30 mg/kg (n=6)	Bepranemab 60 mg/kg (n=6)	Bepranemab 120 mg/kg (n=6)
Cmax (µg/mL)	GeoMean (GeoCV (%))	0.04804 (46.5)	0.2011 (33.9)	0.7123 (43.5)	1.421 (61.8)	2.852 (21.6)
tmax (hours)	Median (min, max)	144.0 (144, 672)	143.7 (144, 144)	144.2 (144, 145)	143.5 (143, 144)	143.8 (144, 144)
CSF/serum ratio Day 7/144 hours	GeoMean (GeoCV (%))	0.0008882 (41.9)	0.001443 (37.3)	0.001627 (37.3)	0.002041 (63.5)	0.001959 (29.0)
CSF/serum ratio Day 28/648 hours	GeoMean (GeoCV (%))	0.001424 (41.4)	0.002247 (33.8)	0.002388 (40.9)	0.002512 (27.8)	0.002399 (31.5)
CSF/serum ratio Day 84/1992 hours	GeoMean (GeoCV (%))	0.001813 (34.2)	0.002322 (47.4)	0.002728 (30.8)	0.002591 (39.9)	0.002483 (27.3)

C_max and t_max values in CSF should be evaluated considering the sparse sampling time.

Note: Only performed on cohorts 3–7 (figure 1A).

Note: GeoMeans and GeoCVs were only calculated if at least two-thirds of the parameters were properly determined parameters (ie, non-calculated and non-flagged).

Note: CSF:serum ratio was derived as the sum of the CSF values divided by the sum of serum values at the corresponding three scheduled time points.

C_max, maximum observed concentration; CSF, cerebrospinal fluid; GeoCV, geometric coefficient of variation; GeoMean, geometric mean; max, maximum; min, minimum; n, number of participants; PK, pharmacokinetic; PK-PPS, PK-per protocol set; t_max, time to maximum concentration.

Free tau species targeted by bepranemab in CSF

Mean change from baseline in free tau levels was assessed in CSF samples from participants in UP0047 (figure 3). A dose response effect of bepranemab on free tau levels was observed across the doses (3–120 mg/kg). The greatest reduction from baseline in mean free tau levels was observed at day 7 after administration of the highest dose of bepranemab (120 mg/kg) and ranged from 86.2 to 81.4%. In the 120 mg/kg group, reduction from baseline in mean free tau levels remained >60% at day 28 postdose. This suggested that monthly administration would maintain high TO during the entire dosing period. At the highest dose of bepranemab (120 mg/kg), a reduction from baseline in mean free tau levels in CSF was still observed at day 84 postdosing.

Figure 3. Change from baseline in free tau levels in CSF of participants from UP0047. Box and whisker plots (median; IQR; 1.5× IQR). Change from baseline in free tau was measured postdose on days 7, 28 and 84 and stratified per dose level. Results from the placebo treatment are also included for reference. Individual participant data are shown as black and grey dots. CSF, cerebrospinal fluid.

Data on the dose response effect of bepranemab on mean change from baseline in free tau levels in CSF were consistent with those predicted in a physiologically based PK/PD (PBPK/PD) model (online supplemental figure 4). For more information on the modelling and simulated TO in people with PSP, please see the online supplemental material.

Discussion

Bepranemab was generally well tolerated and had an acceptable safety profile in Caucasian participants and participants of Japanese descent at the single ascending doses assessed (0.3–120 mg/kg). There were no serious TEAEs reported in either of the two studies and no participants discontinued either study due to TEAEs. There was one severe TEAE in UP0047, considered not related to study treatment by the investigator, of varicose vein ulceration in a participant with a history of varicose veins and varicose vein stripping, who received bepranemab 120 mg/kg. One participant had two TEAEs deemed treatment-related by the investigator, which were headache of moderate intensity, and nausea and vomiting of mild intensity, reported within a few hours of administration of 60 mg/kg bepranemab.

In both studies, the geometric mean serum concentrations of bepranemab increased proportionally with increasing doses of bepranemab. The terminal phase of the concentration-time profiles in serum declined in a parallel manner, indicating linearity in PK. In UP0047, the mean CSF:serum ratio of bepranemab was comparable across the dose range (0.0009‒0.002 at day 7).

These results are consistent with CSF:serum ratios previously reported for the anti-tau IgG antibody BMS-986168 (0.002),¹⁰ and for rituximab in systemic non-Hodgkin’s lymphoma (0.001).¹¹ In UP0047 and UP0065, there were no differences in any PK parameters across ethnicities/between Caucasian participants and participants of Japanese descent.

Free tau in CSF was measured in UP0047, and the maximum reduction from baseline in mean free tau levels was observed at day 7 after administration of the highest dose of bepranemab (120 mg/kg). Smaller reductions from baseline in free tau levels were observed in the 60 and 30 mg/kg dose groups. The small sample size and high variability in tau concentrations did not allow for the determination of any significant reduction in free tau versus placebo. Bepranemab PK in serum and CSF showed an extended half-life, which corresponded to a prolonged effect on free tau levels. Reductions from baseline in mean free tau levels were also observed at days 28 and 84 with the highest bepranemab dose. The reduction of >60% at day 28 supports a repeated monthly dosing regimen to maintain TO during the dosing period.

The model-informed drug development paradigm¹² was applied to these phase 1 data and population PK and PBPK/PD models were developed to describe the serum and CSF PK/PD data (online supplemental figure 5). The PBPK/PD model supported a dose of 90 mg/kg every 4 weeks to achieve a reduction in mean change from baseline in free tau levels of up to 90% in the majority of participants.

Previous studies assessing the efficacy and safety of anti-tau therapies used mAbs targeting the N-terminal region of tau.^{13 14} Both tilavonemab and gosuranemab failed to show signs of efficacy in phase 2 studies in PSP. In contrast, bepranemab targets a mid-region of tau within the central proline-rich region, adjacent to the MTBR, which is believed to be essential for aggregation and is less likely to be lost due to post-translational cleavage than epitopes in proximity to the terminals.^{2 3} This is consistent with preclinical studies showing that bepranemab prevented both in vitro seeding of human tau and in vivo spread of pathological tau throughout the brain.²

The limited sample size in UP0047 and UP0065 was typical and appropriate for phase 1 studies, and available data show no evidence of any effect of ethnicity on the safety, tolerability, PK and PD of bepranemab.

The safety profile, dose response on PK parameters (C_max and AUC; fitted to a PBPK/PD model that supported a dose of 90 mg/kg every 4 weeks for use in future studies) and TO of up to 90% further supports continued clinical development of bepranemab, and this includes a phase 1b study in people with PSP (NCT04185415) and a phase 2 study in AD (NCT04867616).

The study sponsor was involved in the study design; in the collection, analysis and interpretation of the data; in the writing of the report; and in the decision to submit the paper for publication. The funder did not influence the results/outcomes of the study despite author affiliations with the funder.

Data availability statement

No data are available.