Tau-targeting antisense oligonucleotide MAPTRx in mild Alzheimer’s disease: a phase 1b, randomized, placebo-controlled trial

Tau plays a key role in Alzheimer’s disease (AD) pathophysiology, and accumulating evidence suggests that lowering tau may reduce this pathology. We sought to inhibit MAPT expression with a tau-targeting antisense oligonucleotide (MAPTRx) and reduce tau levels in patients with mild AD. A randomized, double-blind, placebo-controlled, multiple-ascending dose phase 1b trial evaluated the safety, pharmacokinetics and target engagement of MAPTRx. Four ascending dose cohorts were enrolled sequentially and randomized 3:1 to intrathecal bolus administrations of MAPTRx or placebo every 4 or 12 weeks during the 13-week treatment period, followed by a 23 week post-treatment period. The primary endpoint was safety. The secondary endpoint was MAPTRx pharmacokinetics in cerebrospinal fluid (CSF). The prespecified key exploratory outcome was CSF total-tau protein concentration. Forty-six patients enrolled in the trial, of whom 34 were randomized to MAPTRx and 12 to placebo. Adverse events were reported in 94% of MAPTRx-treated patients and 75% of placebo-treated patients; all were mild or moderate. No serious adverse events were reported in MAPTRx-treated patients. Dose-dependent reduction in the CSF total-tau concentration was observed with greater than 50% mean reduction from baseline at 24 weeks post-last dose in the 60 mg (four doses) and 115 mg (two doses) MAPTRx groups. Clinicaltrials.gov registration number: NCT03186989.

Tau plays a key role in Alzheimer's disease (AD) pathophysiology, and accumulating evidence suggests that lowering tau may reduce this pathology. We sought to inhibit MAPT expression with a tau-targeting antisense oligonucleotide (MAPT Rx ) and reduce tau levels in patients with mild AD. A randomized, double-blind, placebo-controlled, multiple-ascending dose phase 1b trial evaluated the s af et y, p ha rm ac okin etics and target engagement of MAPT Rx . Four ascending dose cohorts were enrolled sequentially and randomized 3:1 to intrathecal bolus administrations of MAPT Rx or placebo every 4 or 12 weeks during the 13-week treatment period, followed by a 23 week post-treatment period. The primary endpoint was safety. The secondary endpoint was MAPT Rx p ha rmacokinetics in cerebrospinal fluid (CSF). The prespecified key exploratory outcome was CSF total-tau protein concentration. Forty-six patients enrolled in the trial, of whom 34 were randomized to MAPT Rx and 12 to placebo. Adverse events were reported in 94% of MAPT Rx -treated patients and 75% of placebo-treated patients; all were mild or moderate. No serious adverse events were reported in MAPT Rx -treated patients. Dose-dependent reduction in the CSF total-tau concentration was observed with greater than 50% mean reduction from baseline at 24 weeks post-last dose in the 60 mg (four doses) and 115 mg (two doses) MAPT Rx groups. Clinicaltrials.gov registration number: NCT03186989.
Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by cognitive and functional decline resulting in substantial disability 1 . Onset of pathology is marked by progression of neuroimaging and fluid biomarker measures (preclinical phase) before the appearance of subtle cognitive changes, known as mild cognitive impairment (MCI). The eventual progression to dementia occurs over a variable period of time and is characterized by cognitive and behavioral symptoms that impair an individual's ability to Article https://doi.org/10.1038/s41591-023-02326-3 axonal transport or sensory, motor or cognitive behavior tasks 22,25,26 . Moreover, complete tau knockout mice had normal development and cognition with only a minor motor phenotype developing later in life [27][28][29][30] . These data mitigate potential safety concerns of lowering tau as a therapeutic approach for AD and other tauopathies.
In this study, we aimed to evaluate the safety and pharmacokinetics (PK) of MAPT Rx in patients with mild AD and explore the hypothesis that precisely targeted degradation of MAPT mRNA using an ASO would result in lowering of t-tau and phosphorylated tau (p-tau) levels in the central nervous system (CNS). We report the results of a first-in-human phase 1b clinical trial evaluating a tau-targeting ASO administered intrathecally as a bolus in adults with mild AD.

Patients
From August 2017 through February 2020, 102 participants were screened for eligibility and 46 underwent randomization according to the protocol (Fig. 1). All participants received all scheduled doses of the study drug (MAPT Rx or placebo) during multiple ascending dose (MAD) part 1. Three participants (6.5%) voluntarily withdrew from the study during the post-treatment period: one each from the placebo, 60 mg MAPT Rx cohort (four total doses administered monthly) and 115 mg MAPT Rx cohort (two total doses administered quarterly). Participants completing MAD part 1 were eligible to participate in the open-label long-term extension (LTE) part 2. Participants randomized to 10 mg or 30 mg MAPT Rx (four total doses administered monthly) cohorts experienced a variable gap between completion of the 13 week treatment period of the MAD in part 1 and day 1 of LTE part 2 since the protocol was amended to add the LTE after participants in these cohorts had begun the study. Transition to LTE part 2 was seamless after a 23 week post-treatment period for participants in the 60 mg and 115 mg MAPT Rx cohorts.
The characteristics of participants at baseline were representative of relatively younger (mean age of 66 years in both placebo and MAPT Rx groups) patients with mild AD and were generally similar across trial groups (Table 1). MAPT Rx groups and placebo group had similarly elevated CSF levels of mean t-tau (405.6 ± 132.7 and 387.3 ± 120.9 pg ml −1 , respectively) and p-tau181 (40.7 ± 14.2 and 38.7 ± 13.0 pg ml −1 , respectively) concentrations at baseline. The mean Clinical Dementia Rating (CDR) Sum of Boxes score at baseline was numerically lower for MAPT Rx 60 mg and 115 mg treatment groups due to an amendment during the study, which allowed inclusion of participants with a CDR Global Score of 0.5 and Memory Score of 1 in addition to participants with a CDR Global Score of 1.0.

Primary endpoint of safety
Adverse events (AEs) were reported in 94% of participants treated with MAPT Rx and 75% of participants treated with placebo; all events were considered mild (88%) or moderate (12%) in severity by investigators (Table 2; for AE incidence and frequency by treatment group, see Extended Data Table 1). A greater percentage of participants receiving MAPT Rx experienced mild AEs compared with those receiving placebo; the incidence of moderate AEs was similar.
The most reported AE in participants treated with MAPT Rx was post-lumbar puncture (LP) headache, which was generally considered mild (N = 13); two participants reported post-LP headache considered moderate. Post-LP headache considered potentially related to study procedure occurred after 20% of LPs. There was no evidence of an increased risk of post-LP headache with successive LPs. All post-LP headaches resolved (median duration, <1 day), and no blood patches were required to resolve.
AEs considered potentially related to study drug by Investigators were reported in 15 participants (44%) treated with MAPT Rx and no participants treated with placebo. Investigators were blinded to treatment assignment. Most participants (80%) with AEs considered potentially function in daily life 2 . Symptom onset typically occurs in patients aged 65 years and older, while symptom onset before age 65 accounts for less than 5% of all patients with AD 3 . Historically, diagnosis of AD has been primarily focused on clinical criteria 4 , but accumulating evidence has demonstrated that cerebrospinal fluid (CSF) biomarkers, including amyloid-β 42 (Aβ42) and tau (total tau (t-tau) and phosphorylated tau181 (p-tau181)) as well as positron emission tomography (PET)-amyloid and PET-tau, are reliable surrogate measures of neuropathologic change enabling more robust characterization of patients across the AD continuum [5][6][7] . For most patients with AD, treatment remains limited to multidisciplinary management of symptoms, including pharmacological therapies that have no disease-modifying impact. The recent US Food and Drug Administration accelerated approvals of aducanumab and lecanemab provide the first treatment targeting a key disease mechanism in AD, the accumulation of amyloid plaques, for patients with MCI or mild AD. There are over 50 million people worldwide currently living with dementia mostly due to AD, and this number is expected to double every 20 years 8 ; therefore, additional disease-modifying treatments to prevent or slow progression of this disease remain a significant unmet need.
Growing evidence suggests that aggregated, hyperphosphorylated tau may be a key driver of neurodegeneration in AD. Tau protein is encoded by the microtubule-associated protein tau (MAPT) gene and is a microtubule-associated protein primarily expressed in neurons 9 . Under pathogenic conditions, hyperphosphorylated tau accumulates intracellularly, aggregating into oligomers and fibrils resulting in intraneuronal neurofibrillary tangles, and is associated with cognitive decline in AD 10,11 . Tau is also secreted from neurons, spreading through specific neural networks via a trans-synaptic route causing propagation of tau pathology that is associated with further synaptic dysfunction and neuronal loss [12][13][14][15] . Preclinical evidence has demonstrated that tau reduction prevents specific Aβ-mediated deficits, supporting a central role of tau in mediating Aβ toxicity in the early pathogenesis of AD. Intracerebroventricular injection of purified Aβ oligomers into adult rodents has been shown to impair long-term potentiation, and this impairment of long-term potentiation in hippocampal slices of wild-type mice is prevented in tau knockout mice 16,17 . Evidence from amyloid precursor protein (APP) mouse models of AD have shown that both hetero-and homozygote tau deficiency rescued premature mortality and prevented memory deficits in transgenic mice expressing familial AD mutations in human APP (hAPP), which appeared to be conferred by reduced susceptibility to excitotoxicity in tau knockout mice 18,19 . In addition, knocking out tau rescued memory impairments, loss of synapses and premature death in hAPP mice expressing human mutant PS1 (ref. 20). Given the important role of tau in AD pathophysiology and the accumulating evidence that lowering tau may reduce this pathological effect, we sought to inhibit MAPT expression and thus reduce tau levels, directly targeting a key disease effector mechanism in patients with AD.
MAPT Rx (ISIS 814907/BIIB080) is an antisense oligonucleotide (ASO) designed to reduce concentrations of MAPT messenger RNA. MAPT Rx is a chemically modified synthetic oligomer that is complementary to an 18-nucleotide stretch of MAPT pre-mRNA. MAPT Rx binds within intron 9 of the MAPT pre-mRNA through Watson-Crick base pairing, with hybridization resulting in endogenous ribonuclease H1-mediated degradation of the MAPT mRNA, inhibiting translation of the tau protein. ASO-mediated selective reduction of MAPT mRNA leads to lowered tau protein levels and sustained amelioration of disease-associated phenotypes in transgenic animal models of tauopathy and hyperexcitability [21][22][23][24] . For example, MAPT mRNA-targeting ASOs in a mouse model of tauopathy resulted in a 50% reduction of endogenous intracellular tau, reduced cell-to-cell spread of oligomerized tau, markedly reduced neuronal and cognitive impairments and was not associated with adverse consequences 15 . Knockdown of tau in animal models and primary neurons did not impair microtubule assembly, Article https://doi.org/10.1038/s41591-023-02326-3 related to study drug experienced mild AEs. Safety magnetic resonance imaging (MRI) was performed 6 months post-baseline, and no clinically meaningful changes were observed on qualitative neuroradiological review. There were no deaths, dose-limiting AEs or discontinuations of dosing regimens during the trial. One participant experienced a delay of approximately 2 months in study drug administration due to coronavirus disease 2019 restrictions.
Two serious AEs occurred in two participants receiving placebo: hospitalization due to diverticulitis and an emergency room visit due to a minor stroke from which both participants recovered. Neither suicidal behavior nor serious suicidal ideation emerged in any participant during the trial. A mildly increased CSF leukocyte count (26-28 cells mm −3 , >90% lymphocytes) without any associated symptoms was observed in one participant, a 64-year-old female, 16 weeks after administration of the second MAPT Rx 115 mg dose; MRI with contrast and electroencephalographic results were normal. The participant did not transition to the LTE, but follow-up safety MRI and CSF collection performed post-MAD part 1 completion showed that the pleocytosis had completely resolved (5 months after initial finding).
Two participants receiving quarterly 115 mg MAPT Rx experienced mild confusional state and restlessness 1-2 days after their first and second doses, which resolved within 2-4 days of onset. Both participants had a medical history of anxiety and were treated with psychotropic medications before their enrollment in the study.

Secondary endpoint
MAPT Rx was measurable in the CSF in all participants receiving MAPT Rx (Fig. 2a). Pre-dose or trough concentrations increased from the 10 mg monthly dose to those observed at the 30 mg and 60 mg monthly doses. Similar trough CSF concentrations were observed after the 30 and 60 mg monthly doses. It is unclear why there is no apparent difference in MAPT Rx CSF trough concentration between the 30 mg and 60 mg MAPT Rx groups. Only six participants in each group received 30 mg or 60 mg MAPT Rx , and CSF is not a well-mixed compartment with variable results observed previously 31 . The lowest mean trough CSF concentrations observed after the initial 115 mg quarterly dose were expected considering the longer time for elimination between MAPT Rx doses and sample collection (84 days versus 28 days). Mean trough MAPT Rx concentrations in CSF generally increased during the treatment period in all dose groups probably due to the slow clearance and long elimination half-life of MAPT Rx relative to the dosing interval. The increase over time was less in the 115 mg quarterly cohort compared with the cohorts dosed every month.

Exploratory endpoints
Plasma concentrations of MAPT Rx . The median peak plasma concentrations of MAPT Rx were achieved within 4 h after intrathecal (IT) administration and declined to less than 30% of the peak concentration by 24 h after administration. The concentration of MAPT Rx in plasma increased approximately proportionally to the dose over the explored dose range (Fig. 2b). There was no evidence of accumulation of concentration in plasma 24 h after dose administration over the course of the trial, and there was a minor increase (<20%) in the peak concentration at the 115 mg dose level.

Concentration of t-tau in CSF.
In participants receiving MAPT Rx , there were dose-dependent decreases in the concentration of t-tau in CSF. Steady-state maximal reduction of the concentration of CSF t-tau was not reached during the 13 week treatment period, and t-tau concentrations continued to decline during the post-treatment period (Fig. 3a). The mean percentage change from baseline in t-tau concentration at 8 weeks post-last dose was −30%, −40%, −49% and −42% in MAPT Rx 10 mg, 30 mg and 60 mg monthly and 115 mg quarterly groups, respectively (Fig. 3b). In the higher-dose groups with seamless entry into the LTE, CSF t-tau continued to decline at 24 weeks (day 1 LTE part 2) post-last dose in the 60 mg monthly (N = 7) and 115 mg quarterly (N = 10) MAPT Rx groups (−56% and −51% mean percentage change from baseline, respectively; Fig. 3c). The 60 mg and 115 mg MAPT Rx groups received almost identical cumulative doses of 240 mg and 230 mg, respectively, over the 13 week treatment period, which may account for the similar t-tau reduction observed in both groups. Participants randomized to 10 mg and 30 mg MAPT Rx monthly groups had a variable gap between completing MAD part 1 and starting LTE part 2; the duration between the last dose in MAD part 1 and CSF collection before the first dose in LTE part 2 ranged from 22 to 28 months for the 10 mg group and from 10 to 16 months for the 30 mg group. Despite the prolonged gap, a durable reduction in t-tau concentration was observed in the 30 mg MAPT Rx group (−31% mean percentage change from baseline; N = 5) on day 1 of the LTE. T-tau levels had returned to baseline levels in the 10 mg MAPT Rx group (N = 3). In participants receiving placebo, the mean percentage change from baseline ranged from −1% to −2.4% among all post-baseline visits (Fig. 3c).

Additional exploratory outcomes.
Reductions similar to those observed for t-tau were observed for p-tau181 concentration and the ratio of t-tau to Aβ42 in CSF (Fig. 4). In participants receiving MAP-T Rx , there were dose-dependent decreases in the concentration of p-tau181 in CSF 8 weeks post-last dose with mean percentage change from baseline of −35%, −44%, −52% and −49% in MAPT Rx 10 mg, 30 mg and 60 mg monthly and 115 mg quarterly groups, respectively. CSF p-tau181 continued to decline in participants treated with MAPT Rx in 60 mg monthly and 115 mg quarterly groups 24 weeks (day 1 LTE part 2) post-last dose with mean percentage change from baseline of −56% and −46%, respectively.
Performance on functional, cognitive, psychiatric and neurologic clinical outcomes slightly declined as expected for participants with mild AD over the duration of the treatment and post-treatment periods. While no consistent trends were observed across change   Table 1). CSF NfL levels decreased from baseline in the placebo and 10 mg MAPT Rx groups, and a slight increase from baseline was observed in the 30, 60 and 115 mg MAPT Rx groups. All groups experienced a slight increase from baseline in CSF NfH with the 30 mg MAPT Rx group experiencing the greatest increases in both NfL and NfH. Decrease from baseline in CSF YKL40 was observed in all MAPT Rx treatment groups, whereas no change from baseline was observed in the placebo group. CSF NRGN levels decreased from baseline in the 10, 30 and 60 mg MAPT Rx groups, and no change from baseline was observed in the placebo or 115 mg MAPT Rx groups. The mean change from baseline in ventricular volume (VV) as a percentage of total intracranial volume 6 months post-baseline was greater in the 10 mg, 30 mg, 60 mg and 115 mg MAPT Rx dose groups (0.5%, 0.7%, 0.7% and 0.6%, respectively) than that observed in the placebo group (0.2%; Extended Data Table 2). Ventricular enlargement (VE) was not observed in qualitative neuroradiological review of safety MRIs from participants in MAPT Rx or placebo groups. Clinical findings potentially associated with VE were not observed during the treatment or post-treatment periods. Whole-brain volume declined slightly from baseline in all groups and did not differ between participants who received placebo and those who received MAPT Rx.

Discussion
In this first in human phase 1b study, bolus IT administrations of four monthly doses of MAPT Rx at 10 mg, 30 mg and 60 mg or two quarterly doses at 115 mg to adults with mild AD were not accompanied by any severe or serious AEs during the 13 week treatment period or 23 week post-treatment period. The proportion of participants experiencing AEs was greater in those receiving MAPT Rx versus placebo (94% versus 75%, respectively), and this was mainly due to an increased incidence of mild AEs in the MAPT Rx treatment group (62% versus 42% in placebo). AEs considered potentially related to study drug by investigators were reported more frequently in with MAPT Rx -treated participants compared with placebo-treated participants (15 (44%) versus 0, respectively). The most reported AE in participants receiving MAPT Rx was post-LP headache, which was generally mild in severity. All study participants were white, and it will be important in larger, later-phase clinical studies of MAPT Rx to include a diverse patient population to adequately evaluate both efficacy and safety. Overall MAPT Rx treatment was generally well tolerated, with all participants completing the treatment period and over 90% of participants completing the post-treatment period.
MAPT Rx administration resulted in dose-and time-dependent reduction in the concentration of CSF t-tau and p-tau181 with approximately 50% mean reduction from baseline observed 24 weeks post-last dose. Further characterization of MAPT Rx PK and PD data from the LTE will be important for selection of the optimal dose level and frequency for future clinical studies. It is not feasible to directly quantify the reduction of MAPT mRNA or tau protein in cortical tissue in a clinical trial; however, it is possible to index PD activity with CSF protein assays 31,32 . Tau is a long-lived protein in the CNS, and thus CSF tau in this study will be a lagging indicator of the reduction of MAPT mRNA and newly synthesized tau in the CNS 33 . The predicted MAPT Rx concentrations in brain tissue at all dose levels in this study are sufficient to provide >50% reduction in t-tau production in the cerebral cortex 31 . Therefore, it is not surprising that all doses assessed in this study demonstrated a similar trajectory of CSF tau lowering (Figs. 3 and 4), a trajectory that probably reflects potent reductions in new tau synthesis at all doses  . Each AE was rated as mild, moderate or severe, corresponding to grades of 1, 2 and 3, respectively. In addition, serious AEs were rated as life-threatening (grade 4) or not life-threatening. At each level of summation (overall and according to system organ class or preferred term), patients for whom more than one AE was reported were counted only once for the incidence according to the most severe grade, and if there was a missing severity for the same subject, then the non-missing severity, if available, was chosen for the same subject. b Post-LP headache indicates both post-LP syndrome and headache that were potentially related to study LP procedure. Related was defined as 'related', 'possible' or missing relationship to LP procedure.
Many of the tau-targeting antibody and vaccine approaches currently in development aim to reduce spread of specific extracellular tau species and are not predicted to have a major impact on intracellular p-tau 34 . MAPT Rx prevents tau protein production, and should lower the levels of all tau species and subsequent posttranslational modifications with the potential to reduce both pathological spreading of extracellular tau and neuronal dysfunction due to intracellular tau accumulation. Limitations of evaluating CSF t-tau and p-tau to index target engagement include the clearance of existing tau protein and reliance on tau transport to the CSF. Physiological and pathological conditions may impact the rate of synthesis of tau, passive and active release into the extracellular space, and the clearance of tau including degradation by microglia 33,35,36 . At baseline, CSF tau represents previously synthesized tau. However, at steady state with respect to both ASO distribution and the production and elimination of tau, lowered CSF tau levels should index the reduction of newly synthesized tau 21,22,37 . In the ongoing LTE, steady-state reductions in CSF t-tau protein levels should emerge that reflect the degree to which newly synthesized tau protein is reduced in the CNS. While measuring CSF tau protein can index PD activity, it cannot inform on regional differences in MAPT mRNA reduction in CNS tissues.
Quantitative assessment of VV showed greater increases in MAPT Rx treatment groups compared with placebo. Importantly, VE was not evident on qualitative neuroradiological review of safety MRI scans, and there were no clinical correlates. Quantitative increases in VV have been observed in treatment groups relative to placebo in clinical studies of patients with AD 38-41 while other clinical studies of AD have not reported treatment related-increases [42][43][44][45] . The etiology of greater quantitative VE relative to placebo in these studies remains unclear, and     Error bars indicate the standard error of the mean. Q4W and Q12W indicates dosing every 4 or 12 weeks, respectively. *Participants assigned to cohort A or B did not seamlessly transition to LTE part 2 and experienced a variable gap ranging from 5 to 19 months between completion of MAD part 1 at day 253 and start of LTE part 2 (D1P2). χ Placebo group was pooled. Subjects assigned to cohorts A or B and randomized to placebo had a variable gap between completion of MAD part 1 and start of LTE part 2 (D1P2). Article https://doi.org/10.1038/s41591-023-02326-3 associated changes in clinical outcomes were generally not reported in these studies. Slow, progressive whole-brain atrophy (that is, irreversible loss of brain tissue) and VE are characteristic features of AD 46,47 , and neuroinflammation is a known phenomenon in AD 48 . Although 'pseudoatrophy' (that is, VE due to resolution of inflammatory edema and gliosis) has been described in clinical studies of multiple sclerosis and AD, it has been challenging to differentiate between treatment-induced pseudoatrophy and disease-related atrophy 38,39,42,[49][50][51] . There were no apparent differences in whole brain volume in MAPT Rx treatment groups versus placebo group in this study. Further work is needed to assess the effect of MAPT Rx treatment on inflammation or gliosis in humans or animal models. Comparing the rate of VE observed in our study with that observed in the Alzheimer's Disease Neuroimaging Initiative is challenging since our patient population was younger (mean age 66 years versus 75 years in the Alzheimer's Disease Neuroimaging Initiative) 46,47,52 . Nearly half of the participants in our study were diagnosed before age 65, representing a much higher proportion of participants with early onset AD compared with the general AD population (46% versus 5%). Disease severity, age and genetic status may influence the degree and rate of increase in VV and requires further evaluation in future studies of the drug.
MAPT Rx is the first ASO treatment evaluated in a clinical study of patients with AD. The results from this first in human study demonstrate that MAPT Rx engaged its target, as evidenced by the marked dose-dependent and sustained reductions in the concentration of CSF t-tau, and had an acceptable safety profile in participants *Participants assigned to cohort A or B did not seamlessly transition to LTE part 2 and experienced a variable gap ranging from 5 to 19 months between completion of MAD part 1 at day 253 and start of LTE part 2 (D1P2). χ Placebo group was pooled. Subjects assigned to cohorts A or B and randomized to placebo had a variable gap between completion of MAD part 1 and start of LTE part 2 (D1P2). This first-in-human study of MAPT Rx has several limitations mainly due to its small size (N = 46) as is typical of phase 1 studies. The MAP-T Rx doses evaluated in this study achieved the target t-tau reduction of ~50%, but determining whether this reduction is efficacious will require further evaluation in larger, well-controlled trials. Similarly, MAPT Rx -treatment related effects on exploratory biomarkers will require further evaluation in larger trials. Changes in NfL, Nfh, YKL40 and NRGN levels did not appear to be dose responsive and were not concordant, but interpretation of the clinical meaningfulness of these results is limited due to the combination of assay variability and small sample size. Lastly, this study was conducted in relatively young participants with mild AD, and it will be important that future studies of MAPT Rx evaluate safety and efficacy in an older population, which may be more representative of late-onset AD. A randomized, double-blind, placebo-controlled phase 2 study of BIIB080 (MAPT Rx ) is currently underway and includes patients aged 50-80 years with an estimated enrollment of over 700 participants with MCI due to AD or mild AD (Clinicaltrials.gov registration number NCT05399888).
These results demonstrate that antisense-mediated suppression of tau protein synthesis in the CNS of participants with mild AD is possible and warrant further evaluation of the effect of MAPT Rx on the clinical course of patients with AD and in other tauopathies.

Study design and participants
In this randomized, double-blind, placebo-controlled, multicenter, MAD phase 1b trial with an open-label LTE, we evaluated the safety, PK and target engagement of MAPT Rx in participants with mild AD. This study was divided into two parts: MAD part 1, was completed in September 2020, and LTE part 2 was completed in May 2022. Participants had the option to enroll in the open-label LTE upon completion of MAD part 1.
Eligible participants were between the ages of 50 and 74 years and had mild AD, defined by a CDR 58 Overall Global Score of 1 or Global Score of 0.5 with a Memory Score of 1, Mini-Mental State Examination score (MMSE 59 ) of 20-27 inclusive (scores range from 0 to 30, 20-27 may represent mild AD); CSF pattern of low Aβ42 and elevated t-tau and p-tau consistent with diagnosis of AD; and diagnosis of probable AD based on National Institute of Aging-Alzheimer Association criteria 60 (for further details, see Supplementary Information).
MAD part 1 was conducted at 12 centers in Canada, Finland, Germany, the United Kingdom, the Netherlands and Sweden. A centralized automated randomization system assigned participants 3:1 to receive bolus IT injections of MAPT Rx or placebo (artificial CSF) within each of four dose cohorts during the 13 week treatment period: cohort A, 10 mg monthly, cohort B 30 mg monthly or cohort C, 60 mg monthly (total of four doses each); or cohort D, 115 mg quarterly (total of two doses). Twenty milliliters of CSF were removed before administration of 20 ml of study drug. There was a 23 week post-treatment period during which no study drug was administered. Participants assigned to cohort A or B did not seamlessly transition to LTE part 2 and experienced a variable gap ranging from 5 to 19 months between completion of MAD part 1 at day 253 and registration for LTE part 2. Participants assigned to cohort C or D seamlessly transitioned to LTE part 2. CSF samples were obtained before each administration of study drug (MAPT Rx or placebo), 4 and 8 weeks post-last dose for cohorts A and B, and 8 and 16 weeks post-last dose for cohorts C and D (Fig. 1). Investigators, participants and the sponsor were blinded to trial-group assignments for the trial duration.
The primary objective was evaluation of the safety of MAPT Rx . Safety evaluations included collection of AEs, physical and neurologic examination, Columbia Suicide Severity Rating Scale, laboratory assessments, vital signs, electrocardiograms and safety MRI sequences. At each trial visit, participants were queried for other changes in health status and concomitant medications in an open-ended fashion.
The secondary endpoint was the characterization of the PK of MAPT Rx in CSF.
The key exploratory endpoint was CSF t-tau concentration. Additional exploratory endpoints included characterization of MAPT Rx PK in plasma; exploration of the effects of MAPT Rx on PD biomarkers including CSF concentrations of p-tau, Aβ42 and NfL; and clinical, cognitive and neuroimaging assessments relevant to AD (Supplementary Table 1).

Study drug
MAPT Rx is a second-generation 2′-O-(2-methoxyethyl) ASO complementary to a nucleotide sequence in the human MAPT pre-mRNA transcript. The sequence of MAPT Rx is (5′ to 3′) ccogttTTCTTACCacocct, where capital letters represent 2′-deoxyribose nucleosides, and small letters 2′-(2-methoxyethyl)ribose nucleosides. Nucleoside linkages represented with a subscript o are phosphodiester, and all others are phosphorothioate. Letters represent adenine, 5-methylcytosine, guanine and thymine nucleobases. Hybridization of MAPT Rx to the cognate pre-mRNA via Watson and Crick base pairing results in ribonuclease H1-mediated degradation of the MAPT pre-mRNA, thus selectively preventing production of the tau protein 61 . Dose selection was guided by a preclinical model in mouse and monkey relating dose level to reduction in MAPT mRNA (model described by Tabrizi et al. 31 ).

Study oversight
The trial was conducted in accordance with the Declaration of Helsinki. The trial protocol and all documentation were approved by the institutional review board or independent ethics committee at each investigational site (for list of ethics committees approving the study, see Supplementary Information). All participants provided written informed consent. The trial was sponsored by Ionis Pharmaceuticals, which provided the study drug (MAPT Rx and placebo). Personnel from Ionis Pharmaceuticals designed the trial in conjunction with collaborators from Biogen, academic investigators and other disease experts. A Formal Safety Monitoring Group, composed of sponsor personnel with medical and clinical trial expertise and independent from the conduct of the study, authorized each dose escalation after unblinded review of safety data. The investigators collected the data, which were held and maintained by the sponsor. Data were analyzed by personnel from the sponsor and were interpreted by all authors. The investigators vouch for the fidelity of the trial to the protocol and protocol amendments. The authors vouch for the completeness and accuracy of the data. The authors and sponsor made the decision to submit the manuscript for publication.

Measurement of brain volumes
We obtained three-dimensional T1-weighted structural MRI scans of the head and transferred these data, blinded to trial-group status, to an independent image-analysis provider that performed quality control, processing and volumetric analyses according to established methods. Whole-brain and regional volume changes were calculated using a validated pipeline implemented in VivoQuantTM, composed of a preprocessing module and a multi-atlas segmentation module, followed by visual inspection and manual editing if needed 62 .

Statistical analysis
The primary objective of the trial was the evaluation of the safety of MAPT Rx treatment. Safety data were summarized according to treatment group. Quantitative assessments were summarized using descriptive statistics including number of patients, mean, median, standard deviation, standard error of mean, interquartile range (25th percentile, 75th percentile) and range (minimum, maximum). Qualitative assessments were summarized using frequency counts and percentages. All safety analyses were performed in the safety population (all randomized participants receiving at least one dose of study drug). PK parameters were assessed for MAPT Rx in CSF and plasma. Analyses of PD biomarkers and exploratory and clinical endpoints were summarized according to treatment group, and MAPT Rx -treated groups were compared with the pooled placebo group. While there is no statistical rationale for the sample size, it has been selected on the basis of prior experience with generation 2.0 ASOs 31 given by IT bolus injection to ensure that the safety, tolerability, PK and exploratory pharmacodynamics will be adequately assessed while minimizing unnecessary patient exposure. Statistical analyses were performed using Statistical Analysis System (SAS) Version 9.4.

Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability
To request access to data, please visit Vivli. The individual participant data collected during the trial and that support the research proposal Nature Medicine Article https://doi.org/10.1038/s41591-023-02326-3 will be available to qualified scientific researchers, in accordance with Biogen's Clinical Trial Transparency and Data Sharing Policy on www. biogentrialtransparency.com. Data requests are initially reviewed by Vivli and Biogen for completeness and other parameters (relating to scope and meeting sponsor policies) and are then reviewed by an Independent Review Panel. Deidentified data, study protocol and documents will be shared under agreements that further protect against participant reidentification, and data are provided in a secure research environment further protecting participant privacy.