Background

Gantenerumab is a fully human IgG1 antibody designed to bind with subnanomolar affinity to a conformational epitope on Aβ fibrils. It encompasses both N-terminal and central amino acids of Aβ. The therapeutic rationale for this antibody is that it acts centrally to disassemble and degrade amyloid plaques by recruiting microglia and activating phagocytosis. Gantenerumab preferentially interacts with aggregated brain Aβ, both parenchymal and vascular. It binds to cerebral amyloid angiopathy fibrils more avidly than some other Aβ antibodies (Soderberg et al., 2024). The antibody elicits phagocytosis of human Aβ deposits in AD brain slices co-cultured with human macrophages. It also neutralizes oligomeric Aβ42-mediated inhibitory effects on long-term potentiation in rat brains. In APP/PS-1 transgenic mice, gantenerumab binds to cerebral Aβ, reduces small plaques by recruiting microglia, and prevents new plaque formation. Gantenerumab does not alter plasma Aβ (Bohrmann et al., 2012). It has been studied as a potential combination therapy with the Roche BACE inhibitor RG7129 in mouse models of Aβ amyloidosis (Apr 2013 conference news).

Findings

Four Phase 1 trials conducted internationally in a total of 308 patients have tested the safety, tolerability, pharmacokinetics, and pharmacodynamics of single and multiple doses of infused and subcutaneous gantenerumab in healthy controls and people with Alzheimer's disease, respectively. Gantenerumab was generally safe and well-tolerated, but amyloid-related imaging abnormalities (ARIA) are a concern. For example, in one published study, two of six patients in the highest-dose group had focal areas of inflammation or vasogenic edema on MRI scans in brain areas with the most amyloid reduction (Ostrowitzki et al., 2012). The imaging finding was transient, but ARIA are being monitored closely with MRI in subsequent trials. One dosing study with 16 participants indicated an 11 percent difference in the amount of amyloid change between placebo and the higher dose at six months.

In 2010, Roche started a Phase 2 trial of 105 or 225 mg gantenerumab injected subcutaneously once a month into 360 participants, and in 2012 expanded the study to a Phase 2/3 registration trial of 799 people. Called SCarlet RoAD, this multinational, 159-center study of gantenerumab's effect on cognition and function in prodromal Alzheimer's disease delivered treatment for two years with the option of a two-year extension. Co-primary endpoints were to be the Clinical Dementia Rating Sum of Boxes (CDR-SOB) and change in brain amyloid levels as measured by PET. The latter was dropped mid-study, and instead the trial included a PET sub-study of 90 participants. SCarlet RoAD enrolled people aged 50 and older whose memory function tested below normal on the Free and Cued Selective Reminding Test (FCSRT-IR), whose MMSE was 24 or above, whose CDR was 0.5, and who were positive on amyloid PET. This was one of first times the new diagnostic criteria by the International Working Group were applied in a large clinical trial. At the November 2014 CTAD conference, data were reported to suggest that this screening process worked to enroll a homogenous population of early symptomatic patients whose memory deficit was likely due to underlying Alzheimer's pathology (Dec 2014 conference news).

On December 19, 2014, Roche stopped dosing in SCarlet RoAD based on an interim futility analysis (Dec 2014 news). In July 2015, Roche reported no efficacy on primary or secondary endpoints in this trial, but a trend toward a benefit in the fastest progressors based on post hoc analysis (Aug 2015 conference news). Subsequent reports noted biomarker and efficacy signals on the high dose among fast progressors, and the data was formally published (Nov 2015 conference news; Ostrowitzki et al., 2017). SCarlet RoAD participants who entered the open-label extension study were titrated up to 1,200 mg subcutaneous gantenerumab. The slower this titration, the less ARIA-E they experienced (Dec 2017 conference news). The open-label extension was completed in September 2020.

In March 2014, Roche started a Phase 3 trial of monthly subcutaneous injections of gantenerumab in what was anticipated to be 1,000 patients with clinical diagnoses of mild AD. This trial, called Marguerite RoAD, used the ADAS-Cog and ADCD-ADL as co-primary and various biomarkers and clinical/neuropsychiatric measures as secondary outcomes. On October 10, 2016, the clinicaltrials.gov entry for this study was updated to reflect that this trial had stopped enrolling at 389 participants. Roche later reported that, like SCarlet RoAD, Marguerite RoAD had failed an interim futility analysis. It was switched to an open-label extension study, with participants titrated up to 1,200 mg gantenerumab.

At the 2018 AAIC, Roche reported that two years of high-dose gantenerumab in the SCarlet and Marguerite RoAD extension studies lowered brain amyloid by an average of 59 centiloid on florbetapir PET, with half of the 28 participants who reached this timepoint falling below the threshold for amyloid positivity, and the rest on trajectory to do so. About one-third of participants in the extension studies developed ARIA-E; the majority were asymptomatic (Aug 2018 conference news; Klein et al., 2019). In a subsequent paper, the company reported continued amyloid reductions in the third year of the extension. Of 30 participants with three-year PET scans, 80 percent reduced their amyloid load below the positivity threshold (Klein et al., 2021). At the 2022 AAIC, biomarker data was presented. Consistent with other trials (see DIAN-TU results below), treatment increased plasma Aβ40 and Aβ42, and decreased p-tau217 and p-tau181. Compared to historical controls, the OLE cohort declined more slowly on the CSDR-SB, ADAS-Cog13, and MMSE (Aug 2022 conference news).

Gantenerumab, in parallel with Eli Lilly's solanezumab, was evaluated by the Dominantly Inherited Alzheimer Network Trials Unit (DIAN-TU) in a Phase 2/3 trial aimed at preventing dementia in 210 people who are on the path to Alzheimer’s disease due to an inherited autosomal-dominant mutation in APP, presenilin-1, or presenilin-2. This trial began in December 2012 as a two-year, Phase 2 biomarker study (Oct 2012 news). It later converted into a Phase 3 registration trial with a cognitive endpoint measured after four years of treatment. Gantenerumab dosing was increased fivefold, from 225 mg to 1,200 mg monthly, midway through the trial. This trial, completed in November 2019, missed its primary endpoint. Gantenerumab produced no significant treatment-related change on the DIAN Multivariate Cognitive Endpoint, a composite developed for this trial (Feb 2020 news). 

Additional data were presented at the 2020 AAT-AD/PD Focus Meeting. Of 194 participants, 52 received gantenerumab and 39 completed the trial. Gantenerumab treatment led to a large reduction in amyloid plaque as per PiB PET scans, and it normalized CSF Aβ42. Gantenerumab reduced toward normal elevated levels of CSF total tau and p-tau181, and slowed the rise of CSF neurofilament light (Apr 2020 conference news; Apr 2020 conference news). Analysis of the cognitive data revealed that people who were symptomatic at baseline declined over the course of the study, while those who were asymptomatic remained stable. This dichotomy lessened the power of the study to detect treatment effects. Peer-reviewed results were published (Jun 2021 news; Salloway et al., 2021, Joseph-Mathurin et al., 2022; Wang et al., 2022). In a subsequent biomarker analysis, treatment improved markers of synaptic function and inflammation, with greater benefits observed in presymptomatic participants (Wagemann et al., 2024).

Based on the biomarker data, DIAN-TU and Roche invited participants from all arms of this trial to an open-label extension with high-dose gantenerumab (1,200 mg monthly) for up to three years. The dose was further increased to 3,000 mg monthly for the last two years. This OLE started in the fall of 2020, and finished in August 2023. According to preliminary results presented at CTAD that year, asymptomatic mutation carriers who took gantenerumab for eight years were half as likely to progress to symptoms, or developed symptoms six years later than their expected age of onset (Nov 2023 conference news). More data was presented at the 2024 AAIC, comparing 22 initially asymptomatic mutation carriers to 74 untreated people. In some participants, gantenerumab treatment slowed or stopped tau tangle accumulation; in others, it improved mild memory symptoms. A few participants remained cognitively normal up to 10 years after their expected age of onset (Aug 2024 conference news; Aug 2024 conference news). Despite the use of high doses, ARIA incidence was similar to other trials (Aug 2024 conference news).

In 2016, Roche started two new Phase 1 trials, both investigating subcutaneous administration of higher doses of gantenerumab, in a total of 98 healthy participants. According to published results, the procedure was safe and injection pain tolerable, opening the possibility of at-home administration (Portron et al., 2020).

On March 6, 2017, MorphoSys, which partners in the development of gantenerumab, announced Roche would start two new Phase 3 trials for prodromal or mild, amyloid-confirmed AD. Graduate 1 and 2 began enrolling in June 2018, each at 216 different sites worldwide, with a goal of 760 participants each. Participants were titrated up to 1,020 mg subcutaneous gantenerumab, given as 510 mg every two weeks, or placebo and treated for two years, with an option to continue on open-label. The primary outcome measure was change on the CDR-SB. In March 2020, target enrollment for each trial was increased to 1,016. According to baseline data presented at the November 2021 CTAD, enrollment was complete at 1,966, with data readout expected in late 2022 (conference news).

At the 2019 AD/PD conference in Lisbon, Portugal, Roche showed further analyses of the two-year open-label extension of the SCarlet and Marguerite RoAD studies. It indicated that ARIA tends to occur in hotspots of rapid amyloid removal, but is not required for amyloid clearance across the brain. Roche also showed how comparing titration schemes in the combined open-label study informed the Graduate trial design of a single uptitration scheme independent of participants' ApoE genotype (May 2019 news).

In May 2020, the company began a second open-label continuation for participants in the Scarlet and Marguerite RoAD studies. The 116 participants receive antibody every four weeks, at the dose they were on previously. At the same time, Roche registered a rollover study for people who completed either the double-blind or open-label phases of Graduate 1 and 2. In this open-label continuation, an expected 2,032 participants will receive 510 mg of gantenerumab every two weeks. The trial started enrolling in February 2021 and was to run through 2024.

In November 2020, Roche began a Phase 2 trial testing once-weekly subcutaneous gantenerumab injections in 192 people with prodromal to mild AD. Starting with monthly shots of 120 mg, participants in this two-year study ramped up to a target dose of 255 mg weekly. Administration was at home by a care partner or in the doctor’s office by a health care professional. The primary endpoint was change in amyloid deposition from baseline; secondary outcomes included safety, pharmacokinetics, and participant experience with home injection. Roche ended the trial in January 2023, when the company discontinued development of gantenerumab. At that time, 149 patients had one year PET data; 12 had two-year scans. According to published results, amyloid reduction averaged 26 and 35 centiloids after one or two years, respectively. Most care partners found home administration easy and convenient, and no new safety issues were identified (Boess et al., 2024).

In October 2021, the FDA designated subcutaneous gantenerumab a Breakthrough Therapy, offering an accelerated review and approval process. 

In March 2022, Roche began SKYLINE, a Phase 3 secondary prevention trial to evaluate gantenerumab in cognitively unimpaired people with CSF or PET evidence of brain amyloid. It was to enroll 1,200 participants between 60 and 80 years old, and test four years of gantenerumab or placebo against a primary outcome of the Preclinical Alzheimer’s Cognitive Composite-5 (Mar 2022 conference news). Participants were to receive a dose of 255 mg weekly, or 510 mg every other week, and to have the option to self-administer the antibody at home. Secondary outcomes included progress to cognitive impairment, functional measures, structural MRI, blood biomarkers of amyloid, tau, and neurodegeneration, and safety. PET and CSF substudies were planned. Starting in six locations in the U.S., the trial was to run in 17 countries through 2028.

On November 14, 2022, Roche and Genentech announced that gantenerumab failed to slow cognitive decline on the CDR-SB in the Graduate trials (Nov 2022 news). At the December 2022 CTAD conference, Roche presented more results on the trial, which enrolled 1,965 participants from 30 countries (Dec 2022 conference news). The data were subsequently published (Bateman et al., 2023). Gantenerumab cleared only half as much plaque as expected, and fewer participants became amyloid-negative on PET scans than in previous trials. Clinical measures trended to improvement, but fell short of statistical significance. People who cleared amyloid below the positivity threshold did the best clinically. Biomarkers of tau, synaptic loss and neuronal death dropped or slowed their increase significantly on antibody compared to placebo. The company announced it had stopped all gantenerumab trials including SKYLINE, but said it would continue to provide antibody for the ongoing DIAN-TU open-label extension study. For a review of gantenerumab's development history, see Bateman et al., 2022.

In December 2021, DIAN-TU investigators had launched a primary prevention trial testing gantenerumab in people as young as 18 with familial AD mutations but no or little brain amyloid. The study plans to enroll approximately 220 participants for four years and measure amyloid accumulation. An open-label extension will run for an additional four years, with a primary endpoint of odds of lower disease progression based on modeling of the change in six CSF and MRI biomarkers, compared to an observational control group and the placebo group from the previous gantenerumab study. After Roche stopped gantenerumab development, DIAN-TU investigators announced that they will choose a different drug for this trial (20 Dec 2022 release). In May 2024, the trial was relaunched with remternetug (press release).

Roche is developing a new formulation of gantenerumab that uses “brain shuttle” technology to increase the antibody’s ability to enter the brain (RO7126209).

New analyses using gantenerumab trial data continue to be published, for example to develop quantitative disease models, validate ARIA rating scales, to compare radiotracer response in drug trials, and to develop new bioassays (Mazer et al., 2022; Bracoud et al., 2023; Chen et al., 2023; Wang et al., 2023). DIAN-TU data was used to validate amyloid reduction as a surrogate endpoint for clinical improvement (Wang et al., 2024).

For all gantenerumab trials, see clinicaltrials.gov.

Clinical Trial Timeline