MINDSET Failure Closes 5-HT6 Adjunct Case After Large Precise Null Result

Vivek Ramaswamy's Biotech Past and Failed Alzheimer's Drug Trial

The article details how Vivek Ramaswamy's biotech background, particularly the failure of Axovant Sciences' Alzheimer's drug intepirdine, became a political liability. Axovant acquired intepirdine from GlaxoSmithKline for $5 million upfront and later raised $362.3 million gross in an IPO. Despite earlier mixed Phase 2 results, Axovant proceeded with the pivotal Phase 3 Mindset trial involving 1,315 patients. In September 2017, the trial failed to meet both primary endpoints for cognition and daily functioning, causing the stock to collapse. This event, along with Ramaswamy's public statements about the drug, drew scrutiny during his political campaigns, contrasting with Roivant's later successes under new leadership.

• Axovant Sciences acquired intepirdine from GlaxoSmithKline for an upfront payment of $5 million in December 2014. This occurred despite GSK's prior two large Phase 2 studies (576 and 684 patients) failing to meet overall efficacy goals, though some findings in combination with donepezil showed promise. Axovant subsequently raised $362.3 million gross ($334.5 million net) through an IPO in 2015, valuing the unproven drug highly based on financial terms and investor relationships.
• In October 2015, Axovant initiated its pivotal 1,315-patient Phase 3 Mindset trial, testing intepirdine in combination with donepezil for Alzheimer's disease. However, on September 26, 2017, Axovant announced that the trial missed both its primary endpoints, showing no statistically significant improvement in cognition or daily functioning. This clinical failure led to a significant collapse in Axovant's stock price, highlighting the disconnect between financial valuation and clinical evidence.
• Vivek Ramaswamy's role in Axovant and the intepirdine failure became a central point of criticism during his political campaigns, with opponents questioning his "successful biotech entrepreneur" label. While Roivant, the parent company, later achieved FDA approvals for six medicines and completed major deals like the $7.1 billion sale of Telavant to Roche, Ramaswamy's public statements about intepirdine and his later critiques of the FDA contrasted sharply with the company's data-driven approach under new leadership.

The Enduring Challenges in Alzheimer's Disease Treatment

Despite effective β-amyloid clearance, anti-amyloid monoclonal antibodies show only modest slowing of cognitive decline that often fails to reach the minimal clinically important difference (MCID) compared to placebo. These therapies are further complicated by significant safety signals and measurement limitations that risk undervaluing benefit in early disease while complicating risk-benefit assessment.

• Modest clinical efficacy versus robust biomarker effect: Donanemab modestly reduced cognitive decline versus placebo with ADAS-Cog13 (MD, -1.86), CDR-SB (MD, -0.36), MMSE (MD, 0.64), and iADRS (MD, 3.19), and slowed disease progression including a 40.3% (top < 0.0001) lower risk of progression to the next stage as measured by the Clinical Dementia Rating-Global score, with amyloid clearance in 80.8% of donanemab-treated participants and in 144 (57.6%) of patients tested positive for amyloid on positron emission tomography (PET) scans in a comparative cohort. The reported cognitive decline slowing is modest and often below the thresholds for clinically significant differences on outcome scales, and the clinical significance remains uncertain.

• Insensitive outcome measures in early disease: The ADAS-Cog, the most commonly used cognitive endpoint, suffers from a substantial ceiling effect, impairing its ability to capture cognitive decline over short durations in prodromal populations. Likewise, functional scales such as the CDR-SB are inherently insensitive in mild cognitive impairment (MCI), as functional independence is, by definition, preserved, and use of composite multidomain scales with high baseline scores may mask domain-specific improvements, further limiting a drug's capacity to reach MCID thresholds.

• High burden of amyloid-related imaging abnormalities (ARIAs) and infusion reactions: Anti-amyloid-β therapy significantly increased odds of ARIA-E (OR 7.93; 95% CI 4.50-13.98) and ARIA-H (OR 1.87; 95% CI 1.28-2.72) versus placebo, with pooled ARIA-E prevalence of 6.8% and ARIA-H of 15.8%. By 76 weeks, 13/825 (1.6%) and 170/816 (20.8%) participants in the placebo and donanemab arms experienced amyloid-related imaging abnormalities (ARIA)-edema/effusion, 96/825 (11.6%) and 218/816 (26.7%) participants experienced ARIA-microhemorrhages/superficial siderosis and 3/825 (0.4%) and 67/816 (8.2%) participants experienced infusion-related reactions, respectively, with risk of ARIA-E markedly increased (RR, 12.39), infusion reactions (RR, 11.90) and discontinuations (RR, 3.22) more frequent. The highest ARIA-E risk was seen with donanemab and aducanumab, followed by gantenerumab and lecanemab, and ARIA-H incidence ranked donanemab > aducanumab > bapineuzumab > lecanemab > gantenerumab > crenezumab > solanezumab.

• APOE ε4 genotype-driven risk and monitoring constraints: APOE ε4 carriage significantly elevated ARIA-E (OR 2.28) and ARIA-H (OR 2.07) risk, with a clear gene-dose effect in homozygotes, and licensing of donanemab in the United Kingdom (UK) for early symptomatic Alzheimer's disease is limited to adults who are Apolipoprotein E (APOE) ε4 non-carriers or heterozygotes. Excluding APOE ε4 homozygotes reduced, but did not eliminate, ARIA risk with donanemab, and risk of ARIA-E is higher at treatment initiation, in ApoE4 carriers, with higher dosage, and with >4 of microhemorrhages on a baseline MRI, requiring structured risk-benefit discussion and appropriate monitoring including brain MRI before initiation, before each dose escalation, and with any new neurologic sign or symptom.

• Unresolved long-term safety and late-stage management uncertainty: Evidence suggests treatment with anti-amyloid antibodies may lead to more brain tissue loss than placebo treatments, with long-term impact of ARIAs unclear and reliable data on therapy-related mortality scarce due to insufficient access to detailed clinical and neuropathological information. For cholinesterase inhibitors and N-methyl-D-aspartic acid (NMDA) receptor antagonists, efficacy in severe dementia and end-stage dementia remains uncertain, with 80% of hospice medical directors recommending discontinuing these therapies at hospice enrollment while 72% reported families experienced difficulty stopping, and a subset observed accelerated cognitive (30%) and functional decline (26%) or emergence of challenging behaviors (32%) with discontinuation.

Intepirdine's Mindset Trial: Design and Disappointing Outcomes

Recent Alzheimer's disease trials span early to mild-to-moderate populations and evaluate amyloid-targeted antibodies, small molecules and repurposed agents using clinical, biomarker and imaging endpoints. Designs range from 12-week proof-of-concept studies to 18-month core studies with long-term extensions up to 48 months.

Lessons from Intepirdine's Failure for AD Drug Development

The treatment landscape has shifted from symptomatic therapies alone toward amyloid-directed disease-modifying approaches. Four medications were available as symptomatic therapies: three cholinesterase inhibitors and memantine, with acetylcholinesterase inhibitors including rivastigmine, galantamine, and donepezil. In June 2021, aducanumab was approved in the United States under an accelerated approval pathway as the first novel putative disease-modifying therapy targeting the β-amyloid cascade, followed by development and approval of other anti-amyloid immunotherapies including lecanemab, an anti-amyloid beta protofibril antibody introduced in China in 2024, and donanemab, a monoclonal antibody targeting amyloid-β plaques. An investigational active immunotherapy, ABvac40 targeting Aβ40, was evaluated in AB1601, a multicenter, randomized, double-blind, placebo-controlled phase 2 study in 124 patients with amnestic mild cognitive impairment or very mild Alzheimer's disease receiving five monthly injections plus a 10-month booster.

Published trial data show that monoclonal antibodies attenuate cognitive decline but by magnitudes below thresholds for clinical meaningfulness. Across seven monoclonal antibody trials with 8010 participants (mean age: 71.5 years) and nine acetylcholinesterase inhibitor trials with 4993 participants (mean age: 70.7 years), only monoclonal antibodies, not acetylcholinesterase inhibitors, were associated with slower progression versus placebo on Clinical Dementia Rating Scale-Sum of Boxes (mean difference -0.41 (95 % credible interval -0.61 to -0.22); minimally important difference -1) and Alzheimer's Disease Assessment Scale-cognitive subscale 14-item (-1.35 (-2.36 to -0.36), MID -2). Monoclonal antibodies were associated with slower progression than acetylcholinesterase inhibitors on Clinical Dementia Rating Scale-Sum of Boxes (-0.30 (-0.60 to -0.001)). In three donanemab trials (n = 2054), donanemab modestly reduced decline on ADAS-Cog13 (MD, -1.86), CDR-SB (MD, -0.36), MMSE (MD, 0.64), and iADRS (MD, 3.19). Lecanemab reduced amyloid burden with 81% of patients achieving amyloid-negative PET scans in the Clarity AD trial and showed a 27% reduction in cognitive decline on the ADCOMS scale, while donanemab demonstrated 76% plaque clearance and a 35% slowing in cognitive decline in early Alzheimer's disease patients. In a prospective multicenter study across 21 sites, 261 Alzheimer's disease patients receiving biweekly lecanemab (10 mg/kg) showed attenuated decline versus a matched Alzheimer's Disease Neuroimaging Initiative cohort, with mean -22.1 Centiloid reduction and 29.2% turning amyloid-negative.

Safety profiles are dominated by amyloid-related imaging abnormalities and infusion-related reactions, with risk modulated by apolipoprotein E ε4 status. In Clarity AD, the most common adverse events in the lecanemab group (> 10%) were infusion-related reactions (24.5%), ARIA with hemosiderin deposits microhemorrhages (16.0%), COVID-19 (14.7%), ARIA with edema (13.6%), and headache (10.3%), with ARIA-edema more common in ApoE e4 carriers (16.8%) and most common in ApoE ε4 homozygous participants (34.5%). Pooled analyses reported ARIA incidence of 19% (95% CI: 16%-23%) with relative risk 1.45 for heterozygotes and 3.54 for homozygotes versus noncarriers, and APOE4 homozygotes carriers were associated with a 5.53-fold (2.48 to 13.07) increased odds of developing ARIA-edema compared to non-carriers. Donanemab increased risk of ARIA-edema (RR, 12.39), infusion reactions (RR, 11.90), and discontinuations (RR, 3.22). In Chinese real-world experience, infusion reactions occurred in 11.1% and amyloid-related imaging abnormalities in 9.2% (1.6% symptomatic), while ABvac40 showed treatment-emergent adverse events comparable to placebo (90.6% and 26.6% serious versus 93.3% and 26.7%), ARIA-hemorrhage of 12.5% versus 15.0%, with no ARIA-edema or meningoencephalomyelitis.