spec sheet10 rows
Milameline was Parke-Davis and Roussel-Uclaf's shot at treating Alzheimer's disease by stimulating muscarinic receptors directly instead of propping up acetylcholine with a cholinesterase inhibitor. It is a partial agonist with roughly equal affinity at all five muscarinic subtypes, which was the deliberate choice: no subtype selectivity, just enough intrinsic activity to signal without saturating the system. The preclinical package was strong for its era. It reversed spatial memory deficits in rats with lesioned forebrain cholinergic neurons, raised cortical blood flow, desynchronised the EEG in both rats and rhesus monkeys in the pattern that reads as increased arousal, and rescued scopolamine-impaired attention in monkeys. Then the dose window closed. In Alzheimer's patients, 1 mg every six hours was fine, 2 mg was tolerated by most people, and 2.5 to 3 mg brought sweating, hypersalivation, nausea, diarrhoea, hypotension and, notably, parkinsonian signs including cogwheeling, tremor and a shuffling gait; that study was stopped after the fourth 3 mg dose. The pivotal 52-week trial across 38 centres was then terminated early when an interim analysis projected it would not work. It never worked, and it was never approved.
- cognition-activating effects in preclinical and primate models
- combined with tacrine to reverse scopolamine-induced performance deficits in primate studies
- early example of using MRI progression markers alongside cognitive endpoints in an Alzheimer's trial
- reversed scopolamine-induced deficits in water-maze performance in rodents
- increased cortical blood flow and improved basal-forebrain-lesion-impaired maze performance in animal studies
- produced measurable cerebral blood-flow changes during a verbal recognition task in early Alzheimer's patients
- Reversed spatial memory deficits in rats with lesions of forebrain cholinergic neurons
- Reversed scopolamine-induced attention impairment in rhesus monkeys, alone and combined with tacrine
- Produced EEG desynchronisation consistent with increased cortical arousal in rats and monkeys
- Increased cortical blood flow in rats
- Clean pharmacology: no cholinesterase inhibition and no meaningful activity at about thirty other neurotransmitter sites
- Reversed scopolamine-induced memory and attention deficits in rats and rhesus monkeys
- Improved water maze performance in rats with lesioned forebrain cholinergic neurons
- Raised cortical blood flow and shifted the electroencephalogram toward an aroused pattern in rats and monkeys
- Broadened the effective dose range when combined with tacrine in monkeys without adding side effects
- cholinergic side effects typical of muscarinic agonists (GI upset, sweating)
- increased gastrointestinal motility, a classic peripheral muscarinic side effect
- reduced spontaneous locomotor activity at higher doses
Overview
The plainest illustration in the whole muscarinic story of why the orthosteric approach could not work. A partial agonist with equal affinity at all five subtypes has no way to reach a central effect without a peripheral one, and the clinical record says exactly that.
- Milameline's 2003 Alzheimer's trial is remembered almost as much for its method as its drug effect; it was an early example of pairing serial MRI scans with cognitive testing to track disease progression during a drug trial.
- Milameline carried two separate development codes, CI-979 from Warner-Lambert/Parke-Davis and RU 35926 from a co-development link to Roussel-Uclaf, a common sign of cross-company Alzheimer's research partnerships in the 1990s.
- The failed milameline trial produced one of the most influential datasets in Alzheimer's research. Its MRI arm was continued after the drug arm was stopped, and it showed that measuring hippocampal volume needed about 21 patients per group to detect a treatment effect where the standard cognitive scale needed 320.
- Roussel-Uclaf and Parke-Davis both filed their own code numbers for the same molecule, which is why it turns up in the literature as CI-979, PD-129409 and RU-35926 depending on which company's paper you are reading.
- The most cited paper to come out of milameline's development is not about milameline. Its 52-week trial was halted for projected futility, but the imaging arm was allowed to run on and became a landmark demonstration that hippocampal volume detects a treatment effect with about 21 patients per arm where the standard cognitive scale needs 320.
- A muscarinic agonist produced parkinsonian signs; cogwheeling, tremor and a shuffling gait appeared at the top doses tested in Alzheimer's patients, a reminder that striatal muscarinic tone opposes dopamine.
Mechanism
Milameline is a partial with essentially no subtype selectivity; it binds in the nanomolar range against an agonist radioligand and the micromolar range against radioligands, with roughly equal affinity at human M1 through M5 [1]. In transfected cells it does what a does at every subtype: it drives phosphatidylinositol hydrolysis at M1 and M3, inhibits forskolin-stimulated accumulation at M2 and M4, and it reduces potassium-stimulated release from rat cortical slices, which is presynaptic M2 autoreceptor activation working directly against the drug's own purpose [1].
That -to- radioligand gap is the classic partial agonist fingerprint, and in a microphysiometry comparison across the five cloned human subtypes milameline showed a slight potency preference for M2 over M1, which is the wrong way round for a cognition drug [8]. The trouble showed up in native tissue: it did not measurably stimulate phosphoinositide hydrolysis in rat cortical slices at concentrations up to 100 micromolar [9], nor in mouse brain in vivo, where it instead produced the full non-selective cholinergic effect profile [10]. Central effects in rats and monkeys appeared only slightly below the doses that stimulated peripheral cholinergic receptors, and that narrow window is what eventually closed on it in patients [1].
receptor fingerprint
M1 (CHRM1)Partial agonist
M2 (CHRM2)Agonist
M3 (CHRM3)Agonist
M4 (CHRM4)Agonist
M5 (CHRM5)Agonist
receptors, rat cerebral Partial agonist binding signature
Dosingtypical ranges, not medical advice
interested in protocols and clinical dosages? make an account to see them! ^_^
Safetyrisks and cautions, not medical advice
The dose-limiting toxicity was straightforward cholinergic overflow plus something less expected. In a rising-dose study in ten men with Alzheimer's disease, 1 mg every six hours was tolerated by everyone and 2 mg every six hours by most, but 2.5 and 3 mg brought diaphoresis, hypersalivation, nausea, diarrhoea, hypotension, chills, headache, flatulence and urinary frequency, together with signs suggestive of parkinsonism including cogwheeling, tremor, pill rolling, posturing and a shuffling gait; five of the eight men on drug stopped for adverse events and the study was terminated after the fourth 3 mg dose [7].
In rats, sustained plasma concentrations produced corneal opacities and urinary tract disease, including hydronephrosis, pyelonephritis and inflammation and necrosis of the kidney, bladder, urethra and urinary papilla; both the corneal and the urinary findings were prevented by scopolamine, which established them as consequences of muscarinic stimulation rather than of off-target chemistry [7]. In healthy volunteers, single doses of 0.002 to 1 mg were well tolerated and cholinergic symptoms such as hypersalivation and sweating appeared at 2 to 4 mg [5].
History
Developed at Parke-Davis, part of Warner-Lambert, with Roussel-Uclaf, which is why it carries three code names: CI-979, PD-129409 and RU-35926. Preclinical and phase 1 characterisation was published in 1995, showing linear pharmacokinetics from 0.1 to 4 mg, an elimination half-life of two to five hours and extensive metabolism, with cholinergic symptoms appearing at 2 to 4 mg in healthy volunteers.
A rising-dose tolerability study in Alzheimer's patients that same year set the ceiling at about 2 mg every six hours and was halted early for adverse events. Monkey work in 1999 showed milameline combined with tacrine reversed a scopolamine-induced attention deficit, and the full biochemical and in vivo profile was published in the Journal of Pharmacology and Experimental Therapeutics that year. The 52-week phase 3 trial was discontinued mid-course after an interim analysis projected a lack of efficacy. Development ended there.
Reputation
Remembered less for what it did than for what it demonstrated: that a partial agonist with equal affinity at all five muscarinic subtypes cannot be dosed to a central effect without a peripheral one. Its afterlife is mostly methodological, through the imaging substudy of its failed trial, and pharmacological, as the standard example of the orthosteric approach that the M1 modulator programmes were built to replace.
Subjective profileweighing the evidence above
A textbook version of how the muscarinic agonist era ended: good animal data, a real mechanism, and a therapeutic window so narrow that the peripheral side effects arrived almost as soon as the central ones did. The trial that killed it turned out to be more valuable than the drug.
Resources
This entry is here for reference.
Research
- 1995first citedSafety and tolerability of CI-979 in patients with Alzheimer's disease.
- 1999most active year4 papers
- 2003controlled trialMRI as a biomarker of disease progression in a therapeutic trial of milameline for AD
- 2006most recentCombined cerebral blood flow effects of a cholinergic agonist (milameline) and a verbal recogni…
- 1.Milameline (CI-979/RU35926): a muscarinic receptor agonist with cognition-activating properties: biochemical and in vivo characterization
- 2.MRI as a biomarker of disease progression in a therapeutic trial of milameline for AD
- 3.Combined cerebral blood flow effects of a cholinergic agonist (milameline) and a verbal recognition task in early Alzheimer's disease
- 4.Safety and tolerability of CI-979 in patients with Alzheimer's disease.
- 5.Preclinical and phase 1 clinical characterization of CI-979/RU35926, a novel muscarinic agonist for the treatment of Alzheimer's disease.
- 6.Combining tacrine with milameline reverses a scopolamine-induced impairment of continuous performance in rhesus monkeys.
- 7.Toxicological comparison of a muscarinic agonist given to rats by gavage or in the diet
- 8.Functional comparison of muscarinic partial agonists at muscarinic receptor subtypes hM1, hM2, hM3, hM4 and hM5 using microphysiometry
- 9.In vitro characterisation of the muscarinic receptor partial agonist, sabcomeline, in rat cortical and heart membranes
- 10.Xanomeline compared to other muscarinic agents on stimulation of phosphoinositide hydrolysis in vivo and other cholinomimetic effects
- 11.Antagonism of scopolamine-induced memory impairments in rats by the muscarinic agonist RU 35,926 (CI-979)
11 listed here; entry last updated August 2026
Reviews
My notesprivate to this device
FAQ
Why does milameline have three different names?
It was studied under a Parke-Davis code (CI-979/PD 151832) and a parallel Roussel Uclaf number (RU 35926) reflecting the era's co-development arrangements between pharma companies.
Is milameline still being researched?
No. Development stopped after Phase II Alzheimer's trials failed to show a clear benefit; it is a historical compound only.
Why did milameline stop at broad receptor activity instead of being selective?
It was designed as a nonselective partial agonist across M1-M4 receptors; that broad activity gave reliable cholinergic effects in animal models but also brought along peripheral side effects like increased GI motility that selective M1 agonists were meant to avoid.
Did milameline ever show benefit in Alzheimer's patients?
Small imaging studies showed it altered cerebral blood flow during a memory task in early Alzheimer's disease, but development was discontinued before larger efficacy trials were completed.
Did milameline ever help anyone with Alzheimer's disease?
No completed trial showed a benefit. The pivotal 52-week study was stopped part way through because an interim analysis projected it would not reach efficacy, so the drug never produced a positive clinical result.
Why did it cause parkinsonian symptoms?
Muscarinic and dopaminergic systems balance each other in the striatum, and pushing muscarinic tone up across all subtypes tips that balance the same direction Parkinson's disease does. The rigidity, tremor and gait changes seen at higher doses were a predictable consequence of non-selective muscarinic agonism.
How is milameline different from alvameline or xanomeline?
Milameline deliberately hit all five muscarinic subtypes about equally. Alvameline and xanomeline both tried to be selective, aiming at M1 and sparing the subtypes responsible for peripheral effects. All three failed in Alzheimer's disease, though xanomeline's M1 and M4 profile later found a second life in schizophrenia.
Why did milameline fail?
On two counts. Tolerability closed the dose window first: cholinergic effects and parkinsonian signs were dose-limiting at 2.5 to 3 mg every six hours, so efficacy trials were capped at 2 mg. Then the pivotal 52-week trial was stopped early because an interim analysis projected that it would not show a benefit at the doses that could be given. The underlying reason is in the pharmacology; with roughly equal affinity at all five muscarinic subtypes, there is no dose that stimulates M1 in the cortex without also stimulating M2 in the heart and M3 in the gut and sweat glands.
Was milameline selective for M1?
No, and it was never meant to be. Its affinities at human M1 through M5 are approximately equal, and the design bet was that partial agonism alone would create a therapeutic window. It did not. In a comparison across the five cloned human subtypes it even showed a slight potency preference for M2 over M1, the presynaptic receptor whose activation reduces acetylcholine release.
Is milameline available anywhere?
No. It was never approved and development ended after the pivotal trial was halted for projected lack of efficacy.
How does it compare with TAK-071 or the other M1 PAMs?
Milameline is the problem those compounds were designed to solve. It binds the acetylcholine site itself, so activating M1 means activating everything else too. TAK-071 and VU0486846 bind a separate site on M1 and only amplify the acetylcholine that is already there, and both were tuned to amplify weakly so that the peripheral receptors never see enough signal to cause sweating or diarrhoea. Same target, opposite strategy, and the first human tests of the newer strategy are tolerated in a way milameline never was.
Limitations of the evidence
- no confirmed clinical efficacy signal over placebo in its Phase II readout
- decreased core body temperature in animal studies
Adverse effects
- cholinergic side effects typical of muscarinic agonists (GI upset, sweating)
- increased gastrointestinal motility, a classic peripheral muscarinic side effect
- reduced spontaneous locomotor activity at higher doses