MEM-1003, also identified by the synonym BAY Z 4406, was an investigational small molecule drug belonging to the dihydropyridine chemical class, specifically optimized for central nervous system (CNS) applications.[1] It was developed by Memory Pharmaceuticals Corp. with the primary therapeutic goals of treating Alzheimer's disease and acute mania associated with bipolar disorder.[3] The core mechanism of action for MEM-1003 centered on its function as an L-type calcium channel modulator, intended to normalize aberrant calcium signaling within neurons, a pathological feature implicated in various CNS conditions.[3]
Preclinical investigations yielded promising results, suggesting that MEM-1003 could enhance cognitive functions in animal models of aging and Alzheimer's disease and exhibited a degree of CNS selectivity by preferentially affecting cerebral vasculature over peripheral systems.[5] Early-phase human trials further established a generally favorable safety and tolerability profile for MEM-1003, with no significant adverse cognitive effects noted in Alzheimer's patients at the doses tested.[7]
Despite these encouraging early findings, the clinical development program for MEM-1003 ultimately encountered insurmountable efficacy hurdles. Two pivotal Phase 2a clinical trials, one in Alzheimer's disease (NCT00257673) and another in bipolar mania (NCT00374920), failed to achieve their primary efficacy endpoints.[2] The Alzheimer's study was notably confounded by a significant placebo response, while the bipolar mania study showed no therapeutic benefit over placebo.[2] Consequently, due to this lack of demonstrated clinical efficacy, the development of MEM-1003 was discontinued.
The trajectory of MEM-1003 underscores the substantial challenges inherent in CNS drug development. Promising preclinical data and acceptable early-phase safety often do not translate into demonstrable efficacy in more extensive clinical trials. This is particularly true for complex neurological and psychiatric disorders where animal models may not fully recapitulate human disease, and clinical trial outcomes can be influenced by factors such as high placebo responses and disease heterogeneity. The acquisition of Memory Pharmaceuticals by Roche, which occurred during this period, was primarily focused on other assets in Memory's pipeline, and MEM-1003 was not pursued further by the acquiring entity.[9]
The investigational compound was primarily known as MEM-1003.[3] It was also referred to by synonyms including MEM 1003, BAY Z 4406, and BAY-Z-4406.[1] Classified as a small molecule drug [1], MEM-1003 belongs to the dihydropyridine chemical class. It was specifically described as a CNS-optimized dihydropyridine, indicating a design tailored for applications within the central nervous system.[2] This optimization aimed to differentiate it from conventional dihydropyridines often used for cardiovascular conditions, potentially by enhancing its ability to penetrate the blood-brain barrier or by minimizing peripheral side effects such as hypotension, which can be dose-limiting for this class of compounds when targeting brain disorders.
MEM-1003 was originated and developed by Memory Pharmaceuticals Corp..[1] This biopharmaceutical company was focused on the discovery and development of innovative drugs for a range of debilitating CNS disorders, including Alzheimer's disease, schizophrenia, depression, and bipolar disorder.[3]
For the development of MEM-1003 in bipolar disorder, specifically for acute mania, Memory Pharmaceuticals Corp. entered into a collaborative agreement with The Stanley Medical Research Institute (SMRI). SMRI provided funding support for the Phase 2a clinical trial in this indication.[2] This collaboration signified external interest and validation for exploring MEM-1003's potential in psychiatric disorders, alongside its primary investigation in neurodegenerative conditions.
In a significant corporate development, Memory Pharmaceuticals Corp. was acquired by Roche. The transaction, valued at approximately $50 million, was primarily driven by Roche's interest in securing the future development of nicotinic alpha-7 agonist drug candidates (R3487/MEM 3454 and R4996/MEM 63908) which were part of partnered programs between the two companies.[9] While MEM-1003 was a component of Memory Pharmaceuticals' portfolio, it was not the principal asset highlighted in the acquisition rationale. The timing of MEM-1003's clinical trial failures, which became apparent around the period of such corporate discussions, likely influenced its perceived value and future within the acquiring entity, solidifying its status as a non-core asset for Roche.
Sources indicate the existence of "100 Patents (Medical) associated with MEM-1003".[1] However, specific details regarding these patents, such as patent numbers for the composition of matter of MEM-1003 (or BAY Z 4406) or its method of use for Alzheimer's disease or bipolar disorder, are not directly accessible from the provided research information.[11] This lack of specific patent data represents an informational gap within the available materials.
MEM-1003's primary pharmacological target was identified as neuronal L-type calcium channels.[1] It was shown to bind with high potency, exhibiting a Ki of approximately 5nM, and selectivity to the dihydropyridine binding site located on L-type calcium channels within the brain.[5] This binding affinity was in a similar range to, though marginally lower than, other known L-type calcium channel blockers like nimodipine and nitrendipine.[5]
The intended therapeutic action of MEM-1003 was to modulate the influx of calcium ions into neurons. The overarching goal was to re-establish normal intracellular calcium concentrations.[3] Dysregulation of calcium homeostasis is a well-recognized pathological feature in several CNS disorders, including neurodegenerative conditions such as Alzheimer's disease and psychiatric illnesses like bipolar disorder.[4] By normalizing calcium flow, MEM-1003 was hypothesized to mitigate downstream detrimental effects associated with calcium dyshomeostasis.
A body of preclinical research provided the foundational evidence for advancing MEM-1003 into clinical trials. These studies explored its effects on neuronal excitability, cognitive functions, neuroprotection, and cerebrovascular dynamics.
Neuronal Excitability:
In studies utilizing rat hippocampal slices, MEM-1003 demonstrated an ability to reduce the amplitude of the slow afterhyperpolarization (sAHP).5 The sAHP is a calcium-dependent phenomenon that influences neuronal firing rates and tends to become more pronounced with aging. By diminishing the sAHP, MEM-1003 was thought to increase neuronal excitability, a mechanism considered potentially beneficial for improving cognitive functions that may be impaired due to reduced neuronal activity in aged or diseased brains.5
Cognitive Enhancement:
Multiple animal models indicated potential cognitive benefits:
Neuroprotection:
Based on its ability to modulate intracellular calcium levels, which can become excitotoxic when excessive, MEM-1003 was described as having neuroprotective properties. It was posited to be useful in restoring the excitability of neurons adversely affected by the processes of aging and Alzheimer's disease.5
Cerebrovascular Effects and CNS Selectivity:
A key differentiating feature of MEM-1003 highlighted in preclinical studies was its apparent CNS selectivity. It was observed to preferentially relax cerebral vascular smooth muscle compared to peripheral vascular smooth muscle.5 Specifically, the concentration required for 50% relaxation (EC50) of the middle cerebral artery was 120 nM, whereas for the coronary artery, it was 43 nM.5 This suggested that MEM-1003 might improve cerebral blood flow without inducing significant systemic hypotension, a common adverse effect of many L-type calcium channel blockers that limits their utility for CNS indications. This CNS selectivity was hypothesized to be related to the voltage dependence of its action at calcium channels.6 While this preclinical vascular profile was promising, the ultimate failure in clinical trials indicates that either this selectivity did not translate effectively to the human in vivo situation or that the fundamental mechanism of L-type calcium channel modulation, even if CNS-selective, was insufficient to treat the complex pathologies of Alzheimer's disease or bipolar mania.
Preclinical data for MEM-1003 were disseminated through scientific forums, including presentations at conferences such as the American College of Neuropsychopharmacology (ACNP) [6], reflecting standard scientific communication practices during drug development.
Animal Model/System | Key Parameter Studied | MEM-1003 Effect | Significance/Implication | Source Snippet(s) |
---|---|---|---|---|
Aged rats (cognitively impaired) | Spatial memory (MWM) | Improved latency to find platform | Potential for cognitive enhancement in aging/AD | 5 |
Aged rats | Attentional set-shifting | Improved performance | Potential for enhancing executive function/attention | 5 |
Rat hippocampal slices | Slow afterhyperpolarization (sAHP) | Reduced sAHP amplitude | Increased neuronal excitability | 5 |
Older rabbits | Eyeblink conditioning | Enhanced learning (at 2.0 mg/kg s.c.; other doses ineffective) | AD-relevant learning improvement, complex dose-response | 1 |
Genetic AD mouse model | Cognition | Potential to improve cognition | Disease modification potential in AD | 6 |
Isolated arteries (cerebral vs. coronary) | Vasorelaxation | Preferential cerebral vasodilation (EC50 120nM vs 43nM) | Potential for improved cerebral blood flow without hypotension | 5 |
The clinical development of MEM-1003 progressed through early-phase safety and tolerability studies in healthy volunteers and Alzheimer's disease (AD) patients, leading to two key Phase 2a proof-of-concept trials, one in AD and another in bipolar disorder.
Healthy Volunteer Studies (Conducted in the UK):
Initial human exposure to MEM-1003 involved studies in 125 healthy volunteers.2 These Phase 1a trials primarily assessed the safety and tolerability of single and multiple doses of the compound. Pharmacokinetic (PK) parameters were also likely evaluated. The key outcome from these studies was that MEM-1003 was generally safe and well-tolerated by healthy individuals.2 Human PK data indicated dose-dependent plasma concentrations. Importantly, unlike some other L-type calcium channel blockers such as nimodipine, MEM-1003 did not appear to reduce blood pressure at the clinical exposure levels tested, supporting its potential for CNS-selective effects.5
Phase 1b Safety and Tolerability Study in Alzheimer's Disease Patients (US IND):
Following the healthy volunteer studies, a Phase 1b trial was conducted in the United States under an Investigational New Drug (IND) application to assess MEM-1003 specifically in patients with Alzheimer's disease.3 This study employed a double-blind, placebo-controlled design and consisted of two segments:
The primary objectives were to evaluate safety and tolerability. MEM-1003 was found to be safe and generally well-tolerated at all dose ranges tested, including 120 mg BID.[7] Headache was the most commonly reported adverse event. There were no concerning trends observed in clinical laboratory safety tests, vital signs, or electrocardiogram (ECG) parameters following treatment with MEM-1003.[7]
A critical aspect of this study was the assessment of cognitive safety. The results indicated that 10 days of exposure to MEM-1003 did not lead to any statistically significant increase or decrease in cognitive function, as measured by the Cognitive Drug Research (CDR) battery and the Alzheimer's Disease Assessment Scale -- cognitive subscale (ADAS-cog).[7] This confirmation of cognitive safety in the target AD patient population was an essential prerequisite for proceeding to larger efficacy trials.
Study Identifier/Type | Population (N, characteristics) | Key Objectives | Dose(s) Administered | Key Safety/Tolerability Findings | Key PK/Cognitive Findings | Source Snippet(s) |
---|---|---|---|---|---|---|
Phase 1a Healthy Volunteers (UK) | 125 healthy volunteers | Safety, Tolerability, PK | Single and multiple doses | Generally safe and well tolerated | Dose-dependent plasma concentrations; No blood pressure reduction at clinical exposures. | 2 |
Phase 1b AD Patients (US) | 81 AD patients (49 dose escalation, 32 multiple dose) | Safety, Tolerability, Cognitive Safety | Up to 120 mg BID | Safe and generally well tolerated; headache most common AE. | No significant change in cognition after 10 days (cognitively safe). | 3 |
The Phase 2a trial aimed to evaluate the efficacy and safety of MEM-1003 in patients with mild to moderate Alzheimer's disease.
Efficacy Findings:
The trial's primary endpoint was the mean change from baseline at twelve weeks in the ADAS-cog score for the overall study population. The trial failed to meet this primary endpoint.8 The company reported that these negative results were largely attributable to an "unusually large placebo response" observed in the subgroup of patients receiving MEM-1003 as monotherapy.8 Such substantial improvements in the placebo group can make it statistically challenging to demonstrate a true drug effect, a common issue in CNS clinical trials.
In a subgroup analysis of patients who were concurrently receiving stable doses of cholinesterase inhibitors, the change in ADAS-cog scores numerically favored MEM-1003 treatment over placebo; however, this difference did not achieve statistical significance. Similar non-significant numeric improvements favoring MEM-1003 were observed for all four secondary endpoints in this subgroup: the Mini-Mental State Exam (MMSE), the Alzheimer's Disease Cooperative Study -- Activities of Daily Living (ADCS-ADL), the Clinician Interview-Based Impression of Change with Caregiver Input (CIBIC-Plus), and the Neuropsychiatric Inventory (NPI).[8] While such trends in subgroups are sometimes explored for future research directions, the lack of statistical significance and the overall primary endpoint failure meant these observations were not sufficiently compelling to support further development for AD based on this trial.
Safety and Tolerability:
MEM-1003 was generally well-tolerated in this Phase 2a AD trial. The rates of treatment-emergent adverse events were reported to be similar across all treatment arms (30 mg MEM-1003, 90 mg MEM-1003, and placebo). There were seven treatment-emergent serious adverse events (SAEs) recorded in five subjects during the study; four of these subjects were in the 30 mg MEM-1003 dose group, and one was in the 90 mg MEM-1003 dose group. These SAEs included two deaths. However, none of the SAEs were deemed by the investigators to be related to the study treatment.8
Publication of Results:
Top-line data from this trial were announced by Memory Pharmaceuticals via a press release.8 A comprehensive, peer-reviewed publication detailing the full results of NCT00257673 is not evident from the provided information sources.22 The ClinicalTrials.gov record for this study was inaccessible or did not list any publications.27
Concurrently, MEM-1003 was evaluated for acute mania in patients with Bipolar I Disorder.
Efficacy Findings:
The primary outcome measure for this trial was a comparison of the percentage of subjects in the MEM-1003 and placebo treatment groups who achieved at least a 50% improvement from their baseline score on the Young Mania Rating Scale (YMRS) at the end of the 21-day treatment period. MEM-1003 did not demonstrate efficacy; this primary endpoint was not met.2
Furthermore, none of the secondary outcome measures were achieved. These included the mean change from baseline in the YMRS score, the Modified Clinical Global Impression - Bipolar Scale (CGI-BP), and the Montgomery-Asberg Depression Rating Scale (MADRS) at 21 days.[2] The complete lack of separation from placebo on any efficacy measure indicated that MEM-1003, at the dose tested, did not provide a therapeutic benefit for acute mania.
Safety and Tolerability:
Despite the lack of efficacy, MEM-1003 was found to be safe and generally well-tolerated in this patient population.2 There were no serious adverse events reported in the group treated with MEM-1003. In contrast, three SAEs, all characterized as worsening of mania, were reported in the placebo treatment group. The most commonly reported adverse events in the MEM-1003 treatment arm were headache, gastrointestinal effects, and pruritus (itching).2 The favorable safety profile, however, was insufficient to outweigh the definitive lack of efficacy.
Publication of Results:
Similar to the Alzheimer's trial, top-line data for the bipolar mania trial were disseminated via a press release.2 A review article published in 2013 mentioned that the results for NCT00374920 were "not available" from ClinicalTrials.gov at that time.30 No full peer-reviewed publication of this trial's results is apparent from the available information.30 The ClinicalTrials.gov record was inaccessible or did not list publications.29
The consistent failure to meet primary efficacy endpoints in these two distinct Phase 2a trials, targeting different CNS indications, was the decisive factor leading to the discontinuation of MEM-1003's development. Phase 2a studies serve as crucial "proof-of-concept" stages, and the absence of a clear efficacy signal at this juncture typically precludes further investment in larger and more costly Phase 2b or Phase 3 trials.
Trial ID | Indication | Sponsor(s) | Phase | Design | Patient Population (N, key characteristics) | Intervention Arms (Drug, Dose, Control, Duration) | Primary Endpoint(s) & Results | Key Secondary Endpoint(s) & Results | Key Safety Findings | Reason for Discontinuation/Failure | Source Snippet(s) |
---|---|---|---|---|---|---|---|---|---|---|---|
NCT00257673 | Alzheimer's Disease | Memory Pharmaceuticals Corp. | 2a | Randomized, DB, PC | 183; mild to moderate AD; monotherapy or on stable cholinesterase inhibitors | MEM-1003 30mg BID, MEM-1003 90mg BID, or Placebo BID for 12 weeks | 12-week mean change in ADAS-cog; FAILED (large placebo response in monotherapy subgroup) | MMSE, ADCS-ADL, CIBIC-Plus, NPI; Numeric improvements in ChEI subgroup, not statistically significant. | Generally well-tolerated; similar AE rates across arms; 7 SAEs (5 subjects, 2 deaths), none deemed treatment-related. | Lack of efficacy | 1 |
NCT00374920 | Bipolar Mania (Acute) | Memory Pharmaceuticals Corp., SMRI (funding) | 2a | Randomized, DB, PC | 84; Bipolar I Disorder, acute manic or mixed episode | MEM-1003 (max allowed dose) or Placebo for 21 days | % subjects with $\geq$50% YMRS improvement at 21 days; FAILED | Mean change in YMRS, CGI-BP, MADRS at 21 days; None achieved. | Safe and generally well-tolerated; No SAEs in MEM-1003 group (3 SAEs in placebo); Common AEs: headache, GI effects, pruritus. | Lack of efficacy | 1 |
DB: Double-Blind, PC: Placebo-Controlled, ChEI: Cholinesterase Inhibitor, ADAS-cog: Alzheimer's Disease Assessment Scale-cognitive subscale, MMSE: Mini-Mental State Examination, ADCS-ADL: Alzheimer's Disease Cooperative Study - Activities of Daily Living, CIBIC-Plus: Clinician's Interview-Based Impression of Change Plus Caregiver Input, NPI: Neuropsychiatric Inventory, YMRS: Young Mania Rating Scale, CGI-BP: Clinical Global Impression - Bipolar Version, MADRS: Montgomery-Åsberg Depression Rating Scale, SAE: Serious Adverse Event, AE: Adverse Event.
Across its clinical development program, MEM-1003 consistently demonstrated a generally favorable safety and tolerability profile.
Early Phase 1 studies involving 125 healthy volunteers in the UK and a subsequent Phase 1b study in 81 Alzheimer's disease (AD) patients in the US established initial safety.[2] In these studies, MEM-1003 was generally well-tolerated at doses up to 120 mg twice daily (BID). Headache was reported as a common adverse event (AE).[7] Importantly, there were no significant adverse impacts on vital signs, electrocardiogram (ECG) parameters, or overall cognitive safety in AD patients treated for 10 days.[7] The finding that MEM-1003 was "cognitively safe" in AD patients was particularly crucial, as any drug intended to treat cognitive impairment must not, as a side effect, worsen cognition. This early safety data, including the lack of significant blood pressure reduction at clinical exposure levels [5], supported its progression into Phase 2a efficacy trials.
In the Phase 2a trial for Alzheimer's disease (NCT00257673), involving 183 patients, MEM-1003 continued to be generally well-tolerated.[8] The rates of treatment-emergent AEs were comparable between the MEM-1003 treatment arms (30 mg BID and 90 mg BID) and the placebo arm. Although seven treatment-emergent serious adverse events (SAEs) occurred in five subjects, including two deaths, none of these SAEs were considered by the investigators to be related to MEM-1003 treatment.[8]
Similarly, in the Phase 2a trial for acute mania in bipolar disorder (NCT00374920), which enrolled 84 subjects, MEM-1003 was reported as safe and generally well-tolerated.[2] No SAEs were reported in the MEM-1003 treatment group. The most commonly reported AEs in subjects receiving MEM-1003 were headache, gastrointestinal effects, and pruritus (itching).[2] Interestingly, this trial saw three SAEs, all instances of worsening mania, in the placebo group, while none occurred in the MEM-1003 group. While this observation does not imply a protective effect of MEM-1003, it underscores the inherent clinical instability of patients with acute mania and the challenges of conducting trials in this population.
Overall, the safety profile of MEM-1003 remained consistent from early-phase studies in healthy individuals to Phase 2a studies in distinct patient populations (AD and bipolar disorder). This consistency suggested that no new or unexpected major safety concerns emerged with more extensive exposure in patient groups. The AEs associated with MEM-1003 were typically mild to moderate in nature. The lack of significant safety signals was a positive attribute of its development program, though this was ultimately overshadowed by the failure to demonstrate clinical efficacy.
MEM-1003 (BAY Z 4406) was a CNS-optimized L-type calcium channel modulator developed by Memory Pharmaceuticals Corp. Its development was underpinned by a plausible scientific rationale derived from preclinical studies, which indicated potential for cognitive enhancement, neuroprotection, and a degree of CNS selectivity.[3] Early clinical development confirmed a generally good safety and tolerability profile in both healthy volunteers and patient populations with Alzheimer's disease and bipolar disorder, at the doses tested.[2]
Despite this promising preclinical and early clinical safety foundation, MEM-1003's journey was halted by its failure to demonstrate efficacy in pivotal Phase 2a clinical trials. The trial in Alzheimer's disease (NCT00257673) did not meet its primary endpoint, a result significantly influenced by a high placebo response in a subgroup of patients.[8] Similarly, the Phase 2a trial in acute mania associated with bipolar disorder (NCT00374920) also failed to achieve its primary efficacy endpoint, showing no discernible therapeutic benefit over placebo on any of the efficacy measures.[2]
These consecutive failures in demonstrating proof-of-concept in two distinct CNS indications led to the discontinuation of MEM-1003's development program by Memory Pharmaceuticals.[1] The negative trial results had a tangible impact, with reports of Memory Pharmaceuticals' shares declining significantly after the announcement of the bipolar trial failure.[36] This situation often places considerable financial strain on smaller biopharmaceutical companies whose valuations are closely tied to the success of their lead clinical assets.
The subsequent acquisition of Memory Pharmaceuticals by Roche was primarily motivated by other assets within Memory's pipeline, specifically nicotinic alpha-7 agonists (R3487/MEM 3454 and R4996/MEM 63908).[9] There is no indication that Roche continued the development of MEM-1003 post-acquisition, which is consistent with the conclusive negative efficacy data from the Phase 2a trials.
The development history of MEM-1003 serves as a salient illustration of the formidable challenges encountered in CNS drug development, often referred to as the "valley of death" where many candidates fail during Phase 2. Factors such as the inherent complexity of neurological and psychiatric disease pathologies, the limitations of preclinical animal models in predicting human responses, and the difficulties in designing and executing clinical trials that can overcome issues like substantial placebo effects, all contribute to the high attrition rate. Even compounds like MEM-1003, which possess a rational mechanistic basis and a good safety profile, can falter when subjected to rigorous efficacy testing in patient populations. Furthermore, the apparent absence of full peer-reviewed publications for the failed Phase 2a trials—a common scenario for studies with negative outcomes—limits the broader scientific community's ability to extract detailed learnings that could inform future research efforts in these challenging therapeutic areas.
Published at: May 12, 2025
This report is continuously updated as new research emerges.