BackgroundFreezing of gait (FOG) is one of the most debilitating motor symptoms in Parkinson’s disease (PD), affecting a substantial proportion of people with PD. Numerous hypothetical mechanisms exist, attempting to explain FOG. Current evidence originated from stationary measures in lying positions, limiting the understanding of neurophysiological changes elicited by FOG episodes during regular locomotion. Neurophysiological aspects of FOG, especially in response to vibrotactile cueing, have been explored only by a few studies. To address this gap, the current study aims to investigate FOG episodes in people with PD and the impact of vibrotactile cueing while walking through virtual environments on a self-paced treadmill.MethodsThirty people with Parkinson’s Disease and FOG will be recruited. They will undergo screening and familiarization before performing different walking tests in virtual reality using the Gait Real-time Analysis Interactive Lab (GRAIL). Participants will perform two experimental blocks, each consisting of four walking trials of approximately 10 min. A break of 10 to 30 min will be provided between blocks. Walking with or without vibrotactile cueing will be performed in alternation. During walking, electrocortical (EEG), hemodynamic (fNIRS), kinematic data (3D motion capture), and kinetic data (force plates) will be recorded. Data will be analyzed to investigate spatiotemporal and frequency characteristics of electrocortical and hemodynamic brain activity related to FOG and the impact of vibrotactile cueing on neural signatures and gait improvements in FOG.DiscussionCombining neurophysiological measurements of EEG and fNIRS paired with kinematics and kinetics will provide insights into the cortical and behavioral changes associated with FOG and vibrotactile cueing, to derive results that can be used for the design of intervention strategies for treating PD.Study protocol registrationhttps://drks.de/search/de/trial/DRKS00034584, identifier DRKS00034584.
Neurophysiological effects of transcranial alternating current stimulation combined with multidisciplinary rehabilitation in Parkinson’s disease
IntroductionMotor impairment in Parkinson’s disease (PD) often becomes inadequately controlled as the disease progresses. Rehabilitation therapy combined with noninvasive neuromodulation has shown benefits, yet the neural mechanisms underlying this remain unclear. This study aimed to identify neurophysiological and brain network features associated with motor improvement following high-intensity transcranial alternating current stimulation (Hi-tACS) combined with multidisciplinary intensive rehabilitation therapy (MIRT).MethodsThis secondary analysis was based on a randomized controlled trial (ChiCTR2300071969). Thirty patients receiving Hi-tACS combined with MIRT were included. Responders were defined as patients achieving a minimal clinically important difference improvement greater than 3.25 points on the Unified Parkinson’s Disease Rating Scale part III. Electroencephalographic spectral analysis, resting-state functional connectivity, and dynamic brain states assessed using hidden Markov modeling were performed. Partial correlation analyses were conducted for group-differentiating measures, controlling for age and disease duration.ResultsCompared with non-responders, responders showed increased temporal high beta frequencies relative power (95%CI 0.85(0.01–1.66), P = 0.047), reduced connectivity between the default mode and sensorimotor networks (95%CI -0.89(-1.71–0.05), P = 0.038), and decreased fractional occupancy of state 3 (95%CI 0.03(0.00–0.28), P = 0.033) and 6 (95%CI 0.02(0.00–0.35), P = 0.038). After Bonferroni correction, only changes in temporal high beta frequencies relative power were negatively correlated with changes in total UPDRS part III scores (R = -0.59, PBON = 0.012) and rigidity subscores (R = -0.74, PBON < 0.001).DiscussionThese results suggest that motor improvement following combined Hi-tACS and MIRT in PD patients is associated with alterations in brain oscillations and network connectivity, particularly in high-beta frequencies. The findings contribute to understanding the neural mechanisms underlying rehabilitation and neuromodulation in PD. Future studies should investigate the long-term effects of this intervention and its applicability in larger, more diverse populations. Moreover, exploring additional biomarkers may help identify individuals who are most likely to respond to this treatment.Clinical Trial Registrationhttps://www.chictr.org.cn, identifer ChiCTR2300071969.
Blood circRNAs Can Predict Alzheimer’s Years Prior to Symptoms Onset
A set of blood-based circular RNAs (circRNAs) could change how we diagnose and monitor Alzheimer’s disease, providing a simple, noninvasive test that can detect the disease with remarkable accuracy and predict its progression years before symptoms appear.
Washington University School of Medicine researchers analyzed blood samples from 1,221 individuals, including people with Alzheimer’s disease and cognitively healthy participants, making it one of the largest investigations of blood circRNAs in Alzheimer’s to date. The findings, published in a Nature Medicine study, identified 34 circRNAs whose combined expression patterns accurately distinguished Alzheimer’s disease from healthy aging.
Unlike conventional RNA molecules, circRNAs form single-stranded closed loops that resist degradation and are abundant in the brain. Their stability and ability to cross the blood-brain barrier make them attractive candidates for blood-based biomarkers that reflect changes occurring in the brain.
The research indicated that a predictive model built from the 34 circRNAs achieved an area under the curve (AUC) of 0.945 for identifying biomarker-confirmed Alzheimer’s disease, outperforming the widely used plasma biomarker pTau217 (AUC 0.877). When circRNA measurements were combined with pTau217, diagnostic performance increased further to an AUC of 0.977.
Beyond diagnosis, the circRNA signature demonstrated exceptional ability to predict disease progression. Individuals with elevated circRNA scores were nearly three times more likely to progress to symptomatic Alzheimer’s disease than those with lower scores. The model also outperformed pTau217 in forecasting progression and remained highly specific for Alzheimer’s, showing limited predictive ability for other neurodegenerative disorders such as Parkinson’s disease, frontotemporal dementia, and dementia with Lewy bodies.
Importantly, the findings were independently replicated in two additional cohorts, including 551 participants from the Knight Alzheimer’s Disease Research Center and 1,767 participants enrolled in the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease (A4) study. This independent validation demonstrates that the circRNA signature is robust across multiple populations and study designs.
The researchers also found evidence that circRNA changes begin approximately two to four years before the onset of clinical symptoms, suggesting that these molecules may capture biological processes closely linked to the transition from silent pathology to cognitive decline. Such biomarkers could become increasingly valuable as disease-modifying therapies enter clinical practice, where monitoring ongoing neurodegeneration is just as important as detecting amyloid pathology.
The work builds on intellectual property protected by patent PCT/US2026/017857, “Blood Circular RNA as a Noninvasive Biomarker of Alzheimer’s Disease,” which Circular Genomics has licensed. The company, based in San Diego, California, is developing next-generation molecular blood biomarker diagnostics for precision neurology. The patent covers the use of blood circRNA signatures for the diagnosis and monitoring of Alzheimer’s disease and supports continued development of clinically deployable blood tests.
While the authors emphasize that larger prospective studies are still needed before widespread clinical implementation, the findings position blood circRNAs as a promising new class of biomarkers for Alzheimer’s disease. By combining high diagnostic accuracy with strong prediction of disease progression using a simple blood sample, circRNA-based testing could help identify patients earlier, improve clinical trial enrollment, and provide physicians with new tools to monitor disease over time.
The post Blood circRNAs Can Predict Alzheimer’s Years Prior to Symptoms Onset appeared first on Inside Precision Medicine.
Modulating autonomic nervous system activity with transcutaneous auricular vagus nerve stimulation in Parkinson’s disease: a proof of concept study
BackgroundCardiovascular dysautonomia is a debilitating non-motor symptom of Parkinson’s disease (PD) that limits exercise capacity and neurorehabilitation outcomes. Transcutaneous auricular vagus nerve stimulation (taVNS) is an emerging non-invasive neuromodulatory therapy that modulates cardiovascular activity and could potentially serve as an adjunct to exercise, yet its physiological effects on cardiovascular function in PD remains unexplored.ObjectiveThis proof-of-concept, sham-controlled crossover pilot study (N = 8) investigated the acute effects of taVNS on cardiovascular autonomic activity in idiopathic PD.MethodsParticipants underwent active taVNS (30 Hz, 250 μs, 0.1–4 mA) or sham stimulation (0 mA) during a 15-min resting phase, immediately followed by the Ewing Battery of cardiovascular reflexes. Acute autonomic shifts were phenotyped using continuous heart rate variability (HRV) monitoring.ResultsThis proof-of-concept protocol was feasible, as all participants completed the randomized crossover stimulation visits and autonomic reflex testing without adverse events. Baseline autonomic burden (COMPASS-31) was associated with the magnitude of heart rate response to active stimulation. Immediately following stimulation and during the deep breathing challenge, active taVNS was associated with directionally consistent changes in vagally mediated HRV metrics including RMSSD, pNN50, and HF power, relative to sham.ConclusionContinuous HRV monitoring and autonomic reflex testing appears feasible for characterizing acute autonomic responses to taVNS in PD. Active stimulation was associated with directional changes in vagally mediated HRV metrics during the post-stimulation period and during deep breathing, supporting the biological plausibility of acute autonomic modulation. These preliminary findings justify larger, adequately powered studies designed to determine whether taVNS can reliably modulate cardiovascular autonomic regulation and inform rehabilitation optimization in PD.
Integrated bulk, single-cell, and spatial transcriptomic analyses prioritize NOTCH1 as a candidate gene associated with neurovascular and immune-related alterations in Parkinson’s disease
IntroductionParkinson’s disease (PD) is classically defined by dopaminergic neurodegeneration in the substantia nigra, yet how immune activation is linked to neurovascular dysfunction in the diseased brain remains incompletely understood.MethodsHere, we integrated bulk substantia nigra microarray expression datasets with single-cell and spatial transcriptomic data to delineate disease-associated neurovascular and immune-related transcriptomic programs in PD.ResultsAcross three independent human microarray cohorts, differential expression and weighted gene co-expression network analyses identified PD-associated genes enriched for synaptic processes together with immune, adhesion, and vascular-related pathways. Network topology analysis and machine-learning feature selection prioritized a five-gene candidate panel, among which NOTCH1 showed the most consistent cross-dataset association and external directional support. Importantly, quantitative real-time PCR (qRT-PCR) validation in the substantia nigra of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced PD model mice further supported dysregulated Notch1 expression. Single-cell mapping placed NOTCH1 expression within neurovascular and glial cellular contexts, including pericytes and endothelial cells, while CellChat and NicheNet analyses nominated transcriptome-derived ligand-receptor relationships involving NOTCH-related, vascular, inflammatory, extracellular-matrix, and growth-factor-associated programs. Spatial transcriptomics from mouse 6-hydroxydopamine (6-OHDA) substantia nigra sections provided model-based anatomical context for spatial proximity between pericyte and microglial signatures, without establishing direct functional communication. In parallel, exploratory in silico perturbation and docking-based screening generated hypotheses regarding the NOTCH1-associated regulatory context and compounds with predicted docking affinity toward NOTCH1.DiscussionCollectively, these analyses prioritize NOTCH1 as a reproducible PD-associated candidate gene and suggest that NOTCH1-related signals may be embedded within broader neurovascular, glial, inflammatory, extracellular-matrix, and immune-associated transcriptomic alterations. These findings provide a computational prioritization framework for future experimental validation rather than evidence of a defined NOTCH-driven mechanism.
Endocannabinoid system modulation in bruxism: a neurobiological hypothesis and translational model of ECS-targeted intervention
Bruxism is a multifactorial motor behavior of predominantly central origin, characterized by repetitive masticatory muscle activity and associated with dysregulation of dopaminergic, serotonergic, GABAergic, and glutamatergic pathways involved in motor control, emotional regulation, and stress responsivity. The endocannabinoid system (ECS) has emerged as a key homeostatic neuromodulator capable of integrating these neurotransmitter systems, thereby influencing pain processing, sleep–wake dynamics, and motor output. This article develops a neurobiological hypothesis based on a narrative integrative synthesis of clinical, experimental, and translational evidence regarding ECS involvement in the pathophysiology of bruxism. Findings from randomized clinical trials suggest that topical cannabidiol (CBD) may modulate motor neuron excitability and reduce pain-related outcomes, while case-based and experimental evidence supports the interaction between cannabinoid signaling and neural circuits implicated in motor control and behavioral regulation. Building on this evidence, we propose a hypothesis-driven translational model in which ECS-mediated neuromodulation may influence central mechanisms underlying bruxism, including motor pattern generation, stress responsivity, and nociceptive processing. Rather than providing prescriptive therapeutic recommendations, this model is intended as a hypothesis-generating construct that integrates current knowledge on ECS signaling within the broader neurobiology of motor control. Although heterogeneity in study design and outcome measures limits definitive conclusions, the available evidence supports the ECS as a plausible modulatory system in bruxism, with potential implications for future mechanistic and clinical research in centrally mediated motor disorders.
Micro- and nanoplastics as environmental modifiers of neuroimmune dysfunction in Parkinson’s disease
Parkinson’s disease (PD) is a progressive neurodegenerative disorder characterized by the loss of dopaminergic neurons and the aggregation of α-synuclein, with increasing evidence implicating environmental factors and neuroimmune dysfunction in its pathogenesis. Micro- and nanoplastics (MNPs), ubiquitous environmental pollutants generated from plastic degradation, have recently emerged as potential biological stressors capable of entering the human body and accumulating in sensitive tissues, including the brain. Due to their small size, environmental persistence, and capacity to carry toxic additives and environmental contaminants, these particles can induce oxidative stress, impair mitochondrial and lysosomal function, and activate both innate and adaptive immune responses. This review summarizes current evidence linking microplastic exposure to neuroinflammatory processes relevant to PD, with a particular focus on microglial activation, astrocyte reactivity, peripheral immune involvement, and dysfunction of the gut–brain axis. Although a direct causal relationship between MNPs and PD has yet to be established, and direct human epidemiological evidence linking MNP exposure to PD is currently absent, the immunotoxic and neuroinflammatory effects of these particles suggest that they may contribute to disease susceptibility and progression. Elucidating the interactions between MNPs and neuroimmune pathways may help refine current frameworks linking environmental exposure, neuroimmune dysfunction, and PD susceptibility.
Insilico, SK Launch Up-to-$2.5B Neuroimmune AI Drug Collaboration
Insilico Medicine will partner with SK Biopharmaceuticals to discover new artificial intelligence (AI)-based drug candidates for disorders affecting the neuroimmune area of the central nervous system (CNS), through a collaboration that could generate up to $2.5 billion for the AI-based drug developer.
Insilico agreed to apply its Pharma.AI platform, which addresses target validation, generative chemistry, and molecule optimization, along with its preclinical drug discovery expertise, to discover, design, and optimize candidates for neuroimmune indications against targets that will originate with SK.
SK will contribute its development and clinical capabilities in neuroimmune disorders, steering the late-stage development and commercialization of all resulting programs.
“Some of the collaborations we do are very focused on target discovery, but here it’s more focused on the delivery of the real drug,” Alex Zhavoronkov, PhD, Insilico’s founder, co-CEO, and chief business officer, told GEN in an interview at his company’s exhibition-hall booth during the Biotechnology Innovation Organization (BIO) International Convention, held recently in San Diego.
“Basically, we are being brought in to develop a drug, to discover and take it to a certain point, after which the partner takes it over. And they usually have a lot of choices to do it with other partners,” Zhavoronkov explained. “But they trust AI. They like AI. They like the way we design drugs and like our speed and efficiency.”
Headquartered in Seongnam, South Korea, SK Biopharmaceuticals is a global biotech focused on the research, development, and commercialization of new therapies for CNS disorders and beyond, including radiopharmaceutical and targeted protein degradation therapies. In 2020, SK became the first Korean pharma to independently develop and commercialize a novel drug in the United States, the epilepsy treatment Xcopri® (cenobamate), after it won FDA approval a year earlier.
Beyond epilepsy
“This collaboration represents an important milestone in expanding our growth beyond epilepsy into new CNS therapeutic areas, building on the deep CNS expertise we have established through the successful development and commercialization of cenobamate,” Donghoon Lee, SK Biopharmaceuticals’ president and CEO, said in a statement. “By combining Insilico’s AI-powered drug discovery platform with SK Biopharmaceuticals’ clinical development and U.S. commercialization capabilities, we believe we can accelerate the discovery of innovative CNS therapies for patients.”
“Beyond a single program, we see this collaboration as a scalable and repeatable growth platform that can be leveraged for future target discovery and development opportunities,” Lee added.
SK Biopharmaceuticals is part of the SK Group, South Korea’s second-largest family-owned chaebol or conglomerate, after Samsung Group, and a chaebol whose holdings include the vaccine developer SK Bioscience, the contract development and manufacturing organization (CDMO) SK Pharmteco, and SK Hynix, a supplier of high bandwidth memory (HBM) chips that power the AI processors of Nvidia and AMD. SK Hynix and a sister chaebol company, SK Telecom, are investors in Rebellions, a Korean dedicated fabless design company specializing in manufacturing AI neural processing units optimized for data centers and large language models.
Insilico’s AI-based drug development background complemented SK’s focus on leveraging AI and digital technologies across drug discovery, development, and treatment, SK Biopharmaceuticals concluded.
“Very difficult space”
“After you have done this,” Zhavoronkov said, pointing to a graphic showing Insilico’s AI-based pipeline, “people know that we can do this. The question is, can we do it in neuroimmunology? That is a very difficult space, one of the most difficult disease areas to tackle, given the need to develop molecules with properties that include high levels of safety and brain penetration.”
Insilico’s pipeline includes one candidate designed to treat CNS disorders—ISM8969, a Phase I oral brain penetrant NLRP3 inhibitor, which the company is co-developing with Hygtia Therapeutics under an exclusive global license and co-development collaboration. Both companies hold 50% worldwide rights to ISM8969, with Insilico eligible to receive up to $66 million in upfront and milestone payments from Hygtia, an incubatee of Shenzhen Pengfu Fund of Fosun Health Capital and Fosun Pharma.
Insilico is leading initial clinical development of ISM8969, from IND submission through execution of the Phase I trial (NCT07581431) for the drug’s initial indication of Parkinson’s disease. Hygtia will lead subsequent global clinical studies, regulatory submissions, and commercialization activities.
Discovered using the company’s generative AI platform Chemistry42, ISM8969 has shown strong efficacy, favorable safety, and robust blood-brain barrier (BBB) penetration, leading to marked anti-inflammatory activity in preclinical studies, according to Insilico.
Unlike other drug developers that concentrate on a few therapeutic areas, Insilico maintains a pipeline of 40+ programs across a wide variety of indications, including idiopathic pulmonary fibrosis (IPF), cancer, obesity and metabolic diseases, pain, and inflammatory diseases, including inflammatory bowel disease.
Longevity focus
“We focus on aging. That’s what we care about,” Zhavoronkov declared. “Most of the programs that we like to work on are focused on longevity.”
Furthest along in clinical studies is rentosertib (formerly called ISM001-055), a small molecule designed to treat idiopathic pulmonary fibrosis (IPF) by targeting Traf2- and NCK-interacting kinase (TNIK), a serine/threonine kinase whose activation plays a crucial role in cellular processes that include signal transduction pathways essential for fibrosis development.
Rentosertib has completed a 12-week Phase IIa trial (NCT05938920) conducted across 22 sites in China, with results published in Nature Medicine, and is in a separate Phase II trial in the United States. In the Chinese trial, rentosertib met its primary endpoint of safety and tolerability across all dose levels, and showed positive results for the secondary efficacy endpoint, wherein a dose-dependent forced vital capacity (FVC) improvement was seen.
“We’re preparing for the next step. When that gets announced, it’s going to be a big deal. Hopefully sooner than later. Like, much sooner than much later,” Zhavoronkov said.
As in later this year?
“It’s in the second half, but maybe closer to the earlier second half,” he replied.
The U.S. trial has not progressed as quickly as the Chinese trial. “We have not seen a trial slower than that in our history. Enrollment is just extremely slow because our criteria for enrollment are very high. Also, there are not that many [IPF] patients compared with China, where it was just much faster,” Zhavoronkov said.
Given the slow speed of the U.S. trial, he said, it would be more worthwhile to just start a Phase IIb or Phase III following more data from China. “It’s a game of chess, so to speak. You need to time it [an additional trial], and you need to properly adjust to the realities of enrollment.”
In April, Insilico received investigational new drug (IND) clearance from China’s Center for Drug Evaluation (CDE) to begin a Phase I study of inhalable rentosertib in IPF—the company’s 13th pipeline program to receive IND clearance. The study will evaluate the safety, tolerability, and pharmacokinetic (PK) profiles of rentosertib inhalation solution—first through a randomized, double-blind, placebo-controlled trial in healthy participants involving single and multiple ascending dose cohorts; then through a non-randomized, open-label evaluation in IPF patients who will receive multiple doses. Approximately 80 people are expected to be enrolled.
“Most promising”
“IPF is the most promising disease for longevity therapeutic testing because the patients are old. And even normal people up to 65, they start losing force valve capacity quite a bit, like the amount of air you can breathe out of your lungs. And it’s like 30, 40 milliliters a year. IPF patients can lose up to 400 milliliters,” Zhavoronkov said. “That’s the critical measure of lung function, and that’s what we measure in the study.”
Insilico researchers chronicled the drug’s discovery and early development in Nature Biotechnology in March 2024, detailing a novel target discovered by Insilico’s target identification engine, PandaOmics, and a novel molecular structure designed by its generative chemistry engine, Chemistry42. Both are specific-function platforms within the company’s AI platform, Pharma.AI.
“We are making massive progress on the AI side,” Zhavornkov said.
Massive enough that users should expect to see either tweaks in the platform or new platforms? “100%, you’re going to see a complete rewall,” he replied, as in a secure, self-contained AI environment or “walled garden” pursued by AI developers during commercial inflection points.
“We have so many new next-generation tools right now that it’s actually very difficult to productize them. Because at the lab level and at the platform level, we see superintelligence already. I’m talking about, we can probably go from prompt to drug in some areas: You basically prompt it, and you could make it and potentially take it,” Zhavoronkov explained. “I think we’re there. It’s just, fortunately, you have to do all the nitty-gritty testing and then clinical studies.”
Insilico’s other Phase II program is ISM5411, a gut-restricted molecule designed to treat inflammatory bowel disease (IBD) by taking aim at another anti-aging target, PHD 1/2. Unlike with rentosertib, clinical studies for the PHD1/2 inhibitor have found it easier to recruit patients in the United States than in China, where fewer patients are diagnosed with the disease.
The program, formerly called ISM012-042, was shown in preclinical studies to restore intestinal barrier function and alleviate gut inflammation in multiple experimental colitis models, while exhibiting favorable safety and pharmacokinetic profiles, according to a 2024 study published in Nature Biotechnology. The program is one of two that target PHD 1/2; the other is a small molecule designed to treat anemia of chronic kidney disease, for which Greater China rights have been outlicensed to TaiGen.
Longevity-linked targets
PHD 1/2, TNIK, and NLRP3 are three of numerous longevity-linked targets for the drug candidates within Insilico’s growing pipeline. Among the others that are targets of candidates in the clinic or IND-cleared:
- ENPP1 (ectonucleotide phosphodiesterase 1), a target of a program designed to treat anti-PD-1/-L1 resistant cancers, and has won IND clearance.
- KAT6 (lysine acetyltransferase 6 ) and KIF18A (kinesin family member 18A), targets of MEN2312 and MEN2501, respectively, are both Phase I cancer-fighting candidates outlicensed to Menarini Group through collaborations launched in 2024 and 2025.
- MAT2A (methionine adenosyltransferase 2α), a target of a Phase I small molecule candidate designed to treat MTAP -/- (methylthioadenosine phosphorylase deficient) cancer.
- QPCTL (glutaminyl-peptide cyclotransferase-like protein), a target of a first-in-class Phase I oral small molecule cancer immunotherapy for cold tumors being co-developed in partnership with Fosun.
- TEAD (transcriptional enhanced associate domain), a target of ISM6631, a Phase I “pan-TEAD” (TEAD 1/2/3/4) inhibitor designed to treat mesothelioma and solid tumors that include epithelioid hemangioendothelioma (EHE), meningioma, glioblastoma, liposarcoma, and pancreatic cancers.
- USP1 (ubiquitin-specific protease 1), a target of a Phase I BRCA-mutated cancer drug outlicensed to Exelixis under a 2023 collaboration.
“Our differentiation from everybody else is novelty—novelty of the target,” Zhavoronkov said. “Nobody I know in our industry has such a large number of absolutely novel targets that have never been in the clinic before or that are novel for indication. But with novelty comes a great risk. And pharma doesn’t want to take that risk up until a certain point.”
“Very often, you need to spend a long time in the process of discovery and then development in order to license a drug,” he added. “Once you license a drug, usually in Insilico’s case, some of the pharma companies actually like to get some access to AI technologies, and then it would be structured as a licensing class collaboration.”
Second multi-billion-dollar collaboration
SK Biopharmaceuticals is the second multi-billion-dollar collaboration announced by Insilico this year. The first was an up-to-$2.75 billion discovery and development partnership with Eli Lilly, to which Insilico granted an exclusive global license to develop, manufacture, and commercialize what the companies described in an announcement only as “potentially best-in-class, novel oral therapeutics in preclinical development for certain indications,” without detailing the therapeutic areas where the companies plan to partner.
“Those are early preclinical drugs that have incredible properties. I like to use the term maximally multi-parameter optimized molecule or MMOMs,” Zhavoronkov said.
Lilly agreed to pay Insilico $115 million upfront, as well as development, regulatory, and commercial milestones plus tiered royalties on future sales. The deal continued and expanded a relationship that began late in 2023, when Lilly inked a licensing agreement allowing it to access Insilico’s Pharma.AI software suite.
The Lilly collaboration will allow Insilico and Zhavoronkov to work with Jiye Shi, PhD, the pharma giant’s senior vice president of discovery technology & platforms and early molecule discovery, who has specialized in research on integrating machine learning and AI into the pharmaceutical pipeline. Previously at UCB, he led a computational biology team that used machine learning and computational design to create bimekizumab, a humanized interleukin-17A and F antagonist hailed as one of the first, if not the first, AI-based dual-targeting monoclonal antibodies to reach the market, where it is sold as Bimzelx® (bimekizumab-bkzx).
In February, Shi and Zhavoronkov co-authored a paper outlining a vision for a “prompt-to-drug” pipeline, where AI not only generates novel hypotheses and designs optimized drug candidates but also orchestrates synthesis, validation, and clinical planning in a closed-loop system.
“The realization of a true ‘prompt-to-drug’ pipeline, in which a natural language request initiates a fully autonomous drug development program, is no longer a distant aspiration. With the development of modular AI platforms, humanoid-in-the-loop robotics, and multi-agent systems, the foundational components for this vision are already operational,” wrote Shi, Zhavoronkov, and co-author David Gennert, PhD, a medical writer who at the time was Insilico’s senior scientific writer and editor.
Insilico’s collaboration with SK, Zhavoronkov said, reflects how AI “has transformed from being a fairy tale or a promise, to being a real tool that is used routinely to discover and develop drugs.”
“This is basically production level,” he added. “We’re not trying to do a pilot here.”
The post Insilico, SK Launch Up-to-$2.5B Neuroimmune AI Drug Collaboration appeared first on GEN – Genetic Engineering and Biotechnology News.
Women with Parkinson’s Have More Amyloid Plaques than Men
A study led by the Mayo Clinic Arizona shows women with Parkinson’s disease have greater amyloid plaque burden than men with the condition, even after controlling for factors like carriage of the APOE4 Alzheimer’s disease susceptibility gene variant.
As reported at the European Academy of Neurology Congress in Geneva this week, 57% of women included in the study had a high amyloid plaque burden versus 40% of the men.
Amyloid-beta is a protein fragment that normally gets cleared from the brain. In Alzheimer’s disease, it misfolds and aggregates into oligomers and plaques between neurons. This disrupts synaptic signaling, activates neuroinflammation, and promotes tau protein hyperphosphorylation into neurofibrillary tangles as the disease progresses.
In contrast, Parkinson’s disease is caused by the misfolding and clumping of a protein called alpha-synuclein into toxic deposits known as Lewy bodies, which build up in and destroy the neurons that produce dopamine in a brain region called the substantia nigra. While Parkinson’s is known for its characteristic motor symptoms, at least 25% also have dementia-like symptoms similar to those seen in Alzheimer’s disease. Amyloid beta plaques are thought to worsen Parkinson’s disease and increase the risk of dementia symptoms.
There are known differences in the prevalence and symptoms shown by men and women with Parkinson’s disease. To investigate this further, 230 people enrolled in the Arizona Study of Aging and Neurodegenerative Disorders and Brain and Body Donation Program were included in this study after death. Amyloid burden in the brain was assessed during autopsy. Other clinical factors such as cognition and symptoms were recorded prior to death.
The study found that amyloid plaque burden in women was higher than in men with Alzheimer’s. For example, mean cortical total plaque score in women was 6.5/15 vs 4.9/15 in men. Neuritic plaque density was also higher in women at 1.7/3 compared with 1.3/3 in men.
After correcting for age at death and APOE4 status, women in the study were more than twice as likely to have a high plaque burden than men.
This did not seem to translate to cognitive differences between men and women in the study though. “Men and women with Parkinson’s disease had similar rates of Alzheimer’s dementia and similar results on cognitive testing. However, women showed a higher amyloid plaque burden compared with men,” explained presenting author Erika Driver-Dunckley, MD, Mayo Clinic Arizona, in a press statement.
Notably, in standard Parkinson’s disease, men are at higher risk of developing dementia than women, so it is possible women have some protection from alpha-synuclein-driven decline but not from damage linked to amyloid accumulation. Women with Parkinson’s also live longer than men with the condition, as well as being more prone to amyloid buildup and Alzheimer’s disease, which complicates understanding the meaning of these results.
“Our findings highlight the need for further research into sex differences in Parkinson’s disease and Alzheimer’s-related pathology,” concluded Driver-Dunckley. “An important next step will be to confirm these findings in additional large clinicopathological studies and better understand the biological mechanisms that may underlie these differences.
The post Women with Parkinson’s Have More Amyloid Plaques than Men appeared first on Inside Precision Medicine.
Remote Assessment of Parkinson Disease Using Deep Learning on Structured Mouse-Trace Data From Suspected Cases: Machine-Learning Pilot Feasibility Study
Background: Parkinson disease (PD) is a pervasive neurodegenerative disorder globally, largely characterized by motor symptoms. Most existing artificial intelligence models for PD detection are trained on participants in well-resourced settings with confirmed clinical diagnoses. However, specialist-confirmed labels are often infeasible in low-resource settings. Objective: We developed a web platform for structured mouse data collection through pattern tracing tests. We sought to assess the feasibility of leveraging data from a community-recruited sample of participants with suspected but undiagnosed PD to train artificial intelligence models that achieve respectable performance in predicting diagnosed PD. We tested whether using weaker diagnostic labels that may be more feasible to collect in community or global health settings, where access to professional neurologists is sparse or nonexistent, can lead to models that learn predictive signals that are diagnostically useful. Methods: 261 participants (73 self-reported PD, 155 non-PD, and 33 suspected PD) were recruited from community organizations in Hawaii and completed 3 pattern tracing tasks on our custom web assessment: straight line, sine wave, and spiral wave. During each task, cursor positions, screen dimensions, and an in-target boolean flag were recorded. From these data, we engineered features and generated mouse trace images. We built 3 categories of classifiers: (1) a feed-forward neural network using engineered features, (2) fine-tuned computer vision deep learning models, and (3) multimodal models concatenating a feed-forward neural network with computer vision models. Performance was evaluated using 1 primary experiment and 2 secondary analyses. The primary experiment involved training on suspected PD versus non-PD and testing on self-reported PD versus non-PD. A secondary analysis evaluated the reverse direction by training on participants with self-reported PD and without PD and then testing on participants with suspected PD versus participants without PD. Additionally, a cross-validation analysis was conducted using participants with self-reported PD versus those without PD with 5-fold cross-validation to establish baseline performance under well-defined diagnostic labels. Results: The best-performing models included a multimodal Vision Transformer in the primary experiment (: mean 0.7619, SD 0.0535), a multimodal ResNet-50 in the secondary analysis (: mean 0.9353, SD 0.0334), and an image-based DenseNet-201 in the cross-validation analysis (: mean 0.9027, SD 0.0332). Training on patients with suspected PD yielded meaningful performance in predicting self-reported PD, supporting the feasibility of using lower-specificity labels for model development. Conclusions: This pilot feasibility study suggests that remotely collected mouse-tracing data can support PD screening models under data labeling conditions of low diagnostic specificity: models trained on suspected PD from a community sample may learn signals that can transfer to predicting actual PD. Future work may consider pretraining using weaker labels and then fine-tuning on stronger clinical labels.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/591064dc81d94c027cbe8ed308f60713" />