LongeCityNews Last Updated: 03 April 2026 - 11:48 PM

Cells in aged tissues suffer a range of biochemical dysfunctions; broken proteins, altered structures, leakage of materials from one compartment to another. Many of these issues provoke the cell into inflammatory reactions. A range of sensors operate in every cell, triggered by different forms of damage and stress characteristics of aging, and converging on the activation of regulators of inflammatory signaling. One example is the interaction between cGAS and STING. cGAS acts to detect the presence of DNA in the cell cytosol, an evolved defense against infectious pathogens. Unfortunately it is maladaptively triggered by leakage of fragments of DNA from the cell nucleus or mitochondria, a feature of cells in aged tissues. cGAS then interacts with STING to produce inflammatory signaling.

Today's open access paper is interesting for the discussion of what exactly might be done about unwanted cGAS-STING interactions in aging tissues. The focus is the aging of the ovary, but this is a problem that occurs throughout the body. The primary challenge in attempting to suppress unwanted inflammation is that control of unwanted inflammation runs through the same pathways as control over desirable inflammation. Known approaches to interfere in the regulation of inflammation and inflammatory signaling shut down both excessive and necessary inflammation, resulting in undesirable side effects. But perhaps there can be better ways forward, the ability to better distinguish between these modes of activation. As yet there are only hints in early stage research that this can be possible, however.

The inflammatory clock: how cGAS-STING ticks in the aging ovary

Premature ovarian insufficiency (POI) is more than a fertility issue; it's a silent epidemic of accelerated systemic aging in young women, with current treatments failing to address its root cause. For too long, the relentless decline of ovarian function has been viewed as an inevitable mystery. But what if the ovary holds an internal "inflammatory clock," ticking away with each cellular insult and dictating the pace of its own decline? Here, we spotlight a surprising culprit: the cGAS-STING signaling pathway. Far beyond its day job in antiviral defense, this pathway emerges as a master integrator of ovarian aging.

We reveal how stresses like DNA damage and mitochondrial dysfunction leak genetic material into the cell's interior, where cGAS-STING sounds a relentless alarm. This alarm does not just trigger inflammation; it initiates a vicious, self-amplifying cycle of cellular senescence, tissue fibrosis, and follicle destruction - a cycle that may explain why ovarian aging often feels like a one-way street.

Therapeutically, we move beyond mere symptom management to explore strategies for resetting this inflammatory clock. We dissect both direct "brakes" - novel small molecules that silence cGAS or STING - and upstream "shields" that protect mitochondria and genome integrity. Most provocatively, we introduce the concept of "signal reprogramming": not just shutting down the pathway, but cleverly rewiring its output to favor repair over destruction. By repositioning cGAS-STING from a simple sensor to the central processor of ovarian aging, this review charts a course for a new class of therapeutics aimed at preserving ovarian function, not just managing its loss.

In oncology models, persistent STING activation has been shown in certain settings to promote an immunosuppressive microenvironment; notably, co-administration of a TLR2 agonist was reported to "reprogram" STING downstream signaling by enhancing NF-κB activity while attenuating IRF3-dependent interferon responses, thereby overcoming therapeutic resistance. This oncology-informed framework provides a conceptual basis for cautiously exploring whether selective downstream signaling modulation, rather than global pathway inhibition, could theoretically attenuate chronic inflammation while permitting adaptive tissue responses in ovarian aging models.

Building on two previous deals between the companies, this new agreement is potentially worth up to $2.75 billion and involves Lilly licensing assets from Insilico’s pipeline.

More than software

Earlier this week, the AI-driven drug discovery company Insilico Medicine announced a large-scale collaboration with the pharmaceutical giant Eli Lilly. The deal is worth up to $2.75 billion, with $115 million upfront, plus milestones and tiered royalties, making it one of the largest deals of its kind in the longevity space. However, numbers are not the whole story. What makes this notable is the apparent shift from Lilly using Insilico’s tools to licensing its drug programs.

The core of Insilico’s business is its software suite for drug discovery automation, the Pharma.AI platform, which the company touts as the most comprehensive out there, covering the entire process starting with target identification. Insilico, founded by Alex Zhavoronkov in 2014, claims that it collaborates with 13 of the top 20 largest global pharma companies by 2024 sales and that Pharma.AI can speed up drug discovery times significantly. For instance, the company says that its IPF program, now known as rentosertib, which it describes as the first wholly AI-discovered and AI-designed small-molecule drug, went from project start to preclinical candidate in about 18 months and to Phase 1 in under 30 months.

The relationship between Insilico and Lilly started in 2023 with a software licensing deal and was followed by a larger 2025 research collaboration. In parallel, Insilico has been developing its own drug pipeline, where it either advances programs itself or partners them out. Apparently, part of the current deal is for Lilly to buy into assets from Insilico’s pipeline, which would be a stronger form of validation and a marker of growing trust between the two companies.

“The agreement grants Lilly an exclusive worldwide license for the development, manufacturing, and commercialization of potentially best-in-class, novel oral therapeutics in preclinical development for certain indications,” the press release says. “In addition, Insilico and Lilly will collaborate on multiple R&D programs focused on targets selected by Lilly, by combining Insilico’s state-of-the-art Pharma.AI platforms with Lilly’s development capabilities and deep disease-area expertise.”

The companies have not publicly disclosed the exact number of licensed assets, the targets, or the disease areas. Some reporting has suggested that a GLP-1-related asset may be part of the deal, but that has not been confirmed publicly.

Experts weigh in

“The Insilico–Eli Lilly deal marks a turning point for AI in drug discovery,” said Garri Zmudze, co-founder of LongeVC, an early investor in Insilico. “Having followed Alex Zhavoronkov for years, his level of commitment and work ethic has been exceptional, and this milestone feels well deserved. This is a landmark moment for AI in biotech, because it proves that AI-driven platforms can consistently translate science into commercial partnerships. Alex’s relentless dedication over many years has played a key role in making this possible.”

Alexey Strygin, longevity entrepreneur and early Insilico team member, shares the same enthusiasm: “This collaboration validates what those of us who were there early have always believed – that AI-driven drug discovery would eventually earn the trust of the world’s largest pharma companies. Deals like this grow Insilico’s war chest and valuation, and Alex is already allocating those resources toward the aging cause, both internally (the company is hiring longevity researchers) and as an angel investor in bold new ventures (including biostasis).”

Karl Pfleger, longevity investor and creator of AgingBiotech.info, has a more nuanced view on whether Insilico is a “true” longevity company: “Insilico is unique among the many AI-driven-drug-discovery (AIDD) companies in having a nontrivial focus on aging and being led by someone clearly passionate about aging. On the one hand, there’s much more money in AI & AIDD than in aging, so considering any AIDD companies to be part of the aging field can skew the numbers, because even Insilico’s pipeline is largely cancer and non-aging related. On the other hand, Insilico really is special, as evidenced again recently by its saving of the ARDD conference. Other aging biotechs have been increasingly making big-pharma deals, including with Lilly, but this new deal is by far the largest in the aging sector if we consider it to be in that sector. It takes the total value of announced deals with aging biotechs from at least $8.5 billion to over $11 billion based on the data in AgingBiotech.info/companies.”

We asked Alex Zhavoronkov a few questions in a flash interview:

What changed between Lilly’s earlier work with Insilico and this larger licensing deal?

Our relationship with Eli Lilly has evolved from tools to collaboration to assets. We began with an AI software licensing agreement in 2023, expanded into a research collaboration in 2025, and now this latest deal reflects a shift to licensing actual drug candidates. This progression demonstrates growing confidence not just in the platform, but in the output of the platform.

Should we see this as validation not just of your AI platform, but of actual drug assets?

Yes – this is validation of both. This agreement gives Lilly exclusive rights to develop and commercialize specific AI-discovered drug candidates, not just access to the technology. That represents a meaningful shift, as large pharma is now betting on AI-generated assets entering the pipeline.

Your release refers to a portfolio of oral therapeutics; can you give details?

While we cannot disclose specific molecules, the deal includes preclinical-stage oral therapeutics across selected disease areas. These were discovered using our end-to-end AI platform and are designed to address high-value, high-unmet-need indications, with Lilly leading downstream development and commercialization.

There’s speculation about metabolic or GLP-1-related assets; can you comment?

We do not comment on specific assets, but public reporting suggests a GLP-1–related program may be part of the broader portfolio licensed. More broadly, we are active across metabolic disease, and our platform is well suited to identifying targets relevant to multiple diseases simultaneously.

More broadly, where do you think AI-driven drug discovery stands today – what is working, what is still overhyped, and what should we realistically expect in the next few years?

We are transitioning from AI hype to real-world execution. AI can now generate viable targets and molecules and move them into pipelines, significantly compressing early discovery timelines. However, fully autonomous drug development remains overhyped. In thecoming years, we expect more AI-designed drugs entering clinical trials and more partnerships shifting toward asset-level deals.

Researchers here present an interesting view of heart failure and atrial fibrillation, providing evidence for both to be manifestations of reduced TBX5 expression. TBX5 is a transcription factor, and thus influences expression of a very large number of genes; transcription factors are thus often central points of regulation for cell and tissue behavior. The evidence suggests that the present quite distinct treatments for atrial fibrillation and heart failure could potentially be replaced in the future by forms of therapy that increase TBX5 expression, a point of intervention that is upstream of present targets.

Heart failure occurs when the heart muscle is damaged and unable to pump enough nutrient-rich blood to meet the body's needs for oxygen. Heart failure is usually evaluated in the heart's lower chambers, called ventricles, which provide most of the pumping power. Atrial fibrillation is an arrhythmia - an irregular heart rhythm - that originates in the heart's upper chambers, known as the atria. During atrial fibrillation, the heart beats too fast, resulting in a lower blood flow to the body and a higher risk for clots or stroke. Epidemiologists have observed that these two conditions aren't independent of one another: People with heart failure are much more likely to have atrial fibrillation, and vice versa. Patients' outcomes also tend to be worse when they have both conditions.

New research was guided by a past study in which scientists created a mouse model by "turning up" a gene linked to human heart failure in the mouse heart. "We expected to get a heart failure mouse model, but instead we got an atrial fibrillation model. That observation put us on the right path." This focused attention on a gene called TBX5. TBX5 is a transcriptional regulator - a protein in the cell nucleus that controls which genes are turned on or off at a given time. When TBX5 levels are decreased in the atrium, it disrupts the normal gene expression needed to maintain a stable heart rhythm.

Zeroing in on transcriptional responses, the researchers compared different mouse models of heart failure and atrial fibrillation, finding that an atrial fibrillation model created by removing TBX5 from the atria actually creates gene expression changes almost identical to those seen in heart failure. "That made us think that diminished TBX5 may be important in heart failure. So, we looked at human gene expression data, and lo and behold, TBX5 was very downregulated in the atria of patients with heart failure, but not the ventricles."

Further analysis revealed that over 100 other transcription factors - proteins that regulate gene expression - were altered in both the heart failure and TBX5-deficient atrial fibrillation models. Almost all the key transcription factors changed in the same direction in both conditions. Researchers argue that the results should prompt a fundamental shift in how atrial fibrillation is understood. The rhythm disorder seen in atrial fibrillation may be a symptom of underlying atrial muscle dysfunction similar to the ventricular dysfunction in heart failure.

Link: https://www.uchicagomedicine.org/forefront/heart-and-vascular-articles/heart-failure-atrial-fibrillation-same-disease

As for the gut microbiome, the composition of the oral microbiome appears to change with age. The oral microbiome receives less attention than the gut microbiome, but the same scientific tools can be used to correlate specific changes with specific age-related conditions. Here, researchers correlate abundance of specific microbial species with the existing known relationship between periodontal disease and manifestations of neurodegeneration, such as loss of cognitive function. One mechanism that likely contributes to these associations is the contribution of the oral microbiome to chronic inflammation, when microbes and microbial metabolites leak into circulation via damaged gums, but the size of the effect remains debated, and there are other possible mechanisms to consider as well.

Emerging evidence implicates the oral-brain axis in neurodegeneration, yet large community-based studies remain limited. This study aimed to examine associations between periodontal health, oral microbiome, and cognitive performance, and to explore potential biological pathways underlying these relationships. We conducted a cross-sectional analysis of 1,157 participants from the community-based Taizhou Imaging Study, all of whom underwent comprehensive periodontal examinations, salivary microbiome profiling, and cognitive assessments. Periodontal health and microbiome features were treated as exposures, and cognitive performance as the outcome.

Five clinical periodontal indices were found to be inversely associated with cognitive performance. Ten microbial genera (e.g., Haemophilus), 21 functional pathways (e.g., FoxO signalling), and two co-abundance modules, including a Treponema module, were significantly related to cognitive function. Mediation analysis suggested that 11 features, including nitrate-reducing taxa and a Treponema-driven inflammatory module, may partially mediate the relationship between periodontal health and cognition. These community-based findings reveal microbiome-mediated links along the oral-brain axis and highlight periodontal health and oral microbial homoeostasis as potential targets for early prevention of cognitive decline.

Link: https://doi.org/10.1016/j.ebiom.2026.106231