A new study maps existing drugs to the hallmarks of aging
A new study maps existing drugs to the hallmarks of aging by placing 2,358 longevity-linked genes on the human protein interactome and scoring 6,442 DrugBank compounds for both network proximity and a directionality metric called pAGE. The Northeastern and Harvard team published the SHARP pipeline in Nature Aging, offering a faster route to longevity repurposing candidates.
Key Takeaways
- Researchers mapped 1,250 aging genes onto hallmark modules within a network of 500,000-plus protein interactions.
- The SHARP pipeline combines network proximity with a pAGE metric to flag drugs that may help or harm specific aging hallmarks.
- Of 6,442 screened compounds, 370 sit close enough to hallmark modules to be candidates, including 83 indirect network drugs.
- Validation against mouse longevity trials and human aging studies supported the approach, though clinical proof remains years away.
- Related work on primate-specific RNA LINC01021 shows why senescence direction matters when screening drugs.
Why does a study that maps existing drugs matter for longevity?
Testing whether a drug slows human aging would take decades, and regulators still expect single-disease indications. Yet thousands of approved medicines already exist. A method that predicts longevity effects from existing biology could unlock hidden potential without starting from scratch.
That is the promise behind work led by Bnaya Gross and Albert-László Barabási at Northeastern, with colleagues including Vadim Gladyshev and Joseph Loscalzo, published in Nature Aging. Their network-medicine framework asks which drug targets sit near aging gene neighborhoods and whether those drugs push gene expression in a youthful or aged direction.
How does the SHARP pipeline score drugs against aging hallmarks?
The team started with the OpenGenes database, linking 2,358 genes to aging and longevity at varying confidence levels. Of those, 1,250 could be assigned to at least one hallmark of aging, from DNA stability to intercellular communication. Many genes span multiple hallmarks; TP53 touches seven.
Those genes were placed on the human interactome, a map of more than 500,000 experimentally supported protein interactions. Genes for each hallmark clustered into distinct modules, much like disease neighborhoods in prior network-medicine work on asthma, heart disease, and COVID-19.
Researchers then screened 6,442 DrugBank compounds, measuring each drug's network proximity, the shortest-path distance from its protein targets to hallmark genes. Proximity alone was not enough: some nearby drugs worsened hallmarks such as cellular senescence. So the team added pAGE, a transcription-based score that checks whether drug-induced expression shifts counteract age-related changes.
Together, proximity and pAGE form SHARP, the Systematic Hallmark-based Aging Repurposing Pipeline. The method flagged 370 drugs proximal to at least one hallmark, including 83 network drugs that do not directly target any known aging gene but still perturb hallmark modules through network topology.
Which drugs did the analysis highlight, and what comes next?
Validation offered early support. All eight Intervention Testing Program mouse lifespan winners with interactomic data showed positive pAGE for at least one hallmark, while fewer than half of ITP failures did. Among 17 compounds in human longevity trials, 11 had significant proximity. Aspirin mapped to six hallmarks; dasatinib to five; rapamycin to one, intercellular communication.
The researchers also tested predictions against 10 compounds from a parallel Gladyshev-led study that appeared after their forecasts, the closest thing to a prospective check. Mechanistic follow-up on oxymetazoline, a decongestant and rosacea treatment, showed how SHARP traces specific longevity-related pathways.
Some pro-longevity candidates helped certain hallmarks while harming others, a reminder that trade-offs matter. The same week, separate research on the primate-specific lncRNA LINC01021 showed how senescence can worsen when non-coding RNA disrupts RBMX and p53 signaling, underscoring why directionality counts.
None of this means a longevity pill is imminent. SHARP generates testable hypotheses, not prescriptions. For more context on aging science and biohacking trends, see our Longevity and Biohacking coverage. The real win is speed: screening thousands of existing drugs against eleven interconnected hallmarks in silico, then letting labs and trials decide which candidates deserve a closer look.