Many diseases are heterogeneous by nature. When designing a drug target, it is extremely difficult to treat a big population with a single targeted drug or even a combination of drugs. The most recent data shows that the aging population (≥50 years old) is accountable for 77% of deaths globally. The major cause of death was related to age-associated diseases such as cardiovascular diseases, cancer, diabetes, and other neurodegenerative disorders. This suggests that aging plays an important role in different diseases.

A substantial percentage of the human clinical trials, including those evaluating investigational anti-aging drugs, fail in Phase II where the efficacy of the drug is tested. This poor rate of success is in part due to inadequate target choice and the inability to identify a group of patients who will most likely respond to specific agents. This challenge is further complicated by the differences in the biological age of the patients, as the importance of therapeutic targets varies between age groups. Unfortunately, most targets are discovered without considering patients' age and are tested in a relatively younger population (average age in phase I is 24 years). Hence, identifying potential targets that are implicated in multiple age-associated diseases which also play a role in the basic biology of aging, may have substantial benefits.

The ultimate goal of geroscience is the extension of lifespan by extending healthspan, the period of lifetime free from diseases and disabilities caused by aging. As a result, it is important to look for dual-purpose targets that are implicated in aging and disease at the same time to extend healthspan and delay age-related health issues, a philosophy envisioned by the CEO of Insilico Medicine. Even if the target is not the most important for a specific patient, the drug would still benefit that patient.

PandaOmics is an AI-enabled biological target discovery platform. It utilizes advanced deep learning models and AI approaches to predict the target genes associated with a given disease through a combination of Omics AI scores, Text-based AI scores, Finance scores, and Key opinion leader (KOL) scores, and is currently being employed in both academic and industry settings. The algorithm also allows the prioritization of protein targets for their novelty, confidence, commercial tractability, druggability, safety, and other key properties that drive target selection decisions. The integrated omics database consists of a vast amount of published systems biology data, spanning over 1,500 diseases and 10,000 disease subtypes. The database includes approximately 1.9 trillion data points derived from over 10 million samples using microarrays, RNA sequencing, proteomes, and methylomes, among other data types. PandaOmics' text database embeds information from over 40 million documents, including patents, grants (that amount to over $2 trillion in funding), publications, clinical trial results, and company reports, among other text-based sources. PandaOmics has various competitive advantages with respect to the integrated deep learning-based algorithms, time machine validation approach, comprehensive database, and intuitive user interface. It enables biologists and clinicians to identify confidence targets for drug repurposing and nominate novel targets for further investigation in a short period of time.

To identify potential dual-purpose targets that are implicated in both aging and age-associated diseases, Insilico Medicine used PandaOmics to perform target identification for 14 age-associated diseases (AADs) and 19 non-age-associated diseases (NAADs) across multiple disease areas. By comparing the Top-100 targets discovered from AADs to those from NAADs, age-associated diseases targets and common targets were identified. Targets associated with aging were allocated into corresponding aging hallmark(s) based on their biological functions and roles in regulating important aging pathways from literature. In addition, targets that were consistently dysregulated in two or more disease classes (fibrotic, inflammatory, metabolic, and neurologic) in a unidirectional manner were selected for further druggability and safety assessment for target selection. This approach was validated by the significant enrichment of identified aging-associated genes in the current study when compared with a pool of well-known aging-associated genes curated from clinical trials, publications, geroprotectors, and GenAge.

Study Workflow. Transcriptomic datasets of age-associated diseases (AADs) and non-age-associated diseases (NAADs) were retrieved from public repositories and processed by PandaOmics. Genes were ranked in the corresponding disease with over 20 artificial intelligence and bioinformatics models ranging from Omics AI scores, Text-based AI scores, Finance scores to KOL scores. A list of potential targets with dual-purpose for aging and age-associated diseases was proposed by evaluating their association with aging hallmarks, expression profiles, mechanism of action, and safety.

Upon the hallmarks of aging assessment, 145 genes were considered as potential aging-related targets and mapped into corresponding aging hallmark(s). As shown in the target wheel, there are 69 high confidence targets with high druggability, 48 medium novel targets with high or medium druggability, and 28 highly novel targets with medium druggability. A set of well-known aging-associated genes identified are part of the mTOR, insulin/IGF, and sirtuin family signaling, counteracting multiple aging hallmarks to delay the aging process or to extend lifespan, strongly supporting the validity of this approach for aging target identification. The most frequently associated aging hallmark was inflammation, which is in line with the view that inflammation is associated with multiple age-related diseases and is an intrinsic and major component of the aging process. In addition, a recently proposed hallmark of aging, extracellular matrix stiffness, was associated with 30 target genes identified by PandaOmics suggesting it might play an important role in aging. Further pathway enrichment analysis revealed the AI-derived targets crosstalk with multiple key aging-associated pathways, such as the MAPK, PI3K-AKT, and FOXO signaling networks. A list of potential therapeutic dual-purpose aging targets for drug discovery was disclosed in the paper.

AI-derived targets associated with hallmark(s) of aging. Top-ranked targets in age-associated diseases were extracted under 3 novelty settings (high confidence, medium, and high novelty). The loss of novelty would be a trade-off for high confidence, the abundance of evidence connecting a target to a disease. Druggability indicates the accessibility of a target protein by small molecules. Target names are labeled in blue for age-associated targets, and black for common targets. Targets annotated as cancer driver genes in the NCG7.0 database are underlined.

Developing interventions that target aging and multiple age-associated diseases could result in unprecedented health benefits through extending not only lifespan but also healthy life expectancy. It highlights the importance of dual-purpose targets. The current study also demonstrated the power of PandaOmics to identify novel dual-purpose targets not only for specific disorders but across multiple types of diseases in a cost-saving and time-efficient manner.