Healthtech News and Headlines
  • All Technology
  • AI
  • Autonomy
  • B2B Growth
  • Big Data
  • BioTech
  • ClimateTech
  • Consumer Tech
  • Crypto
  • Cybersecurity
  • DevOps
  • Digital Marketing
  • Ecommerce
  • EdTech
  • Enterprise
  • FinTech
  • GovTech
  • Hardware
  • HealthTech
  • HRTech
  • LegalTech
  • Nanotech
  • PropTech
  • Quantum
  • Robotics
  • SaaS
  • SpaceTech
AllNewsDealsSocialBlogsVideosPodcastsDigests
NewsDealsSocialBlogsVideosPodcasts
HealthtechNewsEvery Cure and Computational Pharmacophenomics: A New Field of Medicine
Every Cure and Computational Pharmacophenomics: A New Field of Medicine
HealthTechPharmaBioTech

Every Cure and Computational Pharmacophenomics: A New Field of Medicine

•February 24, 2026
0
healthcare.digital
healthcare.digital•Feb 24, 2026

Why It Matters

By unlocking therapeutic potential of existing medicines, Every Cure can deliver affordable treatments for millions of patients overlooked by the blockbuster drug model, reshaping drug discovery economics and global health equity.

Key Takeaways

  • •75% of diseases lack FDA‑approved treatments
  • •Every Cure uses AI to repurpose FDA‑approved drugs
  • •MATRIX predicts efficacy for 75 million drug‑disease pairs
  • •Human‑in‑the‑loop review refines algorithmic predictions
  • •$48.3M ARPA‑H and $60M TED funding support scaling

Pulse Analysis

The pharmaceutical landscape today is defined by a stark imbalance: while genomic tools can identify roughly 18,000 distinct human diseases, more than three‑quarters of them have no FDA‑approved therapy. Traditional de‑novo drug discovery costs up to $2 billion and spans a decade, prompting companies to chase blockbuster drugs for affluent markets. Computational pharmacophenomics, pioneered by Every Cure, flips this model by applying artificial‑intelligence and graph‑based analytics to the existing pharmacopeia. By treating every approved molecule as a potential cure, the approach promises to slash R&D expenditures to less than one percent of conventional budgets.

The core of this strategy is the MATRIX platform, a cloud‑native pipeline that ingests billions of data points from knowledge‑graphs such as RTX‑KG2 and ROBOKOP, normalizes them with PySpark, and stores relationships in Neo4j. Machine‑learning models learn vector embeddings from known drug‑disease pairs and extrapolate to the 75 million untested combinations, producing probabilistic efficacy scores. A human‑in‑the‑loop review team evaluates the top predictions, feeding clinical insights back into the model for continuous improvement. Strategic alliances with Google Cloud, Elsevier, and major funders like ARPA‑H and the TED Audacious Project provide the compute power and capital needed to scale this effort.

Early results illustrate the model’s clinical relevance: sirolimus induced long‑term remission in idiopathic multicentric Castleman disease, a carfilzomib‑cyclophosphamide‑dexamethasone cocktail rescued a POEMS‑syndrome patient from hospice, and lidocaine is being tested as an intra‑operative anti‑cancer adjunct. By publicly releasing all 75 million drug‑disease scores by 2026, Every Cure aims to democratize discovery, enabling researchers, physicians, and patient groups to prioritize affordable therapies without patent constraints. If successful, this paradigm could force the broader industry to reconsider the economics of drug development and accelerate equitable access to treatments for rare and neglected diseases.

Every Cure and Computational Pharmacophenomics: A New Field of Medicine

Read Original Article
0

Comments

Want to join the conversation?

Loading comments...