Longevity, genes,
and disease-modifying AI.

FDA approves Ultragenyx Fayuvi for Sanfilippo syndrome Type A

First approved therapy for MPS IIIA — one-time IV gene therapy delivering functional SGSH; FDA notes treated children maintained or improved cognitive function vs untreated; list price $3.95M; U.S. specialized-center availability expected in 30–60 days.

Caveat: Approved rare-disease gene therapy — not a general longevity or aging indication; outcomes and access/pricing are disease-specific. (17 Sep 2026.)

Prime assembly enables large targeted genomic integrations

Boston Children’s / Bauer lab CRISPR method stitches medium-to-large DNA into precise genomic sites without relying on HDR or DSB-driven donors — potential path toward mutation-agnostic / “universal” gene therapies.

Caveat: Preclinical / tool paper — not an approved therapy or human efficacy result; delivery and safety still open. (16 Sep 2026.)

Adenine base editing rescues hereditary motor neuron disease (HMSN-P) in models

CiRA / Tokushima team corrects TFG mutation via AAV-delivered adenine base editing; treated mice showed delayed onset, preserved motor neurons, longer survival; patient iPSC organoids showed less aggregation and neuronal loss.

Caveat: Mouse + organoid evidence only — not human clinical efficacy; safety package still required before clinic. (18 Sep 2026.)

Stanford “virtual biotech” — tens of thousands of AI agents for end-to-end drug discovery

Science paper: agentic biotech analyzed ~50k trials, found cell-type-specific + switch-like (bimodal) targets linked to better Phase advancement / fewer AEs, and independently designed a B7-H3 ADC strategy later mirrored by industry.

Caveat: Computational / retrospective + independent industry coincidence — not a bedside tool or cleared clinical AI; wet-lab validation still required. (17 Sep 2026.)

Anthropic confirms Bay Area wet lab as it ramps AI drug / rare-disease ambitions

Beyond in-silico Claude Science tooling (incl. pharma partnerships), Anthropic confirms physical biology lab work and preclinical focus on conditions industry often skips — AI directing lab automation with human oversight.

Caveat: Company / operational reporting — not a clinical efficacy claim, approved drug, or disclosed disease program with trial data; spokesperson clarified lab is not solely for drug discovery. (18 Sep 2026.)

Research AI ≠ clinical cure claims

Approved rare-disease gene therapy, tool papers, preclinical models, computational discovery, and company wet-lab reporting are different evidence classes. This site keeps them labeled so a headline never collapses into “AI cured aging.”