The Year That Changed Everything
2025 wasn't just another year of incremental progress in biotech. Things actually broke through. Real patients got real treatments. Tools that had been "five years away" for a decade suddenly weren't.
The first gene-edited baby received personalized CRISPR treatment for a rare liver disease. Baby KJ was born with a deficiency that prevented his liver from processing ammonia, a condition that's usually fatal. He got custom-designed base editing therapy delivered through lipid nanoparticles. He's walking and thriving now. That's not a projection or a pilot study. That happened.
Bridge recombinases emerged as a serious alternative to CRISPR for large DNA insertions. CRISPR is great at cutting and small edits, but inserting big sequences has always been spotty. These natural bacterial enzymes can insert sequences up to a million base pairs long, entire gene clusters, not just single genes. If that holds up in human cells, it opens doors CRISPR can't.
Roche set a world record by sequencing and analyzing a complete human genome in under four hours. We went from needing months in the 2000s to days in the 2010s to hours now. That's not a novelty. It means hospitals can actually use genome sequencing as a diagnostic tool rather than a research exercise.
AI-designed proteins moved from papers to reality. Tools like RFDiffusion and AlphaFold are being used to create entirely new enzymes: proteins that don't exist in nature but catalyze reactions we want. And weight loss pills entered late-stage trials, with Eli Lilly's orforglipron showing people losing over 27 pounds with a daily pill instead of injections.
What's Coming in 2026
Personalized medicine is the obvious next step. Baby KJ's treatment isn't a one-off. It proved that N-of-1 therapies (treatments designed for a single patient) can actually work. The FDA is setting up frameworks to speed approval for these approaches. If you have a rare genetic disease, this might be the year custom gene therapy becomes a real option.
AI-driven drug discovery is going to accelerate hard. AlphaFold 3 and similar models are already in use at every major pharmaceutical company. They predict how drugs interact with proteins, DNA, and RNA. Expect the first AI-designed drug candidates to enter clinical trials this year. Companies are racing to prove AI can actually deliver new medicines, not just papers.
Genome sequencing is moving from research labs into clinical reality. With sub-4-hour turnaround times, hospitals can diagnose genetic diseases in real time. Newborns with mysterious symptoms can get answers before it's too late, cancer patients can have their tumors sequenced to find the exact mutation driving their disease. This changes how medicine actually gets practiced.
CAR-T therapies are expanding beyond cancer. Originally a blood cancer breakthrough, researchers are now testing CAR-T for autoimmune conditions like lupus and multiple sclerosis. Trials for stiff person syndrome could produce the first FDA-approved CAR-T therapy for a non-cancer condition.
If FDA approvals go through, daily weight loss pills could hit the market in 2026. Not just orforglipron: multiple candidates from different companies. This could reshape how we treat obesity, diabetes, and metabolic disease at scale.
Spatial biology tools from Illumina and others are letting researchers see which genes are active in specific cells within tissues. In cancer, where the location of cells matters as much as their genetics, this matters a lot. And multi-omics, combining genomics, proteomics, and metabolomics into unified models, is becoming standard practice. Single data types miss things. Combined data types catch disease mechanisms you'd never find otherwise.
My Take
2025 was genuinely exciting. I don't say that lightly. Most "breakthrough" years in biotech are mostly hype. But baby KJ's base editing therapy actually worked in a human. Bridge recombinases actually inserted meaningful amounts of DNA. Roche actually sequenced a genome in four hours. These aren't press releases. They happened.
But I think the harder problems are the ones nobody wants to talk about. Funding for global health programs got slashed in 2025, and that disrupted clinical trials and treatment programs worldwide. The biotech sector is still recovering. Breakthrough therapies cost hundreds of thousands of dollars. Making them accessible, not just possible, is the real challenge, and almost nobody is working on that with the same urgency.
The ethical questions around gene editing, AI-designed drugs, and personalized medicine are still unresolved. As these tools get more powerful, the gap between what we can do and what we should do gets wider. I think 2026 is going to be defined less by what's technically possible and more by whether we can build the infrastructure (regulatory, financial, ethical) to actually deliver on the promise.