Inside a South San Francisco lab, a molecule is learning restraint.
It is 2026, and somewhere on a benchtop in South San Francisco, an siRNA is being held back. Not stored. Held back. The molecule is technically active - it knows exactly which gene to silence - and yet it does nothing. It is waiting for a signal. A specific cell-state. A particular zip code in the body. Only then will it switch on.
That waiting is the entire idea behind Switch Therapeutics. For two decades, RNA interference - the Nobel-winning method of telling a cell to stop making a specific protein - has been one of biology's most precise tools. And one of its most stubbornly stuck. Stuck in the liver, mostly. Stuck in a handful of tissues that an siRNA can be persuaded to reach. The rest of the body has been, depending on who you ask, either off-limits or unsafe.
Switch wants to change the verb. Not "reach." Activate. Their proprietary platform - Conditionally Activated siRNAs, or CASi - threads a kind of molecular logic gate into the RNA itself. The drug is administered. The drug diffuses. The drug encounters cells. In most of them, nothing happens. In the cells where a biomarker matches, the molecule rearranges and the silencing begins.
If that sounds like a software trick, it is. Si-ping Han, the company's co-founder and CTO, came up through the Caltech labs where nucleic-acid nanotechnology grew up. Dee Datta, the CEO, has done this before - he co-founded Allozyne, a multiple-sclerosis biotech, in a previous life. The two of them, plus a roster of academic founders from Harvard Medical School and City of Hope, are betting that the next decade of RNA medicine won't be about delivery. It will be about discretion.
The first proof of that bet has a name. CASi-APOE. Announced in December 2024, it is the company's first development candidate: a liver-sparing RNAi therapy aimed at APOE, the gene whose APOE4 variant is the largest known genetic risk factor for late-onset Alzheimer's disease. "Liver-sparing" is the polite phrase. The honest phrase is "everywhere RNAi has historically been stuck, this one isn't."
“Switch combines advantageous properties of both single and double-stranded RNAs in a single molecule - allowing for cell-selective RNAi activity.”
— Switch Therapeutics, on the CASi platformWhat, exactly, is a conditionally activated siRNA?
Most siRNAs are double-stranded - sharp little hairpins of RNA, always-on the moment they slip inside a cell. Switch's CASi molecules carry the same payload but in a folded, dormant form. Inside the wrong cell, the structure holds. Inside the right cell - one expressing a specific biomarker - the structure unfolds. Now it's a working siRNA. Same chemistry as nature's. Better aim.
Get in by itself
The single-stranded geometry lets CASi enter cells without lipid nanoparticles - the delivery scaffold that traps most siRNAs in the liver.
Only fire when called
A biomarker inside the cell flips the molecule from dormant to active. Wrong cell, no activation. Right cell, silence the target.
Aimed at the brain
The pipeline targets microglia, astrocytes, and neurons - the cell populations where unconditional knockdown has always been too risky.
Four bets, one platform.
A $52 million launch - and a syndicate worth reading.
Investor Syndicate (Series A, 2023)
What $52M Buys in Preclinical Biotech
Approximate allocations — illustrative.
Two co-founders. One academic brain trust.
Dee Datta
A repeat biotech founder — previously co-founded Allozyne, a multiple-sclerosis biotech. At Switch, he turned an academic platform into a syndicate-backed pipeline company in under three years.
Si-ping Han, Ph.D.
RNA-nanotechnology researcher whose Caltech work seeded the CASi platform. He designed the conditional architecture that lets one molecule act like both a sensor and a switch.
Academic founders and SAB include researchers from Caltech, Harvard Medical School and City of Hope. In December 2024, Switch added Alzheimer's researcher David M. Holtzman, M.D. — one of the most cited names in neurodegeneration — to its Scientific Advisory Board.
From a Caltech bench to a candidate molecule.
Who Switch is for - and how to engage.
Watch the APOE program
If you follow Alzheimer's research, CASi-APOE is the candidate to track. Selective APOE knockdown without hepatic exposure would be a meaningfully different shape of intervention.
A platform, not a single shot
Switch isn't selling one drug. The CASi chemistry plausibly applies to dozens of targets currently considered undruggable by RNAi.
Read the December 2024 update
The CASi-APOE announcement is the clearest signal yet of where Switch thinks platform meets product. Worth the time.
Where to read further.
Back at the bench.
The molecule is still on the benchtop. It hasn't moved. But the room around it has changed.
Three years ago this lab was a corner of a shared incubator. Now it's a company with its own walls, its own pipeline, and an Alzheimer's researcher of David Holtzman's caliber paying attention. The molecule waits the same way it did before - dormant, patient, conditional. The difference is what waits with it. A candidate name. A regulatory plan. A syndicate. And, eventually, a patient whose own cells will write the trigger that tells this little folded thing to do its work.
Switch Therapeutics is, in the end, a company about timing. The right gene. The right cell. The right moment. Get one of those wrong and RNAi has always punished you for it. Get all three right and you've built something the field has been chasing since 2006.
The bench is quiet. The molecule is ready. That's the whole story, really.