How siRNA Drugs Silence a Gene: The GalNAc-Liver Delivery Story

Small interfering RNA (siRNA) drugs feed a cell a short RNA that guides the cell's own machinery to chop up one specific messenger RNA, silencing a single gene before it can make its protein. A sugar tag called GalNAc lets these molecules bind a receptor found almost exclusively on liver cells, concentrating the drug where the target lives. That pairing, a precise silencing mechanism plus an address label for the liver, is why several of these drugs are now approved and dosed only a few times a year.

The silencing mechanism, step by step

RNA interference is a natural process cells use to regulate their own genes, described by Andrew Fire and Craig Mello in work that earned the 2006 Nobel Prize. A gene is transcribed into messenger RNA (mRNA), and the mRNA is read by ribosomes to build a protein. If you can destroy the mRNA, the protein never gets made.

A therapeutic siRNA is a short double-stranded RNA, roughly 21 to 23 base pairs. Once inside the cell, one of its two strands, the guide strand, is loaded into a protein assembly called the RNA-induced silencing complex, or RISC. The guide strand acts as a search string: it base-pairs with any mRNA carrying the matching sequence, and RISC then cleaves that mRNA so it can no longer be translated. Because RISC is catalytic, a single loaded complex can cut many mRNA molecules over time. That catalytic recycling is the reason a dose given today can keep a gene quiet for months, a point worth holding onto when durability claims come up later.

The appeal of this approach is that it targets the message rather than the protein. A small-molecule drug usually has to fit into a pocket on a protein, and many proteins have no good pocket. An siRNA only needs the target's genetic sequence, which in principle makes previously "undruggable" targets reachable.

Why GalNAc points the drug at the liver

Naked siRNA is a fragile, negatively charged molecule that the body clears quickly and that cannot cross cell membranes on its own. Getting it into the right cells intact was the central obstacle for two decades. Two solutions reached patients. The first approved siRNA drug, patisiran (Onpattro, 2018, for hereditary transthyretin amyloidosis), wrapped the RNA in a lipid nanoparticle. The second and now dominant approach is GalNAc conjugation.

GalNAc, N-acetylgalactosamine, is a sugar that binds the asialoglycoprotein receptor (ASGPR). ASGPR is expressed at very high density on hepatocytes, the main functional cells of the liver, and at negligible levels elsewhere. When several GalNAc groups are chemically stitched onto an siRNA, the conjugate is grabbed by ASGPR and pulled into the hepatocyte by receptor-mediated endocytosis. The receptor then recycles back to the cell surface within minutes, ready to ferry in more drug. The 2022 review in Frontiers in Pharmacology describes this selective ASGPR binding and rapid internalization as the core of GalNAc's liver specificity.

This is also the honest limit of the platform. GalNAc delivers to the liver, and to little else. It is elegant for diseases driven by a protein the liver makes, and largely irrelevant for a target in the brain, muscle, or kidney. Much current research is about finding equivalent ligands for other tissues.

The approved drugs and what they silence

GalNAc-conjugated siRNAs approved to date each shut down a liver-expressed gene: givosiran (Givlaari, 2019) lowers ALAS1 for acute hepatic porphyria; lumasiran (Oxlumo, 2020) lowers HAO1 for primary hyperoxaluria type 1; inclisiran (Leqvio) lowers PCSK9 to reduce LDL cholesterol; and vutrisiran (Amvuttra) lowers transthyretin (TTR). Later entrants such as nedosiran extended the list. The pattern is consistent: pick a harmful protein the hepatocyte produces, and switch off its message at the source.

Inclisiran is the clearest illustration of durability. In the ORION-11 trial, reported in the European Heart Journal, subcutaneous inclisiran given at day 1, day 90, and then every six months produced sustained, placebo-corrected LDL cholesterol reductions of roughly 44 percent. Two injections a year is a direct consequence of catalytic RISC and stable GalNAc chemistry. Vutrisiran, dosed once every three months, was approved for TTR polyneuropathy and, based on the HELIOS-B trial, received an expanded FDA indication in 2025 for transthyretin amyloid cardiomyopathy, where Alnylam reported reductions in a composite of cardiovascular death and events. An expanded label is a neutral legal statement of an approved use, not an endorsement, a recommendation, or a safety all-clear for any individual. These are meaningful outcomes in defined trial populations; effect sizes come from specific enrolled patients and do not automatically transfer to everyone.

How to appraise durability and off-target claims

Two questions separate substance from marketing for this drug class.

First, durability. Long dosing intervals are real, but ask what endpoint the interval is anchored to. A drug can suppress a biomarker like LDL cholesterol or serum TTR for months; whether that translates into fewer clinical events is a separate, harder question answered only by outcome trials such as HELIOS-B. Biomarker durability and clinical benefit are not the same claim.

Second, off-target effects. The guide strand's "seed region," positions roughly 2 through 8, can partially match the untranslated regions of unintended mRNAs and silence them the way a microRNA would. Chemists blunt this with modifications, 2'-fluoro and 2'-O-methyl sugar changes plus phosphorothioate linkages, which also protect the RNA from degradation. When you read that a candidate is "highly specific," the useful follow-up is whether seed-region off-target activity was actually screened, not assumed away.

This article is educational and is not medical advice; decisions about any specific therapy belong to a patient and their own clinician. Read as a class, GalNAc-siRNA drugs are a genuine advance for liver-driven disease, and also a reminder that a clever delivery trick is only as good as the outcome data behind each individual product.