Targeted lipid nanoparticles use a surface ligand or a changed lipid mix to favour particular cells or organs. Enrichment in a tissue is not the same as working protein made only in the intended cell type.
Two strategies are separate. A ligand strategy adds a binding molecule such as an antibody to the particle surface so it recognises a marker on intended cells; binding must still lead to uptake and usable release, and the marker may sit on other cells. A formulation strategy changes the lipids: selective organ targeting (SORT) added a supplemental lipid to otherwise standard LNPs and, after intravenous injection in mice, shifted delivery toward lung, spleen or liver; with a Cre reporter, the lung formulation edited about 40 per cent of epithelial and 65 per cent of endothelial cells, the spleen formulation 12 per cent of B cells and 10 per cent of T cells, and the liver formulation 93 per cent of hepatocytes (Cheng and colleagues, as reported). These are measured results in mice, not a rule connecting a composition to an organ in people.
A CD5-directed example shows both the aim and its limit. CD5-targeted LNPs carrying modified mRNA for a chimeric antigen receptor against fibroblast activation protein were tested in activated mouse T cells in culture (81 per cent expressed a GFP reporter and 83 per cent the receptor, against 7 per cent with isotype-control antibody particles) and then injected intravenously into mice with angiotensin II and phenylephrine induced cardiac injury. Receptor expression appeared in each major T-cell subset by flow cytometry (87 per cent of receptor-positive T cells were CD4-positive), bioluminescence showed spleen targeting only with CD5 particles, and liver expression occurred with both CD5 and control particles, so an organ-level signal and the identity of the expressing cells were measured separately; treatment reduced fibrosis and restored cardiac function in these mice (Rurik and colleagues, author manuscript). The delivery address was CD5 and the encoded receptor recognised a different target. This was a fibrosis model, not a cancer trial, and the authors disclose patents on the platforms.
A later study used targeted LNPs to deliver mRNA to specific T-cell subsets; its abstract reports reprogrammed CD8-positive T cells in samples from healthy donors and people with autoimmune disease, tumour control in humanised mice and B-cell depletion in cynomolgus monkeys (Hunter and colleagues; abstract only, full text not retrievable). Each of those is a different experiment in a different setting, and none is evidence of benefit against solid tumours in people.
Enrichment is not exclusivity. In a Cre-reporter experiment recombination appeared in 81.1 per cent of CD4-positive and 75.6 per cent of CD8-positive splenocytes but also in 15.0 per cent of CD3-negative cells, mainly B cells, dendritic cells and macrophages (Rurik and colleagues), and the 87 per cent figure above is a share of receptor-positive cells, not the share of CD4 T cells that expressed the receptor, so the denominator matters. Cargo type also matters for how long the effect lasts: RNA encoding a receptor produced transient CAR T cells in that study, whereas RNA encoding a genome editor can leave a lasting DNA change, as when SORT lipid nanoparticles delivered Cas9 mRNA and guide RNA to edit cells in mice (Cheng and colleagues) and Cas9 mRNA with an AAV repair template knocked a CAR into T cells in culture (Eyquem and colleagues).
Organ signal, cell identity, protein function and clinical effect need separate measurements, with matched untargeted controls and identified intended and unintended recipient populations.
Showing the technology this term belongs to: In vivo CAR-T.
Shares In vivo CAR-T and the tag delivery-vehicle.
Shares In vivo CAR-T and the tag delivery-vehicle.
Shares In vivo CAR-T and the tag delivery-vehicle.
Shares RNA-lipoplex, mRNA and lipid nanoparticle manufacturing.
Every source this page cites, grouped by who published it. Grouping is read from the web address, not assigned by hand.
A source that has moved or gone is reported by the weekly link check. /audit/