Sure, the best example is the in vivo CAR T paper (also from Penn, below) for targeting heart failure models in mice. Anti-CD5 fuses to LNPs with CAR mRNA. it works because targeting circulating cells with a highly type specific antibody is relatively easy (CD5 is great)
I am not confident the delivery problem will be solved anytime soon. People have been working on this for recombinant proteins,DNA, RNAi, TALENs, CRISPR/Cas systems… since the 80s. Both specificity and efficiency are still struggles for lots and lots of cell types. Solid tumor cancer is probably the worst system by far - poor vascular, reduce or no immune infiltration, massive genomic and proteomic heterogeneity….
Hope I am wrong!