3 ms·

Coupling small molecules to antibodies is tricksy :-P. There have been a lot attempts at doing that (the first I recall is from the late 90's but possibly even

by entee 11y ago

Coupling small molecules to antibodies is tricksy :-P. There have been a lot attempts at doing that (the first I recall is from the late 90's but possibly even earlier), but not many have made it into actual clinical practice. Turns out that the linkage is important, the pharmacokinetics can get complicated and getting the proper concentration in the bloodstream can be difficult. That said I've heard people are getting this kind of strategy to work more recently, particularly in oncology. But of course, better understanding and finding an alternative method around the blood brain barrier is absolutely critical to drug development. It's a huge issue for any small molecule treatment for neurological disease, and regardless of how you conjugate or what you conjugate it to (now that we know something about it, maybe a full antibody isn't required, you could maybe get nearly the same effect with a shorter peptide), this is a breakthrough.

daveguy 11y ago
I was thinking more of larger protein-antibody conjugates and I agree it probably doesn't need to be a whole antibody. Still a lot more to learn, but very exciting for central nervous system pharma development.
- dnautics 11y ago
  If you don't use the whole antibody, the serum half life us capped in the knees because the body gets rid of serum proteins without glycosylation. In the case if antibodies, glycosylation is on the heavy chain FC region. The first protein antibody conjugates were to couple ricin to antibodies in the late 80s. Ricin's warhead polypeptide chain conveniently has a single cysteine residue that they tried to conjugate to antibodies using disulfide exchange. Of course this was not kind to the antibodies, and the structural disruption is very likely one of the reasons this strategy failed miserably. In general the challenge for ADC is that attaching your drug to an antibody is nontrivial, and the titer is tricky. How many of your molecule does it require to take down a cell? 100? 1,000,000? Can you deliver that number of molecules to each cell, that start reliably conjugated to a large molecule, after yield losses as a result of molecular decoupling?

by entee 11y ago

daveguy 11y ago

I was thinking more of larger protein-antibody conjugates and I agree it probably doesn't need to be a whole antibody. Still a lot more to learn, but very exciting for central nervous system pharma development.

dnautics 11y ago
If you don't use the whole antibody, the serum half life us capped in the knees because the body gets rid of serum proteins without glycosylation. In the case if antibodies, glycosylation is on the heavy chain FC region. The first protein antibody conjugates were to couple ricin to antibodies in the late 80s. Ricin's warhead polypeptide chain conveniently has a single cysteine residue that they tried to conjugate to antibodies using disulfide exchange. Of course this was not kind to the antibodies, and the structural disruption is very likely one of the reasons this strategy failed miserably. In general the challenge for ADC is that attaching your drug to an antibody is nontrivial, and the titer is tricky. How many of your molecule does it require to take down a cell? 100? 1,000,000? Can you deliver that number of molecules to each cell, that start reliably conjugated to a large molecule, after yield losses as a result of molecular decoupling?