Maleimide conjugate stability
Maleimide is the default thiol chemistry because it is fast, selective at mild pH and commercially available on everything. The bond it forms is also reversible, and that reversibility is the main stability liability in a cysteine-linked ADC.
The retro-Michael problem
The thiosuccinimide can undergo retro-Michael elimination, regenerating the maleimide and releasing the thiol. In plasma the free maleimide is immediately captured by something else, and the something else is usually albumin, which circulates at high concentration with a reactive free cysteine at position 34.
The consequence is payload transfer from the antibody to albumin. The conjugate loses DAR over days, the payload is redistributed to a carrier with a long half-life and no targeting, and the therapeutic index narrows from both directions at once. This is the mechanism behind much of the off-target toxicity seen with first-generation cysteine conjugates.
Rates depend on the local environment of the conjugation site. A site in a solvent-exposed, positively charged neighborhood exchanges faster. A site in a partially buried or negatively charged pocket is markedly more stable, which is one of the reasons site selection in engineered-cysteine formats is a real design decision rather than a convenience.
Hydrolysis is the cure, not the disease
The thiosuccinimide ring can hydrolyze to the open succinamic acid. Once it opens, retro-Michael is no longer available, and the linkage is locked. A hydrolyzed conjugate is stable.
So the goal is to drive hydrolysis deliberately rather than let it happen slowly in the vial. Two routes.
A mild base treatment after conjugation, holding at around pH 8 to 9 for a defined period, pushes the ring open. It works and it needs care, because the same conditions promote deamidation and disulfide scrambling in the antibody.
Better is to design the linker so it self-hydrolyzes. Placing an electron-withdrawing group adjacent to the maleimide nitrogen accelerates ring opening by orders of magnitude, so the conjugate hydrolyzes within hours of formation under normal conditions. The self-hydrolyzing maleimides are the standard answer now, and they cost nothing in handling.
Confirming it actually happened
Ring opening adds eighteen daltons. That is resolvable by intact or subunit mass spectrometry, and it is the direct readout of whether your hydrolysis step worked. Do not assume it from the reaction conditions.
The functional test is a serum stability incubation. Hold the conjugate in human or species-matched serum at 37 degrees, sample over several days, and measure DAR over time by immunocapture followed by mass spectrometry. A conjugate losing more than a small fraction of its payload over a week has a linkage problem regardless of what the chemistry was supposed to do.
Run albumin explicitly as the acceptor in a simplified version of the same experiment. It isolates the exchange mechanism from general degradation and tells you quickly whether retro-Michael is the route.
When to use something else
If the site is intrinsically exchange-prone and hydrolysis is not enough, the alternatives are real and available.
Bromo- and chloroacetamide form thioether bonds that are not reversible. They react more slowly and need a higher pH, which is the tradeoff, and for a stable antibody that is usually acceptable.
Bromomaleimides and the related dibromo reagents bridge two thiols from a reduced disulfide, which both conjugates and structurally re-staples the chain. That addresses the chain dissociation problem inherent in reducing interchain bonds.
Phenyloxadiazole sulfones and vinyl sulfones give stable thioethers with different selectivity profiles.
None of these are drop-in. Each changes reaction kinetics, optimal pH and the impurity profile, so switching chemistry means redeveloping the conjugation step rather than substituting a reagent. Where maleimide with a self-hydrolyzing design gives adequate serum stability, and for most sites it does, the simpler path is worth keeping.