Antibody Drug Conjugate Metabolism Related to Efficacy

Understanding ADC metabolism is an important requirement in selecting appropriate efficacy and toxicity models and in addressing species comparability. The information gained from in vitro metabolic stability studies will ultimately help guide clinical monitoring. The metabolism of ADCs has been shown to mimic that of radioimmunoconjugates (vide supra) in that the major metabolite tends to be an adduct of the drug still linked to the amino acid to which it was originally conjugated. For instance, Doronina et al. demonstrated both in vitro cytotoxicity and in vivo therapeutic efficacy for monomethyl auristatin F (MMAF)-conjugated anti-CD30 ADCs, and the major metabolite was identified by high-performance liquid chromatography (HPLC) and mass spectrometry as the cysteine-linked adduct of MMAF (136). This is consistent with the idea that ADC metabolism occurs by the same lysosomal degradation responsible for catabolism of naked Abs into their substituent amino acids. Further supporting the idea of lysosomal metabolism, Sutherland et al. have demonstrated that an auristatin-conjugated ADC based on a chimeric anti-CD30 Ab, cAC10, trafficked to the lysosome, and cysteine protease metabolism was implicated to play a role in drug release (146).

Metabolism studies have also helped to explain what we observed as "bystander effect," that is, the antigen-negative cells may be killed by administration and subsequent metabolism of an ADC. This concept was illustrated by Kovtun et al. who showed that the N2-deacetyl-N2-(3-mercapto-1-oxopropyl)-maytansine (DM1) or DC1 conjugate of anti-CanAg (cancer antigen) (huC242) killed antigen-positive cells and neighboring antigen-negative cells in culture, and that in vivo, the same conjugates effectively eradicated tumors containing both antigen-positive and antigen-negative cells (149,150). Similar conclusions were drawn by Erickson et al. at ImmunoGen, Inc. in relating the in vitro potency and in vivo activity of huC242-maytansinoid conjugates; however, in this case, the choice of linker heavily influenced the outcome (145). A disulfide-linked ADC yielded an uncharged, lipophilic drug metabolite that was found to be highly toxic to cells; however, a charged lysine-maytansinoid adduct resulting from metabolism of the thioether-linked conjugate had a very low cell-killing potency, presumably due to its inability to cross plasma membranes to exit/enter cells (145).

Summary of Antibody-Drug Conjugates

ADCs represent a special case of Ab therapeutics, combining many important concepts discussed throughout this chapter. Numerous examples of IVIVCs involving naked Abs were outlined in the preceding sections. Similar IVIVCs also apply for the Ab component of ADCs. When the cytotoxic drug is released from the ADC, IVIVC principles as established for small-molecule drugs remain applicable and will not be further discussed as they are beyond the scope of this presentation. Instead, the discussions herein address the specific details of IVIVCs for ADCs in the context of entire macromolecular ADC constructs.

ADCs are associated with additional risks compared with Abs. The highly cytotoxic nature of the free drug and the possibility of its detachment through chemical or metabolic processes are both complications that must be approached with great caution. One must constantly assess whether in vitro cytotoxicity or in vivo efficacy is due to the intact ADC or results from a metabolite/catabolite, and the same questions must also be applied to any normal organ toxicity observed in vivo. Examples described here and throughout the literature indicate that initial in vitro/in vivo studies provide critical information on identifying the determinant factors in optimizing ADCs. The behavior of these complex macromolecules is often target, antibody, linker, and drug specific; therefore, a combination of both in vitro and in vivo methods will be necessary to dissect the appropriate factors in their development.

0 0

Post a comment