ADCs: Pipeline and Progress

Terry Chapman, University of Bath PhD Student, Bath ASU Researcher

Introduction

This is the fifth article in a series of five blog posts that are putting ADCs in the spotlight. This article will describe ways in which third generation ADCs may improve upon current ADCs. Please read the introductory post ‘ADCs: What are they and why do they matter?’ before this post, if you are unfamiliar with ADCs.

Future Development

ADCs, like trastuzumab emtansine and brentuximab vedotin, are a clinical success and demonstrate ADCs are worth developing beyond their current limitations and expanding on their advantages.

A key problem for ADCs is that they have heterogeneous DAR and DOP. Current ADCs have an average DAR of 3 or 4, an ideal DAR for pharmacokinetics, efficacy & stability but range from 0 to 12. High DAR negatively impacts on pharmacokinetics and a low DAR is less efficacious. Heterogeneity, due to variable linker attachment, results in more expensive, less effective therapies . One way being investigated to tackle this problem is site specific conjugation (SSC). Linkers are attached to engineered cysteines or non-natural amino acids. These are engineered into the mAb which allows control of location and number of possible attachment sites. Research into SSC reports: improved serum stability, pharmacokinetic properties and comparable efficacy/toxicity profiles. SGN-CD33A is a site specifically conjugated ADC currently in phase 1 clinical trials for the treatment of acute myeloid leukaemia.

Taken From: May, et al,. 2013. SGN-CD33A: A novel CD33 targeting ADC. Blood. 122 (8).

A second problem for ADCs is that they must release their payloads inside cancer cells. This involves finding a cancer specific or selective receptor that is internalised upon binding. These requirements clearly limit the number of possible receptors to target. Target identification previously used for mAbs included therapeutic activity assays. ADCs don’t require  a therapeutic response from receptor binding, just that the receptor is internalised. This means that receptors previously dismissed for mAbs may become viable ADC targets.
Other approaches are also being used to increase the number of targets such as initiating localised non-specific cytotoxicity. ADCs like epratuzumab-SN-38, that continuously release their payloads at a slow rate, are being investigated, if successful ADC internalisation wouldn’t be necessary as payloads are released in the micro tumour environment and can diffuse into and out of nearby cells. This approach could overcome heterogeneous solid tumour receptor expression by ensuring tumour cells that no longer express the extracellular receptor are still exposed to the cytotoxic warhead.

ADCs are huge molecules, they suffer from poor penetration in solid tumours so can’t treat them effectively. One way to improve their penetration is reduce their size. Rather than using a complete antibody, single chain antibody fragments (scFv) or fragment antigen binding region (Fab) have been tested and are able to selectively target and deliver cytotoxic payloads to tumour cells in vivo. These have their own drawbacks such as reduced serum half-life and no Fc mediated effects, but are a viable way to develop ADCs into effective therapeutics for solid tumours.

Figure 2 – Size and structure comparison of mAb, Fab and scFv

Many ADCs in development and in addition to the licensed ADCs use payloads that are substrates for drug-efflux pumps. This is problematic for ADCs and helps explain why the payloads must be so potent. Scientists have already however, made payloads that are not substrates for common efflux pumps. DM4 has comparable therapeutic activity to the trastuzumab emtansine warhead DM1, but is unaffected by the common drug efflux pump p-glycoprotein pump. The linker can also affect whether the payload will be effluxed by a pump as demonstrated by Kovtun (2010) who compared non polar to highly hydrophilic linkers while retaining the same payload. Kovtun found that the hydrophobic linker reduced efflux and therefore increased efficacy of the ADC. If we can prevent efflux affecting ADC payloads we would be able to use less ADC or less potent payloads, reducing off target effects and increasing the therapeutic window. SGN-CD33A is resistant to the most common multi drug resistance pump, MDR1, as well as having site specific conjugation.

Summary

ADCs are past a critical point in their development. They have been demonstrated as more effective than conventional therapies, but still have obvious limitations, such as heterogeneous conjugation chemistry, restrictive target receptor characteristics, poor penetration, or rapid drug efflux. There are many proposed ways to overcome these limitations, such as site specific conjugation, novel drug release mechanisms, use of Fabs or scFvs instead of complete antibodies and efflux pump resistance payloads and linkers respectively. With over 40 ADCs in phases one to three, expect to see far more licensed ADCs in the near future, and as the drug class improves I expect it will begin being used to treat diseases other than cancer.

References

  1. Kovtun YV, Audette CA, Mayo MF, Jones GE, Doherty H, Maloney EK, Erickson HK, Sun X, Wilhelm S, Ab O, Lai KC, Widdison WC, Kellogg B, Johnson H, Pinkas J, Lutz RJ, Singh R, Goldmacher VS, Chari RV. Antibody-maytansinoid conjugates designed to bypass multidrug resistance. Cancer Res. 2010 Mar 15;70(6):2528-37.