Significant advances have been made in recent years to develop therapeutic agents beyond the “blunderbuss” approach of broadly targeting all rapidly dividing cells, to ones that better target tumor cells; specifically to improve the killing of tumor cells while reducing collateral damage to healthy tissues.

The first generation of such targeted therapies include the development of monoclonal antibodies such as Herceptin for HER2-positive breast cancer and Rituxan for CD20-positive lymphomas. These types of therapeutics work by adhering, or binding, to proteins found on the outer surface of the tumor cell. Once bound, tumors are flagged for killing and removal by the immune system.

However, while antibodies are able to target tumors with great specificity, leaving healthy tissue unharmed, and thus generally have an improved toxicity profile compared to conventional untargeted chemotherapy, they can sometimes be limited in efficacy.

These observations have spurned new efforts to improve targeted therapeutics by preserving the specificity of the antibody while enhancing their ability to kill the tumor cell. Such approaches include linking a cytotoxic drug (also called a chemotherapeutic, or drug that kills cells) to the tumor-targeting antibody, which are called antibody-drug conjugates, or ADCs.

Antibody-drug conjugates are frequently described as “empowered antibodies” or “smart bombs”, as they deliver the highly potent chemotherapy drug, the “warhead”, directly into the targeted cell and therefore are able to destroy tumor cells specifically while avoiding other healthy cells. The antibody-drug conjugate approach has achieved significant milestones recently as the technology used to link the antibody and the chemotherapy drug has advanced. The FDA approved Seattle Genetics’ Adcetris in 2011 for the treatment of rare lymphomas, and this year the agency approved Genentech’s Kadcyla for the treatment of metastatic breast cancer patients that are HER2 positive.

New Technologies will Further Advance Development

The approvals of Adcetris and Kadcyla were significant advances for the antibody-drug conjugate space. Past attempts to develop antibody-drug conjugates were hindered by the ability to produce stable antibody-drug combinations, and the first drugs of this type frequently had the chemotherapy drug or “warhead” break off before reaching the target cell, which resulted in toxic, off-target effects. The success of Adcetris and Kadcyla were due to advances in linking technologies that enabled the creation of antibody-drug conjugates that are stable in the blood stream yet, once internalized into the tumor cell, release the drug which then destroys the cell.

However, designing antibody-drug conjugates remains an imprecise science. For example, Pfizer announced recently the discontinuation of a very late-stage, Phase 3 study evaluating the safety and efficacy of its antibody-drug conjugate, inotuzumab ozogamicin.

But what makes the design of antibody-drug conjugates so complicated? One reason is the multiple moving parts that need to be optimized in order to create a successful antibody-drug conjugate. Choosing the appropriate protein target on the surface of the tumor cell, also called the tumor antigen, designing the right antibody that recognizes it, choosing the appropriate warhead for the type of tumor and putting the correct linkages between those components are all fundamentally important in order to see optimal response rates in patients, a measure to determine the percentage of patients whose tumor shrank or disappeared, in the correct patient populations. Above and beyond these requirements, the designed antibody-drug conjugate has to have sufficient affinity for the tumor antigen, be internalized into the tumor cell at the right rate, have appropriate separation from the cytotoxin once internalized and also have the correct cytotoxin to balance highly effective tumor killing while minimizing collateral damage to healthy cells. Such design complexities are further complicated by the imprecise nature of the chemistry used to link the antibody and the drug. In sum, there are multiple parameters that must be adjusted in order to create a truly optimized antibody-drug conjugate.

Although the efficacy and tolerability for Kadcyla is impressive, a significant proportion of patients still suffer serious adverse events, which limits amounts of those drugs that can be dosed and therefore limits the effectiveness at the tumor and potential benefit for patients. Adcetris and Kadcyla are both products that consist of many species of antibody-drug conjugate molecules within the product, differing in the amount and location of cytotoxin linked to the antibody. The total number of individual species of molecules within the product is enormous and may result in species of antibody-drug conjugate molecule with unwanted characteristics, such as lack of binding to the tumor cell, less stability, poor internalization into the tumor cell, or the premature release of the warhead in the blood before it can get to the tumor. All of these possibilities probably explain why only a very small portion of the cytotoxic agent ever makes it to the tumor cell, which has been reported to be only 1-1.5% of the total systemic dose.

To overcome this limitation, the industry is moving towards more precise and optimal positioning of the link between the cytotoxin and the antibody, resulting in a “homogeneous” product that might assure more efficient delivery of cytotoxin to the tumor cell, resulting in better efficacy and less toxicity. Preclinical studies with precise linking technologies that can result in a single species of antibody-drug conjugate molecules, for example, Genentech’s ThioMab technology, have already shown that the therapeutic index, the ratio between efficacy and tolerability, can be improved.

Emerging technologies that can specifically position the site of linking of the cytotoxin to the antibody as well as produce more stable conjugations at those sites include several approaches that can incorporate non-natural amino acids into the antibody, providing a site/anchor for very stable attachment of the cytotoxin. The most advanced of these technologies are emerging from Sutro Biopharma [Note: author is the Chief Scientific Officer of Sutro Biopharma] and Ambrx.

In order to develop an antibody-drug conjugate where all properties are optimized, production of many hundreds of antibody-drug conjugates with varying sites of linkage must be created and subsequently tested. Currently, however, this process is limited by production technologies, which typically utilize cell-based expression systems and is very laborious.

A new, promising and powerful approach to overcome these limitations is through use of cell-free protein synthesis technology. At Sutro Biopharma, methods have been developed to produce non-natural amino acid containing antibodies, in as little as several hours. This technology enables simultaneous development of hundreds of antibody versions, and thus the parallel generation of many variations of antibody-drug conjugate molecules that can be tested. After testing all of those variants, the antibody-drug conjugate with optimal characteristics can then be selected.

Interest in Antibody-Drug Conjugates Will Only Grow in the Future

While the efficacy and tolerability of antibody-drug conjugates will undoubtedly improve as control over the design of antibody-drug conjugates is achieved, new technologies will result in new therapeutics that will combine a number of early design concepts.

Clearly, there is plenty of opportunity for improvement in the design of antibody-drug conjugates, and now, technologies are available to make these “smart bombs” smarter.

Images: Sutro Biopharma