Antibody–drug conjugates: Targeted Delivery and Future Prospects

Keywords:

• ADC
• antibody–drug conjugate
• auristatin
• calicheamicin
• maytansine

Treating cancer patients with cytotoxic chemicals, based on the premise that rapidly dividing cancer cells would be more sensitive than normal cells to their lethal effects, became standard therapy in the latter half of the twentieth century, following the pioneering work of physicians such as Sidney Farber [1]. However, the effects were short-lived, toxicity to normal cells limiting the dose that could be administered thereby limiting the degree of cancer cell kill achievable. Combining different anticancer drugs having different mechanisms of action and nonoverlapping toxicity profiles resulted in improved antitumor activity [2], and such combination chemotherapy remains the mainstay of cancer treatment today. However, systemic toxicity limits what can be achieved by this modality and long-term remissions, or cures, are rarely seen except in a small subset of patients with cancers that are particularly sensitive to chemotherapy.

The possibility of exploiting the specific binding properties of antibodies for targeted delivery of a cytotoxic chemical to cancer cells was a concept that required the invention of monoclonal antibodies in 1975 [3], and the further inventions over the subsequent 10–15 years of technologies for developing nonimmunogenic antibody molecules [4], for the concept to begin to be realized. Attaching cytotoxic effector molecules to an antibody provides a mechanism for selective delivery of the cytotoxic payload to cells to which the antibody binds, while at the same time conjugation of the cytotoxic agent to a protein restricts penetration of the small molecular weight hydrophobic cytotoxin across cellular membranes of normal cells to which the antibody does not bind. This simple concept is a particularly attractive solution to the challenge of increasing the therapeutic index of a cytotoxic chemical agent.

The simple concept turned out to be not so simple to execute. The early work in the field of antibody–drug conjugates (ADCs) sought to increase the specificity of cytotoxic drugs already in use in cancer chemotherapy. Drugs with very different mechanisms of actions such as melphalan, methotrexate, doxorubicin or vinblastine were chemically linked to antibodies for preclinical and clinical evaluation. However, despite early optimism generated by some of the preclinical results [5], the results of clinical trials of these conjugates were disappointing. By the early 1990s, there emerged a greater appreciation of the in vivo biodistribution properties of antibodies, and it was realized from clinical dosimetry studies with radiolabeled antibodies in patients that only about 0.01% of the injected dose of antibody per gram of tumor tissue could be localized to a solid tumor mass, 24 h after infusion (approximately the peak delivered concentration), irrespective of tumor type or antibody target [6]. Thus, while there may be a favorable reduction in systemic toxicity by giving to the small molecule cytotoxic agent the in vivo distribution properties of an antibody, one must also take into account the constraints imposed by conjugation to antibodies. Besides the slow rate of diffusion of antibodies from blood plasma into tumor tissue relative to small molecules, retention of ADC at the tumor is limited by the number a target cell surface antigens (perhaps ~10⁵ receptors/cell), and by the rate of release of active cytotoxic agent (depends on internalization, intracellular trafficking, release mechanism and linker chemistry) [7]. Thus, in order to be successful as an anticancer agent, a key requirement for an ADC is that the cytotoxic potency of the conjugated agent should be sufficiently high to effect a robust killing of the targeted tumor cells at the amounts that can be delivered by antibody retention in the tumor. This notion has guided much of the subsequent research in the field.

The first ADC to be approved by the US FDA was gemtuzumab ozogamicin [8], an anti-CD33 antibody conjugated to the highly potent DNA-targeting antibiotic, calicheamicin. However, it was withdrawn from the US market in 2010, 10 years after its initial approval, following an unsuccessful confirmatory Phase 3 trial [9]. By then, the development of two classes of highly potent tubulin-acting antimitotic agents as payloads for ADCs reinvigorated the promise of the ADC approach for treating cancer, and have resulted in two (so far) ADCs approved for clinical use. The first, brentuximab vedotin (Adcetris^TM), is an anti-CD30 antibody conjugated with the auristatin, MMAE, which received accelerated approval from FDA in August, 2011, for treatment of patients with relapsed Hodgkin lymphoma or with anaplastic large cell lymphoma [10]. The second compound, ado-trastuzumab emtansine (Kadcyla^TM), an ADC made by conjugation of the anti-HER2 antibody, trastuzumab, to the maytansinoid, DM1, received approval from FDA in February, 2013, for treating patients with HER2-positive metastatic breast cancer whose disease had progressed on or after a trastuzumab-containing treatment regimen [11], the first ADC to receive full approval based on a randomized Phase 3 study. These two compounds met the long-awaited goal of the ADC field, to make highly active, well tolerated, anticancer agents. Their successful development has sparked much research by the biopharmaceutical industry to develop ADC molecules that may improve the treatment of many cancers, as evident by the large number of ADCs in clinical trials–more than 50 at the time of writing (February 2016).

What are the future prospects and opportunities for ADCs? First, with regard to ADC design and synthesis, the lack of cytotoxicity of antimitotic microtubule-disrupting agents toward nondividing normal cells may contribute to the tolerability profile of ADCs made using these payloads, given that target antigens are rarely completely tumor-specific and, in any case, most of the administered antibody or ADC is ultimately eliminated by catabolism via the reticuloendothelial system [6]. Thus, it is not surprising that the first ADCs to have a significant place in the armamentarium of treatments for the cancers they target used payloads of this type. Several other ADCs made with either an auristatin or a maytansinoid (about 75% of ADCs currently in development utilize one of these classes of compound), have shown promising anticancer activity in early phase clinical trials, in both hematologic cancers as well as solid tumors. As more clinical data emerge over the next 2–3 years from the >35 ADCs in development that utilize this type of payload, we can anticipate several compounds moving into pivotal clinical trials, while others will not, allowing the field to gain knowledge about which targets/target diseases can best be addressed by ADC compounds of this design. Recently, a maytansinoid-ADC that targets mesothelin, anetumab ravtansine, advanced into a pivotal study for treating patients with mesothelioma.

Research into ADC chemistry and design is now expanding the scope of the linker-payload beyond the potent tubulin agents, especially to address those targets/target diseases toward which tubulin-agent ADCs are insufficiently active. There is great interest in utilizing highly potent (≤pM range) DNA-acting cytotoxic agents which may expand the potential of ADCs to cell surface targets that, while specific, are expressed at low antigen density, or to target tumor types that are generally insensitive to tubulin-acting agents (e.g., colon cancer). Several such ADCs in development showing promising antitumor activity in clinical trials utilize calicheamicin [12]: improvements in process biochemistry and analytical methods over the last 15–20 years have greatly improved the design and performance of calicheamicin-ADCs relative to the ‘pioneer’ compound, gemtuzumab ozogamicin. Other classes of DNA-acting agents newly adapted for incorporation into ADCs include DNA-crosslinking compounds based on pyrrolobenzodiazepine dimers [7,13], DNA-alkylating agents based on indolinobenzodiazepines [14] and duocarmycin derivatives [7,15], all of which are being utilized as payloads for ADCs currently in early clinical development.

Second, clinical investigators are beginning to explore the activity of ADCs in combination with other agents for increased clinical benefit. For example, indatuximab ravtansine has been reported to show interesting levels of activity in combination with lenolidomide plus low-dose dexamethasone in patients with multiple myeloma [16]. Brentuximab vedotin is being evaluated in a number of combination regimens [17]. An emerging prospect of great potential is the opportunity for combining ADCs armed with potent microtubule-disrupting agents (auristatins or maytansinoids) with check-point inhibitors. Recent preclinical studies have shown that these ADCs may stimulate a tumor-specific adaptive immune response, and that combination of an ADC with the immunomodulatory antibodies targeting CTLA-4 and PD-1 can increase antitumor activity [18,19]. These data provide a strong rationale to assess such combinations in clinical trials.

Brentuximab vedotin and ado-trastuzumab emtansine are already providing treatment options for patients with cancers which express high levels of the target antigens, CD30 and HER2, respectively. For solid tumors, several more ADCs in development show promising activity in early clinical trials, anetumab ravtansine in mesothelioma and mirvetuximab soravtansine in ovarian cancer [20]. Nevertheless, there is still much to learn about the chemical design of ADCs, especially with novel payloads, that may further improve their therapeutic index. There is also much more to learn with respect to the optimal clinical application of ADC technology, in particular in combination with current or newly emerging therapeutic modalities, including targeted agents and immuno-oncology approaches. However, it is safe to say that the addition of ADCs to the arsenal of therapeutic options for treating cancer patients has the prospect of improving treatment outcomes for many patients.

Acknowledgements

The author thanks C Bennett for her critical review of this editorial.

Financial & competing interests disclosure

The author is an employee of ImmunoGen, Inc., the company that developed the maytansinoid-ADC platform utilized in ado-trastuzumab emtansine, and in several other ADCs in clinical development. ImmunoGen has also developed the indolinobenzodiazepine-ADC platform used in the ADC compound, IMGN779, which will begin clinical evaluation in 2016. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.