Formulating an ADC Development Solution

vpanteon - stock.adobe.com

The concept of bioconjugation has been recognized for many decades, as has the specific idea of antibody-drug conjugates (ADCs). Technology for the production of ADCs did not advance sufficiently until the late 20th century. Even then, the first ADC approved by FDA-gemtuzumab ozogamicin (Mylotarg) from Pfizer in 2000-was voluntarily withdrawn in mid-2010 due to its association with the serious liver condition called veno-occlusive disease.

Gemtuzumab ozogamicin was ultimately re-approved in 2017 at a lower recommended dose and schedule in a different patient population. There are now six additional FDA-approved ADCs on the market: brentuximab vedotin (Adcetris, Seattle Genetics/Takeda); (ado-) trastuzumab emtansine (Kadcyla, Genentech/Roche); inotuzumab ozogamicin (Besponsa, Pfizer); polatuzumab vedotin-piiq (Polivy, Genentech/Roche); Fam-trastuzumab deruxtecan-nxki (Enhertu, Daiichi Sankyo and AstraZeneca); and Enfortumab vedotin (Padcev, Astellas Pharma/Seattle Genetics), the latter of which received approval in February 2020. And, there are approximately 80 ADC candidates undergoing nearly 600 clinical trials (1).

Why have more not already been approved? The issue of toxicity has been a challenge, as has the complexity of ADCs and the consequent manufacturing challenges.

Crafting the right molecule

Finding the right target, protein, linker, and payload all belong to the discovery phase. “This phase is very challenging, as the correct definition of the drug is key to move it into the clinic successfully. ADCs represent a product class that is also highly complex and difficult to predict,” comments Iwan Bertholjotti, director of commercial development for bioconjugates with Lonza. It is, in fact, one of the key questions in the development of ADCs since the rules are still being written, as data from clinical trials are used to guide ADC technology improvements, according to Steve Coats, global project leader at AstraZeneca.

Finding the right targets and crafting an effective molecule has proven to be much more difficult than many researchers had expected at the outset of ADC research, adds Justin Sweeley, senior technology manager, biologics at Novasep. “It was initially understood that finding a good target antigen for an ADC-based compound would require a target with rapid internalization once it had reached the cell,” he notes. “The result,” he says, “was that in many cases targets that were ineffective for monoclonal antibody (mAb) therapies because of rapid internalization became perfect targets for ADC therapies.”

This picture became much more complex upon the addition of payloads to these mAb candidates, however. This issue can be seen most clearly in the fact that the clinical trial landscape includes well over 50 different payload candidates, but only seven have successfully made it into commercial production, according to Sweeley.

“As a result, there has been a recent shift in the industry to recognize that when trying to predict ADC effectiveness, researchers must take into account the target epitope, the physical attributes of the payload on the mAb once conjugated, and also the effect of the payload mechanism of action on the specific kind of cancer being targeted,” Sweeley observes.

AstraZeneca takes an empirical approach to target and payload selection, keeping the patient population top-of-mind, using data from clinical trial results and non-clinical data to help validate target and payload selection. “With potency being driven by the strength of the warhead, we optimize the therapeutic index of our molecules by selecting the most favorable drug-to-antibody ratio (DAR) in designing our ADCs,” Coats says. “We believe,” he adds, “that establishing target rationale and strategy early on are key to determining the right payload.”

Is site-specific conjugation important?

Site-specific conjugation is looked at as the second generation of ADC conjugation techniques. “The homogeneity of these molecules allows for much tighter control at both the manufacturing and characterization levels,” Sweeley explains. Better definition for an ADC means that its therapeutic index can be further improved, adds Bertholjotti.

Additionally, the technology allowing for site-specific conjugation has evolved dramatically from the Thiomab concept piloted by Genentech to current techniques allowing site-specific conjugation with any native mAb, such as Synaffix’s glycogen modification techniques or Ajinomoto’s AJI-cap technology, Sweeley remarks. Many of the ADCs in preclinical and clinical stages are based on site-specific conjugation technology of one form or another, Bertholjotti adds.

On the other hand, Sweeley notes that even though more site-selective ADCs are expected in the future, ADCs with stochastic conjugation can still be successful. Coats agrees that site-selective conjugation does not seem to demonstrate significant differences in terms of clinical activity and safety when compared to classic non-site-selective conjugation. In fact, the seven commercially approved ADCs are not based on site-specific conjugation.

It is possible, according to Sweeley, that the lack of a commercially available site-specific conjugated molecule is just a result of the head start that stochastic conjugation has had, but it is also possible that simplicity in manufacturing and homogeneity in analytic testing methods doesn’t directly result in better outcomes for patients.

Linker technology matters

Finding an ADC that is better than standard therapy or that provides a solution to an unmet need definitely represents a challenge, just as with other drugs based on different approaches, according to Bertholjotti. The linker chemistry is an important component of ADC and has a significant impact on performance.

“Linker chemistry plays a critical role in in-vivo stability, but initially was assumed to have a passive role with the exception of either being stable or labile in the acidic cytosolic cellular environment,” observes Sweeley. More recently, however, he notes that there has been growing recognition that the linker plays an active role in conjugate hydrophobicity and therefore stability of the ADC as a whole.

The technology, according to Coats, has advanced and matured during the past 20 years to a point where linker technology is stable and molecules in development demonstrate low levels of deconjugation in patients. Both protease cleavable and non-cleavable linkers are in development and on the market.

Classic conjugation chemistry approaches include maleimide and N-hydroxysuccinamide-ester moieties (Seattle Genetics, Roche, and Pfizer). There are also various newer technologies for linkers, with some, Bertholjotti comments, at the proof-of-concept stage and others already in preclinical evaluation.

An example pointed out by Sweeley is the use of non-natural amino acid-based click-chemistry (Ambrx and Sutro). “To my knowledge, these technologies have clearly simplified the conjugation process, but have not shown a clear improvement in product effectiveness in the clinic,” he states.

Undesired immune responses

ADCs are designed to be more selective than traditional chemotherapy agents. The mAb enables targeting of cancer (or other disease) cells where the payload is delivered with high selectivity, thus reducing the systemic toxicity in comparison with standard chemotherapeutic drugs.

Unfortunately, undesired immune responses have presented a problem that has been difficult to overcome. These responses are largely due to the massive number of variables being examined and the relatively small number of clinical trials going on, according to Sweeley. “Many of the warheads being investigated for ADCs induce immunogenic cell death, which may enhance anti-tumor immunity,” adds Coats. He also observes that there are recent examples demonstrating clinical activity with a combination of an ADC with PD-1 inhibition.

“In the process of taking any ADC through the approval process, it is necessary to look at the mAb, the linker, and the payload chemistries individually, the ADC as a whole, and the payload and linker residues after the ADC has been internalized and digested within the cell. The end result is that for each ADC trial, researchers must monitor all of the normal undesired immune responses that occur during any oncology trial, but then try to attribute the cause of any that are observed to one of the five different potential sources,” Sweeley explains.

“Factoring these additional considerations into the normal variables of any trial, such as the specific cancer type, patient population, and level of pretreatments, etc., the picture becomes incredibly complex. Even after 20 years of clinical testing, the issue is still not fully understood,” he concludes.

Lower-than-expected therapeutic window

The primary issue affecting ADC approvals today, according to Sweeley, is the lower-than-expected therapeutic window of these therapies. “If the therapeutic window of ADCs were as large as people expected when ADCs first came to light, then many more approvals would already be on the market,” he asserts. This window has proven to be much more complicated than originally hoped, however, and therefore the clinical impact less significant than originally imagined.

Here again, the complexity of ADCs themselves is the main problem. For instance, Sweeley notes that if an innovator company wants to test a new conjugation method, they might choose a mAb and payload pair that have already shown success in the clinic. Similarly, a mAb company with a new antigen target is likely to use an established conjugation approach (e.g., that for Adcetris) to lower their risk. “The best molecule, however,” Sweeley explains, “might be a combination of multiple new technologies that are too risky to investigate for a small company with a limited budget.”

Complex supply chain

The supply chain for ADCs is highly complex as well. ADCs are unique in that their manufacture is an amalgamation of classical small-molecule production and traditional mAb manufacturing with the added complexity of highly potent drug manufacturing. “All three characteristics are complex on their own, and combining them brings the complexity to a level where it is no surprise that the vast majority of ADCs are manufactured by outsourcing partners,” Sweeley says.

For the most part, he notes that innovator companies treat the manufacturing of an ADC as three separate processes:

• mAb manufacturing using traditional mAb techniques

• Payload synthesis using traditional highly potent synthesis techniques

• ADC conjugation using specialty contract manufacturing organizations capable of both working with both biologics and highly potent payloads.

“Taken individually, all three of these steps are actually defined quite well and have been orchestrated successfully for a long time. But because ADC manufacturing requires all three to happen together, any delays or issues in one process will necessarily affect the other two. The reality, therefore, is that ADC supply-chain management is one of the most complex processes in the pharmaceutical industry and must be managed by an experienced team who can mitigate risks whenever possible,” observes Sweeley.

Bertholjotti agrees. The supply chain for ADCs is complex and requires specialized companies to manage different steps in a safe way and with the necessary quality. “If the supply chain is not properly managed, delays and supply issues may arise. Therefore, compromises related to the supply chain can result in critical impacts on timelines and costs to bring a drug to market,” he asserts.

Support from FDA

FDA has shown willingness to work closely with sponsors to ensure that ADCs that truly benefit patients are brought to market expeditiously, says Coats. For example, he comments that AstraZeneca’s ADC Enhertu (developed in collaboration with Daiichi Sankyo) demonstrated significant benefit in a high unmet-need cancer population. FDA approved Enhertu four months prior to the FDA goal date for the treatment of adult patients with unresectable or metastatic HER2-positive breast cancer who have received two or more prior anti-HER2 based regimens in the metastatic setting. Enhertu is manufactured using site-selective conjugation and contains a cleavable linker technology that releases a warhead with bystander activity. It has also shown encouraging clinical activity in other HER2-driven cancers.

Prudence and integration are warranted

Over the past decade, awareness of these issues has increased dramatically, leading to the development of more optimized molecules. Even so, Sweeley emphasizes that the complexity of the space means that it is still prudent to go slowly and only tackle one challenge at a time. “Progress is therefore slow, but continually happening. Each success creates a new platform for companies to reach out a little bit farther. And as the industry matures, the rate of success will increase, just as it has in every other area of pharmaceuticals,” he asserts.

Novasep has strong expertise in highly potent compound synthesis and for the past three years has brought this expertise into the realm of ADC conjugation, Sweeley notes. “Our focus has always been on producing the highest quality materials by developing the simplest possible procedures and coupling them with world-class analytics. The end result of these efforts is a simplification of the payload synthesis and conjugation steps along the supply chain, enabling rapid development of the two most complicated parts of the manufacturing of ADCs,” he states.

Since 2006, Lonza has been establishing a center of excellence for the development and manufacture of ADCs. Today the company offers an integrated solution, which helps to reduce the challenges presented by such a complex supply chain, according to Bertholjotti. “Integrated offerings such as that offered by Lonza represent a competitive advantage. They afford ADC developers with better predictability of timelines and costs, along with substantially reduced complexity and supply risk,” he says.

AstraZeneca is developing novel ADCs and building a library of payloads using its antibody engineering expertise for site-specific conjugation and next-generation ADCs. “ADCs form one of AstraZeneca’s key oncology scientific platforms, along with immuno-oncology, DNA damage response, and tumor drivers and resistance mechanisms. Within these platforms, multiple technology and scientific options offer great potential to yield effective medicines for cancer patients,” Coats says.

Growth expected despite the challenges

As more ADCs are approved commercially, Sweeley believes a roadmap will be established for other companies to follow, leading to more success in the near future. “In the last 12 months alone, the number of approved ADCs on the market has more than doubled, and I would expect that this track record will be followed by a sizable increase in the amount of research being put into the molecule discovery and preclinical trial stage,” he observes. “With the healthy increase in candidates being brought into the preclinical stage, it is only a matter of time until some of those candidates make it through the rigorous demands of clinical testing and reach the market in the not-so-distant future,” Sweeley concludes.

The increased number of ADC candidates in the pipeline in combination with ever improving knowledge and the further development of new conjugation technologies will result in an increased rate of approvals, agrees Bertholjotti. Lonza expects that the number of ADC molecules will in fact rise at a 10% compound annual growth rate until 2029.

Reference

1. S. Coats et al., Clin. Cancer Res., 25(18) 5441-5448 (2019).

Article Details

Pharmaceutical Technology
Vol. 44, No. 5
April 2020
Pages: 22–27

Citation

When referring to this article, please cite it as C. Challener, “Formulating an ADC Development Solution," Pharmaceutical Technology 44 (5) 2020.