THE AGE OF ADCs
The year 2021 saw the approval of the eleventh Antibody Drug Conjugate (ADC) by the US Food and Drug Administration: Zynlonta. The recent approvals and more than one hundred and fifty of other ADCs currently undergoing clinical trials, demonstrate the great interest and work ongoing in this field (1-3).
Nevertheless, literature and clinical trials results shows that the success for an ADC is a consequence of correct selection of its components: the monoclonal antibody, the cytotoxic drug and the linker (4). In this, the history of Mylotarg is a good example of the difficulties behind the development of an ADC: approved in 2000, withdrawn from market in 2010 and then re-approved in 2017 with a different dose regime.
EVOLUTION OF ADCs
Researchers are pursuing new epitopes to enlarge the applicability of ADCs to new tumor classes, but aside the search for new targets for mAb, for which development strategies and industrial production technologies are well established, a great load of research is currently devoted to the linker – payload part of the ADC (5).
In fact, the linker – payload has been recognized to be the most important component of the ADC, for it influences the bioavailability, in-vivo plasma stability, the formulation stability, therapeutic window and therefore determines the overall ADC probability of approval success.
NEXT GENERATION PAYLOADS
Indeed, the improvements on the pharmacological proprieties of the ADC have been mainly linked to the optimization of the linker – payloads.
Today, several chemistries are available for the conjugation, coupled with an even larger number of linker types aimed at increasing solubility, improving stability in the biological environment, reducing aggregation and fine tuning the biological activity. A recent review on these topics gives a good overview of the current research trends and a glimpse of the complexity of the payload – linker related chemistry (6).
Nevertheless, such improvements from the pharmacological point of view, has led to more and more complex molecules (7, 8) which pose huge challenges to the chemist who has to develop the industrial production process.
Molecules such as the ones reported in the cited literature, with dozen of stereocenters, polyethylene glycols moieties, tens of hydroxylic groups, overall high molecular weight, sensitive functional groups and, of course, high toxicity, represent a hard development challenge even to the seasoned process chemist. A challenge which can become a nightmare, when it comes to scale up from the initial few milligrams obtained through medicinal chemistry route, to the commercial kilograms scale GMP process.
The same grade of effort is requested for the analytical development chemist who must develop sound analytical methods for in process control, stability testing and drug release of the desired compound.
In our experience, the analytical development, and following qualification, is an extremely complex and resource demanding task due to the necessity to track the fate of impurities during the process development and support the evolving chemical synthesis process during the development phases. Moreover, as the ADC development program progress from phase I to more advanced stages, impurities preparation and qualification can become a separate project due to the complexity of the impurity profile and therefore the efforts needed.
OUTSOURCING AND ADC
For the abovementioned reasons, it is common for innovator companies to outsource the development and manufacturing activities to CDMO to avoid costly internal development and production departments structures. This enables the innovator companies to contain costs and focus on the core research. Moreover, most of the smaller innovator companies, most of them being start-ups born from institutional research centers, do not have the internal competences, nor infrastructures, to successfully scale up and produce in GMP their products. Therefore, relying on an experienced external CDMOs company is the best solution to reduce time from the proof of concept to IND submission; this allows to reduce costs, allow faster return on investment, and satisfy the investors.
After the linker – payload selection and initial medicinal chemistry route of synthesis for producing the first few milligrams needed for the in vitro studies, the next step is to develop a reliable process for the toxicology and clinical studies. In our experience, the medicinal chemistry process is seldom transferrable to the production lines; most of the time, adjustments for reducing reaction volumes, even of order of magnitude reduction, substitution of dangerous or exotic chemicals or optimization of the synthetic path are needed to fit with the available industrial equipment and technologies.
Nevertheless, the tight timelines to get the ADC to the clinical stage are ill-suited for a complete development of such complex payloads following the “small molecule” approach. A more effective strategy is required.
ENTER QUALITY BY DESIGN
A quick and robust development strategy is therefore needed to prioritize the development activities to tackle the most critical steps without sacrificing timings and quality: Quality by Design (QbD) is the approach we employ to achieve this. A recent oral presentation at PACIFICEM 2021 by Dr. Thomas Matt (9) testify how the challenges encountered during the development of a six chemical step process for a pyrrolobenzodiazepine dimer payload, were approached and solved by applying QbD.
Dr. Matt’s team started with a throughout familiarization work, followed by definition and standardization of the chemical route, which was then optimized through a QbD approach. After defining the Quality Target Product Profile and determining the potential Critical Quality Attributes, Interaction matrixes and risk prioritization allowed the team to focus on the critical parts of the process. Design of Experiments based optimization then allowed our experienced chemists to rapidly improve the process, assess its robustness and define the critical process parameters. Finally, by analyzing the interactions between the critical process parameters and the critical material attributes, the team was able to define a robust control strategy to reach the desired quality. At the end of the journey, the development team doubled the overall process yield (17% to 33%), improved purity (92% to 96.3%) and most important, timely delivered a robust process which was successfully transferred to GMP plant.
DEALING WITH COMPLEXITY
Although QbD permit to focus on the critical problems to solve and gives powerful weapons to the development chemist, the extremely complex nature of new generation payloads requires innovative approach compared to the “small molecule” process development. As previously mentioned, to improve the pharmacological profile of ADCs, the next generation payload – linkers are designed to have high hydrophilicities, high solubilities and as result, also high molecular weights.
It turns out that, the impurities of such large and complex molecules share very similar physical-chemical properties with the desired molecule. As consequence, reverse phase preparative high-performance liquid chromatography (prep-HPLC), becomes the only viable control strategy in these cases. Nevertheless, even if preparative chromatography process development is a science itself, in our experience, separation is seldom the real issue with chromatography, since it is always possible to separate two compounds by tweaking the purification conditions.
As many practitioners know, reverse phase prep-HPLC is a powerful purification technique, amenable especially for water soluble compounds, but recovering the desired product from water (the main component of the reverse phase prep-HPLC mobile phases) may be less straightforward than a crystallization process. The issue is exacerbated when the compound being recovered is heat sensitive or does not crystallize due to high water affinity or high molecular flexibility. Adding complexity to complexity, sometimes mobile phase modifiers (salts, organic modifiers) employed during chromatography must be removed from the mixture before proceeding to the next steps. Indeed, the recovery is the real issue most of the times, but, due to the chemical and physical properties of such large molecules, some technologies borrowed from the biotech world provide solutions to the recovery issues.
In many cases, our R&D group has successfully solved the recovery issues employing high pressure tangential flow filtration (TFF), which permitted us to remove most of the solvents and modifiers without heat stressing a very complex hydrophilic molecule.
Nevertheless, heat sensitive, hydrophilic molecules with flexible backbone are hardly recovered from water solutions; in such cases, therefore, we employed the most straight forward and general approach for this kind of molecules: lyophilize the aqueous solution. For these reasons Cerbios has invested in dedicated prep-HPLC capability, high pressure TFF, lyophilization and, of course, state-of-the-art containment systems.
Figure 1. General workflow of process development through QbD.
Figure 2. Example of risk prioritization table employed during development via QbD.
CONCLUSIONS
No “one-fits-all” solution is available for the development and production of toxic complex linker – payloads, this is the lesson we learnt in our experience in the field. Nevertheless, a development strategy based on QbD in addition to creative problem solving, is what we have found to be the most effective in terms of time saved and quality of the solutions obtained.
As final remark, the trend which one can observe examining patents and literature, clearly shows that complexity in the world of ADC linker – payloads is here to stay, so will special technologies be needed for their development and QbD.
REFERENCES AND NOTES
- Tong, J., Harris, P., Brimble, M., & Kavianinia. (2021). An Insight into FDA Approved Antibody-Drug Conjugates for Cancer Therapy. Molecules, 26, 5847. doi:10.3390/molecules26195847
- ClinicalTrials.gov. (2022, January 13). Retrieved from clinicaltrials.gov.
- EU Clinical Trials Register. (2022, January 13). Retrieved from clinicaltrialsregister.eu.
- Powers, R. (2019). Advanced Design of ADCs: principles and Applications with Next Generation Linkers and Site-Specific Technology. San Diego.
- Mckertish, C., & Kayser, V. (2021). Advances and Limitations of Antibody Drug Conjugates for Cancer. Biomedicines, 9(8), 872. doi:10.3390/biomedicines9080872
- Kostova, V., Désos, P., Starck, J.-B., & Kotschy, A. (2021). The Chemistry Behind ADCs. Pharmaceuticals, 14, 442. doi:10.3390/ph14050442
- Kim, Y. Z., Oh, Y. S., Chae, J., Song, H. Y., Chung, C.-w., Park, Y. H., . . . Lee, S. I. (2016). Patent No. WO2017089890.
- Widdison, W. C. (2018). Patent No. US10792372.
- Matt, T. (2021). CPC-1406 From Medicinal Chemistry to GMP-Production of a PBD-Dimer Payload. PACIFICHEM 2021.
