Science Pool

Blood microsampling is a less invasive and simplified alternative to traditional venipuncture for PK/TK sampling, used mainly in small-animal studies. The purpose of this work was to evaluate the possibility of using microsampling technique also to support PK/TK studies in non-human primates.
A comparison of plasma PK parameters was conducted by traditional blood collection from the femoral vein and microsampling from the tail vein of six non-naïve cynomolgus monkeys. Four drugs were selected for this comparison, based on acid-base properties and volume of distribution. 
The results obtained in this work, supported by robust statistics, demonstrated the suitability of microsampling in supporting PK/TK studies in non-human primates. 
The plasma exposures of the tested drugs are comparable for both sampling techniques and are not influenced by acid-base characteristics and volume of distribution. 
Microsampling used in non-human primates avoids the occurrence of hematomas at the animal sampling site and can also refine practices to limit pain and distress to which animals are exposed (refinement of 3Rs) and, as a result, may reduce the impact of animal stress on PK/TK readouts; moreover, it also provides significant advantages for animal technicians during in life handling.

To request a copy of the article, contact the authors. For Evotec: massimo.breda@evotec.com

Katie Haughan, Drug Transporter Sciences Team, Cyprotex 

Drug transporters play a pivotal role in drug-drug interactions (DDI’s), as such regulatory bodies such as the FDA and EMA recommend the study of specific transporters that are known to cause clinical DDI’s. One of these recommendations for orally administrated drugs, unless waivered based on the BCS classification, is the victim (substrate) potential at Breast Cancer Resistance Protein (BCRP) due to evidence that inhibition of intestinal BCRP in DDI can increase the absorption and therefore exposure of sensitive substrate drugs such as rosuvastatin and topotecan. Pharmacogenetics also need to be considered when developing drugs that are BCRP substrates as this can impact the BCRP expression levels between individuals and within the global population and can therefore result in differing pharmacokinetic profiles. Polymorphisms associated with BCRP may or may not impact on the expression or functionality of the transporter, however the prevalence of these can vary dependent on the ethnicity. One example that is of importance for understanding the pharmacogenetic impact is BCRP SNP C421A, which is associated with lower BCRP protein expression, and leads to the ABCG2-Q141K polymorphism which has a higher frequency in Japanese and Chinese populations compared to Caucasian populations [Birmingham et al. 2015, Hua et al. 2012]. As a result, the plasma levels of BCRP substrates such as rosuvastatin in these populations is increased due to greater absorption and therefore this would need to be considered for Cmax estimates and dosage strategies.  Differences in BCRP expression amongst the global population can be a result of many factors such as ethnicity, genetic polymorphisms and disease states.  

The industry gold standard, and regulatory recommendation, for BCRP substrate identification is the use of a polarised cell monolayer system to determine the bidirectional flux of the investigational drug in the absence and presence of a selective inhibitor. To do so there are two cell lines which are favoured across the industry; BCRP over expressing transfected Madin-Derby Canine Kidney cell line (MDCK-BCRP), or the immortalised colorectal adenocarcinoma cell line Caco-2, both of which differentiate and polarise allowing for the expression on a range of proteins and display the in vivo like characteristics such as tight junctions. 

In order to select which cell line to utilise in these regulatory studies there are many aspects to consider, each cell line comes with its own advantages and disadvantages and can favour the outcome of certain classes of compounds. Whilst MDCK-BCRP has advantages over Caco-2 such as short culturing times and higher expression of the transporter of interest, there is a risk that data interpretation may be clouded due to an intrinsic limitation of the MDCK cell line which, for efflux transporter substrate determination, may result in false negatives. This potential error in the reported classification can have implications downstream as being a substrate may reduce the oral absorption and bioavailability of the drug, which may result in therapeutic dose not being achieved.  Whilst transfected MDCK cells have been used for many years in order to assess an investigational drug’s BCRP substrate status, it’s applicability may be limited due to the cells lacking expression of the relevant basolateral uptake transporters that allow polar substrates to enter the cell and in turn interact with BCRP transporter on the apical membrane. Without this uptake mechanism, and in combination with the compounds poor permeability, the basolateral to apical flux would be negligible resulting in an efflux ratio (ER) less than 2 (B-A/A-B).  The initial classification by the FDA is dependent on this ratio with a BCRP substrate having an efflux ratio greater than 2. 

Whilst Caco-2 cells have a considerably longer culture time compared to that of MDCK cells, they have the benefit of being able to correctly identify the efflux compounds that rely upon the interplay between apical and basolateral transporters. Organic solute transporter alpha (OST-α) is expressed on the basolateral membrane of epithelial cells in the small intestine, kidney, liver, and colon amongst other organs aswell as on Caco-2 cells, and this trans-membrane protein allows passive facilitative transport of endogenous and exogenous molecules across the membrane, including polar structures. A primary function of OST-α is the transport of bile acids, and therefore lends itself to the transport of other polar substrates such as the prototypical BCRP substrates rosuvastatin and estrone 3-sulfate across the membrane and into the cell. Once inside the cell the compound then has access to the binding site of the BCRP transporter and the opportunity to be effluxed if it is indeed a substrate. For cells that lack this uptake mechanism, such as MDCK-BCRP, polar compounds have restricted entry to the cell and do not have the opportunity to be effluxed regardless of if they are a substrate of BCRP or not. 

Rosuvastatin is one of the most widely prescribed statins and implicated in BCRP DDI’s due to its victim classification. Rosuvastatin is a polar compound with a logP of 0.13 and therefore has low permeability and is dependent upon uptake mechanisms to enter cells such as MDCK. The flux of rosuvastatin, apical to basolateral, basolateral to apical and the subsequent efflux ratios can be seen in Table 1 [Li et al. 2012] for three cell lines. The B-A secretory apparent permeability (and derived efflux ratio) is significantly higher (and therefore more sensitive) in Caco-2 cells compared to that in the two MDCK cell lines even though the protein expression of the BCRP transporter in Caco-2 cells would be considerably lower than in the transfected MDCK-BCRP cells, demonstrating the crucial requirement of the basolateral uptake mechanism that is present in Caco2, but absent in MDCK-BCRP, to facilitate the interaction of rosuvastatin with intracellular BCRP on the apical membrane.  This factor needs to be taken into consideration when it comes to assessing the efflux potential of an investigational drug using MDCK-BCRP cells. 

Table 1: Bidirectional (apical-to-basolateral (A-B) & basolateral-to-apical (B-A)) apparent permeability (Papp) and efflux ratios for rosuvastatin in different cell test systems [1]  

*P<0.001, significance level of the difference from the B-to-A transport in Caco-2 

As shown the use of Caco-2 cells for BCRP substrate identification removes a complication seen in MDCK-BCRP cells that is dependent on the physicochemical properties of the compound, however Caco-2 cells come with their own complexities. As Caco-2 are human derived cells they express a range of endogenous human transporters including human P-glycoprotein (P-gp). BCRP and P-gp have a similar substrate profile and therefore the efflux seen in Caco-2 cells may be due to BCRP, Pgp efflux or a combination of both. In order to decipher between the two main efflux transporters a selective P-gp inhibitor such as verapamil can be added to the test system buffer to chemically knockdown P-gp activity. The residual efflux is then considered to be due to BCRP transport however this can be confirmed using a selective BCRP inhibitor, fumitremorgin C (FTC) as shown in Table 2. 

Table 2: Bidirectional (apical-to-basolateral (A-B) & basolateral-to-apical (B-A)) apparent permeability (Papp) and efflux ratios for test compounds and the substrate classifications given per assay condition 

Choosing the right in vitro cell test system for BCRP substrate identification is critical for achieving the correct classification. The two gold-standard approaches have their own pros and cons, each of which can be mitigated if these test system limitations are understood. If a chemical series has a reasonable degree of lipophilicity then MDCK-BCRP should correctly identify a substrate of BCRP. However, as the industry moves towards more metabolically stable molecules and chemical series that have similar low intrinsic permeability and polarity to the clinically relevant BCRP substrate rosuvastatin, the use of Caco-2 cells would be required for the correct identification of BCRP substrates in order to avoid false negatives.

References

1. Birmingham BK, Bujac SR, Elsby R, et al. Impact of ABCG2 and SLCO1B1 polymorphisms on pharmacokinetics of rosuvastatin, atorvastatin and simvastatin acid in Caucasian and Asian subjects: a class effect? Eur J Clin Pharmacol. 2015 Mar;71(3):341-55.  
2. Li J, Wang Y, Zhang W, Huang Y, Hein K, Hidalgo IJ. The role of a basolateral transporter in rosuvastatin transport and its interplay with apical breast cancer resistance protein in polarized cell monolayer systems. Drug Metab Dispos. 2012 Nov;40(11):2102-8.  
3. Hua, W.J., Hua, W.X. and Fang, H.J, The Role of OATP1B1 and BCRP in Pharmacokinetics and DDI of Novel Statins. Cardiovascular Therapeutics,. 2012, 30: e234-e241. 

Rapidly growing tumor cells need a lot of oxygen and nutrients to proliferate, which requires access to the bloodstream. Therefore, tumors induce the formation of new blood vessels through a variety of molecular mechanisms. In normal tissue, there is a delicate balance of pro- and anti-angiogenic factors. In cancers, a process known as the "angiogenic switch" is initiated. As a result, pro-angiogenic signaling becomes dominant, allowing tumors to induce anarchic blood vessel formation. This switch is a critical step in the rapid growth of malignant cells, accompanied by the formation of new blood vessels.

One of the most important initiators of angiogenesis is the family of pro-angiogenic vascular endothelial growth factors (VEGF) and their receptors (VEGFR), which play an important role in both physiological and cancer angiogenesis. All members of the VEGF family stimulate cellular responses by binding to specific tyrosine kinase receptors, the VEGFRs, on the cell surface, causing them to dimerize and become activated. While VEGF-A regulates angiogenesis and vascular permeability by activating VEGFR-1 and VEGFR-2, VEGF-B seems to play a role in the maintenance of newly formed blood vessels under pathological conditions, while VEGF-C and VEGF-D and their corresponding receptor VEGFR-3 regulate lymphangiogenesis, i.e., the formation of lymphatic vessels.

This critical involvement of VEGFRs and their associated signaling pathways in the orchestration of (lymph)angiogenesis makes VEGFRs attractive targets for the treatment of tumors and the prevention of metastasis. However, existing therapies targeting VEGFRs are not very specific and inhibit a broad spectrum of receptor tyrosine kinases including the entire VEGFR family, causing many side effects such as hypertension, proteinuria, hand-foot syndrome, anorexia, and fatigue. While they show good response rates and short-term efficacy, their impact on overall survival is limited, in part because side effects limit the effective dose. This also results in only partial or transient inhibition of VEGFR-3, allowing lymphangiogenesis to serve as a tumor escape mechanism.

Evotec and Kazia therapeutics has therefore started to develop more selective VEGFR-3 receptor tyrosine kinase inhibitors. Its lead candidate, EVT801, an orally available VEGFR-3 inhibitor, is not only highly selective for VEGFR-3, but also the only inhibitor known to inhibit both VEGFR-3 homodimers and VEGFR-3:VEGFR-2 heterodimers. The compound shows low nanomolar inhibitory activity and high selectivity over kinases, various receptors, and ion channels. In November 2022, Evotec scientists reported in Cancer Research Communications that EVT801 showed potent antitumor activity in various in vitro and in vivo models. Moreover, the compound's in vivo efficacy was at least as good as that of the marketed pankinase inhibitors sorafenib and pazopanib. However, unlike sorafenib, EVT801 did not increase blood pressure in monkeys during regulatory toxicology studies or in a rat model of hypertension at doses up to 500 mg/kg, an order of magnitude higher than the pharmacological doses.

The investigators also observed that EVT801 reduced tumor (lymph)angiogenesis, apparently affecting small tumor vessels more significant than larger ones. 

The efficacy of EVT801 is expected to depend on the level of VEGFR-3 expression. Interestingly, expression level of VEGFR-3 did not appear to be affected by sorafenib treatment, suggesting that EVT801 could be used in patients previously treated with any VEGFR tyrosine kinase inhibitor. However, these findings need to be confirmed in clinical trials to determine the minimum threshold of VEGFR-3 expression for effective clinical application of EVT801 and for future patient stratification. 

The proposed mechanism of action of EVT801 involves three sequential anti-cancer mechanisms, all of which contribute to the inhibition of tumor growth and metastasis:

• It prevents tumor growth by impairing both tumor angiogenesis and (lymph)angiogenesis, thereby stabilizing the tumor vasculature, reducing metastasis, and reducing hypoxia in the tumor microenvironment.
• It enhances anti-cancer immunity as reflected by a decrease in immunosuppressive cytokines and cells in the circulation and tumor environment.
• It promotes T-cell infiltration into the tumor, ultimately supporting an enhanced and long-lasting anti-tumor immune response.

Taken together, these studies demonstrate that EVT801 is a novel anti(lymph)angiogenic agent that selectively targets VEGFR-3, modulates the tumor microenvironment to induce tumor vasculogenesis (i.e., fewer and overall larger vessels), and enhances immunotherapy.

Based on these promising results, EVT801 was selected to enter clinical trials. It is currently being evaluated as a single agent in a Phase I trial (NCT05114668) sponsored by Kazia and managed by Evotec Clinical Operations. The first stage of this Phase I study is designed as an open‑label, dose escalation trial to assess the safety, tolerability, and pharmacokinetics of EVT801 in up to 48 patients with advanced solid tumors. Details of the study were presented at the AACR Annual Meeting 2023 (Orlando, FL) (Abstract #1015). This dose escalation part will be followed by a biomarker and pharmacodynamics expansion cohort (second stage), including patients with high VEGFR-3 expressing cancers. These study sections may then be followed by a second dose escalation study, in combination with cancer immunotherapies.

The plan is to establish a patient stratification based on VEGFR-3 expression assessed by immunohistochemical imaging and immunofluorescence on tumor tissues before and after treatment. To enable this patient stratification analysis in a clinical setting, Evotec has established and validated a highly specific protocol for VEGFR-3 immunohistochemistry labeling and scoring strategy that is readily transferable to clinical sites.

To refine the VEGFR3 immunohistochemistry signature and to improve patient characterization, Evotec has developed a VEGFR3 mRNA gene signature consisting of 23 genes highly correlated with VEGFR-3 expression. The gene signature will be analyzed using the Fluidigm platform on matched FFPE patient samples. The relationship of key markers at protein and mRNA level will be investigated to potentially establish biomarkers for patient stratification and selection. 

We expect that the ambitious biomarker strategy will help to better understand the effects of EVT801 in humans and may also help to select the most responsive patients and provide early indications of clinical efficacy as a monotherapy (e.g. clear renal cell carcinoma, soft tissue sarcoma and ovarian cancer) or in combination with standard of care (e.g.immune checkpoint therapies). Evotec and Kazia recently presented a scientific poster on this topic.

Download the Full Poster

Our Drug Transporter Sciences team at Cyprotex are delighted to announce publication of a review article in Expert Opinion on Drug Metabolism and Toxicology entitled ‘Studying the Right Transporter at the Right Time: An In Vitro Strategy for Assessing Drug-Drug Interaction Risk during Drug Discovery and Development’.

As our understanding of the role of drug transporters in clinical drug-drug interactions (DDIs) has developed, the list of transporters requiring in vitro study by regulators has grown to accommodate assessment of risk for new drugs.  Currently, ten transporters require routine study prior to regulatory NDA submission.  Getting the timing wrong for these investigations could result in in vitro data being generated either 1) too early in the drug discovery/development timeline and potentially becoming surplus to requirements if the investigational drug fails for reasons of poor pharmacokinetics (and efficacy) or toxicity, or 2) too late to influence finalisation of the clinical development plan resulting in perhaps unnecessary comedication exclusions that impact patient recruitment and thus delay clinical trials.  In either case, there will be a cost and resource penalty, with the overall impact being considerably cheaper for the former compared with the latter.  To minimize these development risks, project teams should study the right transporters at the right time for their investigational drug and the authors (Dr’s Robert Elsby, Hayley Atkinson, Philip Butler and Rob Riley) have tried to address this in their review article by proposing in vitro strategies that could be employed to either mitigate/remove transporter DDI risk during development through frontloading certain studies, or to manage (contextualize) DDI risk to patients in the clinical setting.

In the article, an overview of clinically relevant drug transporters and observed DDIs is provided, alongside presentation of key considerations/recommendations for in vitro study design when evaluating drugs as inhibitors or substrates of transporters.  Guidance on identifying critical victim comedications and their clinically relevant disposition pathways, and using mechanistic static equations for quantitative prediction of DDI (demonstrating a 97% predictive accuracy for 28 statin DDIs) is also compiled.  To truly alleviate or manage clinical risk, the industry would benefit from moving away from current regulatory qualitative basic static equation approaches to quantitative mechanistic DDI prediction, thereby contextualising risk to ascertain whether a transporter DDI is simply pharmacokinetic or clinically significant requiring intervention.  Furthermore, such a mechanistic approach can be used towards either mitigating perpetrator DDI risk early during candidate selection, or managing clinical risk and aiding patient recruitment by informing labels and potentially providing an alternative to conducting costly clinical interaction studies with co-medications in the future. 

Read the Full Publication

The importance of viral filtration studies

Biological therapeutics need to meet strict safety criteria. Virus safety is ensured through complementary manufacturing and quality control measures. Virus filtration is a critical element in this process, and viral filtration studies have become a key step in bioprocessing over the past decades. They are required by most regulators to bring a biologic to the market. The purpose is to assure that the final medicinal product is safe from the potential risk of viral contamination.

The quality of the design of such studies is key: when inappropriately designed, they may lead to undesired results such as non-representative filter fouling or virus breakthrough.

The characteristics of end-to-end continuous filtration

Continuous processing is a modern manufacturing method for biologicals, e.g. monoclonal antibodies (mAb). Consequently, also continuous viral clearance has to be investigated. Compared to batch filtration, continuous end-to-end processes present new challenges to the operation and validation of the viral filter:

• Higher loadings are required to maximize filter utilization and decrease filter swap outs, thereby reducing process risk;
• Higher loadings and continuous operation can lead to viral clearance assessments lasting several days and potentially overloading the filters with virus causing either non-representative filter fouling or non-representative virus breakthrough;
• Novel filter evaluation and assessment strategies are needed to maximize filter utilization while still demonstrating safety.

Consequences for assessing viral clearance in continuous end-to-end filtration

This, in turn, means that assessment of viral clearance in continuous viral filtration may require studies spanning over several days and also novel virus-spiking strategies. This type of assessment is needed to adequately demonstrate a high level of viral clearance while ensuring the filters are not overloaded with virus and are consistent with real-life manufacturing conditions.

A team of scientists at Just- Evotec Biologics has performed studies to identify a robust virus filter that retains virus despite high load challenges and low operation pressures. Key findings were:

• There are multiple filter options suitable for continuous processing;
• The use of surrogates for virus particles can give clues to filter behavior, however, the assay still needs optimization in terms load challenge.
• Alternatives to traditional virus spiking techniques such as a bracketed, integrity test approach can simplify viral clearance assessment while still demonstrating virus safety.

Summary and outlook

While bioprocessing technologies have evolved rapidly during the past decades, multiple factors such as increased cost, quality and production pressures are calling for further advances. At the same time, regulatory requirements are evolving and sophisticated safety testing has become a key prerequisite for market approval of novel biologics.

The team at JUST - Evotec Biologics has therefore kicked off an initiative to spearhead the development of next-generation bioprocessing technologies such as continuous end-to-end viral filtration.

Future work of the group will include optimization of the surrogate virus spiking and testing techniques as well as testing other virus filters for robustness of virus clearance at low flux/pressure conditions.

Interested in further information about Just-Evotec Biologics? Please visit our site.

For more information download the full poster.

Download Now

Chinese Ovary Hamster (CHO) cells are the most common mammalian cell lines used for the mass manufacturing of therapeutic proteins as they can produce recombinant proteins on the scale of 3-10 grams per liter of culture. However, the expression of these recombinant proteins rely on random genomic integration events, which typically result in a widely heterogeneous cell population. Therefore, cell line development is taking up much time for the extensive pooling and clone screening to identify clones with high expression, growth, and product quality.

Also, the random integration precludes experiments such as variant library screening as this screening method needs stably overexpressing pools or libraries of molecules in a single cell culture. With random integration, the assessment of yield, degree of library enrichment, etc. is very complicated as it is difficult to determine whether resulting library members differ due to inherent characteristics of each variant or merely to due to variable genomic integration site(s).

To address this problem and to provide a time- and cost-efficient solution, Just-Evotec Biologics’ scientists have developed two targeted integration systems by generating two clonal CHO cell lines stably expressing enhanced green fluorescent protein (eGFP) reporter landing pads in genomic hot spots. The goal of this study was to compare two options to evaluate the one that might be most suitable for in-house usage.

For the study, the team combined several approaches to circumvent random chromosomal insertions, resulting in the precision and reproducibility associated with site-specific recombinases as well as the biased selection of genomic hotspots associated with a certain transposon.

Recombination was carried out either by Cre or PhiC31 recombinase. Subsequently, genes for the expression of three therapeutic protein molecules were used to test targeted integration. The cell lines were then assayed for yield and productivity as well as characterized for landing pad copy number and integration fidelity by targeted locus amplification (TLA) and PCR. Both cell lines expressed high levels of the respective recombinant protein. The scientists additionally tested for enrichment of cell subpopulations with fully saturated landing pads with ganciclovir (GCV) counterselection.

The results of these experiments were quite compelling: Genetic characterization of the altered cell lines showed correct targeting of landing pads. Post‐integration enrichment for fully saturated landing pads using GCV counterselection increased recombinant protein titer by 2–2.5‐fold and specific productivity by ∼3.4‐fold.

Finally, the team developed a small antibody library of ~100 variants through random pairing of 10 unique light chains and 10 unique heavy chains by transfecting this library into a cell line containing a single copy landing pad wherein each cell line would express a single variant. Puromycin selection was used to identify cells that had successfully taken up one of the variants and cell-sorting for variants that successfully paired and expressed. Finally, the identity of successful chain pairs was determined with next generation sequencing.

As a result, Just-Evotec Biologics has demonstrated proof-of-principle of targeted integration systems in the CHO host cell line, with consistent genome integration into expected landing pad sites and high productivity. Moreover, test cases using three antibody or antibody-fusion therapeutic molecules showed similar levels of productivity. Finally, the team demonstrated that library screening or CHO display is feasible with the 100-member variant library. The study also reveals preliminary data from ongoing work to build upon these targeted integration systems, which includes isolating a single-copy landing pad cell line and the development of a CHO display platform.

Although additional work and optimization is still needed, the great advantage of this approach lies in the predictability with regards to the chromosomal integration of transgenes of interest. Among others, a key advantage of this approach is its ability to combine eGFP as a reporter gene and transposon-mediated integration to establish high-expression landing pad cell lines. With a significant decrease in heterogeneity between clones, it is possible to be able to reduce extensive pool screening, large scale cloning, and clone screening efforts.

All in all, defining and targeting predefined locations that promote high expression of an exogenous protein allows to develop cell lines expressing different recombinant protein therapeutics with a high degree of specificity and reproducibility - which, in turn, and will save a lot of time and costs.

For more information, please read the study or have a look at our poster.

According to a study from 2020, a total of 133 drugs were withdrawn from the market due to safety reasons between 1990 and 2010. Major causes were hepatotoxicity (27.1%), cardiac disorders (18.8%), hypersensitivity (12.8%), and nephrotoxicity (9.8%), accounting for 69.2% of all drugs withdrawn. In most cases, these withdrawals were initiated because of spontaneous reports and/or case reports. Another study looking into drug withdrawals between 1953 and 2013 revealed that 18% of drug withdrawals from the market in this period were due to liver damage.

Add to these withdrawals of marketed drugs the attrition rate of drug candidates in clinical trials: 90 percent of all drug candidates fail in clinical trials, and 30 percent of these failures are due to unmanageable toxicity issues.

These failures occur despite thorough preclinical work and intensive animal studies. It is estimated that only 50% of the compounds that cause liver toxicity in humans are detected by animal studies. Furthermore, some adverse reactions or idiosyncratic toxic effects are typically not detected until the drug in question has gained large exposure in a broad patient population.

Interestingly, a study evaluating the attrition of drug candidates from AstraZeneca, Eli Lilly and Company, GlaxoSmithKline and Pfizer came to the conclusion that there is a strong link between physicochemical properties of compounds and clinical failure due to safety issues. The results also suggest that further control of physicochemical properties is unlikely to have a significant effect on attrition rates and that additional work is required to address safety-related failures.

These failures are not only costly (according to the FDA, drug development takes over 10–15 years with an average cost of over $1–2 billion for each new drug to be approved), but are also putting the health and the life of patients in danger.

Consequently, Cyprotex and its parent company Evotec are very focused on assessing toxicology issues from the very beginning of its drug R&D process and have invested a significant amount of time and resources to expand its technologies for the toxicological evaluation of drug candidates.

“The idea is to make better informed decisions earlier in your discovery campaign when you can select potentially safer compounds, rather than finding a safety liability later on,” says Paul Walker PhD, Vice President, Head of Toxicology at Cyprotex, in Cheshire, UK.

This improved discovery and selection is implemented by Cyprotex by using the unbiased view of transcriptomics and its potential to predict drug-induced toxicity. Transcriptomics involves sequencing thousands of mRNA molecules to identify which processes are active in the cell and allows for a better understanding of the cell’s reaction to known and novel drugs.

This is by no means a purely academic endeavour. As an example, the Cyprotex team demonstrated via transcriptomics it was able to identify problems in liver cells treated with fasiglifam, a promising diabetes drug candidate, which was withdrawn from late-stage clinical trials by its developer, following signs of liver damage in trial participants. This example proves that transcriptomics could have raised a red flag during preclinical development and might have saved hundreds of millions of dollars.

“Our studies have found potential effects on mitochondrial function, which were previously missed in preclinical studies” says Walker.

Therefore, transcriptomics has the potential to supplement or reduce in vivo toxicology studies by effectively identifying safety issues early in drug development, saving time and money — and animal testing.

Sophisticated Human Cell-Based Models

A key advantage of transcriptomics is its use of human cells and Evotec as well as Cyprotex are not just looking at 2D cell cultures, but investigating 3D organoids. These structures formed of thousands of cells that mimic organ-specific tissues are much closer to the real organ and have valuable features: For example, 3D-organoids of the heart exhibit regular contractions, beating like a living heart, and liver organoids secrete typical liver enzymes for days.

“On top of that, a 3D system allows repeat dosing, mimicking dosing regimens in vivo and potentially helps to detect effects due to toxic metabolites,” says Walker.

As they are small, the organoids can be placed in 384-well plates and individually molecular barcoded for simultaneous sequencing. This combination of miniaturization and high-throughput screening is implemented in Evotec’s EVOpanOmics platform and allows a wider adoption of transcriptomics in preclinical toxicology studies allowing for the repeat testing of dozens or even hundreds of compounds at several doses and in multiple organs.

“People have thought about using transcriptomics for toxicology before, but it was always a numbers game,” explains Rüdiger Fritsch PhD, Principal Scientist and Project Lead for EVOpanOmics. “For any compound that’s a real troublemaker, the evidence will show up in the transcriptomics data if you profile it in a relevant model. You just need to test appropriate dosing scenarios with the breadth of genome-wide off-target effects so that you have a chance to find it.”

Complex Analysis of Transcriptomics Data

Evotec, in conjunction with Cyprotex, offers transcriptomics services to drug developers and carries out the entire process in-house, from growing the organoids to sequencing and analysis. This streamlined process allows its researchers to screen hundreds of compounds a day, each delivering tens of thousands of data points on RNA levels. To analyze all of these vast amounts of data, Evotec has developed a software platform called EVOpanHunter that allows among others the analysis of these transcriptomics in an interactive manner.

“We want to democratize data analysis for the biologists who know the biological pathways and processes, without them needing to rely on additional experts from the bioinformatics department for routine tasks,” says Carla Tameling PhD, Head of Sales and Application for EVOpanHunter at Evotec.

On top of the interactive multi-omics analysis platform machine learning is used to trawl through this immense amount of data in order to find specific patterns hinting for toxicological effects and alert the researchers to dig deeper. “The more data we get, the harder it is for a human to dig through it all,” adds Tameling. “Transcriptomics is an unbiased view. You don’t need to define what to look at prior to your studies — you get all the data, and you might see things that you didn’t think would be relevant initially.”

From publically available sources, Cyprotex has compiled a broad and highly valuable transcriptomics reference database for drug-induced liver injuries.. Machine learning is being applied to predict whether a compound is likely to have issues by comparing the observed pattern of gene activity to the activity patterns of known toxic molecules. Furthermore, this is not restricted to hepotoxicity. Cyprotex is already building databases of other organs, such as heart, kidney and brain, using publicly available drug development trial results to select a broad space of reference comounds. “We’re running reference compounds from all kinds of sources where we know there are either late-stage clinical findings or withdrawals from the market,” states Walker.

Given the rapid advancements of the technology, it may be only a matter of time before transcriptomics and other omics technologies become a regulatory standard approach for preclinical toxicity testing.

Contact Us for More Information

Learn More

Gene therapy based on adeno-associated virus (AAV) has come a long way and has been used for more than 20 years in over 2,000 patients. While AAV based gene therapy is regarded as generally safe, the field has experienced several setbacks recently: some trials were halted for safety reasons and product approvals were delayed.

The reasons for these developments are complex, but gene therapy is nowadays not only applied to a narrow spectrum of rare diseases with no other treatments available but is also beginning to be used as a therapeutic option in more common diseases. Consequently, many more patients are being treated and several limitations have surfaced:

• The quality of the clinically used vector material is often, despite cGMP manufacturing requirements, insufficient to prevent adverse events
•  For difficult to access tissues, site-specific administration and delivery of gene therapy vectors can be an important barrier
• The database to demonstrate long lasting therapeutic effects in clinical trials is often thin and the interpretation of results can be difficult
• Patients experience serious adverse events after escalating to high AAV doses

The lack of AAV materials with high and reproducible quality, particularly the presence of substantial amounts of empty capsids is, at times, making the interpretation of controlled clinical trials difficult. Insufficient product quality in combination with very high AAV vector doses often leads to a massive presence of capsids in the liver, causing inflammatory responses and raising safety concerns. In addition, AAV vectors with a desired tropism towards specific tissues or organs and a high yield manufacturability are scarce, thereby limiting the applicability of AAV gene therapy. Yet another complication is that current AAV-based gene therapy technology cannot offer a once in a lifetime treatment in many cases, requiring a repeated dosing (if feasible) in case of therapeutic effects diminishing over time.

Adding to these issues is the fact that about 50 percent of the population already has some immunity to AAV so that not all affected patients will benefit from AAV-based gene therapies.

How to improve AAV-based therapies?

Evotec has identified and addressed a number of issues which were presented in our recent webinar titled “AAV Gene Therapy: Closing the Translational Gap”. According to Hanspeter Rottensteiner, VP, Head of In Vitro Gene Therapy of Evotec Austria, the most important needs are to:

1. Improve the delivery vectors by developing capsids with better tissue tropism and translating into improved efficacy and safety. Also, promoters and the respective transgene need to be optimized to improve cell-type specificity, controllability, and durability
2. Gain better insights into the effects of gene therapy by developing better biomarkers

Improve tracking and analyzing of side effects, for instance by tracking protein expression and metabolic fluxes with proteomic and metabolomic platforms and via high-throughput sequencing of RNA and DNA in target and non-target tissue and cells

Improved specificity

Dirk Grimm, Professor for Viral Vector Technologies at Heidelberg University, explains in the webinar that - thanks to new powerful technologies - knowledge about the host factors that influence AAV vector transduction has increased tremendously. In addition, many new technologies allow for the rational design of next-generation AAV capsids. These novel vectors not only avoid neutralizing anti-AAV antibodies circulating in the human population but also outperform wild-type capsids in terms of potency and / or specificity. Grimm demonstrated as an example that rationally designed myotropic AAV capsids with a tropism exclusively for muscle cells did not accumulate in the liver at all. They demonstrate a substantial improvement in terms of specificity as compared to capsids in use today. The latest technology advancement – barcoding of capsids and nucleic acids - allows for tracking capsids not only to specific organs and tissues but can also tell whether a functional transduction has taken place. Grimm added that the increased use of artificial intelligence will also make animal experiments largely obsolete. In his view, it will be possible in the future to design capsids tailormade to individual patients and their disease.

Filling the gaps

Werner Höllriegl, VP, Head of In Vivo Gene Therapy at Evotec GT GmbH, detailed in the webinar the current gaps in translational efficacy and safety that Evotec is addressing when preparing first-in-human trials. Great efforts are made to assess toxicity in liver, cardiovascular and nervous systems, but also for evaluation of potential oncogenic effects.
Höllriegl also mentioned gaps that cannot be filled by Evotec alone but require a concerted effort by the gene therapy community. Among them are lacking industry standards for AAV titers and dose assessments and guidance from the regulatory agencies on the design of preclinical studies that can best inform IND decisions.
He stressed, however, that even more research at basic and translational science level is needed to tackle specific efficacy or safety issues.

Comprehensive safety assessment

Evotec also brings its already long-established strong expertise in prediction of drug-induced liver injury to the table – an experience gained with small molecules, as explained by Rüdiger Fritsch, Principal Scientist, Metabolic Diseases at Evotec. Among other safety assessments, the Company uses a wealth of technologies and expertise to identify transcriptomic fingerprints and applies its proprietary and continuously growing database of tox-related data from hundreds of tested compounds. AI-based modeling can predict mechanisms and probabilities for AAV-induced liver toxicity. Moreover, using single nucleic RNA-sequencing tools, Evotec can study and profile the expression and activity of the transgene at single cell level – not only in liver cells and organoids but in a growing portfolio of tissues from brain, muscle, lung, heart, and kidney. Evotec thereby is successfully translating its toxicology expertise from the small molecule space to gene therapy research and development, offering a 21st century omics-approach to provide for safe and efficacious gene therapy vectors.

Stream the webinar

How to knock down proteins driving disease processes in a cell

Many diseases are caused by the overproduction of certain proteins. The traditional approach to interfere with these proteins is based on small molecules or antibodies blocking these proteins or their corresponding target, e.g., receptors. Thanks to the recent progress of nucleic acid research, there are several new approaches today which intervene at different stages, from gene regulation to transcription to translation: CRISPR-Cas9 methods to target the DNA, zinc finger repressors targeting gene transcription, or RNA-molecules (antisense oligonucleotides, RNA interference, micro RNAs, etc.) to inactivate the mRNA or to suppress the translation. All approaches come with advantages and disadvantages. The main problem with DNA- and RNA-based medicine is delivery, followed by off-target-effects.

There are, however other new knockdown strategies as well, e.g., enhancing the protein clearance pathways to speed up the degradation of unwanted proteins, such as the autophagy-lysosome pathway and the ubiquitin-proteasome system (UPS).

Transcriptomics, data analysis, and AI/ML platforms as basis for partnership with BMS in targeted protein degradation

Since 2018, the latter technology of targeted protein degradation is also being used in a cooperation with Bristol Myers Squibb (BMS) to identify first-in-class drug candidates in oncology to treat solid tumors. For this collaboration, Evotec uses its PanOmics platform, EVOpanOmics, which combines enhanced throughput proteomics, high-throughput transcriptomics, and cell imaging with the integrated data analysis platform EVOpanHunter and Evotec’s AI/ML-based drug discovery and development platforms.

This research has led to the discovery of novel first-in-class molecular glue degraders. These small, drug-like compounds induce interactions between an E3 ubiquitin ligase and a molecular target, leading to ubiquitination and subsequent degradation of the recruited protein. The resulting therapeutic effect is long-lasting as the molecular glue degraders themselves are not degraded in the process and can initiate the degradation process through several iterations. BMS is a leader in this field based in particular on its unique library of cereblon E3 ligase modulators (CELMoD®) with specific protein-binding properties. Based on the needs of this project, Evotec focused on the development of dedicated and innovative software solutions that greatly helped to accelerate not only the project’s progress but also contributed to the overall progression of Evotec’s PanHunter platform.

The approach has generated a pipeline of novel first-in-class programs, two of which have transitioned successfully into lead optimization after completing respective validation processes on Evotec’s platforms. In this context, Evotec´s integrated data analysis platform panHunter and the Company’s AI and machine learning tools are used to quickly screen, share, and validate results – not only by Evotec, but also by BMS scientists. In May 2022, the partnership was expanded even further for an additional 8 years with the goal to once again broaden and deepen the strategic alliance.

Targeted protein degradation is not only useful in oncology - a number of other diseases, e.g. Alzheimer’s, bacterial and viral infections lead to the presence of unwanted proteins inside cells that may be marked for destruction by this powerful technology. Evotec therefore is welcoming partners interested in exploring this approach in collaborations.

To learn more about the BMS collaboration and the use of targeted protein degradation technology read the official press release.

Learn More

Antibodies generated in the lab are important as potential treatments for a broad spectrum of diseases, in particular infectious diseases caused by viruses. They can be obtained either by animal-derived B cells or from antibody library display platforms. Evotec’s strategy for the optimal path to obtain lead candidates is offering access to both sources of antibodies for discovery, coupled with the exploitation of state-of- the-art technologies to ensure success for a broad range of targets and disease states. In addition, selected lead candidates can be further optimized using powerful computational platforms to enhance productivity, manufacturability, and formulation stability. This is the end-to-end J.Design biologics platform, which is fueled by the front-end discovery platform, J.HAL™ (Just Humanoid Antibody Library) and associated data-driven, company-wide machine learning methodology.

By using artificial intelligence (AI) and machine learning (ML), J.HAL can generate novel, humanoid antibody sequences that both represent natural repertoires and are biased towards desirable features. To enable properties such as broad target and epitope engagement, focused efficacy, and suitable developability, Just-Evotec Biologics has devised an Antibody-GAN (Generative Adversarial Network), a new synthetic approach to designing a novel class of antibody therapeutics, which is termed humanoid antibodies.

At the conferences International Conference on Antiviral Research (ICAR) 2021 and Antibody Engineering & Therapeutics Europe 2022, researchers from Evotec and Just-Evotec Biologics introduced results obtained by using GAN to generate novel sequences, which mimic natural human response and provide the necessary diversity and developability features.

Competing Neural Networks

GAN is based on competing, deep layer neural networks that learn and produce the features of the mature human antibody repertoire, including sequence characteristics and structure properties, allowing for the encoding of key properties of interest into diverse libraries for a feature-biased discovery platform. It works to:

• capture the complexity of the entire variable region of the standard human antibody sequence space,
• provide a basis for generating novel antibodies that span a larger sequence diversity than standard in silico generative approaches, and
• incorporate transfer learning, a critical feature for antibody discovery to bias the physical properties of the generated antibodies towards broader efficacy traits such as CDR lengths and surface properties, improved developability (e.g., improved thermal and pH stability), and diverse chemical and biophysical properties.

The GAN network is trained by using hundreds of thousands of human antibody sequences to recognize legitimate human v-genes. The generator network generates random sequences to fool the discriminator while continually receiving feedback from the discriminator on sequence validity. Over time, the two networks get progressively better at their tasks. After full training, the Antibody-GAN generator is eventually able to produce fully human, novel antibody sequences for the germline for which the GAN was trained.

Antibodies targeting SARS-CoV-2

To demonstrate the usefulness of this platform, the researchers used their newly constructed, 1 billion theoretical diversity phage Fab library with the intent to discover antibodies to the SARS-CoV-2 spike protein. Candidates that specifically bound SARS-CoV-2 spike protein and did not bind an irrelevant antigen were further characterized for dose-dependent binding using AlphaLISA technology. In the primary “yes/no” binding screen a total of 73 unique antibody sequences specific for SARS-CoV-2 spike protein were identified. The researchers then performed binding assays using unpurified transfection supernatants and later reproduced the results with purified material. The candidate antibody supernatants that specifically bound SARS-CoV-2 spike protein were subsequently tested for their ability to block binding of this protein to human ACE-2 receptor. The team identified multiple antibodies that effectively blocked spike human ACE2 receptor interaction, demonstrating the feasibility to screen unpurified transfection supernatants for functional activity. After further rounds of panning, the top candidates expressed at flask scale were purified and tested for SARS-CoV-2 neutralization ability across multiple strains. The researchers identified multiple candidates with neutralizing activity against several strains of SARS-CoV-2. Nine of these antibodies exhibited blocking activity of the spike protein to the ACE2 receptor in an in vitro functional assay. Of note, all antibody data shown here were from native library candidates without any affinity maturation.

The presentation demonstrates that applying machine learning algorithms in antibody discovery “promotes efficient learning from the least expensive and most abundant data encoded in the DNA of antibodies, to validation of this learning through less abundant, more expensive, but most relevant data from GMP manufacturing at full commercial scale,” stated James N. Thomas, retired Executive Vice President, Global Head of Biotherapeutics and President U.S. Operations at Just - Evotec Biologics. “This is a systems approach to platform definition and continuous improvement, and it is unique in the industry, made possible by a number of factors that will be difficult for others to replicate."

To learn more about Evotec's capabilities read our related poster.

Learn More