Having established a clear link between target and disease, a destination may consist of writing a grant application for further funding based on this increased knowledge between target and disease.

After demonstrating a clear link between a target and a relevant human disease, a potential destination would be the selection of a candidate compound for in vivo testing which does not require extensive safety testing, such as a repurposed drug or compound. At this stage, a partner may be needed for financial support.

In the mid-1980s, chronic myeloid leukemia (CML) was shown to be invariably associated with the Bcr-Abl kinase underlying malignant transformation, i.e. the clinical relevance was clear.



Imatinib acting at this target had no selectivity, pharmacokinetic or toxicity issues and was highly effective on human tumours in mice. Human activity was demonstrated on blood/bone marrow cells from CML patients, used later as a clinical biomarker.



Using kinase engineered in baculovirus, Ciba-Geigy (later Novartis) screened many compounds, leading in 1992 to the identification of imatinib.



As CML is resistant to conventional anticancer drugs, comparative trials were not needed, orphan drug designation could be used and the drug was registered in May 2002.

A novel target for inhibition of inflammatory responses was identified and shown to be relevant for human disease with no associated safety issues and patented as an innovative target. However, the lead compound synthesized by the academic lab was not suitable for drug development.
To make this feasible, the goal is to find an industrial partner to perform extensive or high throughput screening to find a druggable candidate compound.

Having established a clear link between target and disease, that the target is druggable and interaction with it has no safety problems, the destination may consist of a development plan, possibly involving a financially competent partner.

For projects failing to achieve clinical efficacy during clinical trials, major reasons for failure include lack of data for causal linkage of the target with the disease or poor understanding of the role of the target in underlying disease pathophysiology. Thus, it is critical to generate evidence showing that modulation of the target of interest leads to clinically relevant physiological effects by analysing its role in a particular disease.

Guiding Questions to establish a link between target and disease include:

• Have you established a clear, relevant link between the target and the disease?
• Is this just association or causation?
• Are you using a relevant model system/organism?
• Have you considered the desired therapeutically relevant magnitude of change? Statistical differences do not necessarily represent biologically meaningful differences!

Example 1:
The interleukin (IL)-2-inducible tyrosine kinase (Itk) was examined as a novel target for the therapy of inflammatory skin diseases. Itk is expressed selectively in diseased tissues. It is present mainly in T cells and is increased in lesional skin from patients with dermatitis. RNA silencing was used as a target-validation strategy, followed by studies in an Itk knockout mouse model. An available tool compound (Itk inhibitor) confirmed the modifiability of this kinase in disease models. [1]

Example 2:
The correlation of high HIF levels with poor outcomes in cancer may reflect the fact that HIF upregulates genes for proteins such as vascular endothelial growth factor A (VEGFA) and matrix metalloproteinases (MMPs) which promote angiogenesis and tumour invasion. However, HIF also upregulates the expression of genes encoding proteins that suppress protein synthesis and promote autophagy, a process that can be tumour suppressive. The effects of HIF on tumour growth are actually highly context dependent, HIF promoting tumour growth in some models and suppressing tumour growth in others. [2]

[1] von Bonin A, et al., Zügel U. Inhibition of the IL-2-inducible tyrosine kinase (Itk) activity: a new concept for the therapy of inflammatory skin diseases. Experimental Dermatology (2011)
Other PMID: 21158938

[2] Kaelin WG. Common pitfalls in preclinical cancer target validation. Nature Reviews Cancer (2017)
Target identification and validation PMID: 28524181

Requirements for a successful drug discovery program require that aspects like 'druggability' and 'assayability' be taken into account early on in the process:

• Biomarker development: Are biomarkers available to monitor target modulation and/or target engagement?
• Assayability: to support a later screening program for a suitable lead compound, biochemical and/or cellular assays for binding and function are needed
• To qualify as ‘druggable’, a target must be accessible to the therapeutic molecule (small molecule compound and/or biologics), and modifiable (i.e. provoking a measurable biological reaction when the drug interacts with the target)
• Development of high-quality reference compounds and tool modulators for manipulation of the target

Please be aware 1 - Tool Compound Use:
Note that use of a reference tool compound may prove misleading as its activity in a disease relevant test may prove to be due to action at another unidentified target.

Please be aware 2 - Assayability:
Selection of an assay may need revision if new findings reveal important target active binding sites not assessed in the initial assay.

One key question is how to favourably position validated targets for the large-scale (external) investment required to develop new therapeutic products?. Thus, it is important to analyse strategic options for managing and protecting IP associated with validated targets, the competitor landscape, and medical as well as commercial needs. The aim is to generate valuable data for business and commercialization plans, attractive licensing propositions for commercial partners, as well as for potential investors.

Guiding Questions which can help to evaluate the Degree of Innovation include (CRITICAL PATH GENERATOR):

• Is there an unmet medical need?
• Current standard of care analysis: Is the target either a) totally novel or b) addressed using novel technology that promises to be advantageous over previous approaches?
• Should Intellectual Property be generated and protected?
• Can all activities be conducted without infringing any IP rights (i.e. is there freedom-to-operate?)?
• Is the treatment approach superior to the competitor’s product?
• Have commercial needs be addressed?

Example 1 - Unmet medical need (Linagliptin):
Linagliptin is an inhibitor of the dipeptyl-peptidase isoenzyme DPP-4 and was developed to treat diabetes mellitus type 2. It was approved for medical use in the United States in 2011. In Germany, however, the Institute for Quality and Efficiency in Health Care (IQWiG) could not see a clear difference from the current standard of care (metformin and sulfonylurea treatment) in an early assessment and therefore, did not certify additional benefit with linagliptin. Moreover, as several gliptins (sitagliptin, saxagliptin and vildagliptin) have been in commercial use since 2008, the ability of linagliptin to address an unmet medical need is questionable.

Example 2 - Unmet medical need (Hypertension):
There are many effective treatments for arterial hypertension, and thus the unmet medical need in this area seems small. Given the competitive landscape and the high commercial risks, truly innovative antihypertensive agents have rarely been introduced in recent years.

On-target or target-related toxicity refers to exaggerated and adverse effects, that result from manipulating a different function to the therapeutically desired inherent biological function of the target of interest. In contrast, off-target toxicity refers to adverse effects based on modulation of other targets, related or not to the biological target of interest. [1,2]
Early identification of potential target-related safety risks and increased understanding of underlying molecular mechanisms can guide drug development and help derive mitigation strategies to facilitate project progression. [3] This justifies spending further money on the TV project.

There are many approaches to evaluation of target-related safety:

• reviewing human genetic databases
• reviewing published clinical trials or regulatory approval documents
• signalling pathway analysis
• analysis of target distribution and expression in diseased vs healthy samples
• using robust safety biomarkers
• analysis of transgenic KO/KI animals or knockdown cell lines
• ...and many more

In addition, a risk-benefit analysis helps identify a significant therapeutic window between onset of toxicity and pharmacological effects and to identify patient populations likely to benefit from the novel target-based treatment approach (patient stratification strategy).

Example:
Inhibitors of the mitochondrial enzyme dihydroorotate-dehydrogenase (DHODH), such as leflunomide, developed for rheumatoid arthritis are known for their teratogenic potential, as demonstrated in mice. Potential DHODH-related adverse events were assessed using phenotypic information on specific knockout mice. In addition, recent studies have shown that mutations in the DHODH gene are the cause of Miller syndrome, which is characterized by severe craniofacial and limb anomalies. [4,5,6] Thus, early access to gene-disease-association data can contribute to early awareness of potential undesired target-related effects.

[1] Guengerich FP. Mechanisms of drug toxicity and relevance to pharmaceutical development. Drug Metabolism and Pharmacokinetics (2011)
Drug discovery and development & Safety and toxicity PMID: 20978361

[2] Rudmann DG. On-target and off-target-based toxicologic effects. Toxicologic Pathology (2013)
Safety and toxicity PMID: 23085982

[3] Roberts RA. Understanding drug targets: no such thing as bad news. Drug Discovery Today (2018)
Target identification and validation PMID: 29803936

[4] Madak JT, et al., Neamati N. Revisiting the role of dihydroorotate dehydrogenase as a therapeutic target for cancer. Pharmacology & Therapeutics (2019)
Other PMID: 30347213

[5] Fukushima R, et al., Kitagawa H. Inhibiting the teratogenicity of the immunosuppressant leflunomide in mice by supplementation of exogenous uridine. Toxicological Sciences (2009)
Other PMID: 19190124

[6] Ng SB, et al., Bamshad MJ. Exome sequencing identifies the cause of a mendelian disorder. Nature Genetics (2009)
Other PMID: 19915526

The maternal embryonic leucine zipper kinase (MELK) was implicated as a therapeutic target in human cancer based on data such as:

• Elevated levels of MELK RNA expression in human tumours and in mouse models
• MELK targeted RNA interference inhibited proliferation in cultured human cancer cell lines
• Reported association between increased MELK expression and poor clinical prognosis
• Efficacy of a small molecule MELK-inhibitor in cell lines/xenografts (OTS167)

With new techniques such as CRISPR-CAS, deleting MELK in different human cancer cell lines had no beneficial effect. In cells depleted of MELK, OTS167 still had effects, suggesting that the observed cytotoxicity following OTS167 treatment reflected off-target, MELK-independent mechanisms. [1]

[1] Settleman J, Sawyers CL, Hunter T. Challenges in validating candidate therapeutic targets in cancer. eLife (2018)
Target identification and validation PMID: 29417929

Cherry picking and exclusion of outlier results (e.g. including only sensitive cell lines in the characterization of a new target) leads to incomplete/erroneous understanding of target function and invalidity of the robustness of the data. A result that is correct, but within only under an extremely narrowly defined set of conditions, is not robust (i.e. the degree to which a finding holds true over a range of experimental conditions). Failure to characterize the heterogeneity of the model system (e.g. heterogeneous responses to different tool compounds acting at the same target) and a lack of robustness will delay the drug discovery process, as robust results are more predictive under real-world conditions, such as those that occur in human clinical trials.

Development of a CB1 antagonist for the treatment of drug abuse (Solvay Pharmaceuticals). Compounds were shown to cause adverse effects, such as epileptic attacks in rats. Combined with the fact that compounds (e.g. Rimonabant) acting at the same target caused suicidal ideation in humans, the development was stopped. [1]
The side-effects were target-related. This example demonstrates the need to investigate target-related toxicity as early as possible once a new target has been identified - for example by studying target-deficient animal models. (LIST OF PUBLICATIONS)

[1] Sam AH, Salem V, Ghatei MA. Rimonabant: From RIO to Ban. Journal of Obesity (2011)
Other PMID: 21773005

Should a target in oligodendrocytes be identified which is relevant for a human CNS disorder, there appears to be (currently) no means of delivering a compound because of the impermeability of the blood-brain-barrier. In such a situation, a drug development project would need to be terminated as the target is not accessible. A potential bypass strategy could be to establish an in vitro permeability assay using primary or derived cells from the human cerebral vasculature.