by Maria Schacker
The study of human genetics can be classed into several categories, ranging from human evolution and studying genetic variations between and within populations to human genetic disease inheritance. The genetics of disease inheritance is becoming more and more important for the pharmaceutical industry.
Until the early 2000s, the only way to study genetic diseases was through family studies and linkage analysis in families with high occurrences of disease. However, these studies are extremely tedious as they can only be done on a small scale and they are by and large only applicable to diseases that follow mendelian inheritance, not multifactorial diseases.
The field of human genetics really changed and thrived with the human genome project which was completed in 2003 . We now have a complete map and sequence of the human genome. This is incredibly useful for the study of diseases as we can now sequence whole genomes for less and less money (ca. $1000/genome) and then analyze which mutations correlate with specific diseases. Genome wide association studies (GWAS) for example are used to identify single nucleotide polymorphisms (SNPs) that are associated with a certain disease or phenotype. All this has helped tremendously to identify genes and biological pathways that are implicated in disease susceptibility.
In the past few years, there have not only been major advances in the identification of disease genes but also in the way that diseases are classified. Whereas the general consensus used to be to classify diseases by their symptoms (i.e. their phenotype), the tendency now goes towards classifying them by their genotype.
But how does all of this affect the process of drug development?
Traditionally, target discovery was often done in a phenotypic approach using animal or cell culture models of disease to identify compounds that have an effect on the phenotype, for example something that kills cancer cells (Figure 2). Only afterwards, the corresponding target was identified. There are two main issues with this approach: Firstly, a number of drugs actually affect the same target, effectively narrowing down the number of disease treatments. Secondly, since most of these studies were based on animal or cell culture models, many drug targets may not be physiologically relevant for human disease.
Using human genetics studies, a different approach to target discovery can be taken (Figure 2). Now that the human genome sequence is available, a number of incredible tools can help with this - the general idea always being to first identify molecular changes between healthy and diseased individuals or tissues. The main advantage this has over traditional target discovery is that any changes that are identified are by their very nature directly relevant to human diseases. Additionally, these studies are often quicker to perform than experiments using animal or cell culture models for target discovery.
RNA sequencing, GWAS to identify SNPs and copy number variation (CNV) analysis are just a few of the techniques that can now be used to discover genes and biological pathways that are affected in different diseases. This data can then help to identify suitable targets. Having identified a potential drug target, the next step involves identifying a compound that is effective against this target. This could be done using a traditional compound screen, which screens targets against a library of chemical compounds. However, using computational biology approaches and the knowledge that we have gained from the human genome project, an “in silico” screen is also an option. The sequencing data could be screened for paralogous genes to the target gene of interest as some of these may already be targeted by drugs which can then be repurposed. The same approach can also be taken for structurally similar targets. Repurposing drugs can save a great amount of money as some of the preclinical and clinical studies can be avoided or shortened.
Furthermore, by understanding the genetic causes of diseases and together with help of innovative gene editing techniques, it is now possible to directly modify genes, effectively expanding the list of “druggable targets” from mainly proteins to protein and DNA targets.
Overall, using knowledge from human genetics studies can lead to the quicker identification of new drug targets, specifically targets that are physiologically relevant for human diseases, potentially leading to better success rates in early stages of clinical trials. Additionally, a completely new field of pharmaceuticals has emerged – gene therapies.
Another process of the drug development pipeline that can benefit from advances in human genetics studies is patient stratification for clinical trials (Figure 3). Especially phase II clinical trials have an extremely high rate of compound failure due to issues with efficacy and safety.
Traditionally, patients for clinical trials were chosen based on their symptoms (i.e. their phenotype), but it can be argued that patient stratification by genotype has the potential to be much more successful, both in terms of safety and efficacy. This approach is becoming more and more common practice. Very often, “one disease” is in reality a number of diseases which all happen to have the same phenotype but are caused by different underlying genetic and molecular mechanisms. Therefore, if patients were screened for specific biomarkers before the beginning of the clinical trial (or during phase II the latest) to only select patients with an appropriate genetic background, this can increase the chances of a drug passing clinical trials (Figure 3). If suitable genetic material is still present, this could even be done in hindsight for drugs that have previously failed clinical trials due to low efficacy.
The same principle applies to safety. It is well known that many drugs are metabolized differently by patients, depending on their genetics. If this is known for a compound of interest this should obviously be taken into consideration from the start. However, it may also be possible to genotype patients after a trial has failed during phase I to analyze whether there is a correlation between drug safety and the patients’ genetic background, in which case the clinical trial may be able to proceed to phase II for a genetic subset of patients.
Overall, patient stratification by genotype can lead to a higher success rate of clinical trials, better efficacy results and higher safety for the patients. Additionally, there is the potential to rescue drugs that have failed previously.
One great example of a drug where patient stratification by genotype has been proven to be successful is the cancer immunotherapy Pembrolizumab (Keytruda®) which targets the PD-1 receptor . It inhibits the cancer’s immune system evasion and activates an anti-tumor immune response. The clinical trials for this drug were limited to patients with specific biomarkers (PD-1 ligand expression) indicating a likely response. This is the first FDA approved drug that is based purely on tumor genetics and has no limitations on the tumor site or the tissue type. This drug is supplied together with an immunohistochemistry test for PD-L1 which serves as a companion diagnostic.
In summary, insights from human genetics have led to a more genotype (rather than phenotype) focused disease classification which can - together with technical advances in the field of genetics - help with quicker and more efficient target discovery and better patient stratification, potentially resulting in more successful clinical trials and a higher number of approved drugs.
Biopharma Excellence has in-depth expertise in the development of novel biopharmaceuticals and in the preparation of integrated development strategies. Please contact us for the support of your development program.
 I. Human Genome Sequencing Consortium, “Finishing the euchromatic sequence of the human genome,” Nature, vol. 431, no. 7011, pp. 931–945, Oct. 2004.
 L. Khoja, M. O. Butler, S. P. Kang, S. Ebbinghaus, and A. M. Joshua, “Pembrolizumab.,” J. Immunother. cancer, vol. 3, p. 36, 2015.