Plenge Lab
Date posted: April 25, 2013 | Author: | No Comments »

Categories: Uncategorized

p1

While most consider pharmacogenomics (PGx) the study of response to drugs in the clinic (e.g., efficacy and toxicity), PGx is also an amazing tool to understand fundamental biology of human disease.  Drugs perturb human physiology in a way that cannot be accomplished in the resting state.  

Most would agree that complex traits such as rheumatoid arthritis (RA) are more than just one disease.  In fact, some advocate using the term “syndrome” rather than “disease” for RA, as syndrome emphasizes the complex and heterogeneous etiology.  However, what are the underlying subsets?

Genomic technology promises to deconstruct complex traits such as RA.  The problem that I have seen, however, is how to classify the subsets of disease.  On one hand, we could take an unsupervised approach, and allow the data to form phenotypic subclassifications.  In the study of RA synovial tissue (the primary site of pathology), data suggest that there are histological categories of disease depending upon the predominant cell type.  One the other hand, what is ultimately important is how to translate disease subsets into clinical care.  And for this to occur, there must be a correlation with clinical findings.

Here is where drug exposure can help translate an unsupervised approach into a clinically actionable discovery.

Drug exposure in humans is a form of perturbation.  Like RNAi in cell culture or genetic knock-outs in the mouse, drugs perturb molecules in a model system – the ideal model system, humans.  Many patients respond differently to drug perturbations, either in the form of efficacy or toxicity.  By studying physiological processes in humans before and after drug perturbations, it should be possible to gain a better understanding of the critical determinants of disease.

For RA, the relative importance of inflammatory cytokines represents a good example.  Over a decade ago, there was debate as to which inflammatory cytokine(s) was most responsible for clinical symptoms in RA.  In particular, debate surrounded IL1 and TNF.  Drugs were developed to inhibit both inflammatory cytokines (anakinra for anti-IL1 and infliximab / etanercept for anti-TNF).  Both categories of drugs were effective in animal models of RA…and anakinra showed more promise than anti-TNF therapy.  However, in humans the winner was clear: anti-TNF therapy was far more effective than anti-IL1 therapy.  Thus, even though pre-clinical models favored IL1 over TNF as the most important inflammatory cytokine, human perturbations with drugs demonstrated that TNF was more relevant to signs and symptoms of patients with RA.  [Note that this distinction is somewhat artificial, as there is no single cytokine that drives the disease.]

Thus, as we consider the importance of PGx as a discipline, it is important that we not forget the concept of disease deconstruction: drug response as a phenotype to identify disease subsets and genetic determinants of disease.  PGx has relevance far beyond predicting efficacy and toxicity in a clinical setting.  It is an invaluable tool to understand basic human physiology.

Leave a Reply