The key is to find targets with novel mechanism of action (MOA) and an increased probability of success to differentiate in the clinic.
The pharmaceutical industry is in desperate need of new therapies with “unambiguous promotable advantage” that address unmet clinical need (see here, here and here). Of course, this is a laudable goal in drug development. In fact, given the current health care climate, we have no other choice (see here). If we are to have a sustainable industry, we must change the way we do discovery science. According to the Bernstein Report on BioBusiness: “Differentiation or Bust: Drug companies must start creating the case for value differentiation in discovery and then steadily build a body of evidence throughout the product development process.”
This means that dedicated drug hunters have a steep challenge ahead: to identify targets with novel mechanisms of action that have an increased probability to differentiate in the clinic. This will take creativity, hard work and innovation.
There is a lot written about “innovation”. [See here for a collection of articles from the HBR Insight Center; you can test your “innovation quotient” here.] Most comments about innovation involve creating a climate of risk taking balanced with accountability. I enjoy reading these commentaries.
The pharmaceutical industry needs an innovative scientific strategy to match the innovative culture that is being created in the biopharmaceutical ecosystem. A culture of innovation alone will not deliver new targets with unambiguous promotable advantage.
My argument – as you may have guessed! – is that human genetics can cause disruptive innovation in target identification and validation (affectionately refered to as TIDVAL in some companies). The argument goes like this
1. What is the fundamental problem in TIDVAL? We need to know which targets, when perturbed, have the desired effect on human physiology. This is difficult to do. Fortunately, nature has done the experiment for us through genetic mutation and human evolution. We just need to test human mutations for associations with clinical phenotypes that are good surrogates for drug efficacy. For more on this concept, see here. Also, I will develop the concept that “phenotype matters” in my next blog.
2. Are we open-minded yet disciplined? It is easy to say we will explore new targets that appear to alter pathways that are orthogonal to the pathways targeted by existing therapies (i.e., targets with new mechanism of action [MOA]). However, it is far more challenging to actually do this type of exploration. First, it takes an open mind to advance targets that we think have new MOA. Human genetics offers an unbiased anchoring point to open our minds, as human evolution does not carry the same baggage that we carry in terms of understanding the biology of human disease. Second, it takes discipline to build from a scientific foundation that we think holds promise. If we truly believe in the power of human genetics, then we will pursue targets that we have not previously considered as relevant to disease. We will need to be disciplined in adhering to this strategy, as it is easy to revert back to our previous understanding of human disease. In some instances, this may mean doing away with existing pre-clinical models that may be poor surrogates for human efficacy.
3. Has this been done before in the pharmaceutical industry? I would argue no, at least not in a systematic and comprehensive fashion. There is a clear trend, however, at many biopharmaceutical companies (see here, here, here), including Merck (which is where I work). One reason for this is that until recently, we did not have the density of data that created the necessary link between human disease and human mutation, or genotype-phenotype maps. While the human genome was mapped in the year 2000, this only served as a reference sequence. We now know that most complex diseases such as diabetes, heart failure, Alzheimer’s, and systemic lupus erythematosus are highly polygenic (see here). It is only today that we can use genotype-phenotype maps to begin to unlock the relationship novel targets and human disease.
4. Is human genetics the only way? Absolutely not! In fact, for some phenotypes, human genetics will not be helpful at discovering new targets (see next blog). And even for those phenotypes that are relevant to TIDVAL, human genetics is only the starting point…and often a very difficult starting point. I do think, however, that human genetics represents a unique strategy that differentiates cause from consequence in the ideal model organism, humans. To be clear: there are other types of human data that are also highly relevant, especially longitudinal profiling in samples with rich clinical data (see Snyderome, GoogleX’s “Baseline Study”) and patients treated with approved drugs of known MOA.
In summary, human genetics addresses the fundamental challenge in TIDVAL; it provides an anchoring point to prosecute targets with new MOA; and while it is not the only way, it does represent an innovative strategy (at least as I define it, see next blogs) that has been underutilized, until recently, for drug discovery.
In the next three blogs, I will describe how to execute on a TIDVAL strategy anchored in human genetics. In brief:
Part 2: Phenotype matters – not all phenotypes are relevant for genetics and drug discovery…which ones are?
Part 3: One gene, one target, one drug – human genetics can find genes with an allelic series to estimate dose-response curves at the time of target ID and validation
Part 4: Many genes to pathways to targets – a systems biology approach, anchored in human genetic, is a powerful complement to the one gene “allelic series” TIDVAL approach