The Hype, the Hope, and the Science Behind Precision Oncology

Anya Plutynski (Washington University in St. Louis)

In 2015, President Obama launched his “Precision Medicine Initiative,” an ambitious effort at funding research that would ensure that patients receive care “tailored to them.” One “short-term goal” of the program was – effectively – to provide continued funding to research in cancer genomics. The Cancer Genome Atlas Project was launched in 2005, and was (perhaps not coincidentally) just wrapping up in 2015-2016. In this talk, I will consider the promises made on behalf of TCGA, and the reality. 
The the first 5 years of the project was largely a pilot project – one that created a vast multi-institutional infrastructure of university hospitals, in relationship with private biotech, and an array of specialists working in a new kind of “big data” research. Big data research of this type faces particular types of methodological challenges. I will consider one particular challenge - a variant of the “curse of dimensionality” - how it arose during TCGA, and in what sense it was resolved. I will then consider how we ought to define and measure success in scientific contexts like this one, and associated challenges facing honest advertising for biomedical research. False hope is impossible to avoid altogether. However, many oft trumpeted successes of precision oncology were developed and approved (in some cases) decades before the launch of cancer genomics, suggesting that TCGA had less impact than is often advertised. Nonetheless, I will argue that TCGA has succeeded; however, “success” in this context is rather different from how it is typically understood in philosophy of science. Sequencing is cheaper, analysis of cancer genomics is more sophisticated, and there is a much wider appreciation of both the heterogeneity of cancer genomes, and the complexity of cancer causation than prior to 2005. If anything, TCGA has complicated and challenged many of the initial presuppositions behind how cancer genomics would or could inform precision oncology.