At his State of the Union address, President Obama announced the creation of a Precision Medicine Initiative. While exact details of its implementation are still being decided, what has been described centers around the creation of a million individual database, along with broad characterization of variation among these individuals, so that genomic (and other -omic) variation can be correlated with disease incidence or outcomes. Additional funding will be used to more extensively characterize the genetic variation of cancers.
What has been promised from these efforts is nothing short of a transformative change in drug effectiveness—a future in which genetic testing identifies the perfect drug for you: an infinitely effective drug we would never discover without taking into account the unique person you are.
Because of my line of work, I am most familiar with the optimism surrounding precision medicine in cancer. This optimism is largely based on a common observation as targeted therapies1 have entered the clinic. For most patients, these drugs only provide temporary relief from the disease and for some no relief whatsoever.
However, a minority of patients respond dramatically and these patients ultimately benefit most in the long-term from the treatment. From studies looking at responders versus non-responders, we know that responders generally have greater dysregulation of the drug target, and also generally have “simpler” cancers, meaning that they have fewer genetic abnormalities overall.
Optimism for “precise” therapies assumes that because some patients respond very well to these therapies, we can find an effective target for non-responders. I wish for this to be the case, but this seems like a foolhardy assumption from what we know.
Cancers that respond to targeted therapies are cells that have managed to break just a few pathways in order to survive, and because of that absolutely rely on those pathways. “Complex” cancers that do not respond to these therapies tend be an entirely different problem. These cancers have developed multiple redundant mechanisms of survival, constantly adapt to become entirely different when treated, coax survival cues from normal cells2, and even diversify into essentially separate tumors with distinct mutations.
A Deep-dive Into Specialty Pharma
A specialty drug is a class of prescription medications used to treat complex, chronic or rare medical conditions. Although this classification was originally intended to define the treatment of rare, also termed “orphan” diseases, affecting fewer than 200,000 people in the US, more recently, specialty drugs have emerged as the cornerstone of treatment for chronic and complex diseases such as cancer, autoimmune conditions, diabetes, hepatitis C, and HIV/AIDS.
Thus, genetic testing will probably help predict when a patient will benefit from a drug that would already be prescribed, but will not provide new answers for patients who are out of options.
The same distinction–between genetically simple and complex diseases–is relevant beyond cancer. Genetic testing has been effective in understanding diseases with a strong genetic basis (e.g. screening for Down’s syndrome during pregnancy).
However, genetics only plays a small part in most common diseases and that genetic basis is likely to be complex and spread across hundreds of genetic changes. That is why it is already possible for companies such as 23andMe to identify individuals at increased risk of diabetes, for example. Such information lacks any understanding of how those genetic differences result in diabetes though, and so is useless for treating patients.
The darling example of precision medicine, Vertex’s Kalydeco, is actually a perfect example of how genetic data is only valuable after a mechanism-based understanding of a disease has developed.
Certainly, the drug is a notable scientific accomplishment and enormous development for people suffering from a horrible disease, but hardly provides justification for a massive expansion of sequencing efforts. The precision aspect of Kalydeco (that it is only effective against the particular mutation) is the result of targeted drug development efforts and protein structure understanding, and in no way arose out of unbiased patient sequencing. While genetic testing is used to identify whether a patient will benefit from the drug, only a small piece of DNA needs to be sequenced, and the information contained in the entire rest of the genome is unlikely to be helpful. This just highlights that what we need more of is a mechanism-based understanding of diseases, not more data for data’s sake3.
Centralized development of a large-scale human and tumor variation resource will be valuable for basic science, and I am sure will lead to important discoveries. However, overselling it as a harbinger of a medical revolution threatens public confidence in the value of science when these promises fail to be delivered.
- Targeted therapies are ones that selectively inhibit a particular dysregulated protein, such as Herceptin targeting HER2 in breast cancer or Iressa targeting EGFR in lung cancer. ?
- Notably, genetic tests would entirely miss this source of resistance. ?
- See also: The Scientific Drunk and the Lampost. ?
