23 March 2017
Uncertainty surrounds us, and while this has always been the case, it feels to me that recent political upheavals have led to a greater sense of uncertainty about our futures, and even uncertainty about what we know, or do not know, to be true.
We are all used to dealing with uncertainty in our daily lives, including when it comes to our health, but individual attitudes vary widely, from craving absolutely certainty about our current and future health status to having a strong desire to ‘not know’ what fate may befall us. The recent Nuffield Council on Bioethics report on non invasive prenatal testing (NIPT) wonderfully illustrates how varied expectations and demands for certainty about health can be, in this case the health of your future child.
NIPT is just one example of a group of tests, often but not always based on genomics, emerging into clinical practice and presented as delivering ‘precision’ medicine. But what do patients and clinicians expect these ‘precision’ tests will offer in terms of certainty about our current and future health?
Underlying the precision / personalised medicine paradigm is the idea that higher resolution and more precise characterisation of a person’s physiology or pathology (using genomics or other molecular methods) will enable more accurate description of their current and future health and more accurate determination of the best way to manage it. This has prompted a common misconception, compounded by the language of precision medicine, that genomic information is deterministic, that with precision will necessarily come certainty.
But is precision medicine actually reducing uncertainty? And is it realistic to expect it to do so now or in the future?
To take an example from our current work on circulating tumor DNA based testing for lung cancer , a prototype of precision medicine in action, it is clear that in some cases the test can provide a great deal of certainty e.g. by identifying a mutation in a patient’s tumour that strongly predicts that they will respond to a targeted cancer therapy. However a negative result (i.e. one where that mutation is not detected in the patient’s tumour) from the same test produces uncertainty as it cannot be definitively interpreted to mean the patient would not respond to the target drug now or in the future, as the technology and our understanding of tumour biology are not well enough developed to be certain of this.
Experience shows that while metagenomics can be great for detecting rare ‘unknown unknown’ organisms causing disease, it also generates enormous uncertainty for clinicians
Another great example of the double edged sword of genomic testing is provided by the field of metagenomic sequencing. Here, scientists are able to simultaneously detect and characterise (at the genome level) the many microorganisms that might be present in a sample e.g. from an infected wound. This method is potentially transformative in that it allows all organisms in a sample to be detected in a single test. Importantly it does not require prior knowledge of which organisms ‘ought’ to be present. However, experience shows that while metagenomics can be great for detecting rare ‘unknown unknown’ organisms causing disease, it also generates enormous uncertainty for clinicians by revealing the presence of organisms (and genes associated with drug resistance or virulence) whose significance it is currently impossible to interpret.
So what to do? If we cannot always provide a definitive interpretation of the result of a test, should we do the test at all? Or do we decide that access to what would certainly be better care for one patient outweighs the risk of an uncertain outcome for another and after all, clinicians deal with uncertainty and patient expectations all the time, so what’s to worry about?
Either way, there is a clear need to manage expectations with respect to the certainty offered by genomic tests, whether they are applied to our infections, our tumours or our even own genomes.
In my view, the current uncertainty associated with genomic test interpretation across a range of clinical areas should not be seen as an impediment to the implementation of these tests into healthcare. Quite the opposite. From a scientist’s perspective, uncertainty is the wellspring of creativity and discovery, offering motivation to uncover knowledge that in time will be used to improve health. But we have to accept that while scientists may get a kick out of the uncertainty arising from genomic medicine, citizens are likely to be less welcoming.
From a scientist’s perspective, uncertainty is the wellspring of creativity and discovery, offering motivation to uncover knowledge that in time will be used to improve health.
As ever, the key will be to develop dialogue, foster understanding and a shared stake for citizens, scientists and clinicians in both the risks and benefits associated with embracing uncertainty in these early days of precision medicine. As participants in a health system that encompasses us all, we could choose to embrace uncertainty - we could be more open with the use of our data and ourselves as part of the research and discovery process. The sooner we are, the sooner we will reap the benefits of this is exciting phase in the development of healthcare.
With the shift towards a more ‘citizen centred’ approach to health, it really is down to us to decide how fast we want to go.